Green Growth or Economic Trade-offs? Economic Costs of Environmental Degradation in ECOWAS. A New Perspective from Artificial Neural Network and SEM Analysis

Green Growth or Economic Trade-offs? Economic Costs of Environmental Degradation in ECOWAS. A New Perspective from Artificial Neural Network and SEM Analysis | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Green Growth or Economic Trade-offs? Economic Costs of Environmental Degradation in ECOWAS. A New Perspective from Artificial Neural Network and SEM Analysis Seth Acquah Boateng, Andy Asare, William Godfred Cantah, Joshua Sebu This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-5675754/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Environmental degradation creates unfriendly conditions for economic growth and public health, especially in developing regions like West African countries, where fast industrialization increases these risks, along with precarious environmental legislation. The study aims to analyze the direct and indirect relationships between environmental degradation, health expenditure, and economic growth within the ECOWAS region. The research adopted structural equation modelling (SEM) and Artificial Neural Network (ANN) analysis to examine the impact of carbon dioxide emission, nitrous oxide emissions, and air quality levels on economic growth, taking health expenditure as the mediating variable. Results of SEM show that carbon emission; and nitrous oxide emission positively influence the economy, while poor air quality negatively affects it, health expenditure mediates the influence of nitrous oxide emission on economic growth with an indirect effect. However, it has an insignificant mediating effect between carbon emissions and air quality. Also, ANN analysis confirms the SEM results that indicate carbon emission has the highest predictive importance. The study, therefore, recommends increased stringency of environmental regulations in the West African region, investment in clean energies, and health infrastructural improvement as ways through which environmental degradation may be minimized to allow the attainment of economic development sustainably. JEL: H51, P18, Q51, Q53, Q56 Environmental degradation ECOWAS region SEM ANN Health expenditure CO2 emissions Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Figure 8 1. Introduction As countries pursue economic growth, they also have to confront the twin problem of fighting environmental degradation and improving public health (Boateng et al., 2024). It requires a comprehension to develop appropriate policies that strike a balance between development objectives as well as environmental issues. Environmental degradation is expressed through air pollution and greenhouse gas emissions (Lima & Hamzagic, 2022). This idea has been associated with economic activities. According to Gwangndi et al. (2016), environmental degradation has a crucial impact on the health of humans. The importance of this relationship is because, in most cases, public resources are scarce, and the health systems of the countries in that region are under immense pressure (Anwar et al., 2022). Despite their significance to economic progress, these factors have also led to various ecological problems. Rapid urbanization, industrialization, and rising energy consumption have been features of the economic growth within the ECOWAS region (Wu et al., 2021). The Economic Community of West African States (ECOWAS) is constituted of fifteen member nations, which comprise Benin, Burkina Faso, Cape Verde, Côte d'Ivoire (Ivory Coast), The Gambia, Ghana, Guinea, Guinea-Bissau, Liberia, Mali, Niger, Nigeria, Sierra Leone, Senegal, and Togo (ECOWAS, 2019). Liu and Raven (2010) and Luo et al. (2022) have shown that China’s pursuit of rapid economic expansion resulted in serious deterioration of the environment and warned other developing nations not to make a similar mistake for fear of long-term repercussions. Therefore, countries need to learn from this and get onto more workable development tracks lest they lag in the heel of the flourishing economies around the earth. Xing and Liu, 2022 established that regulatory quality and the rule of law have significant effects on healthcare spending and environmental outcomes. This calls for a reason why strong institutions are to be built up for resilience and application of the environmental regulations in the countries of the ECOWAS region for economic growth while protecting the environment and public health. This is because the aim of sustainable development among the ECOWAS communities ensures balanced economic growth with environmental conservation and social welfare. In this regard, Madan and Suri (2023) argue that the thought process has to be changed from "environment versus development" to "environment and development." Only then will the ECOWAS community be able to plan for long-term economic growth without compromising environmental standards and public health. Demir et al. (2022) estimated asymmetry in the impact of health expenditures concerning environmental quality and therefore called for adaptive policies concerning negative and positive responses to environmental shocks. Indeed, the adaptable policies open new avenues for building up resilient healthcare systems and environmental policies against the adverse effects of climate change in the ECOWAS region. That by Kahia et al. (2022) implies that, in mitigation strategies for CO2 emissions, reliance on fossil fuel needs to be done away with by shifting toward renewable sources if economic growth has to be sustained. Changes in multidimensional perspectives also include those of environmental impact, public health, and economic development, among other dimensions over this region of Africa. It has been, therefore, one of the major concerns among environmental degradation, health expenditure, and economic progress within the ECOWAS region and, as such, requires advanced research for full comprehension in terms of evidence-based policy formation. Machine learning models have recently obtained greater recognition in research studies focused on environmental factors and economic growth. In this respect, the study of Jabeur et al. (2021) conducted the forecast analysis of CO2 emissions using Natural Gradient Boosting and estimated the key determinants affecting environmental sustainability. This study applied ANN and SEM to unravel the linkages; hence, it unearths new evidence over critical drivers of environmental performance in the region. Lastly, health expenditure also acts as a mediator in the relationship between environmental degradation and economic growth. For example, countries with emerging economies in SSA are investing an increasingly high share of their public finances in the health sector, and environmental degradation tends to heighten health expenditures (Aboubacar & Xu, 2017). Thus, even as such countries will generally register economic growth, they will be able to invest more in hospitals as they register ecological degradation. This study investigates the interrelated relationships that exist between environmental degradation and health expenditure coupled with economic growth through the SEM and ANN algorithms. The study looks at the influence of carbon dioxide, air pollution, and nitrous oxide on gross domestic product per capita and health expenditures, playing the mediating role. Therefore, this paper tends to bring forth state-of-the-art insights into the Environmental Kuznets Curve hypothesis in West African countries and checks on various direct and indirect influences of growth with implications for healthcare expenditure. The findings from this project raise the knowledge bar relating to trade-offs between environmental quality and public health investment and economic development patterns and contribute to the development of better-designed and more sustainable policy formulation in this region over time. 2. Literature 2.1 Environmental degradation Anthropogenic activities lead to environmental degradation, which is the deterioration of all the natural systems and resources of the earth (Wassie, 2020). These include air and water pollution, and cleanup of oil spills in the ocean. Soil degradation is also another example, along with deforestation, which leads to the loss of biodiversity and climate change (Singh & Singh, 2016). Wu et al. (2021) explained this phenomenon has a huge effect on ecosystems, human health, and even economic building. The very root of this problem is the huge amount of carbon dioxide emitted into the atmosphere due to the combustion of fossil fuel; this gas contributes to the absorption of more heat energy, hence intensifying global warming. Although the Intergovernmental Panel on Climate Change, (2021), impressively identified that the contribution of man-made activities is the cause of the global warming phenomenon, the manifestation is getting acuter in the different geographies by way of increased occurrences of extreme weather events like tornadoes, hurricanes, etc., with unparalleled intensities. Then again, there is air pollution, especially in metropolitan areas, which is another important dimension of environmental degradation. According to the World Health Organization (2021), an astonishing 99% of the global population is breathing air which is even more polluted than allowed by the limits set by WHO standards, with countries in categories of low- and middle-income being mostly exposed. Particulate matter, especially PM2.5, has been associated with a wide range of health problems, including many respiratory and cardiovascular diseases (Demir et al., 2022; Guo et al., 2023). Emissions of nitrous oxide stemming from agricultural practices and industrial operations play a substantial role in exacerbating air pollution as well as contributing to climate change (Aryal et al., 2022). Nitrous oxide is recognized as a highly effective greenhouse gas, possessing a global warming potential that is 300 times greater than that of CO2 over a century-long timeframe (Patel, 2021; Griffis et al., 2017). Neidell (2017) indicated that a degraded environment can lead to lower productivity in sectors like agriculture and fisheries, increased healthcare costs, and damage to infrastructure. Specifically, Ayodotun et al. (2019) observed that countries within West Africa remain among those most vulnerable to slip backwards from development dividends due to climate change, which in turn further entrenches poverty. On this note, there is a need to address environmental concerns toward the sustainability of ecosystem integrity and laying a foundation for economic growth and human development in the area. 2.2 Health expenditure Health expenditure is defined by the World Health Organization (2021) as spending on health services, family planning activities, nutrition programs, as well as disaster relief meant for health care only. In the ECOWAS region, health expenditure varies across countries. In the context of Sub-Saharan Africa, Novignon and Lawanson (2017) discovered that healthcare expenditure has a positive and significant effect on health outcomes, such as reduced infant mortality rates and increased life expectancy. This implies that there are diminishing returns to scale, meaning not all levels require equal additions, as each unit increase, within a given limit of time, is associated with a decrease in another variable. Moolla and Hiilamo (2023) noted that countries with higher health spending were in a better position to deal with disease outbreaks, suggesting that investment in health infrastructure is key to thwarting such outbreaks. Nevertheless, increasing spending for health within the ECOWAS region faces numerous constraints, including inadequate resources compared to other sectors, competing against development projects (Uzochukwu et al., 2015). This undertaking is imperative for enhancing general well-being while at the same time boosting overall economic growth. 2.4 Hypothesis Development 2.4.1 Environmental Deterioration and Economic Development The "Environmental Kuznets Curve" or EKC comes into play to describe the relation between degradation in the environment with economic growth. It states that as the nations start to grow, the environment will deteriorate first before it improves. Singh and Yadav, 2021, present evidence that the relationship between economic growth and environmental quality is inverted U-shaped. This generally suggests that during the early stages of growth, environmental degradation takes place, and thereafter, with income increases, it permits a shift toward cleaner modes of production and the adoption of efficient abatement practices. Notwithstanding various criticisms raised against EKC generalizations, especially within developing contexts, Setyari and Kusuma (2021) debunked the expectations by presenting an "N" shaped pattern, where after further growth, there could be another degradation in environmental standards. This necessitates sustainable environmental management, even during periods of economic expansion, involving substantial funds. Economic development causes environmental degradation, which is not always straightforward. Carpio-Thomas and Christian (2020) observed a rapid economic transformation in China, which had profound environmental impacts, warning other developing countries against choosing between environmental preservation and economic growth at the expense of environmental protection laws. This issue is particularly relevant for ECOWAS, as many states aim to achieve quick progress. Zaidi and Saidi (2018) discovered that economic growth had a positive impact on health expenditure in Sub-Saharan African countries, while it was negatively affected by environmental pollution, showing a complicated relationship between economic growth, environmental quality, and public health investments. One critical issue concerning the state of the environment today is air quality, which the World Health Organization (2021) has reported as a major health hazard, especially in developing regions suffering from air pollution. Low air quality not only has public health consequences but also inhibits economic productivity and increases healthcare costs. Both pollution and global warming are exacerbated by the discharge of nitrous oxide. On the other hand, Alrais et al. (2024) pointed out that nitrous oxide has a severe greenhouse effect with implications for long-term environmental impairment. Consequently, the following hypothesis will be developed. H1a: There is a significant relationship between Carbon emissions and Economic Growth in the ECOWAS region. H1b: There is a significant relationship between Air Quality Index and Economic Growth in the ECOWAS region. H1c: There is a significant relationship between Nitrous oxide emissions and Economic Growth in the ECOWAS region. 2.4.2 Health Expenditure as a Mediating Factor It is hypothesized that health expenditure mediates the relationship between economic development and environmental deterioration. This role of mediator has implications for environmental outcomes, as well as growth paths. In this regard, it has been observed by Zaidi and Saidi 2018 that whereas economic growth increased health expenditure positively, it was affected negatively by pollution. This suggests that as economies grow, they may allocate more resources to healthcare, but environmental degradation simultaneously increases healthcare costs, creating a challenging dynamic for policymakers. Anwar et al. (2021) examined the effect that both non-renewable and renewable supplies of energy have on the environmental nexus for health expenditure. It could be even worse in the ECOWAS region; incongruous environmental policies see health expenditure levels rising due to increased health bills. Further research is perhaps needed to see if such trends are indeed repeated in other developing countries. In this context, while improved public health can result in an increase in economic productivity on the one hand (Artekin & Konya, 2020; Wang, 2015), on the other hand, there is the potential that once people are healthier, they may become less productive, which could reduce economic gains (Bu & Ali, 2018). However, it was revealed that countries do spend more on health services when their citizens are ill with diseases (World Health Organization, 2021). The same research proved that more money is spent on maintaining and improving the health of the population when there is a low level of environmental pollution (Bu & Ali, 2018). However, without proper environmental investments alongside health initiatives, long-term sustainability cannot be assured. This implies that neglecting environmental concerns while increasing health expenditure is self-defeating in the long run. H2a: Health expenditure significantly mediates the relationship between Carbon emissions and Economic Growth in the ECOWAS region. H2b: Health expenditure significantly mediates the relationship between Air Quality Index and Economic Growth in the ECOWAS region. H2c: Health expenditure significantly mediates the relationship between nitrous oxide emissions and Economic Growth in the ECOWAS region . 2.5 Research Framework Fig. 1 presents the conceptual framework which highlights a model of the relationship between environmental degradation, economic development, and health expenditure within the ECOWAS region at large. This framework at its core asserts that environmental variables- represented by carbon dioxide emissions, PM 2.5 air quality index as well nitrous oxide emissions- have either direct or indirect impacts on economic progress as measured by GDP per capita. Indirectly, it touches upon healthcare spending thereby maintaining health expenditure as a pivotal intervening variable. Herein lies an avenue through which an environmentalist can understand how GDP growth might be affected not just directly but also indirectly through public health outcomes such that its effects on public health and subsequent health-care spending are evident. 3. Methodology 3.1 Data Description The study utilized panel data from multiple reputable international sources to ensure comprehensive coverage of the ECOWAS region. The World Development Indicators database provided the primary economic data, including GDP per capita and health expenditure figures. Environmental data, specifically CO2 emissions and nitrous oxide emissions, were sourced from the UN database, which offers standard environmental metrics across countries. The World Health Organization contributed data on PM2.5 levels, providing a reliable measure of air quality. The study used yearly period data which span from 2000 to 2021. Table 1 Variable Definition Variable Description and Measurement Source GDP per capita Annual gross domestic product divided by midyear population, measured in current US dollars World Development Indicators (World Bank) Health expenditure Current health expenditure as a percentage of GDP World Development Indicators (World Bank) CO2 emissions Carbon dioxide emissions, measured in metric tons per capita World Development Indicators (World Bank) Nitrous oxide emissions Nitrous oxide emissions, measured in thousand metric tons of CO2 equivalent World Development Indicators (World Bank) Air Quality Index (PM2.5) Annual mean concentration of particulate matter of less than 2.5 microns in diameter, measured in micrograms per cubic meter World Health Organization Word Bank Open Data 3.2 Data Cleaning and Standardization The raw data were consolidated into a single dataset. Missing values were identified and addressed using interpolation for time series data and mean imputation where suitable. Outliers were detected and adjusted. All variables were standardized using z-score normalization, transforming them to have a mean of 0 and a standard deviation of 1, a crucial step for machine learning applications in environmental economics (Ha et al., 2021; Potts & Schmischke, 2022). Finally, a thorough quality check was conducted to verify the consistency and integrity of the cleaned and standardized dataset. 