Grow and eat healthy: Mediating effect of food poverty and dietary diversity with evidence from Tanzania

Grow and eat healthy: Mediating effect of food poverty and dietary diversity with evidence from Tanzania | 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 Grow and eat healthy: Mediating effect of food poverty and dietary diversity with evidence from Tanzania Furaha Rashid, Robert Lihawa This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-3728355/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 Agriculture and health are linked through natural environment, food production, nutrition and production of medicinal plants. The existing literature on the effect of food poverty and dietary diversity as mediating factors for agriculture on household’s health expenditure presents mixed results and are country specific. This study aimed at examining the link between agriculture and household’s health expenditure mediated by food poverty and dietary diversity using the nationally representative Tanzania household budget survey data 2017/18. We employed an instrumental variable generalized method of moment (IV-GMM) as a method of analysis. Results show that participation in agriculture solely has no effect on health expenditure but the effect stems from food poverty incidence and dietary diversity level as mediating factors. Food poor households had 54.1–55.7% lower health expenditure than food secure households. An additional food group consumed by a household leads to 11.3–11.6% increase in household health expenditure. Thus, policies aiming at improving health should go beyond merely fostering agricultural participation and instead place more emphasis on pro-poor and nutrition sensitive agriculture. Agricultural participation health expenditure food poverty dietary diversity Figures Figure 1 Figure 2 1. Introduction Improving human health has been an important global goal due to its linkages to human capital growth, productivity, income generation and economic development [ 28 ]. As a result, global health care expenditure has risen dramatically over the last two decades from 4–9.8% of the global gross domestic product [ 38 , 44 ]. However, high income countries constitute the largest share of the observed health care expenditure (80%) while low income countries mostly found in Sub-Saharan Africa (SSA) and South Asia constitute the lowest share amounting to 0.24% of the growth in health care expenditure [ 38 , 44 ]. To achieve sustainable development goal 3-“achieve healthy lives” in low income countries, it was estimated that an additional 41 US $ per capita per year will be needed from the current 34.9 US $ per capita [ 44 , 46 ]. The observed increase in health care expenditure in low income countries over the last decade has been attributed by factors including an increase in food poverty which affects about 10% of people worldwide whereby 60% of these are found in developing countries [ 15 ]. Malnutrition due to poor diets also affects around two billion people world -wide and has a significant impact on health and health expenditure. Other factors include the emergence of chronic diseases including diabetes and heart disease and partially due to investment in health care facilities [ 38 ]. Food poverty and malnutrition account for 3% of global total health care spending and 45% of deaths occurring to children under five years annually [ 44 ]. Per capita health expenditure in Tanzania as in other low income countries is low at an average of US $ 40.3 relative to an average of US $ 79.4 for SSA, US $ 5638.7 per capita for high income countries and the World average of US $ 1121.9 per capita [ 44 ]. Similarly, food poverty and malnutrition which are linked to health care expenditure are estimated at 8% and 31.8% respectively [ 31 , 42 ]. The rate of malnutrition is higher than the average for SSA (23.2%) thus placing the country in the third position of countries with chronic malnutrition in SSA [ 14 , 23 ]. There is an increase in extant literature linking food poverty, dietary diversity and health care expenditure [ 6 , 10 , 11 , 18 , 28 , 43 ]. Food poverty affects health expenditure through a number of ways including reliance on less health diets relative to the required minimum amount and quality, and trade-off between food purchase and health care expenditure [ 11 ]. It also correlates more with health behaviours and outcomes and health care access than income [ 26 ]. Food poverty strongly predicts mental and physical health among adults [ 6 ]. Similarly, dietary diversity provides macro and micro nutrients necessary for proper functioning of body mechanisms and thereby affecting health [ 28 ]. Food poverty and dietary diversity are in turn affected by household’s agricultural production among other factors including political, institutional, environmental and social factors. Agriculture sector is essential in reducing food poverty and ensuring an increase in dietary diversity in Tanzania since it employs about 61% of its population, generates 24–29% of the country’s gross domestic product (GDP) and 30% of its export earnings [ 42 ]. Agricultural production affects food poverty and dietary diversity through two pathways – 1) direct pathway, 2) indirect pathway. Existing empirical evidence show a direct link between agricultural production, food poverty and dietary diversity through own food production [ 7 , 9 , 30 , 35 , 36 ]. Since about 50% of households in Tanzania sell less than 15% of their annual food produce, their diets are largely determined by the agricultural output produced [ 28 ]. Food poverty and Dietary diversity thus influences health expenditure through the extent of food security and micronutrient provision. An increase in the quantity and diversity of food produced leads to an increase in own food security, affects price level and micronutrient availability which in turn affects health expenditure. The second pathway (indirect pathway) involves income generated from the sale of agricultural produce. With developed market outlets for agricultural produce, the sector generates income that could be used to purchase more food and highly nutrient dense foods to supplement own-food production and thus reducing food poverty and an increase in dietary diversity [ 8 , 13 , 16 , 17 , 34 ]. Income from agricultural production also influences the purchase of health services. However, the causal link between food poverty, dietary diversity and health expenditure is complex. The complexity emanates from whether it is food poverty and dietary diversity that affect health and hence low or higher health expenditure, or it is health status that affects food poverty and dietary diversity through provision of labour force in agriculture (reverse causality). In this study, we examined whether there is a link from agriculture to health expenditure. The pathways from household’s participation in agricultural production to health expenditure mediated by food poverty and dietary diversity are shown in Fig. 1 . Despite existence of evidence on the linkages between agricultural production and health outcomes, most of the previous studies focused on the effects of crop diversity or specific crop production on nutrition and dietary diversity thereby breaking the causal link between agriculture and health care expenditure [ 4 , 9 , 12 , 40 ]. Even few existing studies that tried to link agriculture and health used anthropometric measures of health status and were specific to a certain proportion of population including children, women of reproductive age and adults [ 18 , 28 , 36 , 37 ]. Others examined the effect of specific crop production on health care expenditure [ 10 ]. Furthermore, some studies focused on the effect of food insecurity specifically to developed countries including the United States of America [ 6 , 18 , 37 ]. Similarly some existing studies have presented mixed