3.3 Model Specification 3.3.1 Structural Equation Modeling (SEM) The study applied the SEM to analyze the causal relationships that exist between direct and indirect effects arising from environmental degradation, health expenditure, and economic growth. The analysis was done using SmartPLS 4.0, one of the most advanced software packages that provides for PLS-SEM, well noted for handling complex and multidimensional models with the highest level of efficiency (Chuah et al., 2021; Hair et al., 2018; Khan et al., 2019). The primary reason why PLS-SEM was chosen is that the approach can handle non-normal distribution data, small and medium-sized samples, and both formative and reflective measurement models. SEM allows the analysis of several relations at once and tests both direct and indirect effects in one model. The structural model in the following analysis presents the relationships between exogenous latent variables, environmental degradation indicators, and endogenous latent variables representing economic growth, with the underlying mediation by health expenditure (Ringle et al., 2018). The algorithm of PLS works based on the Varimax method, maximizing the variance explained in the endogenous constructs, thereby improving the predictive relevance of the model (Cheah et al., 2020). Also, for this purpose, the interactions among these factor interactions have been analyzed based on path coefficients, and the significance of the latter has been extracted through the bootstrapping technique included in the PLS-SEM, giving more reliability to the results (Kock, 2019; Ramayah et al., 2018). To add to robustness, the stability of the path estimates and their statistical significance were tested using bootstrapping of 5,000 sub-samples (Kock, 2019). The model fit was evaluated using several goodness-of-fit indices, including Standardized Root Mean Square Residual (SRMR), Normed Fit Index (NFI), and Chi-Square statistics. These scores thus enable an assessment of the accuracy of the model and the degree to which the theoretical structure fits the data (Cheah et al., 2020). More specifically, SRMR focuses on the variance between the observed and variance-predicted correlation, while NFI quantifies the comparative increase in the model fit as opposed to a no-model specification (Ringle et al., 2015). The following is the general model equation used for SEM: General Model: $$\:EG\:=\:\beta\:₁CE\:+\:\beta\:₂NO\:+\:\beta\:₃AQ\:+\:ϵ\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:1$$ Mediation Effect of Health Expenditure $$\:EG\:=\:\beta\:₄HE\:+\:(\beta\:₅CE\:+\:\beta\:₆NO\:+\:\beta\:₇AQ)\:+\:ϵ\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:2$$ Where: EG: Economic Growth (GDP per capita) CE: Carbon Emissions NO: Nitrous Oxide Emissions AQ: Air Quality (PM2.5 levels) HE: Health Expenditure β₁, β₂, β₃, β₄, β₅, β₆, β₇: Path Coefficients ϵ: Error Term Model fit indices, such as the Standardized Root Mean Square Residual (SRMR), Normed Fit Index (NFI), and Chi-Square statistics, were calculated to assess the quality of the model. The formulas for these indices are as follows: SRMR (Standardized Root Mean Square Residual): $$\:SRMR\:=\:\sqrt{\left[\left(\frac{1}{m}\right)\sum\:{ᵢ}^{=1}ᵐ\:{\left(sᵢⱼ\:-\:ŝᵢⱼ\right)}^{2}\right]\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:3}$$ NFI (Normed Fit Index): $$\:NFI\:=\frac{{\chi\:}^{2}ₙᵤₗₗ\:-\:{\chi\:}^{2}ₘₒdₑₗ}{{\chi\:}^{2}ₙᵤₗₗ}\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:4$$ Chi-Square statistic: $$\:{\chi\:}^{2}=\sum\:{ᵢ}^{=1}ⁿ\left[{\left(Oᵢ-Eᵢ\right)}^{2}/Eᵢ\right]\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:5$$ Where: sᵢⱼ: Observed correlation between variables i and j ŝᵢⱼ: Predicted correlation between variables i and j m: Total number of observations χ²ₙᵤₗₗ: Chi-Square value of the null model χ²ₘₒdₑₗ: Chi-Square value of the fitted model Oᵢ: Observed value Eᵢ: Expected value n: Number of data points These indices are essential for evaluating the model's overall goodness-of-fit, ensuring that the model is adequately capturing the underlying relationships among the variables. 3.3.2 Artificial Neural Networks (ANN) Sensitivity analysis using an Artificial Neural Network was done to further improve the outcomes of the previous SEM analysis for further validation. ANN is very good at detecting nonlinear relationships and finding out the relative importance of predictor variables (Kingma & Ba, 2015). The following analysis leverages Scikit-learn, a well-used Python library across machine learning and predictive modelling. ANN application is quite apt in this study because environmental degradation, health expenditure, and economic growth are highly interrelated. The ANN model was developed with carbon dioxide emissions, nitrous oxide emissions, and the air quality index as inputs to predict economic growth (Ramayah et al., 2023). Hidden layers with nonlinear activation functions were included to allow the network to learn the most complex pattern present within the data (LeCun et al., 2015). The hidden layers used the ReLU activation function because this kind of activation function introduces less computation and does not face the problem of a vanishing gradient while handling large data (LeCun et al., 2015). The output layer linearly activates the predicted continuous values of the economy's growth. In the architecture of the ANN model, the design aimed to reduce the estimated and actual differences in values. Also, the performance of the model was enhanced using the Adam optimizer, a sophisticated form of gradient descent, since it possesses adaptive learning rates and is operationally efficient (Kingma & Ba, 2015). The optimizer used is the MSE - Mean Squared Error - because it calculates the average of the squared difference between predictions and actual values (Kingma & Ba, 2015). The MSE formula is as follows: MSE (Mean Squared Error): $$\:MSE=\left(1/n\right)\sum\:{ᵢ}^{=1}ⁿ{\left(\varvec{y}\varvec{ᵢ}-\varvec{ŷ}\varvec{ᵢ}\right)}^{2}\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:5$$ Adam Optimizer: $$\:\theta\:{ₜ}^{+1}=\theta\:ₜ-\alpha\:·mₜ/\left(\sqrt{v}ₜ+ϵ\right)\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:6$$ Where: yᵢ: Actual value ŷᵢ: Predicted value n: Number of data points θₜ: Parameters being optimized at iteration t α: Learning rate mₜ: Estimate of the first moment (mean) of the gradients vₜ: Estimate of the second moment (variance) of the gradients ϵ: Small constant to prevent division by zero The ANN model was trained using an 80:20 train-test split to prevent overfitting and ensure that the model could generalize well to unseen data. Cross-validation was also applied to further verify the robustness of the model. Finally, sensitivity analysis was performed to assess the relative importance of each predictor variable, providing a clearer understanding of which environmental factors most strongly influence economic growth. This hybrid approach, combining SEM and ANN, follows recent studies such as those by Anwar et al. (2020), who demonstrated that integrating machine learning models with SEM enhances the accuracy and robustness of complex environmental-economic analyses. The use of ANN allows for validation of the relationships identified in the SEM model, providing further confidence in the results. 4. Empirical Results and Discussion Table 2 shows that GDP per capita ranges from 354.089 to 3667.057, with a mean of 1084.046 and a standard deviation of 725.298. This indicates a wide spread of economic development levels across the observations. The positive skewness (1.567) suggests that the distribution is right-skewed, with more observations below the mean and some higher values pulling the average up. CO2 emissions show a similar pattern, ranging from 0.051 to 1.074, with a mean of 0.33 and a standard deviation of 0.241. The positive skewness (1.275) indicates that most observations have lower CO2 emissions, with some higher emitters influencing the mean. PM2.5 levels range from 5.897 to 46.668, with a mean of 31.495 and a standard deviation of 10.992. All variables have Shapiro-Wilk p-values of 0.000, strongly suggesting that the variables do not follow a normal distribution. Table 2 Descriptive Statistics Min Max Mean Std. dev Skew Kurtosis Shapiro-Wilk stat p-value GDP_Per_Capita 354.089 3667.057 1084.046 725.298 1.567 1.672 0.782 0.000 CO2 0.051 1.074 0.33 0.241 1.275 1.061 0.865 0.000 pm.2.5 5.897 46.668 31.495 10.992 -0.602 -0.749 0.928 0.000 Nitrus_Oxide 55.534 42693.37 5736.831 8328.243 2.698 7.312 0.64 0.000 Health_Exp 2.279 19.69 5.057 2.424 2.804 11.239 0.739 0.000 Figure 2 shows that in several countries with diversified economic and environmental tendencies GDP per capita normally increases at different rates, very high in Cabo Verde and Nigeria, moderate in Cote d'Ivoire, Senegal, and Ghana, but rather stable yet lower in countries like Benin, Burkina Faso, and Guinea - Bissau. These economic trends are reflected in environmental indicators. In general, CO2 emissions are higher in the more industrial countries, such as Cabo Verde and Nigeria. The levels of PM2.5 are trending down across most countries, except Cabo Verde. The nitrous oxide emissions greatly fluctuate: the most substantial increase is in Nigeria due to their extensive farming and industrial sectors. The less industrialized countries maintain lower and more stable levels. Figure 3 displays the correlation matrix between the GDP per capita rate and all independent variables under study. GDP per capita has a strong positive association with CO2 emission level, indicating high growth rates for the economy have been matched by proportionate amounts spent on carbon production. Conversely, PM2.5 shows a negative relationship with GDP per capita. One could hypothesize that more developed regions could have much stricter environmental regulations leading to clean air policies or deploy more friendly industries that produce less pollution. However, nitrous oxide is weakly positively correlated showing its marginal association between income levels and this greenhouse gas. This suggests that while nitrous oxide emissions tend to increase with economic development, the relationship is not as strong as with CO2 emissions. Health expenditure displays a weak negative correlation with GDP per capita. 4.1 SEM Model Performance Evaluation Table 3 Model Performance Evaluation Q²predict PLS-SEM_RMSE PLS-SEM_MAE LM_RMSE LM_MAE IA_RMSE IA_MAE GDP_Per_Capita 0.857 0.378 0.285 0.378 0.285 1.002 0.775 Health_Exp 0.107 0.949 0.631 0.949 0.631 1.004 0.652 The evaluation of model performance against predictive accuracy and errors with regard to GDP per capita and health expenditure is presented in Table 3 . The high value of Q²predict for GDP per capita of 0.857 indicates strong predictive relevance, combined with very low values of RMSE of 0.378 and MAE of 0.285, which indicates that PLS-SEM serves a good predictive performance. Comparison with LM and IA further supports this result for the variable GDP. Table 4 Model Fit Model SRMR d_ULS d_G Chi-square NFI Saturated model 0.000 0.000 0.000 0.000 1.000 Estimated model 0.000 0.000 0.000 -0.000 1.000 Table 4 shows the Model fit indices of the saturated and estimated model. The fit statistics for both models are perfect: SRMR, d_ULS, d_G, and Chi-square all equate to zero and NFI equals 1.000, proving an excellent fit-meaning the estimates of the model fitted are very close to the empirically observed data. That is a perfect fit, which hardly ever happens; this may indicate that data and model specification need to be cleaned from any sort of irregularity for the estimation procedure. Table 5 Collinearity Check (VIF) Relationship VIF Air Quality Index -> Economic Growth 1.389 Air Quality Index -> Health Expenditure 1.388 Carbon Emission -> Economic Growth 1.402 Carbon Emission -> Health Expenditure 1.342 Health Expenditure -> Economic Growth 1.131 Nitrous Oxide -> Economic Growth 1.106 Nitrous Oxide -> Health Expenditure 1.067 Table 5 represents the VIF values for the relationships specified in the structural model. All the VIFs stand below 1.5, and hence there is no problem with multicollinearity in this model. Multicollinearity occurs whenever a high degree of intercorrelation among predictor variables exists and can be experienced with an increase in standard errors with unreliable estimates. All the VIF values are way below the threshold of 5; hence, we are guaranteed that each predictor adds unique information to the model, which ensures that the estimates of the regression coefficients are appropriate and not over-inflated. Table 6 Direct Effects of Environmental Degradation on Economic Growth Relationship Beta Standard deviation T statistics (|O/STDEV|) P values Air Quality Index -> Economic Growth -0.306 0.029 10.628 0.000 Air Quality Index -> Health Expenditure 0.036 0.040 0.898 0.369 Carbon Emission -> Economic Growth 0.716 0.034 21.244 0.000 Carbon Emission -> Health Expenditure -0.231 0.032 7.159 0.000 Health Expenditure -> Economic Growth 0.045 0.016 2.759 0.006 Nitrous Oxide -> Economic Growth 0.123 0.032 3.802 0.000 Nitrous Oxide -> Health Expenditure -0.186 0.023 8.102 0.000 Presented in Table 6 shows that carbon emissions have a high positive impact on the economic development of the ECOWAS region. This strong positive relationship (t = 21.244) suggests that the kinds of economic activities taking place in the region, in particular, those contributing to carbon emissions, have a deep connection with economic performance overall. Carbon-intensive practices in manufacturing and the generation of energy spur economic development due to the adverse impact it has on the environment. This in itself is a very valid observation underlining the persistent conflict between ecological sustainability and the economic imperatives of development within the region. The Air Quality Index and economic growth are inversely correlated, with the levels of PM2.5 recording a statistically significant negative association: β = -0.306, p < 0.001. This means that poor air quality is negatively impacting economic growth in the region. The higher t-statistic, t = 10.628 supports the result even more. The generally worsening air quality tends to generate significant and increasing costs through lost productivity, adverse health effects, and more generally lower economic performance. It thus follows that environmental degradation is broadly expensive and underlines that clean air is a prerequisite for continued economic growth. Nitrous oxide emissions are positively and significantly related to economic growth, with β = 0.123, p < 0.001. Although the effect size here is smaller than that from carbon emissions, this relationship does hold and suggests that insofar as nitrous oxide emissions reflect industrial and agricultural practices, these have a supporting role in enhancing economic performance within the region. With the t-statistic of 3.802, nitrous oxide pollution, like carbon emissions, is an economic activity fundamental to the current growth trajectory of the ECOWAS nations, though clearly at considerable environmental costs. Table 7 Indirect Effects via Health Expenditure Beta Standard deviation T statistics P values Nitrus Oxide -> Health Expenditure -> Economic Growth -0.008 0.003 2.622 0.009 Air Quality Index -> Health Expenditure -> Economic Growth 0.002 0.002 0.735 0.462 Carbon Emission -> Health Expenditure -> Economic Growth -0.010 0.004 2.355 0.019 The results in Table 7 show that health expenditure mediates the impact of environmental degradation on economic growth due to carbon and nitrous oxide emissions. The mediated effect of carbon emissions on health expenditure has a negative effect on economic growth, and it is statistically significant (β = -0.010, p = 0.019); hence, environmental degradation increases health expenditure and slows down economic growth. Similarly, nitrous oxide emission contributes negatively indirectly at β = -0.008, p = 0.009, meaning health impacts due to pollution slightly offset the economic gains accrued from growth. On the other side, the mediating effect of the Air Quality Index is insignificant, β = 0.002, p = 0.462, implying that health expenditure is not significant in affecting the relationship of air quality-economic growth. 4.2 Artificial Neural Networks (ANN) Results Table 8 Model Performance Neural Networks N Training Training Testing Training SSE RMSE N SSE RMSE Total Sample 1 298 14.792 0.223 32 2.422 0.275 330 2 290 10.458 0.19 40 0.873 0.148 330 3 295 12.92 0.209 35 1.766 0.225 330 4 285 9.354 0.181 45 1.449 0.179 330 5 302 14.279 0.217 28 1.435 0.226 330 6 293 23.891 0.286 37 0.988 0.163 330 7 290 15.923 0.234 40 2.198 0.234 330 8 303 12.326 0.202 27 0.742 0.166 330 9 290 15.899 0.234 40 1.194 0.173 330 10 292 14.095 0.22 38 1.53 0.201 330 Mean 14.394 0.22 1.46 0.199 St. Dev 3.978 0.029 0.55 0.04 Note. N = sample size; SSE = sum of square error, RMSE = root mean square of errors. Table 8 shows that the variation of RMSE in the training lies between 0.181 and 0.286, with an average of 0.220 when the model is trying to predict the target variable at this level. The SSE varies among these networks; Network 6 has the highest with an SSE of 23.891, while Network 4 has the lowest with an SSE value of 9.354, indicating thereby the difference in the model capability of LSE minimization within this training. A low value of RMSE implies perfect performance by the model with regard to fitting the training data. During the testing phase, the generalizing capability of the ANN model is quite good and gives a mean RMSE: of 0.199. A relatively small standard deviation in the RMSE values, 0.040, represents coherence among the various diverse networks during the prediction of unseen data. These are the best test results from, say, Network 2, giving an RMSE value of 0.148, while Network 1 is relatively higher, having a value of 0.275, reflecting network variability, yet still within acceptable limits. The ANN model has substantial predictive power as the error in both training and testing is negligible; hence, it is reliable to study the inter-relationships among the environmental variables, healthcare expenditure, and economic growth in the ECOWAS region. 4.2 Feature Importance Analysis: Output = Economic Growth (GDP) Table 9 Sensitive Analysis: Output = Economic Growth (GDP) NI1 NI2 NI3 NI4 NI5 NI6 NI7 NI8 NI9 NI10 A1 Rank % CO2 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 100% pm.2.5 0.42 0.51 0.54 0.58 0.62 0.55 0.55 0.52 0.43 0.42 0.52 52% Nitrus_Oxide 0.19 0.23 0.17 0.61 0.39 0.12 0.17 0.30 0.19 0.44 0.28 28% Note. NI = normalized importance, AI = average importance, and I = importance/normalized relative importance. Table 9 presents the sensitivity analysis of feature importance for predicting economic growth (GDP) using Artificial Neural Networks (ANN). Carbon emissions (CO₂) show the highest importance with a normalized value of 1.00 across all indicators, representing 100% relative importance. This finding suggests that CO₂ emissions are the most influential variable in determining economic growth in the ECOWAS region. The strong dependence on carbon-intensive activities for economic expansion is clear, reinforcing the critical role that industrial activities and energy consumption, often powered by fossil fuels, play in driving GDP growth. This outcome reflects the complex trade-offs faced by ECOWAS countries where economic progress heavily relies on industries that contribute significantly to carbon emissions. On the other hand, PM2.5, which stands for perceived air quality, has fair significance, with an average normalized importance of 0.52 or 52% of the predictive capability regarding GDP. Evidence that while inadequate air quality does have a perceivable effect on economic growth, its importance is comparatively lower compared to carbon emissions. This moderate rating of PM2.5 points to indirect costs regarding sub-optimality in air quality, which may increase health costs or depress productivity among workers. Nitrous oxide emission was perceived to be of least concern, measuring an average of 0.28 (28% significant); thus, while there might be a role played by nitrous oxide emissions in economic growth, the level of such an influence is much less impressive compared to CO₂ and PM2.5 concentrations. As such, it is a gas that is usually related to agriculture or industry, so its emissions hardly have an importance in determining income. Table 10 Comparison between PLS-SEM and ANN results Output (GDP) Constructs Path Coefficient PLS-SEM Ranking ANN-normalised relative importance (%) ANN ranking Matching PLS-SEM with ANN CO2 0.716 1 100% 1 Match PM.2.5 -0.306 2 52% 2 Match Nitrus_Oxide 0.123 3 28% 3 Match Table 10 presents the comparison between the path coefficients of SEM and the normalized importance values of ANN with regard to GDP. It could be found that the hypothesis that carbon emission is strongly positively related to economic growth could be shared by both models. The carbon emissions show the highest path coefficient of 0.716 in SEM and 100% in ANN, indicating the complete match of these two methods. The consistency of the findings across the different analytical approaches suggests that carbon emissions, irrespective of the approach adopted, are the most crucial determinant factor in influencing GDP growth in the ECOWAS region. Consistency between SEM and ANN suggests that the strong and reliable leading role of carbon emissions is leading the economic path of ECOWAS countries. Both SEM and ANN ranked the level of PM2.5 as the second most important factor that affects GDP, with a negative path coefficient in SEM (β = -0.306) and a normalized importance of 52% in ANN. Such consistency obtained from the two models would thus suggest that though air pollution impairs the linkage of economic growth, it is less important compared with that induced by carbon emissions. For the nitrous oxide emission, SEM also shows a positive but relatively smaller path coefficient, β = 0.123 agreeing with the lower relative importance identified by ANN at 28%. Both nitrous oxide emissions have a smaller impact on GDP; thus, these are less influential variables than carbon emissions and air quality. 