results. For example, [ 26 ] showed that there was a positive relationship between agricultural production diversity and height-for-age Z-score among children in Zambia while [ 39 ] examined the effect of crop diversity on height –for-age Z-score in Nepal and found an insignificant effect. Agricultural production has been shown to positively and negatively affect human health. On a positive stand point, a study by [ 43 ] in Tanzania, using an instrumental variable approach, found that, households participating in urban agriculture increased the consumption of more food groups translating into significant improvement in health of children in both short and long run. This reduces health care expenditure resulting from malnutrition incidences and the budget that could be used in purchasing medication could be used to purchase more food varieties and further improve health [ 6 ]. [ 28 ] studied the effects of household crop diversification on child health using anthropometric measures of health including weight –for-age Z-score (WAZ), and height-for age Z-score (HAZ). Based on the three waves of Tanzania national panel survey, the study findings showed that crop diversification positively affected child health with a marginal effect for HAZ measure even though it resulted into increased dietary diversity among households who diversified their crops cultivation. [ 10 ] examined the effects of participation in oil palm cultivation on household living standard measured in terms of health expenditure as a proxy for health status, education, nutrition, social connectedness and living condition status. The results from this study showed that households cultivating oil palm had higher dietary diversity (6.94) than that of non-cultivators (6.59). Due to higher income generated from oil palm production, cultivating households had 40% more health expenditure than non-cultivating households which implies that participating in oil palm cultivation improves household health. [ 36 ] conducted a study on the effects of household participation in irrigation on an anthropometric measure of health called weight-for-height z-score (WHZ), women’s and household’s dietary diversity in Ethiopia and Tanzania. Through the use of panel fixed effect model, the study findings showed that, participating in irrigation agriculture resulted into higher dietary diversity and was reflected in the improved child health where children of households participating in irrigation had a WHZ of 0.87 standard deviations higher than those who did not participate. This signifies that participating in the irrigation agriculture improved household economic access to food, improved nutrition and health and thus reduces health expenditure on malnutrition related diseases. However, on the negative side, low farm productivity which is linked to food insecurity results into malnutrition and hence high health expenditure. For example, a study by [ 6 ] examined the effect of food insecurity on health care expenditure in the United States using zero inflated negative binomial regression. The results showed that, food insecure households had significantly higher average annual health expenditure of $ 6072 than those who were food secure ( $ 4208). The paper implied that food insecurity is associated with increased health care expenditure. The results from this study are in conformity with those of [ 20 , 21 ] in the United States of America which found that an increase in food insecurity was associated with an increase in health care costs. Food insecurity is linked to a variety of chronic health conditions like diabetes, depression, heart attack and hence leading to increased health care expenditure in curing these diseases [ 19 ]. Similarly, agricultural production negatively affect human health through exposure to harmful agro-chemicals and injuries associated with agriculture. Through a review of literature on the pesticide formulants and their distribution pathways on human health, [ 20 ] showed that occupational exposure, local/imported foods, as well as occurrence of accidental spills of agrochemicals mutates the human genetic make-up and metabolism which in turn affects endocrine and reproductive development. [ 22 , 45 ] examined the environmental and health hazards resulting from agricultural pesticide use in Ethiopia. The study findings showed that household members exposure to pesticides results into negative health impacts including skin irritation, nausea, headache, discomfort and dizziness. These in turn could lead to increased health care expenditure. Similarly, a study by [ 20 ] in China showed that, a one percent increase in pesticide application by rice producers would result in a 0.213 million US $ increase in medication costs in China. Our current study contributes and deviates from existing literature in the following ways. First, our study examines the effect of household’s agricultural participation on health expenditure mediated by food poverty and dietary diversity in a developing country perspective unlike previous studies [ 33 ]. Second, we use country representative data on aggregate crops produced to provide a holistic picture on the relationship between agricultural participation and household’s health expenditure with evidence from Tanzania. The paper is organized as follows. Section 2 presents the methodology explaining the source of data and description of key and control variables used as well as an analytical framework. Section 3 presents the empirical results while the conclusion and policy implications are presented in section 4 . 2. Methodology 2.1 Data and definition of variables This study used the nationally representative Tanzania household budget survey collected by the national bureau of statistics in 2017/18 covering a total of 26 regions found in Tanzania mainland. The survey involved a two stage cluster design in selecting a sample size. First, a total of 796 primary sampling units (PSUs) drawn from the Tanzania population and housing census 2012 frame were selected. Then, a systematic sampling of households followed where a sample size of 9552 was selected. However, the response rate was 99% reducing a sample to 9465. Furthermore, due to missing data for some variables of interest in this study, the sample was further reduced to 9374. The main variables of interest were household’s agricultural participation, food poverty, dietary diversity and health expenditure. Household’s agricultural participation is defined in this study as the involvement of at least one household member in own farm production during at least one seasons in the last 12 months. It was measured as a binary variable taking the value of one if at least one household member participated and a value of 0 if not. The measure is suitable since it reduces the ambiguities inherent in the use of extent of participation or share of farm income as a measure of participation [ 20 , 21 ]. Similarly, food poverty was defined as the existence of insufficient economic access to adequate quality and quantity of food to maintain a nutritionally satisfactory and socially accepted diet [ 21 ]. It was measured as a binary variable whereby the proportion of households whose food consumption expenditure is below the food poverty line took the value of 1 and if not otherwise [ 1 , 2 ]. Furthermore, we measured dietary diversity as the count of average number of food groups consumed at the household level over a given reference period which further measures the household’s economic access to variety of foods following [ 25 ]. In this study, a total of 9 healthy food groups were included to measure the household dietary diversity score (HDDS) as used by [ 10 ]. The score ranged from 0 signifying that a typical household did not consume a given food group to 9 if a household consumed all the food groups during the last seven days before the survey. The food groups included in this study consisted of grains/cereals; tubers and roots; legumes, nuts and seeds; vegetables; meat; fish and other sea foods; fruits; milk and milk products; and oils and fat. Then, we defined health expenditure as the aggregated cost spent by the household for health care aspects including medication, preventive care, out of pocket expenditure, doctor consultations, and the cost of visiting health facilities expressed in Tanzania shillings (TZS) per month. Since health expenditure level of a household could be affected by other variables apart from food poverty and dietary diversity, some control variables including education level and age of household head, household size, off-farm income, location, social protection and dependency ratio informed by previous studies [ 6 , 10 , 11 ]. 