4.3 Feature Importance Analysis: Output = Health Expenditure Table 11 Sensitive Analysis NI1 NI2 NI3 NI4 NI5 NI6 NI7 NI8 NI9 NI10 A1 Rank % CO2 1.00 1.00 1.00 1.00 1.00 0.73 1.00 1.00 1.00 0.66 0.9384 100% pm.2.5 0.52 0.31 0.38 0.19 0.80 0.26 0.17 0.43 0.54 0.08 0.3667 39% Nitrus_Oxide 0.85 0.84 0.83 0.76 0.69 1.00 0.79 0.90 0.99 1.00 0.8645 92% Note. NI = normalized importance, AI = average importance, and I = importance/normalized relative importance. Table 11 shows the sensitivity analysis of the determinants of health expenditure using the ANN methodological approach. Carbon emission (CO₂) has the highest normalized importance of 100%, indicating that it is the most important determinant of health expenditure within the ECOWAS sub-region. The high level of importance of CO₂ is indicative that the pollution resulting from highly carbon-intensive activities such as industrial processes and transportation, is directly affecting the public health for which health expenditures are rising. An increase in CO₂ emissions indicates a rise in respiratory and cardiovascular diseases due to pollution, considering the investments that should be made in health systems. These are, respectively, the second most relevant when projecting health expenditure. Nitrous oxide emissions are at an importance of 92% normalized. The highest sources of nitrous oxide come from agriculture and industrial processes. Because of its gigantic health implications, it is also a very potent greenhouse gas to global warming and deterioration of air quality. This high importance score reflects a strong linkage with environmental degradation and related escalating healthcare costs. On the other extreme, levels of PM2.5 rank the lowest at 39% normalized importance in the ranking scale. While air pollution-quantified through PM2.5 levels and health complications are indeed linked, it is ranking so far below the other two showing it having a relatively less economic impact on health care systems through particulate matter compared to the graver implications through CO₂ and nitrous oxide emissions. Table 12 Comparison between PLS-SEM and ANN results Output (Health Expenditure) Constructs Path Coefficient PLS-SEM Ranking ANN-normalised relative importance (%) ANN ranking Matching PLS-SEM with ANN CO2 -0.231 1 100% 1 Match Nitrus_Oxide -0.186 2 92% 2 Match PM.2.5 0.036 3 39% 3 Match Table 12 compares the results obtained between SEM-ANN on health expenditure prediction. The ranking produced from both models is the same, with CO₂ as the most important variable among the factors. It is indicated that in SEM, the path coefficient for CO₂ emissions is negative, β = -0.231, which infers that health expenditure is negatively affected by increased CO₂ emission. This may probably be because high pollution exposure increases the basic economic burden on healthcare. ANN also confirms this, since the normalized importance of CO₂ emissions is 100%; hence, CO₂ is ranked first according to both approaches. Both models rank NOx emission in the second position in terms of ranking order. It bears a path coefficient of β = − 0.186, and as usual, a high magnitude of this gas promotes health expenditure since both governments and households try to do something to reduce the health effects caused by environmental pollution. This follows then from ANN, where nitrous oxide yields a relative importance of 92%. PM2.5 levels are therefore less influential in either model's determination but retain some level of importance. While air pollution is of a minor magnitude, it has a positive effect on health expenditure; the little positive path coefficient of 0.036 is shown in the SEM. ANN also gives a lower normalized importance of 39% to PM2.5. Weaker in influence, both models affirm that air quality does contribute to health expenditure, but once again at a lesser magnitude than CO₂ and nitrous oxide. Table 13 Hypothesis Testing Hypothesis Path Direction Beta P-value Decision H1a Carbon Emission → Economic Growth Positive 0.716 0.000 Supported H1b Air Quality Index → Economic Growth Negative -0.306 0.000 Supported H1c Nitrous Oxide → Economic Growth Positive 0.123 0.000 Supported H2a Carbon Emission → Health Expenditure → Economic Growth Negative -0.010 0.019 Supported H2b Air Quality Index → Health Expenditure → Economic Growth Positive 0.002 0.462 Not Supported H2c Nitrous Oxide → Health Expenditure → Economic Growth Negative -0.008 0.009 Supported 4.4 Discussion The study results strongly support the first hypothesis: carbon emissions and nitrous oxide emissions have a positive influence on economic growth, whereas poor air quality greatly hinders economic growth. The positive relation between carbon emissions and economic growth yields, as expected from earlier studies conducted in rapidly industrializing parts of the world, such as China, where economic growth usually comes at the cost of environmental quality (Liu & Raven, 2010; Luo et al., 2022). This is in line with the context of many developing nations, such as ECOWAS, which are characterized by carbon-intensive industries that drive economic output but greatly pollute the environment. This is further evidenced in the positive association of nitrous oxide emissions with growth, implying that the contribution of agriculture and industry in the regions acts to sustain economic growth. Other studies account for agricultural intensification and industrial emissions, further supporting this view in developing countries (Aryal et al., 2022; Wu et al., 2021). In contrast, the negative association of air quality-PM2.5 levels with economic growth was inverse: β = -0.306, p < 0.001, indicating negative economic effects due to air pollution, which have also been claimed by the World Health Organization (2021) and Demir et al. (2022). This fact is corroborated by other research showing that air pollution decreases labour productivity and raises healthcare costs, thereby reducing the likelihood of investment and, consequently, limiting economic growth (Demir et al., 2022; Neidell, 2017). In developing countries where healthcare infrastructure is often overburdened, the economic cost of pollution is magnified. The results also indicate that the EKC hypothesis may not be fully applicable to the ECOWAS region. Although coupled with a rise in economic growth, environmental degradation suggests a deterring impact of poor air quality on economic growth, hence deviating from the inverted U-shaped EKC. This therefore supports the other models presented by Shahbaz and Sinha (2019) that because of increased consumption and energy demands, environmental degradation may rise again at higher income levels. The second hypothesis, which was on the mediating effect of health expenditure, was only partially supported. Health expenditure significantly mediates the relationship between nitrous oxide emissions and economic growth, with a β = -0.008 and a p-value of 0.009, accounting for 13.91% of the total effect. This means that environmental degradation, nitrous oxide emissions, and economic growth are partly reduced by the added healthcare costs. These findings are consistent with the observations of Novignon and Lawanson (2017), that increased health expenditure could better some of the negative economic effects of degradation through improvements in public health and productivity. The limit to mediation in the ECOWAS context might be related to regional differences in healthcare systems and environmental policies. The insignificant mediation effect of health expenditure in the relationship between air quality and economic growth would therefore suggest that the economic costs of poor air quality are more direct rather than via increased health expenditure. This result is in line with findings from Uzochukwu et al. (2015), who observe that in regions where healthcare infrastructure is relatively low like in the ECOWAS sub-region, the ability of the healthcare system to absorb and cushion the economic effect of environmental degradation is limited. In such settings, remedying poor air quality based on better healthcare investments may not be feasible. Instead, stronger environmental regulations are needed upfront to reduce pollution and protect public health as well as economic growth (Kutlu, 2021). 5. Conclusion This study has undertaken the analysis of the inter-linkages between environmental degradation, health expenditure and economic growth through SEM and ANN analysis in the ECOWAS region. Carbon emission and nitrous oxide were found to be the positive determinants of economic growth, while poor air quality lowers growth. Health expenditure is mediated between nitrous oxide emission and economic growth. This result would imply that as much as industrial operation and agricultural intensification spur economic growth, the costs of the long-run degradation of environmental and human health related to these activities pose serious risks to sustainable development. The limited mediator function of health expenditure underlines the need for the region to further develop its healthcare infrastructure and to apply more active environmental policy. These results thus imply that there should be strict environmental regulation that limits carbon and nitrous oxide emissions, though, in the same breath, they are investing in cleaner technologies that will not contravene economic growth. Additionally, strengthening public health systems will reduce the health effects of degraded environments, mainly by reducing the burden of industrial emissions. The governments should also promote air quality enhancement policies, given that air quality directly affects economic productivity. This might hint at what the future of research should consider, that is, the impact of the integration of renewable energy on economic development in the long run within the region of ECOWAS, studying other mediating variables that could inform the relationship between environmental and economic growth. Declarations Acknowledgements Not applicable Funding The authors declare that the study was conducted independently and without financial support or funding from any source. Availability of data and materials The datasets analyzed during the current study will be made available from the corresponding author on reasonable request. Informed Consent Statement Not applicable Author information Authors and Affiliations School of Public Policy and Administration, Northwestern Polytechnic University, Xi’an, China. Seth Acquah Boateng Department of Computer Science, University of Calgary, Alberta, Canada. Andy Asare Department of Data Science and Economic Policy, University of Cape Coast, Ghana William Godfred Cantah & Joshua Sebu Contributions SAB conceptualized the key constructs of the research idea. SAB, AA, designed the methodology, conducted the research/investigation/analyses, SAB, AA, wrote the original manuscript, and WC and JS read/approved the final manuscript. Corresponding Author Correspondence to Andy Asare Email: [email protected] Consent for publication Not applicable. Competing interests The authors declare that they have no competing interests. References Aboubacar, B., & Xu, D. (2017). The impact of health expenditure on economic growth in Sub-Saharan Africa. Theoretical Economics Letters, 7 (3), 615-622. African Development Bank. (2023, July 19). Africa's economic growth to outpace global forecast in 2023-2024 – African Development Bank biannual report. https://www.afdb.org/en/news-and-events/press-releases/africas-economic-growth-outpace-global-forecast-2023-2024-african-development-bank-biannual-report-58293 Alrais, G., Godara, J., Godara, S. P., Alnakeb, A., & Khaled, E. (2024). Environmental problem of nitrous oxide in obstetrics: A case review. Global Journal of Research Analysis, 13 (2), 116-118. Anwar, A., Hyder, S., Bennett, R., & Younis, M. (2022). Impact of environmental quality on healthcare expenditures in developing countries: A panel data approach. Healthcare, 10 (1608). Anwar, A., Siddique, M., Dogan, E., & Sharif, A. (2021). The moderating role of renewable and non-renewable energy in environment-income nexus for ASEAN countries: Evidence from method of moments quantile regression. Renewable Energy, 164 , 956-967. Artekin, A. Ö., & Konya, S. (2020). Health expenditure and economic growth: Is the health-led growth hypothesis supported for selected OECD countries? Poslovna Izvrsnost, 14 (1), 77-89. Aryal, B., Gurung, R., Camargo, A. F., Fongaro, G., Treichel, H., & Mainali, B. (2022). Nitrous oxide emission in altered nitrogen cycle and implications for climate change. Environmental Pollution, 314 , 120272. Ayodotun, B., Bamba, S., & Adio, A. (2019). Vulnerability assessment of West African countries to climate change and variability. Journal of Geoscience and Environmental Protection, 7 (2), 13-15. Bloom, D. E., Canning, D., & Sevilla, J. (2004). The effect of health on economic growth: A production function approach. World Development, 32 (1), 1-13. Boateng, S. A., Karikari, F. A., Fumey, M. P., Asmah, E. E., & Winful, S. A. (2024). Understanding global commodity price shocks on exchange rates and inflation in emerging economies: ARDL perspective. Journal of Economics, Management and Trade, 30 (1), 33-47. Carpio-Thomas, C. (2020). The case of China . Digital Commons @ DU. https://digitalcommons.du.edu/cgi/viewcontent.cgi?article=2732&context=etd Demir, S., Demir, H., Karaduman, C., & Cetin, M. (2022). Environmental quality and health expenditures efficiency in Türkiye: The role of natural resources. Environmental Science and Pollution Research, 29 , 1-15. Dhhan, W., Rana, S., Alshaybawee, T., & Midi, H. (2018). The single-index support vector regression model to address the problem of high dimensionality. Communications in Statistics - Simulation and Computation, 47 (10), 2792-2809. Duan, T., Avati, A., Ding, D. Y., Thai, K. K., Basu, S., Ng, A. Y., & Schuler, A. (2019). NGBoost: Natural gradient boosting for probabilistic prediction. arXiv . https://arxiv.org/abs/1910.03225 ECOWAS. (2024, September 1). Member states | Economic Community of West African States (ECOWAS). https://www.ecowas.int/member-states/ Griffis, T. J., Chen, Z., Baker, J. M., Wood, J. D., Millet, D. B., & Lee, X. (2017). Nitrous oxide emissions are enhanced in a warmer and wetter world. Proceedings of the National Academy of Sciences of the United States of America, 114 (45), 12081-12085. Guo, J., Chai, G., Song, X., Xu, H., Li, Z., & Feng, X. (2023). Long-term exposure to particulate matter on cardiovascular and respiratory diseases in low- and middle-income countries: A systematic review and meta-analysis. Frontiers in Public Health, 11 , 1128432. Gwangndi, M. I., Muhammad, Y. A., & Tagi, S. M. (2016). The impact of environmental degradation on human health and its relevance to the right to health under international law. European Scientific Journal, 12 (3), 485. Ha, T. N., Lubo-Robles, D., Marfurt, K. J., & Wallet, B. C. (2021). An in-depth analysis of logarithmic data transformation and per-class normalization in machine learning: Application to unsupervised classification of a turbidite system in the Canterbury Basin, New Zealand, and supervised classification of salt in the Eugene Island minibasin, Gulf of Mexico. Interpretation, 9 (4), T685-T710. Intergovernmental Panel on Climate Change. (2021, August 9). Climate change widespread, rapid, and intensifying. https://www.ipcc.ch/2021/08/09/ar6-wg1-20210809-pr/ International Monetary Fund. (2021). Ghana: 2021 Article IV Consultation—Press release; staff report; and statement by the Executive Director for Ghana (IMF Staff Country Report 2021). https://www.imf.org/en/Publications Jabeur, S. B., Ballouk, H., Arfi, W. B., & Khalfaoui, R. (2021). Machine learning-based modeling of the environmental degradation, institutional quality, and economic growth. Environmental Modeling & Assessment, 27 (4), 953-966. Jones, F. C., Plewes, R., Murison, L., MacDougall, M. J., Sinclair, S., & Davies, C. (2017). Random forests as cumulative effects models: A case study of lakes and rivers in Muskoka, Canada. Journal of Environmental Management, 201 (1), 407-424. Kahia, M., Moulahi, T., Mahfoudhi, S., Boubaker, S., & Omri, A. (2022). A machine learning process for examining the linkage among disaggregated energy consumption, economic growth, and environmental degradation. Resources Policy, 79 , 103104. Kutlu, Ş. Ş. (2021). Türkiye ekonomisinde sağlığa dayalı büyüme hipotezinin geçerliliğine ilişkin ampirik bir analiz. Yaşar Üniversitesi Dergisi, 16 (1), 1808-1822. Lima, L. J. B., & Hamzagic, M. (2022). Greenhouse gases and air pollution: Commonalities and differentiators. Revista Científica Multidisciplinar Núcleo do Conhecimento , 102-144. Liu, J., & Raven, P. H. (2010). China's environmental challenges and implications for the world. Critical Reviews in Environmental Science and Technology, 40 (9), 823-851. Luo, K., Liu, Y., Chen, P. F., & Zeng, M. (2022). Assessing the impact of digital economy on green development efficiency in the Yangtze River Economic Belt. Energy Economics, 112 , 106127. Madan, A., & Suri, A. (2023). The nexus between development and environment. International Journal of Scientific Research in Engineering and Management, 7 (1), 1-5. Moolla, I., & Hiilamo, H. (2023). Health system characteristics and COVID-19 performance in high-income countries. BMC Health Services Research, 23 (1), 242. Neidell, M. (2017). Air pollution and worker productivity. IZA World of Labor , 363. Novignon, J., & Lawanson, A. (2017). Health expenditure and child health outcomes in Sub-Saharan Africa. African Review of Economics and Finance, 9 (2), 96-121. Patel, P. (2021). Nitrous oxide: The unnoticed greenhouse gas. Chemical & Engineering News, 99 (21), 20-23. Potts, D., & Schmischke, M. (2022). Interpretable transformed ANOVA approximation on the example of the prevention of forest fires. Frontiers in Applied Mathematics and Statistics, 8 , 840015. Sarpong, B., Nketiah-Amponsah, E., & Owoo, N. S. (2018). Health and economic growth nexus: Evidence from selected Sub-Saharan African (SSA) countries. Global Business Review, 20 (1), 1-15. Setyari, N. P. W., & Kusuma, W. G. A. (2021). Economics and environmental development: Testing the environmental Kuznets curve hypothesis. International Journal of Energy Economics and Policy, 11 (1), 51-58. Shahbaz, M., & Sinha, A. (2019). Environmental Kuznets curve for CO2 emissions: A literature survey. Journal of Economic Studies, 46 (1), 106-168. Singh, R. L., & Singh, P. K. (2016). Global environmental problems. In Principles and applications of environmental biotechnology for a sustainable future (pp. 13-41). Springer. Singh, S., & Yadav, A. (2021). Interconnecting the environment with economic development of a nation. In Environmental management (pp. 35-60). Elsevier. Towah, W. (2019). The impact of good governance and stability on sustainable development in Ghana (Doctoral dissertation, Walden University). Uzochukwu, B., Ughasoro, M., Okwuosa, C., Onwujekwe, O., Envuladu, E., & Etiaba, E. (2015). Health care financing in Nigeria: Implications for achieving universal health coverage. Nigerian Journal of Clinical Practice, 18 (4), 437-444. Wang, F. (2015). More health expenditure, better economic performance? Empirical evidence from OECD countries. INQUIRY: The Journal of Health Care Organization, Provision, and Financing, 52 , 1-5. Wassie, S. B. (2020). Natural resource degradation tendencies in Ethiopia: A review. Environmental Systems Research, 9 , 33. World Bank. (2020). Ghana country environmental analysis . Washington, DC: World Bank. World Health Organization. (2021, September 22). Air pollution. https://www.who.int/health-topics/air-pollution#tab=tab_1 World Health Organization. (2021, January 1). Health expenditure. https://www.who.int/data/nutrition/nlis/info/health-expenditure Wu, J., Abban, O. J., Yao, H., Boadi, A. D., & Ankomah-Asare, E. T. (2021). The nexus amid foreign direct investment, urbanization, and CO2 emissions: Evidence from energy grouping along the ECOWAS community. Environmental Science and Pollution Research, 24 (12), 10183-10207. Xing, Z., & Liu, X. (2022). Health expenditures, environmental quality, and economic development: State-of-the-art review and findings in the context of COP26. Frontiers in Public Health, 10 , 1005705. Zaidi, S., & Saidi, K. (2018). Environmental pollution, health expenditure and economic growth in the Sub-Saharan Africa countries: Panel ARDL approach. Sustainable Cities and Society, 41 , 833-840. Additional Declarations No competing interests reported. Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-5675754","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":396194087,"identity":"3adc57e8-f191-45fb-821d-1c383704b228","order_by":0,"name":"Seth Acquah Boateng","email":"","orcid":"","institution":"Northwestern Polytechnical University","correspondingAuthor":false,"prefix":"","firstName":"Seth","middleName":"Acquah","lastName":"Boateng","suffix":""},{"id":396194088,"identity":"ef04a3db-aa84-4fd4-aea5-b4b8d5bfbbb8","order_by":1,"name":"Andy Asare","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABBElEQVRIiWNgGAWjYHACAyjNfABIWJCkhS0BSEiQpIXHgDgt/LObt0kw1NTJm7f3fJP4USNh1z/7jAHDjxrcWiTuHCuTYDh22HDOmbPbJHuOSSTPOJdjwNhzDI81N3LMJBjYDjDOkMjdJs3AJpHMcIbHgJmBDbcOebCWf3X2M+TfPJNm+CeRLA/W8g+3FgOQFsY25sQZEjxs0oxtEnYGIC2Mbbi1GN5IK7ZI7DucPIMnzdiyt08iwfAMW8HB3j7cWuRuJG+88eFbne0M9sMPb/z4ZmMvd4Z544Mf3/B4n4GBRSIBiZfYwMBhcACvBmBC+YDMs2dgYH9AQMcoGAWjYBSMMAAAfsVNcbHRXKcAAAAASUVORK5CYII=","orcid":"","institution":"University of Calgary","correspondingAuthor":true,"prefix":"","firstName":"Andy","middleName":"","lastName":"Asare","suffix":""},{"id":396194089,"identity":"635ab29b-6c8d-4754-89f5-8daf4e9b9b03","order_by":2,"name":"William Godfred Cantah","email":"","orcid":"","institution":"University of Cape Coast","correspondingAuthor":false,"prefix":"","firstName":"William","middleName":"Godfred","lastName":"Cantah","suffix":""},{"id":396194090,"identity":"83e97290-41f3-4f7d-84a2-050f37ba5229","order_by":3,"name":"Joshua Sebu","email":"","orcid":"","institution":"University of Cape Coast","correspondingAuthor":false,"prefix":"","firstName":"Joshua","middleName":"","lastName":"Sebu","suffix":""}],"badges":[],"createdAt":"2024-12-19 10:08:13","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-5675754/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-5675754/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":72784609,"identity":"b9171a4e-2aaf-49b6-bbc4-213f52771b30","added_by":"auto","created_at":"2025-01-02 06:46:24","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":56247,"visible":true,"origin":"","legend":"\u003cp\u003eConceptual Framework\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-5675754/v1/9bdf6a5004b921b3bee7b4e5.png"},{"id":72784610,"identity":"9a71d1c7-65d4-468e-9c1a-c0dd80490322","added_by":"auto","created_at":"2025-01-02 06:46:24","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":168762,"visible":true,"origin":"","legend":"\u003cp\u003eTime series Trend of Variables\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-5675754/v1/c5fccb483bd1c8626e337e4a.png"},{"id":72784611,"identity":"a0661a71-bec9-4907-9ee0-4ab49bc467e0","added_by":"auto","created_at":"2025-01-02 06:46:24","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":50010,"visible":true,"origin":"","legend":"\u003cp\u003eCorrelation Matrix of Variables\u003c/p\u003e","description":"","filename":"3.png","url":"https://assets-eu.researchsquare.com/files/rs-5675754/v1/8122afb87f68b4011cd6421e.png"},{"id":72786068,"identity":"c9f75984-ecee-4937-b603-ae212d309716","added_by":"auto","created_at":"2025-01-02 06:54:24","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":76295,"visible":true,"origin":"","legend":"\u003cp\u003eStructural Equation Model Path Analysis\u003c/p\u003e","description":"","filename":"4.png","url":"https://assets-eu.researchsquare.com/files/rs-5675754/v1/23637aa289807428ba845896.png"},{"id":72784617,"identity":"c7e10016-6116-46f4-be02-1a7690325969","added_by":"auto","created_at":"2025-01-02 06:46:24","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":120329,"visible":true,"origin":"","legend":"\u003cp\u003eVisual representation of ANN for GDP\u003c/p\u003e","description":"","filename":"5.png","url":"https://assets-eu.researchsquare.com/files/rs-5675754/v1/f9d6a489363b25081d5ab863.png"},{"id":72784623,"identity":"d534dd83-c6ee-45f7-a514-ef8da92023aa","added_by":"auto","created_at":"2025-01-02 06:46:24","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":60360,"visible":true,"origin":"","legend":"\u003cp\u003eVisual representation of Predictive Value of ANN\u003c/p\u003e","description":"","filename":"6.png","url":"https://assets-eu.researchsquare.com/files/rs-5675754/v1/76b55e3a32870eab385ee156.png"},{"id":72784613,"identity":"30c659b0-1ae6-42d5-9d67-1bb72ae4a75d","added_by":"auto","created_at":"2025-01-02 06:46:24","extension":"png","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":86012,"visible":true,"origin":"","legend":"\u003cp\u003eVisual representation of ANN for Health\u003c/p\u003e","description":"","filename":"7.png","url":"https://assets-eu.researchsquare.com/files/rs-5675754/v1/c4dd3dfd43622cc6c54f9160.png"},{"id":72786385,"identity":"63386b88-90f2-4d67-9380-65763643b95b","added_by":"auto","created_at":"2025-01-02 07:02:24","extension":"png","order_by":8,"title":"Figure 8","display":"","copyAsset":false,"role":"figure","size":64163,"visible":true,"origin":"","legend":"\u003cp\u003eVisual representation of Predictive Value for Health\u003c/p\u003e","description":"","filename":"8.png","url":"https://assets-eu.researchsquare.com/files/rs-5675754/v1/4f8e29bc1776582007ad5e58.png"},{"id":83975970,"identity":"505bd17d-50e6-417c-ad0b-e66f6ea1bdfd","added_by":"auto","created_at":"2025-06-05 09:02:11","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":2094962,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-5675754/v1/79101a8a-36a4-4d92-ba3a-6abc6c489721.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Green Growth or Economic Trade-offs? Economic Costs of Environmental Degradation in ECOWAS. A New Perspective from Artificial Neural Network and SEM Analysis","fulltext":[{"header":"1. Introduction","content":"\u003cp\u003e \u003cdiv class=\"BlockQuote\"\u003e \u003cp\u003eAs countries pursue economic growth, they also have to confront the twin problem of fighting environmental degradation and improving public health (Boateng et al., 2024). It requires a comprehension to develop appropriate policies that strike a balance between development objectives as well as environmental issues. Environmental degradation is expressed through air pollution and greenhouse gas emissions (Lima \u0026amp; Hamzagic, 2022). This idea has been associated with economic activities. According to Gwangndi et al. (2016), environmental degradation has a crucial impact on the health of humans. The importance of this relationship is because, in most cases, public resources are scarce, and the health systems of the countries in that region are under immense pressure (Anwar et al., 2022). Despite their significance to economic progress, these factors have also led to various ecological problems.\u003c/p\u003e \u003cp\u003eRapid urbanization, industrialization, and rising energy consumption have been features of the economic growth within the ECOWAS region (Wu et al., 2021). The Economic Community of West African States (ECOWAS) is constituted of fifteen member nations, which comprise Benin, Burkina Faso, Cape Verde, C\u0026ocirc;te d'Ivoire (Ivory Coast), The Gambia, Ghana, Guinea, Guinea-Bissau, Liberia, Mali, Niger, Nigeria, Sierra Leone, Senegal, and Togo (ECOWAS, 2019). Liu and Raven (2010) and Luo et al. (2022) have shown that China\u0026rsquo;s pursuit of rapid economic expansion resulted in serious deterioration of the environment and warned other developing nations not to make a similar mistake for fear of long-term repercussions. Therefore, countries need to learn from this and get onto more workable development tracks lest they lag in the heel of the flourishing economies around the earth. Xing and Liu, 2022 established that regulatory quality and the rule of law have significant effects on healthcare spending and environmental outcomes. This calls for a reason why strong institutions are to be built up for resilience and application of the environmental regulations in the countries of the ECOWAS region for economic growth while protecting the environment and public health.\u003c/p\u003e \u003cp\u003eThis is because the aim of sustainable development among the ECOWAS communities ensures balanced economic growth with environmental conservation and social welfare. In this regard, Madan and Suri (2023) argue that the thought process has to be changed from \"environment versus development\" to \"environment and development.\" Only then will the ECOWAS community be able to plan for long-term economic growth without compromising environmental standards and public health. Demir et al. (2022) estimated asymmetry in the impact of health expenditures concerning environmental quality and therefore called for adaptive policies concerning negative and positive responses to environmental shocks. Indeed, the adaptable policies open new avenues for building up resilient healthcare systems and environmental policies against the adverse effects of climate change in the ECOWAS region. That by Kahia et al. (2022) implies that, in mitigation strategies for CO2 emissions, reliance on fossil fuel needs to be done away with by shifting toward renewable sources if economic growth has to be sustained. Changes in multidimensional perspectives also include those of environmental impact, public health, and economic development, among other dimensions over this region of Africa. It has been, therefore, one of the major concerns among environmental degradation, health expenditure, and economic progress within the ECOWAS region and, as such, requires advanced research for full comprehension in terms of evidence-based policy formation.\u003c/p\u003e \u003cp\u003eMachine learning models have recently obtained greater recognition in research studies focused on environmental factors and economic growth. In this respect, the study of Jabeur et al. (2021) conducted the forecast analysis of CO2 emissions using Natural Gradient Boosting and estimated the key determinants affecting environmental sustainability. This study applied ANN and SEM to unravel the linkages; hence, it unearths new evidence over critical drivers of environmental performance in the region. Lastly, health expenditure also acts as a mediator in the relationship between environmental degradation and economic growth. For example, countries with emerging economies in SSA are investing an increasingly high share of their public finances in the health sector, and environmental degradation tends to heighten health expenditures (Aboubacar \u0026amp; Xu, 2017). Thus, even as such countries will generally register economic growth, they will be able to invest more in hospitals as they register ecological degradation.\u003c/p\u003e \u003cp\u003eThis study investigates the interrelated relationships that exist between environmental degradation and health expenditure coupled with economic growth through the SEM and ANN algorithms. The study looks at the influence of carbon dioxide, air pollution, and nitrous oxide on gross domestic product per capita and health expenditures, playing the mediating role. Therefore, this paper tends to bring forth state-of-the-art insights into the Environmental Kuznets Curve hypothesis in West African countries and checks on various direct and indirect influences of growth with implications for healthcare expenditure. The findings from this project raise the knowledge bar relating to trade-offs between environmental quality and public health investment and economic development patterns and contribute to the development of better-designed and more sustainable policy formulation in this region over time.\u003c/p\u003e \u003c/div\u003e \u003c/p\u003e"},{"header":"2. Literature ","content":"\u003cp\u003e\u003cem\u003e2.1 Environmental degradation\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eAnthropogenic activities lead to environmental degradation, which is the deterioration of all the natural systems and resources of the earth (Wassie, 2020). These include air and water pollution, and cleanup of oil spills in the ocean. Soil degradation is also another example, along with deforestation, which leads to the loss of biodiversity and climate change (Singh \u0026amp; Singh, 2016). Wu et al. (2021) explained this phenomenon has a huge effect on ecosystems, human health, and even economic building. The very root of this problem is the huge amount of carbon dioxide emitted into the atmosphere due to the combustion of fossil fuel; this gas contributes to the absorption of more heat energy, hence intensifying global warming. Although the Intergovernmental Panel on Climate Change, (2021), impressively identified that the contribution of man-made activities is the cause of the global warming phenomenon, the manifestation is getting acuter in the different geographies by way of increased occurrences of extreme weather events like tornadoes, hurricanes, etc., with unparalleled intensities. Then again, there is air pollution, especially in metropolitan areas, which is another important dimension of environmental degradation.\u003c/p\u003e\n\u003cp\u003eAccording to the World Health Organization (2021), an astonishing 99% of the global population is breathing air which is even more polluted than allowed by the limits set by WHO standards, with countries in categories of low- and middle-income being mostly exposed. Particulate matter, especially PM2.5, has been associated with a wide range of health problems, including many respiratory and cardiovascular diseases (Demir et al., 2022; Guo et al., 2023). Emissions of nitrous oxide stemming from agricultural practices and industrial operations play a substantial role in exacerbating air pollution as well as contributing to climate change (Aryal et al., 2022). Nitrous oxide is recognized as a highly effective greenhouse gas, possessing a global warming potential that is 300 times greater than that of CO2 over a century-long timeframe (Patel, 2021; Griffis et al., 2017). Neidell (2017) indicated that a degraded environment can lead to lower productivity in sectors like agriculture and fisheries, increased healthcare costs, and damage to infrastructure. Specifically, Ayodotun et al. (2019) observed that countries within West Africa remain among those most vulnerable to slip backwards from development dividends due to climate change, which in turn further entrenches poverty. On this note, there is a need to address environmental concerns toward the sustainability of ecosystem integrity and laying a foundation for economic growth and human development in the area.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003e2.2 Health expenditure\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eHealth expenditure is defined by the World Health Organization (2021) as spending on health services, family planning activities, nutrition programs, as well as disaster relief meant for health care only. In the ECOWAS region, health expenditure varies across countries. In the context of Sub-Saharan Africa, Novignon and Lawanson (2017) discovered that healthcare expenditure has a positive and significant effect on health outcomes, such as reduced infant mortality rates and increased life expectancy. This implies that there are diminishing returns to scale, meaning not all levels require equal additions, as each unit increase, within a given limit of time, is associated with a decrease in another variable. Moolla and Hiilamo (2023) noted that countries with higher health spending were in a better position to deal with disease outbreaks, suggesting that investment in health infrastructure is key to thwarting such outbreaks. Nevertheless, increasing spending for health within the ECOWAS region faces numerous constraints, including inadequate resources compared to other sectors, competing against development projects (Uzochukwu et al., 2015). This undertaking is imperative for enhancing general well-being while at the same time boosting overall economic growth.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003e2.4 Hypothesis Development\u0026nbsp;\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003e\u003cem\u003e2.4.1 Environmental Deterioration and Economic Development\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eThe \u0026quot;Environmental Kuznets Curve\u0026quot; or EKC comes into play to describe the relation between degradation in the environment with economic growth. It states that as the nations start to grow, the environment will deteriorate first before it improves. Singh and Yadav, 2021, present evidence that the relationship between economic growth and environmental quality is inverted U-shaped. This generally suggests that during the early stages of growth, environmental degradation takes place, and thereafter, with income increases, it permits a shift toward cleaner modes of production and the adoption of efficient abatement practices. \u0026nbsp;Notwithstanding various criticisms raised against EKC generalizations, especially within developing contexts, Setyari and Kusuma (2021) debunked the expectations by presenting an \u0026quot;N\u0026quot; shaped pattern, where after further growth, there could be another degradation in environmental standards. This necessitates sustainable environmental management, even during periods of economic expansion, involving substantial funds.\u003c/p\u003e\n\u003cp\u003eEconomic development causes environmental degradation, which is not always straightforward. Carpio-Thomas and Christian (2020) observed a rapid economic transformation in China, which had profound environmental impacts, warning other developing countries against choosing between environmental preservation and economic growth at the expense of environmental protection laws. This issue is particularly relevant for ECOWAS, as many states aim to achieve quick progress. Zaidi and Saidi (2018) discovered that economic growth had a positive impact on health expenditure in Sub-Saharan African countries, while it was negatively affected by environmental pollution, showing a complicated relationship between economic growth, environmental quality, and public health investments. One critical issue concerning the state of the environment today is air quality, which the World Health Organization (2021) has reported as a major health hazard, especially in developing regions suffering from air pollution. Low air quality not only has public health consequences but also inhibits economic productivity and increases healthcare costs. Both pollution and global warming are exacerbated by the discharge of nitrous oxide. On the other hand, Alrais et al. (2024) pointed out that nitrous oxide has a severe greenhouse effect with implications for long-term environmental impairment. Consequently, the following hypothesis will be developed.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eH1a: There is a significant relationship between Carbon emissions and Economic Growth in the ECOWAS region.\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eH1b: There is a significant relationship between Air Quality Index and Economic Growth in the ECOWAS region.\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eH1c: There is a significant relationship between Nitrous oxide emissions and Economic Growth in the ECOWAS region.\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003e\u003cem\u003e2.4.2 Health Expenditure as a Mediating Factor\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eIt is hypothesized that health expenditure mediates the relationship between economic development and environmental deterioration. This role of mediator has implications for environmental outcomes, as well as growth paths. In this regard, it has been observed by Zaidi and Saidi 2018 that whereas economic growth increased health expenditure positively, it was affected negatively by pollution. This suggests that as economies grow, they may allocate more resources to healthcare, but environmental degradation simultaneously increases healthcare costs, creating a challenging dynamic for policymakers. Anwar et al. (2021) examined the effect that both non-renewable and renewable supplies of energy have on the environmental nexus for health expenditure. It could be even worse in the ECOWAS region; incongruous environmental policies see health expenditure levels rising due to increased health bills.\u003c/p\u003e\n\u003cp\u003eFurther research is perhaps needed to see if such trends are indeed repeated in other developing countries. In this context, while improved public health can result in an increase in economic productivity on the one hand (Artekin \u0026amp; Konya, 2020; Wang, 2015), on the other hand, there is the potential that once people are healthier, they may become less productive, which could reduce economic gains (Bu \u0026amp; Ali, 2018). However, it was revealed that countries do spend more on health services when their citizens are ill with diseases (World Health Organization, 2021). The same research proved that more money is spent on maintaining and improving the health of the population when there is a low level of environmental pollution (Bu \u0026amp; Ali, 2018). However, without proper environmental investments alongside health initiatives, long-term sustainability cannot be assured. This implies that neglecting environmental concerns while increasing health expenditure is self-defeating in the long run.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eH2a: Health expenditure significantly mediates the relationship between Carbon emissions and Economic Growth in the ECOWAS region.\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eH2b: Health expenditure significantly mediates the relationship between Air Quality Index and Economic Growth in the ECOWAS region.\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eH2c: Health expenditure significantly mediates the relationship between nitrous oxide emissions and Economic Growth in the ECOWAS region\u003cstrong\u003e.\u003c/strong\u003e\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003e\u003cem\u003e2.5 Research Framework\u0026nbsp;\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eFig. 1 presents the conceptual framework which highlights a model of the relationship between environmental degradation, economic development, and health expenditure within the ECOWAS region at large. This framework at its core asserts that environmental variables- represented by carbon dioxide emissions, PM 2.5 air quality index as well nitrous oxide emissions- have either direct or indirect impacts on economic progress as measured by GDP per capita. Indirectly, it touches upon healthcare spending thereby maintaining health expenditure as a pivotal intervening variable. Herein lies an avenue through which an environmentalist can understand how GDP growth might be affected not just directly but also indirectly through public health outcomes such that its effects on public health and subsequent health-care spending are evident.\u0026nbsp;\u003c/p\u003e"},{"header":"3. Methodology","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003e3.1 Data Description\u003c/h2\u003e \u003cp\u003e \u003cdiv class=\"BlockQuote\"\u003e \u003cp\u003eThe study utilized panel data from multiple reputable international sources to ensure comprehensive coverage of the ECOWAS region. The World Development Indicators database provided the primary economic data, including GDP per capita and health expenditure figures. Environmental data, specifically CO2 emissions and nitrous oxide emissions, were sourced from the UN database, which offers standard environmental metrics across countries. The World Health Organization contributed data on PM2.5 levels, providing a reliable measure of air quality. The study used yearly period data which span from 2000 to 2021.