2.2 Statistical Analysis The empirical analysis in this study was preceded by descriptive statistics where the key variable of interest (health expenditure) and some control variables were described by agricultural participation status, food poverty and dietary diversity. A student t-test and a chi-square measure of association were applied to determine whether there was significant difference in health expenditure among the groups. The empirical analysis followed a series of step-wise regression equations. At the first stage, since participation was treated as a dummy variable, the deteminants of household’s agricultural participation were analysed using a probit model (Eq. 1) where \({Z}_{i}^{*}\) is a latent variable for participation influenced by socio-economic and institutional variables ( \({X}_{i}\) ), \(\omega\) is a vector of instruments used while \({\epsilon }_{i}\) is the disturbance term. $${Z}_{i}^{*}= {\alpha }_{0}+ {X}_{i}^{{\prime }}{\alpha }_{i}+{\alpha }_{1}\omega +{\epsilon }_{i}\dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots ..\dots \dots \dots \dots \dots \dots .\dots \dots \dots \dots \dots \left(1\right)$$ $${Z}_{i}=\left\{\begin{array}{c}1 if {Z}^{*}>0\\ 0 if {Z}^{*}\le 0\end{array}\right. , \epsilon ǀ Z∼N\left(0,\left(\begin{array}{cc}1& q\\ q& 1\end{array}\right)\right)\dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \left(2\right)$$ Since participation in agriculture could be influenced by both observable and unobservable factors including entrepreneurial and manageraial capability, it is subjected to endogeneity problem [ 8 ]. Thus, to address the problem, agricultural participation was instrumented by two instruments – access to agricultural extension and ownership of agricultural capital as used by previous studies [ 5 , 12 , 47 ]. For an instrumental variable to be valid, it must not be correlated with the outcome variable (exclusion criteria) but must be significantly and directly related to the endogenous variable – participation, relevance criteria [ 21 ]. At the second stage, since food poverty is also treated as a dummy variable, the effect of household’s agricultural participation on food poverty was estimated by using an instrumental variable probit (Eq. 3) where \({Y}_{1}\) is a binary for food poverty. $${Y}_{1}= {\beta }_{0}+ {X}_{i}^{{\prime }}{\beta }_{i}+{\beta }_{2}Z+ {\mu }_{i}\dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots .\dots .. \left(3\right)$$ Furthermore, the effect of household’s agricultural participation on dietary diversity was estimated by an instrumental variable Poisson since dietary diversity ( \({Y}_{2})\) is a count variable. Since it has been shown by previous studies that food poverty affects household’s dietary diversity, it was included in Eq. 4 as a determinant of household’s dietary diversity. \({\gamma }_{0}\) - \({\gamma }_{3}\) are the coefficients, \(Z\) represents a binary variable for participation while \({e}_{i }\) is a disturbance term. $${Y}_{2}= {\gamma }_{0}+ {X}_{i}^{{\prime }}{\gamma }_{i}+{\gamma }_{2}Z+{\gamma }_{3}{Y}_{1}+ {e}_{i }\dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots .\dots .. \left(4\right)$$ This was followed by estimating the effect of food poverty and dietary diversity as mediating factors of the effect of agricultural participation on household’s health expenditure ( \({Y}_{3}\) ) in Eq. 5. In this equation, \({\varOmega }_{0}\) is the constant term, \({\varOmega }_{1}\) - \({\varOmega }_{3}\) are coefficients that were estimated, \({X}_{i}^{{\prime }}\) are socio-economic characteristics affecting health expenditure, \({Y}_{1}\) and \({Y}_{2}\) represent food poverty and dietary diversity respectively while \({\eta }_{i }\) is the disturbance term. Since the dependent variable is continuous, Eq. 5 was estimated by OLS and an instrumental variable generalised method of moment (IV-GMM). $${Y}_{3}= {\varOmega }_{0}+ {X}_{i}^{{\prime }}{\varOmega }_{1}+{\varOmega }_{2}Z+{\varOmega }_{3}{Y}_{1}+{\varOmega }_{4}{Y}_{2}+ {\eta }_{i }\dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots .. \left(5\right)$$ To test whether participation was endogenous, the Wald test of exogeneity was used while falsification tests were applied to check the strength of instruments [ 29 , 41 ]. 3. Results 3.1 Descriptive results The descriptive statistics are presented in Tables 1 and 2 . From Table 1 , out of the 9374 households included in the sample, 71.6% participated in agriculture while the remaining percent did not participate. About 6% the households were food poor with an average household size of 4.9 people. A large proportion of households (53.1%) and 41.1% consumed at least seven food groups and between four to six food groups respectively while a small proportion (5.8%) consumed at most three food groups. This implies that, most households had higher and medium dietary diversity. The everage age of typical household head was 47.1 years with household’s average monthly health expenditure of 7954.8 TZS and an everage earning from off-farm income activities of 946893.4 Tanzania shillings. Large percent of households sampled in this study resides in rural area and over half of them had no formal education. Table 1 Descriptive statistics Variable Percent (%) Agricultural participation (yes = 0, 0 = no) 71.6 Food poverty (1 = yes, 0 = no) 5.9 Household’s Dietary diversity score (HDDS) Lower HDDS 5.8 Medium HDDS 41.1 Higher HDDS 53.1 Health expenditure (monthly TZS) 7954.8 Age of household head (mean) 47.1 Off-farm income (TZS/annum) 946893.4 Household size (mean) 4.9 Education of household head (1 = no formal education) 50.3 Location of household (1 = rural, 0 = urban) 70.8 Source : Tanzania household budget survey 2017/18 Table 2 presents household’s health expenditure disaggregated by agricultural participation status, food poverty incidence and dietary diversity. Results show that there was no significant difference in health expenditure by agricultural participation status. However, the difference was observed for the case of mediating variables -food poverty incidence and dietary diversity. Food poor households had significantly lower health expenditure than non-food poor households. Similarly, low food diverse households had higher health expenditure than the high and medium diverse households whereby the medium diverse households had the lowest health expenditure. Table 2 Health expenditure by agricultural participation food poverty and dietary diversity Variable Category Health expenditure (Monthly TZS) t/F Agricultural participation Yes 8259.2 -0.99 No 7188.2 Food poverty Yes 1863.2 3.14** No 8333.1 Household dietary diversity (HDDS) Low 10276.8 4.56* Medium 6221.8 High 9034.1 Education of household head No formal 7286.4 1.39 Formal 8629.8 Location Rural 6967.1 3.18** Urban 10349.9 Standard errors in parentheses, * p < 0.05, ** p < 0.01, *** p < 0.001 Source Author computations from the 2017/18 HBS The extent of health expenditure also significantly differed across rural-urban localities. Rural based households had relatively lower health expenditure than their urban counterpart. This is a reflection of the difference in access to health infrastructure, resources and health information flow between rural and urban areas. Health expenditure also exhbits a curve-linear relationship with age of household head and household off-farm income (Fig. 2 ). At first, health expenditure increases with age of household head, reaches maximum at about age 60 years while above this age health expenditure declines. Similarly, initially, health expenditure increases at a decreasing rate with an increase in household’s off-farm income level while at some point it starts declining. Health expenditure is also positively related to household size where at first as the household increases, health expenditure also increases but at decreasing rate until it reaches a maximu of around a household size of 35 members. We also implemented a correlation analysisto measure the degree of association between agricultural participation, food poverty, dietary diversity and health expenditure. Results in Table 3 show a significantly weak and positive association between food poverty and agricultural participation while it was weak and negative for the agricultural participation and dietary diversity. The results further show that there is no direct link between agricultural participation and health expenditure given insignificant correlation coefficient. This study also found significantly weak and negativa correlation between food poverty and health expenditure and household’s dietary diversity. However, household’s dietary diversity score exhibited a positive correlation with health expenditure implying that improved dietary diversity is directly correlated with an increase in household’s health expenditure. Table 3 Correlation analysis Variables (1) (2) (3) (4) (1) Agricultural participation 1.000 (2) Food poverty 0.033** 1.000 (3) Health expenditure 0.010 -0.032** 1.000 (4) Household dietary diversity score -0.056** -0.165** 0.021** 1.000 ** p < 0.05 3.2 Empirical results and Discussion 3.2.1 Determinants of household’s agricultural participation We first present the effect of household’s agricultural participation on intermediate variables (food poverty and dietary diversity) where agriculture is treated as an endogenous variable in Table 4 while access to extension services and agricultural capital were used as instrumental variables. From the Wald test of exogeneity, the results confirm that household’s agricultural participation was potentially endogenous (Wald χ 2 = 571.5, p = 0.0000) which implies that the use of instruments to instrument participation was crucial. Similarly, results from the falsification test on the selected instruments shows that the instruments are jointly significant in the agricultural participation equation but not in the food poverty and dietary diversity equations. This signifies that the instruments used met both the exclusion and relevance restriction criteria. Results from Table 4 show that, household’s agricultural participation is significantly influenced by the age of household head, household size, off-farm income, access to mobile extension services and agricultural capital access. The coefficient of age of household head was positive and significant implying that the likelihood of a household to participate in agriculture increases with an increase in the age of household head. The coefficient of household size was positive and significant suggesting that an increase in household size is also a motivating factor for household’s agricultural participation in the study area. This is explained by the reason that a large household size may imply high supply of family labour to support agricultral activities if and only if the large proportion of family members are in their working age as opposed to many dependents. Off-farm income earned by a household from off-farm activities significantly and negatively affects the likelihood of household’s agricultural participation. Off-farm income demotevates agricultural participation since it acts as a shock absorber during agricultural crop failure, increases purchasing power of a household and makes a household to have choice on investment decisions based on amount of return and risky nature of an activity. Education coefficient was positive and significant implying that households headed by heads with no formal education had higher likelihood of participation in agriculture than those headed by heads with formal education. This suggests that, education creates more employment opportunities, helps a household in accessing information on low and high return activities and thus are more able to make informed decisions, grab more opportunities unlike those with no formal education which are faced with limited income generating activities and information space. The coefficient of the instrumental variables were also significant though with unexpected signs. Access to extension services provides agronomic and technological information that facilitate adoption of newly improved technologies that further enahnce participation in agriculture. Agricultural capital is also an important input in agricultural production through increasing the efficiency of other factors including labour and land. The negative signs might in these instrumental variables could be explained by the low extent nature of agricultural extension and agricultural capital access in Tanzania. Table 4 Effect of agricultural participation on food poverty and dietary diversity IV-Probit IV-Poisson Variable Participation Food poverty Participation HDDS Participation (1 = yes) 2.251*** -4.167* (0.0170) (1.773) Age of household head 0.000803** -0.0019** 0.000766* 0.00204 (0.000302) (0.000684) (0.000298) (0.00196) Household size 0.00414* -0.00183 0.00317 0.0172 (0.00161) (0.00372) (0.00174) (0.00956) LN (off-farm Income) -0.0756*** 0.165*** -0.0761*** 0.284* (0.00638) (0.0145) (0.00659) (0.139) Education (1 = no formal) 0.0762*** -0.165*** 0.0738*** 0.262 (0.00962) (0.0218) (0.00966) (0.140) Mobile exten access(1 = yes) -0.0141*** -0.0203 (0.00243) (0.0115) Ag-capital access (1 = yes) -0.0128*** 0.0257 (0.00271) (0.0151) Constant 1.645*** -3.744*** 1.660*** 8.343** (0.0892) (0.204) (0.0912) (2.950) Athrho -3.527*** (0.148) Lnsigma -0.811*** (0.00744) Wald test of exogeneity χ 2 = 571.5, p = 0.0000 Wald χ 2 17563.3, p > chi2 = 0.0000 N 9374 9374 Standard errors in parentheses, * p < 0.05, ** p < 0.01, *** p < 0.001 Source Author computations from the 2017/18 HBS 3.2.2 Effect of agricultural participation on mediating variables (Food poverty and HDDS) On the first case, results from the instrumental variable probit in Table 4 show that households participationg in agriculture had higher likelihood of being food poor than non-participating households. This can be explained by the reason that due to scarcity of resources including time allocated in the agricultural sector which limits households participation in off-farm activities, depending on own farm production with low productivity and farm gate price and under developed markets leads households participating entirely in the agriculture sector as their main source of income being food poor. Among the control variables, age of the household head, off-farm income and education level of household head had significant effect on household’s poverty incidence. Food poverty decreases with an increase in age of household head. This could be explained by the reason that aged household heads have accumulated life time wealth, have established long run channels of income generating activities, market access information as well as access to remittances and social protection that helps to reduce food poverty incidence. Unexpectedly, the coefficients for off-farm income and education level of household head showed negative signs signifying that an increase in household’s off-farm income led to an increase in food poverty incidence while