\u003c/p\u003e \u003c/div\u003e \u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eVariable Definition\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"3\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eVariable\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eDescription and Measurement\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eSource\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eGDP per capita\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eAnnual gross domestic product divided by midyear population, measured in current US dollars\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eWorld Development Indicators (World Bank)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHealth expenditure\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eCurrent health expenditure as a percentage of GDP\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eWorld Development Indicators (World Bank)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCO2 emissions\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eCarbon dioxide emissions, measured in metric tons per capita\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eWorld Development Indicators (World Bank)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNitrous oxide emissions\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eNitrous oxide emissions, measured in thousand metric tons of CO2 equivalent\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eWorld Development Indicators (World Bank)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAir Quality Index (PM2.5)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eAnnual mean concentration of particulate matter of less than 2.5 microns in diameter, measured in micrograms per cubic meter\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eWorld Health Organization\u003c/p\u003e \u003cp\u003eWord Bank Open Data\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec4\" class=\"Section2\"\u003e \u003ch2\u003e3.2 Data Cleaning and Standardization\u003c/h2\u003e \u003cp\u003eThe raw data were consolidated into a single dataset. Missing values were identified and addressed using interpolation for time series data and mean imputation where suitable. Outliers were detected and adjusted. All variables were standardized using z-score normalization, transforming them to have a mean of 0 and a standard deviation of 1, a crucial step for machine learning applications in environmental economics (Ha et al., 2021; Potts \u0026amp; Schmischke, 2022). Finally, a thorough quality check was conducted to verify the consistency and integrity of the cleaned and standardized dataset.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec5\" class=\"Section2\"\u003e \u003ch2\u003e3.3 Model Specification\u003c/h2\u003e \u003cdiv id=\"Sec6\" class=\"Section3\"\u003e \u003ch2\u003e3.3.1 Structural Equation Modeling (SEM)\u003c/h2\u003e \u003cp\u003eThe study applied the SEM to analyze the causal relationships that exist between direct and indirect effects arising from environmental degradation, health expenditure, and economic growth. The analysis was done using SmartPLS 4.0, one of the most advanced software packages that provides for PLS-SEM, well noted for handling complex and multidimensional models with the highest level of efficiency (Chuah et al., 2021; Hair et al., 2018; Khan et al., 2019). The primary reason why PLS-SEM was chosen is that the approach can handle non-normal distribution data, small and medium-sized samples, and both formative and reflective measurement models. SEM allows the analysis of several relations at once and tests both direct and indirect effects in one model. The structural model in the following analysis presents the relationships between exogenous latent variables, environmental degradation indicators, and endogenous latent variables representing economic growth, with the underlying mediation by health expenditure (Ringle et al., 2018).\u003c/p\u003e \u003cp\u003eThe algorithm of PLS works based on the Varimax method, maximizing the variance explained in the endogenous constructs, thereby improving the predictive relevance of the model (Cheah et al., 2020). Also, for this purpose, the interactions among these factor interactions have been analyzed based on path coefficients, and the significance of the latter has been extracted through the bootstrapping technique included in the PLS-SEM, giving more reliability to the results (Kock, 2019; Ramayah et al., 2018). To add to robustness, the stability of the path estimates and their statistical significance were tested using bootstrapping of 5,000 sub-samples (Kock, 2019). The model fit was evaluated using several goodness-of-fit indices, including Standardized Root Mean Square Residual (SRMR), Normed Fit Index (NFI), and Chi-Square statistics. These scores thus enable an assessment of the accuracy of the model and the degree to which the theoretical structure fits the data (Cheah et al., 2020). More specifically, SRMR focuses on the variance between the observed and variance-predicted correlation, while NFI quantifies the comparative increase in the model fit as opposed to a no-model specification (Ringle et al., 2015).\u003c/p\u003e \u003cp\u003eThe following is the general model equation used for SEM:\u003c/p\u003e \u003cp\u003eGeneral Model:\u003cdiv id=\"Equa\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equa\" name=\"EquationSource\"\u003e\n$$\\:EG\\:=\\:\\beta\\:₁CE\\:+\\:\\beta\\:₂NO\\:+\\:\\beta\\:₃AQ\\:+\\:ϵ\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:1$$\u003c/div\u003e\u003c/div\u003e\u003c/p\u003e \u003cp\u003eMediation Effect of Health Expenditure\u003cdiv id=\"Equb\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equb\" name=\"EquationSource\"\u003e\n$$\\:EG\\:=\\:\\beta\\:₄HE\\:+\\:(\\beta\\:₅CE\\:+\\:\\beta\\:₆NO\\:+\\:\\beta\\:₇AQ)\\:+\\:ϵ\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:2$$\u003c/div\u003e\u003c/div\u003e\u003c/p\u003e \u003cp\u003eWhere:\u003c/p\u003e \u003cp\u003eEG: Economic Growth (GDP per capita)\u003c/p\u003e \u003cp\u003eCE: Carbon Emissions\u003c/p\u003e \u003cp\u003eNO: Nitrous Oxide Emissions\u003c/p\u003e \u003cp\u003eAQ: Air Quality (PM2.5 levels)\u003c/p\u003e \u003cp\u003eHE: Health Expenditure\u003c/p\u003e \u003cp\u003eβ₁, β₂, β₃, β₄, β₅, β₆, β₇: Path Coefficients\u003c/p\u003e \u003cp\u003eϵ: Error Term\u003c/p\u003e \u003cp\u003eModel fit indices, such as the Standardized Root Mean Square Residual (SRMR), Normed Fit Index (NFI), and Chi-Square statistics, were calculated to assess the quality of the model. The formulas for these indices are as follows:\u003c/p\u003e \u003cp\u003eSRMR (Standardized Root Mean Square Residual):\u003cdiv id=\"Equc\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equc\" name=\"EquationSource\"\u003e\n$$\\:SRMR\\:=\\:\\sqrt{\\left[\\left(\\frac{1}{m}\\right)\\sum\\:{ᵢ}^{=1}ᵐ\\:{\\left(sᵢⱼ\\:-\\:ŝᵢⱼ\\right)}^{2}\\right]\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:3}$$\u003c/div\u003e\u003c/div\u003e\u003c/p\u003e \u003cp\u003eNFI (Normed Fit Index):\u003cdiv id=\"Equd\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equd\" name=\"EquationSource\"\u003e\n$$\\:NFI\\:=\\frac{{\\chi\\:}^{2}ₙᵤₗₗ\\:-\\:{\\chi\\:}^{2}ₘₒdₑₗ}{{\\chi\\:}^{2}ₙᵤₗₗ}\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:4$$\u003c/div\u003e\u003c/div\u003e\u003c/p\u003e \u003cp\u003eChi-Square statistic:\u003cdiv id=\"Eque\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Eque\" name=\"EquationSource\"\u003e\n$$\\:{\\chi\\:}^{2}=\\sum\\:{ᵢ}^{=1}ⁿ\\left[{\\left(Oᵢ-Eᵢ\\right)}^{2}/Eᵢ\\right]\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:5$$\u003c/div\u003e\u003c/div\u003e\u003c/p\u003e \u003cp\u003eWhere:\u003c/p\u003e \u003cp\u003e \u003cul\u003e \u003cli\u003e \u003cp\u003esᵢⱼ: Observed correlation between variables i and j\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003eŝᵢⱼ: Predicted correlation between variables i and j\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003em: Total number of observations\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003eχ\u0026sup2;ₙᵤₗₗ: Chi-Square value of the null model\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003eχ\u0026sup2;ₘₒdₑₗ: Chi-Square value of the fitted model\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003eOᵢ: Observed value\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003eEᵢ: Expected value\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003en: Number of data points\u003c/p\u003e \u003c/li\u003e \u003c/ul\u003e \u003c/p\u003e \u003cp\u003eThese indices are essential for evaluating the model's overall goodness-of-fit, ensuring that the model is adequately capturing the underlying relationships among the variables.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec7\" class=\"Section3\"\u003e \u003ch2\u003e3.3.2 Artificial Neural Networks (ANN)\u003c/h2\u003e \u003cp\u003eSensitivity analysis using an Artificial Neural Network was done to further improve the outcomes of the previous SEM analysis for further validation. ANN is very good at detecting nonlinear relationships and finding out the relative importance of predictor variables (Kingma \u0026amp; Ba, 2015). The following analysis leverages Scikit-learn, a well-used Python library across machine learning and predictive modelling. ANN application is quite apt in this study because environmental degradation, health expenditure, and economic growth are highly interrelated. The ANN model was developed with carbon dioxide emissions, nitrous oxide emissions, and the air quality index as inputs to predict economic growth (Ramayah et al., 2023). Hidden layers with nonlinear activation functions were included to allow the network to learn the most complex pattern present within the data (LeCun et al., 2015).\u003c/p\u003e \u003cp\u003eThe hidden layers used the ReLU activation function because this kind of activation function introduces less computation and does not face the problem of a vanishing gradient while handling large data (LeCun et al., 2015). The output layer linearly activates the predicted continuous values of the economy's growth. In the architecture of the ANN model, the design aimed to reduce the estimated and actual differences in values. Also, the performance of the model was enhanced using the Adam optimizer, a sophisticated form of gradient descent, since it possesses adaptive learning rates and is operationally efficient (Kingma \u0026amp; Ba, 2015). The optimizer used is the MSE - Mean Squared Error - because it calculates the average of the squared difference between predictions and actual values (Kingma \u0026amp; Ba, 2015).\u003c/p\u003e \u003cp\u003eThe MSE formula is as follows:\u003c/p\u003e \u003cp\u003eMSE (Mean Squared Error):\u003cdiv id=\"Equf\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equf\" name=\"EquationSource\"\u003e\n$$\\:MSE=\\left(1/n\\right)\\sum\\:{ᵢ}^{=1}ⁿ{\\left(\\varvec{y}\\varvec{ᵢ}-\\varvec{ŷ}\\varvec{ᵢ}\\right)}^{2}\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:5$$\u003c/div\u003e\u003c/div\u003e\u003c/p\u003e \u003cp\u003eAdam Optimizer:\u003cdiv id=\"Equg\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equg\" name=\"EquationSource\"\u003e\n$$\\:\\theta\\:{ₜ}^{+1}=\\theta\\:ₜ-\\alpha\\:\u0026middot;mₜ/\\left(\\sqrt{v}ₜ+ϵ\\right)\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:6$$\u003c/div\u003e\u003c/div\u003e\u003c/p\u003e \u003cp\u003eWhere:\u003c/p\u003e \u003cp\u003e \u003cul\u003e \u003cli\u003e \u003cp\u003eyᵢ: Actual value\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003eŷᵢ: Predicted value\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003en: Number of data points\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003eθₜ: Parameters being optimized at iteration t\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003eα: Learning rate\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003emₜ: Estimate of the first moment (mean) of the gradients\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003evₜ: Estimate of the second moment (variance) of the gradients\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003eϵ: Small constant to prevent division by zero\u003c/p\u003e \u003c/li\u003e \u003c/ul\u003e \u003c/p\u003e \u003cp\u003eThe ANN model was trained using an 80:20 train-test split to prevent overfitting and ensure that the model could generalize well to unseen data. Cross-validation was also applied to further verify the robustness of the model. Finally, sensitivity analysis was performed to assess the relative importance of each predictor variable, providing a clearer understanding of which environmental factors most strongly influence economic growth. This hybrid approach, combining SEM and ANN, follows recent studies such as those by Anwar et al. (2020), who demonstrated that integrating machine learning models with SEM enhances the accuracy and robustness of complex environmental-economic analyses. The use of ANN allows for validation of the relationships identified in the SEM model, providing further confidence in the results.\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e"},{"header":"4. Empirical Results and Discussion","content":"\u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e shows that GDP per capita ranges from 354.089 to 3667.057, with a mean of 1084.046 and a standard deviation of 725.298. This indicates a wide spread of economic development levels across the observations. The positive skewness (1.567) suggests that the distribution is right-skewed, with more observations below the mean and some higher values pulling the average up. CO2 emissions show a similar pattern, ranging from 0.051 to 1.074, with a mean of 0.33 and a standard deviation of 0.241. The positive skewness (1.275) indicates that most observations have lower CO2 emissions, with some higher emitters influencing the mean. PM2.5 levels range from 5.897 to 46.668, with a mean of 31.495 and a standard deviation of 10.992. All variables have Shapiro-Wilk p-values of 0.000, strongly suggesting that the variables do not follow a normal distribution.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eDescriptive Statistics\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"10\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c8\" colnum=\"8\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c9\" colnum=\"9\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c10\" colnum=\"10\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMin\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eMax\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eMean\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eStd. dev\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colspan=\"2\" nameend=\"c7\" namest=\"c6\"\u003e \u003cp\u003eSkew\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c8\"\u003e \u003cp\u003eKurtosis\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c9\"\u003e \u003cp\u003eShapiro-Wilk stat\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c10\"\u003e \u003cp\u003ep-value\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eGDP_Per_Capita\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e354.089\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e3667.057\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e1084.046\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e725.298\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c7\" namest=\"c6\"\u003e \u003cp\u003e1.567\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e1.672\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e0.782\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eCO2\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.051\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1.074\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.33\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.241\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c7\" namest=\"c6\"\u003e \u003cp\u003e1.275\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e1.061\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e0.865\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003epm.2.5\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e5.897\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e46.668\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e31.495\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e10.992\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e-0.602\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c8\" namest=\"c7\"\u003e \u003cp\u003e-0.749\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e0.928\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eNitrus_Oxide\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e55.534\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e42693.37\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e5736.831\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e8328.243\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c7\" namest=\"c6\"\u003e \u003cp\u003e2.698\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e7.312\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e0.64\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eHealth_Exp\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e2.279\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e19.69\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e5.057\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e2.424\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c7\" namest=\"c6\"\u003e \u003cp\u003e2.804\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e11.239\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e0.739\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eFigure\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e shows that in several countries with diversified economic and environmental tendencies GDP per capita normally increases at different rates, very high in Cabo Verde and Nigeria, moderate in Cote d'Ivoire, Senegal, and Ghana, but rather stable yet lower in countries like Benin, Burkina Faso, and Guinea - Bissau. These economic trends are reflected in environmental indicators. In general, CO2 emissions are higher in the more industrial countries, such as Cabo Verde and Nigeria. The levels of PM2.5 are trending down across most countries, except Cabo Verde. The nitrous oxide emissions greatly fluctuate: the most substantial increase is in Nigeria due to their extensive farming and industrial sectors. The less industrialized countries maintain lower and more stable levels.\u003c/p\u003e \u003cp\u003eFigure\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e displays the correlation matrix between the GDP per capita rate and all independent variables under study. GDP per capita has a strong positive association with CO2 emission level, indicating high growth rates for the economy have been matched by proportionate amounts spent on carbon production. Conversely, PM2.5 shows a negative relationship with GDP per capita. One could hypothesize that more developed regions could have much stricter environmental regulations leading to clean air policies or deploy more friendly industries that produce less pollution. However, nitrous oxide is weakly positively correlated showing its marginal association between income levels and this greenhouse gas. This suggests that while nitrous oxide emissions tend to increase with economic development, the relationship is not as strong as with CO2 emissions. Health expenditure displays a weak negative correlation with GDP per capita.\u003c/p\u003e \u003cdiv id=\"Sec9\" class=\"Section2\"\u003e \u003ch2\u003e4.1 SEM Model Performance Evaluation\u003c/h2\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab3\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 3\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eModel Performance Evaluation\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"8\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c8\" colnum=\"8\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eQ\u0026sup2;predict\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003ePLS-SEM_RMSE\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003ePLS-SEM_MAE\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eLM_RMSE\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eLM_MAE\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003eIA_RMSE\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c8\"\u003e \u003cp\u003eIA_MAE\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eGDP_Per_Capita\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.857\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.378\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.285\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.378\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.285\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e1.002\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e0.775\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHealth_Exp\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.107\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.949\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.631\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.949\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.631\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e1.004\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e0.652\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eThe evaluation of model performance against predictive accuracy and errors with regard to GDP per capita and health expenditure is presented in Table\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e. The high value of Q\u0026sup2;predict for GDP per capita of 0.857 indicates strong predictive relevance, combined with very low values of RMSE of 0.378 and MAE of 0.285, which indicates that PLS-SEM serves a good predictive performance. Comparison with LM and IA further supports this result for the variable GDP.