having non-formal education translates into lower food poverty. On the second case, results further show that agricultural participation significantly and negatively affect household’s dietary diversity. The result suggests that households participating in agriculture were less food diverse than non-participating households. This could be explained by the reason that, most of the agricultural participating households reside in rural areas with underdeveloped markets and information infrastructure leading to information asymmetry, over-dependency on a single source of income (agriculture) which is prone to climate variability, low and unstable prices which translates into low income and purchasing power particularly on nutrient dense foods (meat, milk and variety of fruits) they do not produce in their own farms [ 32 , 35 , 40 ]. Off-farm income was also found to positively and significantly affecting household’s dietary diversity. This suggests that dietary diversity increases with an increase in household’s off-farm income. Off farm income increases the purchasing power of a household by supplementing the own produced food with market purchased nutrient-dense food items and thus leading to an increase the number of food groups consumed. However in this stance, off farm income would effectively work out to increase the level of household’s dietary diversity given that there is market access facilitated by developed infrastructure and availability of strong institutions. 3.2.3 Effects of food poverty and dietary diversity on household’s health expenditure Table 5 presents the effect of food poverty and dietary diversity on household’s health expenditure as mediating factors between agricultural participation and health expenditure. Both, the F and Wald chi-square tests were significant (p < 0.01) signifying that both models predicted well the data. Results from both OLS and IV-GMM show that, participating in agriculture does not directly affect household’s health expenditure but its effect can be indirectly observed through food poverty incidence and dietary diversity as mediating factors. The coefficient for food poverty and dietary diversity were significant and with similar signs for both the OLS and the IV-GMM although there was a slight difference in magnitude. Food poverty coefficient had significant and negative effect on household’s health expenditure suggesting that food poor households had relatively lower health expenditure than food secure households. Health expenditure was 54.1–55.7% lower among food poor households than food secure households. This is attributed by the reason that being food poor, there is a trade-off between income directed to food purchase and medication/health care investment. Given that, over 70% of household’s income in Tanzania as in other developing countries is alloacted in food purchase, being food poor forces a household to allocate most of its income earned to purchase more food rather than investing in health and thus reduce health expenditure. Food poverty strongly predict poor mental, physical health and chronic diseases particularly to adults and children under five years [ 11 ]. Table 5 Effect of food poverty and dietary diversity on health expenditure Variable LN (Health expenditure) OLS IV-GMM Participation (1 = yes, 0 = no) -0.0209 -0.748 (0.0453) (2.711) Age of household head 0.00476*** 0.00478* (0.00133) (0.00212) Household size 0.0592*** 0.0589*** (0.00679) (0.00880) LN (off-farm Income) 0.127*** 0.0748 (0.0279) (0.210) Education lev (1 = no formal, 0 = formal) -0.0835* -0.0472 (0.0417) (0.204) Food poverty (1 = yes, 0 = no) -0.814*** -0.779*** (0.0976) (0.122) Household dietary diversity score (HDDS) 0.116*** 0.113*** (0.0126) (0.0206) Constant 5.324*** 6.543 (0.396) (4.649) Wald χ 2 (7) F = 41.95 278.73 Prob > χ 2 Prob > F = 0.0000 0.0000 R-squared 0.0497 0.0056 Standard errors in parentheses, * p < 0.05, ** p < 0.01, *** p < 0.001 Source Author computations from the 2017/18 HBS However, our result deviates from those of previous studies including those of [ 6 , 10 , 11 , 37 ]. These studies altogether concluded that food poor households had reatively higher health care expenditure and utilization than food secure households. The reasons put forward by these studies was that food poor households are prone to outbreak of malnutrition related diseases to children under five years, women of reproductive age and cadio-vascular diseases particularly to elders leading to an increase in visitation to health care facilities and services which ultimately increases household’s health expenditure. The difference emanates in the complex conceptualisation of the health expenditure and health status of a person. Higher health expenditure due to an increase in investment in health care services could be a proxy for improved health as used in our current study and supported by previous studies [ 3 , 24 ]. In contrary, higher health expenditure might also reflect presence of diseases, injuries [ 10 ]. Household’s dietary diversity score had positive and significant effect on household’s health expenditure. This suggests that, households consuming diverse food groups had higher health expenditure than less food diverse households. As the household consumes one additional food group, health expenditure increases by about 11.3–11.6% ceteris paribus. Higher dietary diversity translates into improved nutrition and more economic access to food varieties which in turn protects the household against outbreak of diseases due to malnutrition and ensures food security. With absence or reduced malnutrition and being food secure, household income that could be allocated in purchasing nutrient dense foods is instead allocated in health care improvement and thus leading to an increase in household’s health expenditure. Apart from food poverty and dietary diversity as variables of interest, other control variables included in the model including age and education of household head, off-farm income and household size had also significant effect on household’s health expenditure. As the age of household head increase by one year, household’s health expenditure increases marginally by about 0.5%. Aged people are prone to age-related diseases including cardiovascular diseases caused by a decline in health capital over years which in turn needs continuous health care investment resulting in an increase in health expenditure. The coefficient of household size was positive and significant implying that household’s health expenditure level increases as a household size increases. As one member increases in the household, household’s health expenditure increases by about 6% holding other factors affecting health expenditure constant. This suggests that, an increase in members of a household translates in more demand for health care services and thus leading to an increase in household’s health expenditure level. As expected, the coefficient of household’s off farm income was positive and significant implying that household’s health expenditure level increases as off-farm income increases. With an increase in off farm income by one Tanzania shilling, household’s health expenditure increases by 12.7%. Income facilitates access and utilization of health care services and acts as a motivator to increase more investment in improving health of household members such as regular visit to health facilities for body check-up particularly to highly ranked and expensive health facilities [ 19 ]. An increase in income may also translate in more investment in health insurance services which altogether leads to an increase in household’s health expenditure. Similarly, having no formal education translates in relatively lower health expenditure than having formal education. Households headed by heads with no formal education had 8.01% lower health expenditure than households headed by heads with formal education. The plausible reason for this results is that, with no formal education, households might have limited access to health information and services, low investment in health due to having no health education, lower access to more income generating activities and thus low income that is invested in health care improvement. 4. Conclusion and Policy implications Nutrition sensitive and food pro-poor agriculture have received attention by policy makers and nutritionists particularly in developing countries due due its linkage with person’s health. However, the relationship between agricultural production and health expenditure has received little attention by scholars while the existing literature presents mixed results. This study aimed at examining the effect of household’s agricultural participation on health expenditure mediated by food poverty and dietary diversity. Using the 2017/18 nationally representative Tanzania household budget survey and an instrumental variable Probit, IV-Poisson and IV-GMM, this study found the following results. Agricultural participating househlds were more likely to be food poor and less food diverse than non –participating households. There was no difference in household’s health ependiture between participating and non-participating households in agricultural production. The difference was observed by food poverty incidence and dietary diversity level which were used as mediating variables. Food poor households had about 54.1–55.7% lower health expenditure than non-food poor households. Similarly, an increase in household’s dietary diversity by one more food group leads to an increase in household’s health expenditure by 11.3–11.6%. Our study has the following policy implications. First, agricultural policies in Tanzania and other developing countries should go beyond fostering increased agricushould participation. The focus should be on promoting and investing in poverty reducing and nutrition-sensitive agricultural production which have direct effect on food security and dietary diversity and thereby affecting household’s health status and health expenditure. Second, to more improve investment in health, family planning to take into account the household size, promotion of off-farm income generating activities as well as investment in education particularly health education is inevitable. Declarations Ethics approval and consent to participate We utilized secondary data that is nationally representative, obtained from the Tanzania National Bureau of Statistics website, where it is openly accessible for research purposes. Since the data was acquired from the Tanzania National Bureau of Statistics specifically for research, ethical approval was not necessary for this study. Availability of data and materials Data used in the analysis may be available upon reasonable requests. Please contact Mr. Furaha Rashid [email protected] Competing interests The authors declare that they have no confict of interest Funding This study did not receive any form of financial support Authors' contributions FR was responsible for designing the study and authoring the manuscript. RL carried out the data analysis and its interpretation. All authors reviewed and gave their approval to the final version of the manuscript. References Addai KN, Ng’ombe JN, Temoso O. (2022). Food Poverty, Vulnerability, and Food Consumption Inequality among Smallholder Households in Ghana: A Gender-Based Perspective. Soc Indic Res, 1–29. Assefa T, Abide EB. Determinants of food insecurity in rural households: A case of lemo district, southern Ethiopia. Heliyon. 2023;9(1):1–7. Athukorala W, Wilson C, Robinson T. Determinants of health costs due to farmers’ exposure to pesticides: an empirical analysis. J Agric Econ. 2012;63(1):158–74. Balaji SJ, Jhajhria A, Kumar S, Immanuelraj TK, Kar A. Agriculture–nutrition linkages: A preliminary investigation for rural India. Outlook on Agriculture. 2017;46(4):302–8. Benfica RS, Kilic T. (2015). The Effects of smallholder agricultural involvement on household food consumption and dietary diversity: Evidence from Malawi. In International Conference of Agricultural Economists, Milan Italy . Berkowitz SA, Basu S, Meigs JB, Seligman HK. Food insecurity and health care expenditures in the United States, 2011–2013. Health Serv Res. 2018;53(3):1600. Bernzen A, Mangnus E, Sohns F. Diversify, produce or buy? An analysis of factors contributing to household dietary diversity among shrimp and non-shrimp farmers in coastal Bangladesh. Food Secur. 2022;14(3):741–61. Bui LK, Hoang H. Non-farm employment, food poverty and vulnerability in rural Vietnam. Environ Dev Sustain. 2021;23(5):7326–57. Chegere MJ, Stage J. Agricultural production diversity, dietary diversity and nutritional status: Panel data evidence from Tanzania. World Dev. 2020;129:104856. Chrisendo D, Siregar H, Qaim M. Oil palm cultivation improves living standards and human capital formation in smallholder farm households. World Dev. 2022;159:106034. Dean EB, French MT, Mortensen K. Food insecurity, health care utilization, and health care expenditures. Health Serv Res. 2020;55:883–93. Dillon A, McGee K, Oseni G. Agricultural production, dietary diversity and climate variability. J Dev Stud. 2015;51(8):976–95. Ecker O. Agricultural transformation and food and nutrition security in Ghana: Does farm production diversity (still) matter for household dietary diversity? Food Policy. 2018;79:271–82. UNICEF, WHO and WFP. The state of food security and nutrition in the World 2018. Building climate resilience for food security and nutrition. Rome: Food and agriculture organisation of the United Nations; 2018. WFP and WHO. brief to the state of food security and nutrition in the world 2021. Transforming food systems for food security, improved nutrition and affordable healthy diets for all. Rome: Food and agriculture organisation of the United Nations; 2021. Fraval S, Yameogo V, Ayantunde A, Hammond J, De Boer IJ, Oosting SJ, Van Wijk MT. Food security in rural Burkina Faso: the importance of consumption of own-farm sourced food versus purchased food. Agric Food Secur. 2020;9(1):1–17. Gaillard C, Verger EO, Dury S, Dop MC, Ati E, Medina Study Group. J., and (2022). Farm production diversity and women’s dietary diversity: Evidence from central Tunisia. PLoS ONE, 17(2), e0263276. Garcia SP, Haddix A, Barnett K. Peer reviewed: incremental health care costs associated with food insecurity and chronic conditions among older adults. Prev Chronic Dis. 2018;15:E108. Grossman M. On the concept of health capital and the demand for health. J Polit Econ. 1972;80(2):223–55. Hawkes C, Ruel MT. (2006). Overview: Understanding the links between agriculture and health. 2020 Vis briefs, (13 (1)). Johnston MP. Secondary data analysis: A method of which the time has come. Qualitative and quantitative methods in libraries. 2017;3(3):619–26. Kafle K, Paliwal N, Benfica R. (2018). Who works in agriculture? Exploring the dynamics of youth involvement in Tanzania’s and Malawi’s Agri-food system . A selected paper presented at the 2018 Agricultural and Applied Economics Association Annual meeting, Washington, D.C., August 5–7. Kalyabina VP, Esimbekova EN, Kopylova KV, Kratasyuk VA. Pesticides: formulants, distribution pathways and effects on human health–a review. Toxicol Rep. 2021;8:1179–92. Ke XU, Saksena P, Holly A. The determinants of health expenditure: a country-level panel data analysis. Geneva: World Health Organization. 2011;26:1–28. Kennedy G, Ballard T, Dop MC. Guidelines for measuring household and individual dietary diversity. Nutrition and Consumer Protection Division, Food and Agriculture Organization of the United Nations; 2013. Kumar N, Harris J, Rawat R. If they grow it, will they eat and grow? Evidence from Zambia on agricultural diversity and child undernutrition. J Dev Stud. 2015;51(8):1060–77. Lai W. Pesticide use and health outcomes: Evidence from agricultural water pollution in China. J Environ Econ Manag. 