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab4\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 4\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eModel Fit\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"6\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eModel\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eSRMR\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003ed_ULS\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003ed_G\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eChi-square\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eNFI\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSaturated model\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e1.000\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eEstimated model\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e-0.000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e1.000\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab4\" class=\"InternalRef\"\u003e4\u003c/span\u003e shows the Model fit indices of the saturated and estimated model. The fit statistics for both models are perfect: SRMR, d_ULS, d_G, and Chi-square all equate to zero and NFI equals 1.000, proving an excellent fit-meaning the estimates of the model fitted are very close to the empirically observed data. That is a perfect fit, which hardly ever happens; this may indicate that data and model specification need to be cleaned from any sort of irregularity for the estimation procedure.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab5\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 5\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eCollinearity Check (VIF)\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"2\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eRelationship\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eVIF\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAir Quality Index -\u0026gt; Economic Growth\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1.389\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAir Quality Index -\u0026gt; Health Expenditure\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1.388\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCarbon Emission -\u0026gt; Economic Growth\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1.402\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCarbon Emission -\u0026gt; Health Expenditure\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1.342\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHealth Expenditure -\u0026gt; Economic Growth\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1.131\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNitrous Oxide -\u0026gt; Economic Growth\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1.106\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNitrous Oxide -\u0026gt; Health Expenditure\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1.067\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab5\" class=\"InternalRef\"\u003e5\u003c/span\u003e represents the VIF values for the relationships specified in the structural model. All the VIFs stand below 1.5, and hence there is no problem with multicollinearity in this model. Multicollinearity occurs whenever a high degree of intercorrelation among predictor variables exists and can be experienced with an increase in standard errors with unreliable estimates. All the VIF values are way below the threshold of 5; hence, we are guaranteed that each predictor adds unique information to the model, which ensures that the estimates of the regression coefficients are appropriate and not over-inflated.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab6\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 6\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003e\u003cem\u003eDirect Effects of Environmental Degradation on Economic Growth\u003c/em\u003e\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"5\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eRelationship\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eBeta\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eStandard deviation\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eT statistics (|O/STDEV|)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eP values\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAir Quality Index -\u0026gt; Economic Growth\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e-0.306\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.029\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e10.628\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAir Quality Index -\u0026gt; Health Expenditure\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.036\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.040\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.898\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.369\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCarbon Emission -\u0026gt; Economic Growth\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.716\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.034\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e21.244\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCarbon Emission -\u0026gt; Health Expenditure\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e-0.231\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.032\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e7.159\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHealth Expenditure -\u0026gt; Economic Growth\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.045\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.016\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e2.759\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.006\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNitrous Oxide -\u0026gt; Economic Growth\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.123\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.032\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e3.802\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNitrous Oxide -\u0026gt; Health Expenditure\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e-0.186\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.023\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e8.102\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003ePresented in Table\u0026nbsp;\u003cspan refid=\"Tab6\" class=\"InternalRef\"\u003e6\u003c/span\u003e shows that carbon emissions have a high positive impact on the economic development of the ECOWAS region. This strong positive relationship (t\u0026thinsp;=\u0026thinsp;21.244) suggests that the kinds of economic activities taking place in the region, in particular, those contributing to carbon emissions, have a deep connection with economic performance overall. Carbon-intensive practices in manufacturing and the generation of energy spur economic development due to the adverse impact it has on the environment. This in itself is a very valid observation underlining the persistent conflict between ecological sustainability and the economic imperatives of development within the region. The Air Quality Index and economic growth are inversely correlated, with the levels of PM2.5 recording a statistically significant negative association: β = -0.306, p\u0026thinsp;\u0026lt;\u0026thinsp;0.001. This means that poor air quality is negatively impacting economic growth in the region. The higher t-statistic, t\u0026thinsp;=\u0026thinsp;10.628 supports the result even more. The generally worsening air quality tends to generate significant and increasing costs through lost productivity, adverse health effects, and more generally lower economic performance. It thus follows that environmental degradation is broadly expensive and underlines that clean air is a prerequisite for continued economic growth.\u003c/p\u003e \u003cp\u003eNitrous oxide emissions are positively and significantly related to economic growth, with β\u0026thinsp;=\u0026thinsp;0.123, p\u0026thinsp;\u0026lt;\u0026thinsp;0.001. Although the effect size here is smaller than that from carbon emissions, this relationship does hold and suggests that insofar as nitrous oxide emissions reflect industrial and agricultural practices, these have a supporting role in enhancing economic performance within the region. With the t-statistic of 3.802, nitrous oxide pollution, like carbon emissions, is an economic activity fundamental to the current growth trajectory of the ECOWAS nations, though clearly at considerable environmental costs.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab7\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 7\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eIndirect Effects via Health Expenditure\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"5\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eBeta\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eStandard deviation\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eT statistics\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eP values\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNitrus Oxide -\u0026gt; Health Expenditure -\u0026gt; Economic Growth\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e-0.008\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.003\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e2.622\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.009\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAir Quality Index -\u0026gt; Health Expenditure -\u0026gt; Economic Growth\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.002\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.002\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.735\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.462\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCarbon Emission -\u0026gt; Health Expenditure -\u0026gt; Economic Growth\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e-0.010\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.004\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e2.355\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.019\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eThe results in Table\u0026nbsp;\u003cspan refid=\"Tab7\" class=\"InternalRef\"\u003e7\u003c/span\u003e show that health expenditure mediates the impact of environmental degradation on economic growth due to carbon and nitrous oxide emissions. The mediated effect of carbon emissions on health expenditure has a negative effect on economic growth, and it is statistically significant (β = -0.010, p\u0026thinsp;=\u0026thinsp;0.019); hence, environmental degradation increases health expenditure and slows down economic growth. Similarly, nitrous oxide emission contributes negatively indirectly at β = -0.008, p\u0026thinsp;=\u0026thinsp;0.009, meaning health impacts due to pollution slightly offset the economic gains accrued from growth. On the other side, the mediating effect of the Air Quality Index is insignificant, β\u0026thinsp;=\u0026thinsp;0.002, p\u0026thinsp;=\u0026thinsp;0.462, implying that health expenditure is not significant in affecting the relationship of air quality-economic growth.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec10\" class=\"Section2\"\u003e \u003ch2\u003e4.2 Artificial Neural Networks (ANN) Results\u003c/h2\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab8\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 8\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eModel Performance\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"8\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c8\" colnum=\"8\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003eNeural Networks\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003eN\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eTraining\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eTraining\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eTesting\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003eTraining\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/th\u003e \u003c/tr\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eSSE\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eRMSE\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eN\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eSSE\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003eRMSE\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c8\"\u003e \u003cp\u003eTotal Sample\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e298\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e14.792\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.223\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e32\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e2.422\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e0.275\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e330\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e290\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e10.458\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.19\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e40\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.873\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e0.148\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e330\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e295\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e12.92\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.209\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e35\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e1.766\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e0.225\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e330\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e285\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e9.354\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.181\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e45\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e1.449\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e0.179\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e330\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e302\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e14.279\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.217\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e28\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e1.435\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e0.226\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e330\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e6\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e293\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e23.891\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.286\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e37\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.988\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e0.163\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e330\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e7\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e290\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e15.923\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.234\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e40\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e2.198\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e0.234\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e330\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e8\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e303\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e12.326\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.202\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e27\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.742\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e0.166\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e330\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e9\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e290\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e15.899\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.234\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e40\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e1.194\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e0.173\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e330\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e10\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e292\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e14.095\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.22\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e38\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e1.53\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e0.201\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e330\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMean\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e14.394\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.22\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e1.46\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e0.199\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSt. Dev\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e3.978\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.029\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.55\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e0.04\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003ctfoot\u003e \u003ctr\u003e\u003ctd colspan=\"8\"\u003eNote. N\u0026thinsp;=\u0026thinsp;sample size; SSE\u0026thinsp;=\u0026thinsp;sum of square error, RMSE\u0026thinsp;=\u0026thinsp;root mean square of errors.\u003c/td\u003e\u003c/tr\u003e \u003c/tfoot\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab8\" class=\"InternalRef\"\u003e8\u003c/span\u003e shows that the variation of RMSE in the training lies between 0.181 and 0.286, with an average of 0.220 when the model is trying to predict the target variable at this level. The SSE varies among these networks; Network 6 has the highest with an SSE of 23.891, while Network 4 has the lowest with an SSE value of 9.354, indicating thereby the difference in the model capability of LSE minimization within this training. A low value of RMSE implies perfect performance by the model with regard to fitting the training data. During the testing phase, the generalizing capability of the ANN model is quite good and gives a mean RMSE: of 0.199. A relatively small standard deviation in the RMSE values, 0.040, represents coherence among the various diverse networks during the prediction of unseen data. These are the best test results from, say, Network 2, giving an RMSE value of 0.148, while Network 1 is relatively higher, having a value of 0.275, reflecting network variability, yet still within acceptable limits. The ANN model has substantial predictive power as the error in both training and testing is negligible; hence, it is reliable to study the inter-relationships among the environmental variables, healthcare expenditure, and economic growth in the ECOWAS region.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003e4.2 Feature Importance Analysis: Output\u0026thinsp;=\u0026thinsp;Economic Growth (GDP)\u003c/h2\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab9\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 9\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eSensitive Analysis: Output\u0026thinsp;=\u0026thinsp;Economic Growth (GDP)\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"13\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c8\" colnum=\"8\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c9\" colnum=\"9\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c10\" colnum=\"10\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c11\" colnum=\"11\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c12\" colnum=\"12\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c13\" colnum=\"13\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eNI1\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eNI2\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eNI3\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eNI4\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eNI5\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003eNI6\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c8\"\u003e \u003cp\u003eNI7\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c9\"\u003e \u003cp\u003eNI8\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c10\"\u003e \u003cp\u003eNI9\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c11\"\u003e \u003cp\u003eNI10\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c12\"\u003e \u003cp\u003eA1\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c13\"\u003e \u003cp\u003eRank %\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eCO2\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e1.