2017;86:93–120. Lovo S, Veronesi M. Crop diversification and child health: empirical evidence from Tanzania. Ecol Econ. 2019;158:168–79. Ma W, Zhou X, Renwick A. Impact of off-farm income on household energy expenditures in China: Implications for rural energy transition. Energy Policy. 2019;127:248–58. Muthini D, Nzuma J, Qaim M. Subsistence production, markets, and dietary diversity in the Kenyan small farm sector. Food Policy. 2020;97:101956. NBS. (2019). Tanzania mainland household budget survey 2017-18. Key Indicators report. Dodoma, Tanzania. Retrieved on September 1, 2022 from World Wide Web at https://www.nbs.go.tz/index.php/en/census-surveys/poverty-indicators . Ochieng J, Afari-Sefa V, Lukumay PJ, Dubois T. (2017). Determinants of dietary diversity and the potential role of men in improving household nutrition in Tanzania. PLoS ONE, 12(12), e0189022. O'Connor N, Farag K, Baines R. What is food poverty? A conceptual framework. Br Food J. 2016;118(2):429–49. Ogutu SO, Gödecke T, Qaim M. Agricultural commercialisation and nutrition in smallholder farm households. J Agric Econ. 2020;71(2):534–55. Olabisi M, Obekpa HO, Liverpool-Tasie LSO. Is growing your own food necessary for dietary diversity? Evidence from Nigeria. Food Policy. 2021;104:102144. Passarelli S, Mekonnen D, Bryan E, Ringler C. Evaluating the pathways from small-scale irrigation to dietary diversity: evidence from Ethiopia and Tanzania. Food Secur. 2018;10(4):981–97. Peltz A, Garg A. Food insecurity and health care use. Pediatrics. 2019;144(4):e20190347. Schneider MT, Chang AY, Chapin A, Chen CS, Crosby SW, Harle AC, Dieleman JL. (2021). Health expenditures by services and providers for 195 countries, 2000–2017. BMJ Global Health, 6(7), e005799. Shively G, Sununtnasuk C. Agricultural diversity and child stunting in Nepal. J Dev Stud. 2015;51(8):1078–96. Sibhatu KT, Qaim M. Meta-analysis of the association between production diversity, diets, and nutrition in smallholder farm households. Food Policy. 2018;77:1–18. Signorelli S, Haile B, Kotu BH. (2017). Exploring the agriculture-nutrition linkage in northern Ghana. IFPRI discussion paper 1697. Washington, D.C: Intenational food policy research institute. URT (United Republic of Tanzania). (2021). Integrated Labour force survey 2020/21. Retrieved on September 15, 2022 from World Wide Web https://www.nbs.go.tz/index.php/en/census-surveys/labour-statistics/688-integrated-labour-force-survey-2020-21 . Wagner N, Tasciotti L. Urban agriculture, dietary diversity and child health in a sample of Tanzanian town folk. Int J Shape Model. 2018;39(2):234–51. WHO (World Health Organization). (2021). Global expenditure on health: public spending on the rise? Retrieved on September 16, 2022 from World Wide Web at https://www.who.int/publications/i/item/9789240041219 . Wondimu M, Gezahegn A, Gedefaw Y. (2022). Health and Environmental Hazards Associated with Agricultural Pesticides in Ethiopia. Asian Res J Agric, 14–26. World Bank. (2022). Current health expenditure as a percentage of GDP. Retrieved on September 5, 2022 from World Wide Web at https://data.worldbank.org/indicator/SH.XPD.CHEX.GD.ZS . Zanello G, Shankar B, Poole N. Buy or make? Agricultural production diversity, markets and dietary diversity in Afghanistan. Food Policy. 2019;87:101731. 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-3728355","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":265528434,"identity":"2dacebfb-7cbc-476d-b5e7-4b79c2248f96","order_by":0,"name":"Furaha Rashid","email":"","orcid":"","institution":"Mzumbe University","correspondingAuthor":false,"prefix":"","firstName":"Furaha","middleName":"","lastName":"Rashid","suffix":""},{"id":265528435,"identity":"a7fac767-566a-45a1-bc67-2c676c2491e3","order_by":1,"name":"Robert Lihawa","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA40lEQVRIiWNgGAWjYBACPmYeCMMARHwAYjZ2AlrYkLQwNs4AixDSwoCkpRnMJqiFnffgZ56ae3Lm7L3PH9v82ibPx8zA+OFjDj6H8SVL8xwrNrbsOW7YnNt327CNmYFZcuY2vH4xkM5hS0jccCONsTm35zYjUAsbMy9+Lca/c/4Btdx/xths2XPbnhgtZtK5bSBb2BibGX7cTiRKi/XfvgSgX9IYZ/Y23E5uY2ZsxusXfv4zxjdnfEsAhtgxhg8//ty2nd/efPDDRzxaUAFjG5hsIFY9CPwhRfEoGAWjYBSMFAAAjs1H0HKnz9YAAAAASUVORK5CYII=","orcid":"","institution":"Mzumbe University","correspondingAuthor":true,"prefix":"","firstName":"Robert","middleName":"","lastName":"Lihawa","suffix":""}],"badges":[],"createdAt":"2023-12-09 03:59:09","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-3728355/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-3728355/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":49311670,"identity":"e60f5ce1-5ced-4204-b6e4-303314865286","added_by":"auto","created_at":"2024-01-08 13:46:36","extension":"jpeg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":191591,"visible":true,"origin":"","legend":"\u003cp\u003ePathways from agricultural production to health expenditure\u003c/p\u003e","description":"","filename":"floatimage1.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-3728355/v1/d30206c4639ad42881efc819.jpeg"},{"id":49311669,"identity":"dea21c90-aa91-4024-9cbe-9367dd34a3f3","added_by":"auto","created_at":"2024-01-08 13:46:36","extension":"jpeg","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":311397,"visible":true,"origin":"","legend":"\u003cp\u003eRelationship between health expenditure, age, household size and income\u003c/p\u003e","description":"","filename":"floatimage2.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-3728355/v1/7d64ef985d9c19ac06a54b61.jpeg"},{"id":71261206,"identity":"b0066ce8-62b1-4242-9983-8b989993b88d","added_by":"auto","created_at":"2024-12-12 16:32:01","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1163352,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-3728355/v1/07ac447d-76c2-4511-a03b-6de5906510b1.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Grow and eat healthy: Mediating effect of food poverty and dietary diversity with evidence from Tanzania","fulltext":[{"header":"1. Introduction","content":"\u003cp\u003eImproving human health has been an important global goal due to its linkages to human capital growth, productivity, income generation and economic development [\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e]. As a result, global health care expenditure has risen dramatically over the last two decades from 4\u0026ndash;9.8% of the global gross domestic product [\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e, \u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e]. However, high income countries constitute the largest share of the observed health care expenditure (80%) while low income countries mostly found in Sub-Saharan Africa (SSA) and South Asia constitute the lowest share amounting to 0.24% of the growth in health care expenditure [\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e, \u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e]. To achieve sustainable development goal 3-\u0026ldquo;achieve healthy lives\u0026rdquo; in low income countries, it was estimated that an additional 41 US\u003cspan\u003e$\u003c/span\u003e per capita per year will be needed from the current 34.9 US\u003cspan\u003e$\u003c/span\u003e per capita [\u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e, \u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e46\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eThe observed increase in health care expenditure in low income countries over the last decade has been attributed by factors including an increase in food poverty which affects about 10% of people worldwide whereby 60% of these are found in developing countries [\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e]. Malnutrition due to poor diets also affects around two billion people world -wide and has a significant impact on health and health expenditure. Other factors include the emergence of chronic diseases including diabetes and heart disease and partially due to investment in health care facilities [\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e]. Food poverty and malnutrition account for 3% of global total health care spending and 45% of deaths occurring to children under five years annually [\u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e].