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e1.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e1.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e1.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e1.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e1.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003e1.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c11\"\u003e \u003cp\u003e1.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c12\"\u003e \u003cp\u003e1.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c13\"\u003e \u003cp\u003e100%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003epm.2.5\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.42\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.51\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.54\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.58\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.62\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e0.55\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e0.55\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e0.52\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003e0.43\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c11\"\u003e \u003cp\u003e0.42\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c12\"\u003e \u003cp\u003e0.52\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c13\"\u003e \u003cp\u003e52%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eNitrus_Oxide\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.19\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.23\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.17\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.61\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.39\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e0.12\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e0.17\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e0.30\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003e0.19\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c11\"\u003e \u003cp\u003e0.44\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c12\"\u003e \u003cp\u003e0.28\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c13\"\u003e \u003cp\u003e28%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003ctfoot\u003e \u003ctr\u003e\u003ctd colspan=\"13\"\u003eNote. NI\u0026thinsp;=\u0026thinsp;normalized importance, AI\u0026thinsp;=\u0026thinsp;average importance, and I\u0026thinsp;=\u0026thinsp;importance/normalized relative importance.\u003c/td\u003e\u003c/tr\u003e \u003c/tfoot\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab9\" class=\"InternalRef\"\u003e9\u003c/span\u003e presents the sensitivity analysis of feature importance for predicting economic growth (GDP) using Artificial Neural Networks (ANN). Carbon emissions (CO₂) show the highest importance with a normalized value of 1.00 across all indicators, representing 100% relative importance. This finding suggests that CO₂ emissions are the most influential variable in determining economic growth in the ECOWAS region. The strong dependence on carbon-intensive activities for economic expansion is clear, reinforcing the critical role that industrial activities and energy consumption, often powered by fossil fuels, play in driving GDP growth. This outcome reflects the complex trade-offs faced by ECOWAS countries where economic progress heavily relies on industries that contribute significantly to carbon emissions. On the other hand, PM2.5, which stands for perceived air quality, has fair significance, with an average normalized importance of 0.52 or 52% of the predictive capability regarding GDP. Evidence that while inadequate air quality does have a perceivable effect on economic growth, its importance is comparatively lower compared to carbon emissions. This moderate rating of PM2.5 points to indirect costs regarding sub-optimality in air quality, which may increase health costs or depress productivity among workers. Nitrous oxide emission was perceived to be of least concern, measuring an average of 0.28 (28% significant); thus, while there might be a role played by nitrous oxide emissions in economic growth, the level of such an influence is much less impressive compared to CO₂ and PM2.5 concentrations. As such, it is a gas that is usually related to agriculture or industry, so its emissions hardly have an importance in determining income.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab10\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 10\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eComparison between PLS-SEM and ANN results Output (GDP)\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"6\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eConstructs\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePath Coefficient\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003ePLS-SEM Ranking\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eANN-normalised relative importance (%)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eANN ranking\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eMatching PLS-SEM with ANN\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCO2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.716\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e100%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eMatch\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePM.2.5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e-0.306\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e52%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eMatch\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNitrus_Oxide\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.123\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e28%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eMatch\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab10\" class=\"InternalRef\"\u003e10\u003c/span\u003e presents the comparison between the path coefficients of SEM and the normalized importance values of ANN with regard to GDP. It could be found that the hypothesis that carbon emission is strongly positively related to economic growth could be shared by both models. The carbon emissions show the highest path coefficient of 0.716 in SEM and 100% in ANN, indicating the complete match of these two methods. The consistency of the findings across the different analytical approaches suggests that carbon emissions, irrespective of the approach adopted, are the most crucial determinant factor in influencing GDP growth in the ECOWAS region. Consistency between SEM and ANN suggests that the strong and reliable leading role of carbon emissions is leading the economic path of ECOWAS countries. Both SEM and ANN ranked the level of PM2.5 as the second most important factor that affects GDP, with a negative path coefficient in SEM (β = -0.306) and a normalized importance of 52% in ANN. Such consistency obtained from the two models would thus suggest that though air pollution impairs the linkage of economic growth, it is less important compared with that induced by carbon emissions. For the nitrous oxide emission, SEM also shows a positive but relatively smaller path coefficient, β\u0026thinsp;=\u0026thinsp;0.123 agreeing with the lower relative importance identified by ANN at 28%. Both nitrous oxide emissions have a smaller impact on GDP; thus, these are less influential variables than carbon emissions and air quality.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003e4.3 Feature Importance Analysis: Output\u0026thinsp;=\u0026thinsp;Health Expenditure\u003c/h2\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab11\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 11\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eSensitive Analysis\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"13\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c8\" colnum=\"8\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c9\" colnum=\"9\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c10\" colnum=\"10\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c11\" colnum=\"11\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c12\" colnum=\"12\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c13\" colnum=\"13\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eNI1\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eNI2\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eNI3\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eNI4\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eNI5\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003eNI6\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c8\"\u003e \u003cp\u003eNI7\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c9\"\u003e \u003cp\u003eNI8\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c10\"\u003e \u003cp\u003eNI9\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c11\"\u003e \u003cp\u003eNI10\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c12\"\u003e \u003cp\u003eA1\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c13\"\u003e \u003cp\u003eRank %\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCO2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e1.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e1.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e1.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e0.73\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e1.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e1.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003e1.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c11\"\u003e \u003cp\u003e0.66\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c12\"\u003e \u003cp\u003e0.9384\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c13\"\u003e \u003cp\u003e100%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003epm.2.5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.52\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.31\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.38\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.19\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.80\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e0.26\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e0.17\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e0.43\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003e0.54\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c11\"\u003e \u003cp\u003e0.08\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c12\"\u003e \u003cp\u003e0.3667\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c13\"\u003e \u003cp\u003e39%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNitrus_Oxide\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.85\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.84\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.83\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.76\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.69\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e1.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e0.79\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e0.90\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003e0.99\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c11\"\u003e \u003cp\u003e1.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c12\"\u003e \u003cp\u003e0.8645\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c13\"\u003e \u003cp\u003e92%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003ctfoot\u003e \u003ctr\u003e\u003ctd colspan=\"13\"\u003eNote. NI\u0026thinsp;=\u0026thinsp;normalized importance, AI\u0026thinsp;=\u0026thinsp;average importance, and I\u0026thinsp;=\u0026thinsp;importance/normalized relative importance.\u003c/td\u003e\u003c/tr\u003e \u003c/tfoot\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab11\" class=\"InternalRef\"\u003e11\u003c/span\u003e shows the sensitivity analysis of the determinants of health expenditure using the ANN methodological approach. Carbon emission (CO₂) has the highest normalized importance of 100%, indicating that it is the most important determinant of health expenditure within the ECOWAS sub-region. The high level of importance of CO₂ is indicative that the pollution resulting from highly carbon-intensive activities such as industrial processes and transportation, is directly affecting the public health for which health expenditures are rising. An increase in CO₂ emissions indicates a rise in respiratory and cardiovascular diseases due to pollution, considering the investments that should be made in health systems. These are, respectively, the second most relevant when projecting health expenditure. Nitrous oxide emissions are at an importance of 92% normalized. The highest sources of nitrous oxide come from agriculture and industrial processes. Because of its gigantic health implications, it is also a very potent greenhouse gas to global warming and deterioration of air quality. This high importance score reflects a strong linkage with environmental degradation and related escalating healthcare costs. On the other extreme, levels of PM2.5 rank the lowest at 39% normalized importance in the ranking scale. While air pollution-quantified through PM2.5 levels and health complications are indeed linked, it is ranking so far below the other two showing it having a relatively less economic impact on health care systems through particulate matter compared to the graver implications through CO₂ and nitrous oxide emissions.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab12\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 12\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eComparison between PLS-SEM and ANN results Output (Health Expenditure)\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"6\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eConstructs\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePath Coefficient\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003ePLS-SEM Ranking\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eANN-normalised relative importance (%)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eANN ranking\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eMatching PLS-SEM with ANN\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCO2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e-0.231\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e100%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eMatch\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNitrus_Oxide\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e-0.186\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e92%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eMatch\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePM.2.5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.036\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e39%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eMatch\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab12\" class=\"InternalRef\"\u003e12\u003c/span\u003e compares the results obtained between SEM-ANN on health expenditure prediction. The ranking produced from both models is the same, with CO₂ as the most important variable among the factors. It is indicated that in SEM, the path coefficient for CO₂ emissions is negative, β = -0.231, which infers that health expenditure is negatively affected by increased CO₂ emission. This may probably be because high pollution exposure increases the basic economic burden on healthcare. ANN also confirms this, since the normalized importance of CO₂ emissions is 100%; hence, CO₂ is ranked first according to both approaches. Both models rank NOx emission in the second position in terms of ranking order. It bears a path coefficient of β\u0026thinsp;=\u0026thinsp;\u0026minus;\u0026thinsp;0.186, and as usual, a high magnitude of this gas promotes health expenditure since both governments and households try to do something to reduce the health effects caused by environmental pollution. This follows then from ANN, where nitrous oxide yields a relative importance of 92%. PM2.5 levels are therefore less influential in either model's determination but retain some level of importance. While air pollution is of a minor magnitude, it has a positive effect on health expenditure; the little positive path coefficient of 0.036 is shown in the SEM. ANN also gives a lower normalized importance of 39% to PM2.5. Weaker in influence, both models affirm that air quality does contribute to health expenditure, but once again at a lesser magnitude than CO₂ and nitrous oxide.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab13\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 13\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eHypothesis Testing\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"6\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHypothesis\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePath\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eDirection\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eBeta\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eP-value\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eDecision\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eH1a\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eCarbon Emission \u0026rarr; Economic Growth\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003ePositive\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.716\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eSupported\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eH1b\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eAir Quality Index \u0026rarr; Economic Growth\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eNegative\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e-0.306\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eSupported\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eH1c\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eNitrous Oxide \u0026rarr; Economic Growth\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003ePositive\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.123\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eSupported\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eH2a\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eCarbon Emission \u0026rarr; Health Expenditure \u0026rarr; Economic Growth\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eNegative\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e-0.010\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.019\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eSupported\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eH2b\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eAir Quality Index \u0026rarr; Health Expenditure \u0026rarr; Economic Growth\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003ePositive\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.002\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.462\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eNot Supported\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eH2c\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eNitrous Oxide \u0026rarr; Health Expenditure \u0026rarr; Economic Growth\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eNegative\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e-0.008\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.009\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eSupported\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec13\" class=\"Section2\"\u003e \u003ch2\u003e4.4 Discussion\u003c/h2\u003e \u003cp\u003e \u003cdiv class=\"BlockQuote\"\u003e \u003cp\u003eThe study results strongly support the first hypothesis: carbon emissions and nitrous oxide emissions have a positive influence on economic growth, whereas poor air quality greatly hinders economic growth. The positive relation between carbon emissions and economic growth yields, as expected from earlier studies conducted in rapidly industrializing parts of the world, such as China, where economic growth usually comes at the cost of environmental quality (Liu \u0026amp; Raven, 2010; Luo et al., 2022). This is in line with the context of many developing nations, such as ECOWAS, which are characterized by carbon-intensive industries that drive economic output but greatly pollute the environment. This is further evidenced in the positive association of nitrous oxide emissions with growth, implying that the contribution of agriculture and industry in the regions acts to sustain economic growth. Other studies account for agricultural intensification and industrial emissions, further supporting this view in developing countries (Aryal et al., 2022; Wu et al., 2021). In contrast, the negative association of air quality-PM2.5 levels with economic growth was inverse: β = -0.306, p\u0026thinsp;\u0026lt;\u0026thinsp;0.001, indicating negative economic effects due to air pollution, which have also been claimed by the World Health Organization (2021) and Demir et al. (2022). This fact is corroborated by other research showing that air pollution decreases labour productivity and raises healthcare costs, thereby reducing the likelihood of investment and, consequently, limiting economic growth (Demir et al., 2022; Neidell, 2017). In developing countries where healthcare infrastructure is often overburdened, the economic cost of pollution is magnified. The results also indicate that the EKC hypothesis may not be fully applicable to the ECOWAS region. Although coupled with a rise in economic growth, environmental degradation suggests a deterring impact of poor air quality on economic growth, hence deviating from the inverted U-shaped EKC. This therefore supports the other models presented by Shahbaz and Sinha (2019) that because of increased consumption and energy demands, environmental degradation may rise again at higher income levels.\u003c/p\u003e \u003cp\u003eThe second hypothesis, which was on the mediating effect of health expenditure, was only partially supported. Health expenditure significantly mediates the relationship between nitrous oxide emissions and economic growth, with a β = -0.008 and a p-value of 0.009, accounting for 13.91% of the total effect. This means that environmental degradation, nitrous oxide emissions, and economic growth are partly reduced by the added healthcare costs. These findings are consistent with the observations of Novignon and Lawanson (2017), that increased health expenditure could better some of the negative economic effects of degradation through improvements in public health and productivity. The limit to mediation in the ECOWAS context might be related to regional differences in healthcare systems and environmental policies. The insignificant mediation effect of health expenditure in the relationship between air quality and economic growth would therefore suggest that the economic costs of poor air quality are more direct rather than via increased health expenditure. This result is in line with findings from Uzochukwu et al. (2015), who observe that in regions where healthcare infrastructure is relatively low like in the ECOWAS sub-region, the ability of the healthcare system to absorb and cushion the economic effect of environmental degradation is limited. In such settings, remedying poor air quality based on better healthcare investments may not be feasible. Instead, stronger environmental regulations are needed upfront to reduce pollution and protect public health as well as economic growth (Kutlu, 2021).\u003c/p\u003e \u003c/div\u003e \u003c/p\u003e \u003c/div\u003e"},{"header":"5. Conclusion","content":"\u003cp\u003e \u003cdiv class=\"BlockQuote\"\u003e \u003cp\u003eThis study has undertaken the analysis of the inter-linkages between environmental degradation, health expenditure and economic growth through SEM and ANN analysis in the ECOWAS region. Carbon emission and nitrous oxide were found to be the positive determinants of economic growth, while poor air quality lowers growth. Health expenditure is mediated between nitrous oxide emission and economic growth. This result would imply that as much as industrial operation and agricultural intensification spur economic growth, the costs of the long-run degradation of environmental and human health related to these activities pose serious risks to sustainable development. The limited mediator function of health expenditure underlines the need for the region to further develop its healthcare infrastructure and to apply more active environmental policy. These results thus imply that there should be strict environmental regulation that limits carbon and nitrous oxide emissions, though, in the same breath, they are investing in cleaner technologies that will not contravene economic growth. Additionally, strengthening public health systems will reduce the health effects of degraded environments, mainly by reducing the burden of industrial emissions. The governments should also promote air quality enhancement policies, given that air quality directly affects economic productivity. This might hint at what the future of research should consider, that is, the impact of the integration of renewable energy on economic development in the long run within the region of ECOWAS, studying other mediating variables that could inform the relationship between environmental and economic growth.\u003c/p\u003e \u003c/div\u003e \u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eAcknowledgements\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot applicable\u003cbr\u003e\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors declare that the study was conducted independently and without financial support or funding from any source.\u0026nbsp;\u003cbr\u003e\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAvailability of data and materials\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe datasets analyzed during the current study will be made available from the corresponding author on reasonable request.\u003cbr\u003e\u0026nbsp;\u003cbr\u003e\u003cstrong\u003eInformed Consent Statement\u003cbr\u003e\u0026nbsp;\u003c/strong\u003e Not applicable\u003cbr\u003e\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthor information\u003c/strong\u003e\u003cstrong\u003e\u003cbr\u003e\u0026nbsp;Authors and Affiliations\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eSchool of Public Policy and Administration, Northwestern Polytechnic University, Xi’an, China.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eSeth Acquah Boateng\u003cbr\u003e\u003cstrong\u003eDepartment of Computer Science, University of Calgary, Alberta, Canada.\u003c/strong\u003e\u003cbr\u003e\u0026nbsp;Andy Asare\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eDepartment of Data Science and Economic Policy, University of Cape Coast, Ghana\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWilliam Godfred Cantah \u0026amp; Joshua Sebu\u003cbr\u003e\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eContributions\u003cbr\u003e\u003c/strong\u003eSAB conceptualized the key constructs of the research idea. SAB, AA, designed the methodology, conducted the research/investigation/analyses, SAB, AA, wrote the original manuscript, and WC and JS read/approved the final manuscript.\u003cbr\u003e\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCorresponding Author\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eCorrespondence to Andy Asare\u003cbr\u003eEmail: [email protected]\u003cbr\u003e\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConsent for publication\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot applicable.\u003cbr\u003e\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting interests\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors declare that they have no competing interests.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eAboubacar, B., \u0026amp; Xu, D. (2017). The impact of health expenditure on economic growth in Sub-Saharan Africa. \u003cem\u003eTheoretical Economics Letters, 7\u003c/em\u003e(3), 615-622.\u003c/li\u003e\n\u003cli\u003eAfrican Development Bank. (2023, July 19). Africa\u0026apos;s economic growth to outpace global forecast in 2023-2024 \u0026ndash; African Development Bank biannual report. https://www.afdb.org/en/news-and-events/press-releases/africas-economic-growth-outpace-global-forecast-2023-2024-african-development-bank-biannual-report-58293\u003c/li\u003e\n\u003cli\u003eAlrais, G., Godara, J., Godara, S. P., Alnakeb, A., \u0026amp; Khaled, E. (2024). Environmental problem of nitrous oxide in obstetrics: A case review. \u003cem\u003eGlobal Journal of Research Analysis, 13\u003c/em\u003e(2), 116-118.\u003c/li\u003e\n\u003cli\u003eAnwar, A., Hyder, S., Bennett, R., \u0026amp; Younis, M. (2022). Impact of environmental quality on healthcare expenditures in developing countries: A panel data approach. \u003cem\u003eHealthcare, 10\u003c/em\u003e(1608).\u003c/li\u003e\n\u003cli\u003eAnwar, A., Siddique, M., Dogan, E., \u0026amp; Sharif, A. (2021). The moderating role of renewable and non-renewable energy in environment-income nexus for ASEAN countries: Evidence from method of moments quantile regression. \u003cem\u003eRenewable Energy, 164\u003c/em\u003e, 956-967.\u003c/li\u003e\n\u003cli\u003eArtekin, A. \u0026Ouml;., \u0026amp; Konya, S. (2020). Health expenditure and economic growth: Is the health-led growth hypothesis supported for selected OECD countries? \u003cem\u003ePoslovna Izvrsnost, 14\u003c/em\u003e(1), 77-89.\u003c/li\u003e\n\u003cli\u003eAryal, B., Gurung, R., Camargo, A. F., Fongaro, G., Treichel, H., \u0026amp; Mainali, B. (2022). Nitrous oxide emission in altered nitrogen cycle and implications for climate change. \u003cem\u003eEnvironmental Pollution, 314\u003c/em\u003e, 120272.\u003c/li\u003e\n\u003cli\u003eAyodotun, B., Bamba, S., \u0026amp; Adio, A. (2019). Vulnerability assessment of West African countries to climate change and variability. \u003cem\u003eJournal of Geoscience and Environmental Protection, 7\u003c/em\u003e(2), 13-15.\u003c/li\u003e\n\u003cli\u003eBloom, D. E., Canning, D., \u0026amp; Sevilla, J. (2004). The effect of health on economic growth: A production function approach. \u003cem\u003eWorld Development, 32\u003c/em\u003e(1), 1-13.\u003c/li\u003e\n\u003cli\u003eBoateng, S. A., Karikari, F. A., Fumey, M. P., Asmah, E. E., \u0026amp; Winful, S. A. (2024). Understanding global commodity price shocks on exchange rates and inflation in emerging economies: ARDL perspective. \u003cem\u003eJournal of Economics, Management and Trade, 30\u003c/em\u003e(1), 33-47.\u003c/li\u003e\n\u003cli\u003eCarpio-Thomas, C. (2020). \u003cem\u003eThe case of China\u003c/em\u003e. Digital Commons @ DU. https://digitalcommons.du.edu/cgi/viewcontent.cgi?article=2732\u0026amp;context=etd\u003c/li\u003e\n\u003cli\u003eDemir, S., Demir, H., Karaduman, C., \u0026amp; Cetin, M. (2022). Environmental quality and health expenditures efficiency in T\u0026uuml;rkiye: The role of natural resources. \u003cem\u003eEnvironmental Science and Pollution Research, 29\u003c/em\u003e, 1-15.\u003c/li\u003e\n\u003cli\u003eDhhan, W., Rana, S., Alshaybawee, T., \u0026amp; Midi, H. (2018). The single-index support vector regression model to address the problem of high dimensionality. \u003cem\u003eCommunications in Statistics - Simulation and Computation, 47\u003c/em\u003e(10), 2792-2809.\u003c/li\u003e\n\u003cli\u003eDuan, T., Avati, A., Ding, D. Y., Thai, K. K., Basu, S., Ng, A. Y., \u0026amp; Schuler, A. (2019). NGBoost: Natural gradient boosting for probabilistic prediction. \u003cem\u003earXiv\u003c/em\u003e. https://arxiv.org/abs/1910.03225\u003c/li\u003e\n\u003cli\u003eECOWAS. (2024, September 1). Member states | Economic Community of West African States (ECOWAS). https://www.ecowas.int/member-states/\u003c/li\u003e\n\u003cli\u003eGriffis, T. J., Chen, Z., Baker, J. M., Wood, J. D., Millet, D. B., \u0026amp; Lee, X. (2017). Nitrous oxide emissions are enhanced in a warmer and wetter world. \u003cem\u003eProceedings of the National Academy of Sciences of the United States of America, 114\u003c/em\u003e(45), 12081-12085.\u003c/li\u003e\n\u003cli\u003eGuo, J., Chai, G., Song, X., Xu, H., Li, Z., \u0026amp; Feng, X. (2023). Long-term exposure to particulate matter on cardiovascular and respiratory diseases in low- and middle-income countries: A systematic review and meta-analysis. \u003cem\u003eFrontiers in Public Health, 11\u003c/em\u003e, 1128432.\u003c/li\u003e\n\u003cli\u003eGwangndi, M. I., Muhammad, Y. A., \u0026amp; Tagi, S. M. (2016). The impact of environmental degradation on human health and its relevance to the right to health under international law. \u003cem\u003eEuropean Scientific Journal, 12\u003c/em\u003e(3), 485.\u003c/li\u003e\n\u003cli\u003eHa, T. N., Lubo-Robles, D., Marfurt, K. J., \u0026amp; Wallet, B. C. (2021). An in-depth analysis of logarithmic data transformation and per-class normalization in machine learning: Application to unsupervised classification of a turbidite system in the Canterbury Basin, New Zealand, and supervised classification of salt in the Eugene Island minibasin, Gulf of Mexico. \u003cem\u003eInterpretation, 9\u003c/em\u003e(4), T685-T710.\u003c/li\u003e\n\u003cli\u003eIntergovernmental Panel on Climate Change. (2021, August 9). Climate change widespread, rapid, and intensifying. https://www.ipcc.ch/2021/08/09/ar6-wg1-20210809-pr/\u003c/li\u003e\n\u003cli\u003eInternational Monetary Fund. (2021). \u003cem\u003eGhana: 2021 Article IV Consultation\u0026mdash;Press release; staff report; and statement by the Executive Director for Ghana\u003c/em\u003e (IMF Staff Country Report 2021). https://www.imf.org/en/Publications\u003c/li\u003e\n\u003cli\u003eJabeur, S. B., Ballouk, H., Arfi, W. B., \u0026amp; Khalfaoui, R. (2021). Machine learning-based modeling of the environmental degradation, institutional quality, and economic growth. \u003cem\u003eEnvironmental Modeling \u0026amp; Assessment, 27\u003c/em\u003e(4), 953-966.\u003c/li\u003e\n\u003cli\u003eJones, F. C., Plewes, R., Murison, L., MacDougall, M. J., Sinclair, S., \u0026amp; Davies, C. (2017). Random forests as cumulative effects models: A case study of lakes and rivers in Muskoka, Canada. \u003cem\u003eJournal of Environmental Management, 201\u003c/em\u003e(1), 407-424.\u003c/li\u003e\n\u003cli\u003eKahia, M., Moulahi, T., Mahfoudhi, S., Boubaker, S., \u0026amp; Omri, A. (2022). A machine learning process for examining the linkage among disaggregated energy consumption, economic growth, and environmental degradation. \u003cem\u003eResources Policy, 79\u003c/em\u003e, 103104.\u003c/li\u003e\n\u003cli\u003eKutlu, Ş. Ş. (2021). T\u0026uuml;rkiye ekonomisinde sağlığa dayalı b\u0026uuml;y\u0026uuml;me hipotezinin ge\u0026ccedil;erliliğine ilişkin ampirik bir analiz. \u003cem\u003eYaşar \u0026Uuml;niversitesi Dergisi, 16\u003c/em\u003e(1), 1808-1822.\u003c/li\u003e\n\u003cli\u003eLima, L. J. B., \u0026amp; Hamzagic, M. (2022). Greenhouse gases and air pollution: Commonalities and differentiators. \u003cem\u003eRevista Cient\u0026iacute;fica Multidisciplinar N\u0026uacute;cleo do Conhecimento\u003c/em\u003e, 102-144.\u003c/li\u003e\n\u003cli\u003eLiu, J., \u0026amp; Raven, P. H. (2010). China\u0026apos;s environmental challenges and implications for the world. \u003cem\u003eCritical Reviews in Environmental Science and Technology, 40\u003c/em\u003e(9), 823-851.\u003c/li\u003e\n\u003cli\u003eLuo, K., Liu, Y., Chen, P. F., \u0026amp; Zeng, M. (2022). Assessing the impact of digital economy on green development efficiency in the Yangtze River Economic Belt. \u003cem\u003eEnergy Economics, 112\u003c/em\u003e, 106127.\u003c/li\u003e\n\u003cli\u003eMadan, A., \u0026amp; Suri, A. (2023). The nexus between development and environment. \u003cem\u003eInternational Journal of Scientific Research in Engineering and Management, 7\u003c/em\u003e(1), 1-5.\u003c/li\u003e\n\u003cli\u003eMoolla, I., \u0026amp; Hiilamo, H. (2023). Health system characteristics and COVID-19 performance in high-income countries. \u003cem\u003eBMC Health Services Research, 23\u003c/em\u003e(1), 242.\u003c/li\u003e\n\u003cli\u003eNeidell, M. (2017). Air pollution and worker productivity. \u003cem\u003eIZA World of Labor\u003c/em\u003e, 363.\u003c/li\u003e\n\u003cli\u003eNovignon, J., \u0026amp; Lawanson, A. (2017). Health expenditure and child health outcomes in Sub-Saharan Africa. \u003cem\u003eAfrican Review of Economics and Finance, 9\u003c/em\u003e(2), 96-121.\u003c/li\u003e\n\u003cli\u003ePatel, P. (2021). Nitrous oxide: The unnoticed greenhouse gas. \u003cem\u003eChemical \u0026amp; Engineering News, 99\u003c/em\u003e(21), 20-23.\u003c/li\u003e\n\u003cli\u003ePotts, D., \u0026amp; Schmischke, M. (2022). Interpretable transformed ANOVA approximation on the example of the prevention of forest fires. \u003cem\u003eFrontiers in Applied Mathematics and Statistics, 8\u003c/em\u003e, 840015.\u003c/li\u003e\n\u003cli\u003eSarpong, B., Nketiah-Amponsah, E., \u0026amp; Owoo, N. S. (2018). Health and economic growth nexus: Evidence from selected Sub-Saharan African (SSA) countries. \u003cem\u003eGlobal Business Review, 20\u003c/em\u003e(1), 1-15.\u003c/li\u003e\n\u003cli\u003eSetyari, N. P. W., \u0026amp; Kusuma, W. G. A. (2021). Economics and environmental development: Testing the environmental Kuznets curve hypothesis. \u003cem\u003eInternational Journal of Energy Economics and Policy, 11\u003c/em\u003e(1), 51-58.\u003c/li\u003e\n\u003cli\u003eShahbaz, M., \u0026amp; Sinha, A. (2019). Environmental Kuznets curve for CO2 emissions: A literature survey. \u003cem\u003eJournal of Economic Studies, 46\u003c/em\u003e(1), 106-168.\u003c/li\u003e\n\u003cli\u003eSingh, R. L., \u0026amp; Singh, P. K. (2016). Global environmental problems. In \u003cem\u003ePrinciples and applications of environmental biotechnology for a sustainable future\u003c/em\u003e (pp. 13-41). Springer.\u003c/li\u003e\n\u003cli\u003eSingh, S., \u0026amp; Yadav, A. (2021). Interconnecting the environment with economic development of a nation. In \u003cem\u003eEnvironmental management\u003c/em\u003e (pp. 35-60). Elsevier.\u003c/li\u003e\n\u003cli\u003eTowah, W. (2019). \u003cem\u003eThe impact of good governance and stability on sustainable development in Ghana\u003c/em\u003e (Doctoral dissertation, Walden University).\u003c/li\u003e\n\u003cli\u003eUzochukwu, B., Ughasoro, M., Okwuosa, C., Onwujekwe, O., Envuladu, E., \u0026amp; Etiaba, E. (2015). Health care financing in Nigeria: Implications for achieving universal health coverage. \u003cem\u003eNigerian Journal of Clinical Practice, 18\u003c/em\u003e(4), 437-444.\u003c/li\u003e\n\u003cli\u003eWang, F. (2015). More health expenditure, better economic performance? Empirical evidence from OECD countries. \u003cem\u003eINQUIRY: The Journal of Health Care Organization, Provision, and Financing, 52\u003c/em\u003e, 1-5.\u003c/li\u003e\n\u003cli\u003eWassie, S. B. (2020). Natural resource degradation tendencies in Ethiopia: A review. \u003cem\u003eEnvironmental Systems Research, 9\u003c/em\u003e, 33.\u003c/li\u003e\n\u003cli\u003eWorld Bank. (2020). \u003cem\u003eGhana country environmental analysis\u003c/em\u003e. Washington, DC: World Bank.\u003c/li\u003e\n\u003cli\u003eWorld Health Organization. (2021, September 22). Air pollution. https://www.who.int/health-topics/air-pollution#tab=tab_1\u003c/li\u003e\n\u003cli\u003eWorld Health Organization. (2021, January 1). Health expenditure. https://www.who.int/data/nutrition/nlis/info/health-expenditure\u003c/li\u003e\n\u003cli\u003eWu, J., Abban, O. J., Yao, H., Boadi, A. D., \u0026amp; Ankomah-Asare, E. T. (2021). The nexus amid foreign direct investment, urbanization, and CO2 emissions: Evidence from energy grouping along the ECOWAS community. \u003cem\u003eEnvironmental Science and Pollution Research, 24\u003c/em\u003e(12), 10183-10207.\u003c/li\u003e\n\u003cli\u003eXing, Z., \u0026amp; Liu, X. (2022). Health expenditures, environmental quality, and economic development: State-of-the-art review and findings in the context of COP26. \u003cem\u003eFrontiers in Public Health, 10\u003c/em\u003e, 1005705.\u003c/li\u003e\n\u003cli\u003eZaidi, S., \u0026amp; Saidi, K. (2018). Environmental pollution, health expenditure and economic growth in the Sub-Saharan Africa countries: Panel ARDL approach. \u003cem\u003eSustainable Cities and Society, 41\u003c/em\u003e, 833-840.\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Environmental degradation, ECOWAS region, SEM, ANN, Health expenditure, CO2 emissions","lastPublishedDoi":"10.21203/rs.3.rs-5675754/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-5675754/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eEnvironmental degradation creates unfriendly conditions for economic growth and public health, especially in developing regions like West African countries, where fast industrialization increases these risks, along with precarious environmental legislation. The study aims to analyze the direct and indirect relationships between environmental degradation, health expenditure, and economic growth within the ECOWAS region. The research adopted structural equation modelling (SEM) and Artificial Neural Network (ANN) analysis to examine the impact of carbon dioxide emission, nitrous oxide emissions, and air quality levels on economic growth, taking health expenditure as the mediating variable. Results of SEM show that carbon emission; and nitrous oxide emission positively influence the economy, while poor air quality negatively affects it, health expenditure mediates the influence of nitrous oxide emission on economic growth with an indirect effect. However, it has an insignificant mediating effect between carbon emissions and air quality. Also, ANN analysis confirms the SEM results that indicate carbon emission has the highest predictive importance. The study, therefore, recommends increased stringency of environmental regulations in the West African region, investment in clean energies, and health infrastructural improvement as ways through which environmental degradation may be minimized to allow the attainment of economic development sustainably.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eJEL: \u003c/strong\u003eH51, P18, Q51, Q53, Q56\u003c/p\u003e","manuscriptTitle":"Green Growth or Economic Trade-offs? Economic Costs of Environmental Degradation in ECOWAS. A New Perspective from Artificial Neural Network and SEM Analysis","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-01-02 06:46:19","doi":"10.21203/rs.3.rs-5675754/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"77af67e2-9619-466d-9cc4-bd473f053f92","owner":[],"postedDate":"January 2nd, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2025-06-05T08:54:02+00:00","versionOfRecord":[],"versionCreatedAt":"2025-01-02 06:46:19","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-5675754","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-5675754","identity":"rs-5675754","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}