\u003c/p\u003e \u003cp\u003ePer capita health expenditure in Tanzania as in other low income countries is low at an average of US\u003cspan\u003e$\u003c/span\u003e 40.3 relative to an average of US\u003cspan\u003e$\u003c/span\u003e79.4 for SSA, US\u003cspan\u003e$\u003c/span\u003e 5638.7 per capita for high income countries and the World average of US\u003cspan\u003e$\u003c/span\u003e 1121.9 per capita [\u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e]. Similarly, food poverty and malnutrition which are linked to health care expenditure are estimated at 8% and 31.8% respectively [\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e, \u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e]. The rate of malnutrition is higher than the average for SSA (23.2%) thus placing the country in the third position of countries with chronic malnutrition in SSA [\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e, \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eThere is an increase in extant literature linking food poverty, dietary diversity and health care expenditure [\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e, \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e, \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e, \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e, \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e, \u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e]. Food poverty affects health expenditure through a number of ways including reliance on less health diets relative to the required minimum amount and quality, and trade-off between food purchase and health care expenditure [\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e]. It also correlates more with health behaviours and outcomes and health care access than income [\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e]. Food poverty strongly predicts mental and physical health among adults [\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e]. Similarly, dietary diversity provides macro and micro nutrients necessary for proper functioning of body mechanisms and thereby affecting health [\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eFood poverty and dietary diversity are in turn affected by household\u0026rsquo;s agricultural production among other factors including political, institutional, environmental and social factors. Agriculture sector is essential in reducing food poverty and ensuring an increase in dietary diversity in Tanzania since it employs about 61% of its population, generates 24\u0026ndash;29% of the country\u0026rsquo;s gross domestic product (GDP) and 30% of its export earnings [\u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e]. Agricultural production affects food poverty and dietary diversity through two pathways \u0026ndash; 1) direct pathway, 2) indirect pathway. Existing empirical evidence show a direct link between agricultural production, food poverty and dietary diversity through own food production [\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e, \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e, \u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e, \u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e, \u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e]. Since about 50% of households in Tanzania sell less than 15% of their annual food produce, their diets are largely determined by the agricultural output produced [\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e]. Food poverty and Dietary diversity thus influences health expenditure through the extent of food security and micronutrient provision. An increase in the quantity and diversity of food produced leads to an increase in own food security, affects price level and micronutrient availability which in turn affects health expenditure.\u003c/p\u003e \u003cp\u003eThe second pathway (indirect pathway) involves income generated from the sale of agricultural produce. With developed market outlets for agricultural produce, the sector generates income that could be used to purchase more food and highly nutrient dense foods to supplement own-food production and thus reducing food poverty and an increase in dietary diversity [\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e, \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e, \u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e, \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e, \u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e]. Income from agricultural production also influences the purchase of health services. However, the causal link between food poverty, dietary diversity and health expenditure is complex.\u003c/p\u003e \u003cp\u003eThe complexity emanates from whether it is food poverty and dietary diversity that affect health and hence low or higher health expenditure, or it is health status that affects food poverty and dietary diversity through provision of labour force in agriculture (reverse causality). In this study, we examined whether there is a link from agriculture to health expenditure. The pathways from household\u0026rsquo;s participation in agricultural production to health expenditure mediated by food poverty and dietary diversity are shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eDespite existence of evidence on the linkages between agricultural production and health outcomes, most of the previous studies focused on the effects of crop diversity or specific crop production on nutrition and dietary diversity thereby breaking the causal link between agriculture and health care expenditure [\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e, \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e, \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e, \u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e]. Even few existing studies that tried to link agriculture and health used anthropometric measures of health status and were specific to a certain proportion of population including children, women of reproductive age and adults [\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e, \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e, \u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e, \u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e]. Others examined the effect of specific crop production on health care expenditure [\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e]. Furthermore, some studies focused on the effect of food insecurity specifically to developed countries including the United States of America [\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e, \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e, \u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e]. Similarly some existing studies have presented mixed results. For example, [\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e] showed that there was a positive relationship between agricultural production diversity and height-for-age Z-score among children in Zambia while [\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e] examined the effect of crop diversity on height \u0026ndash;for-age Z-score in Nepal and found an insignificant effect.\u003c/p\u003e \u003cp\u003eAgricultural production has been shown to positively and negatively affect human health. On a positive stand point, a study by [\u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e] in Tanzania, using an instrumental variable approach, found that, households participating in urban agriculture increased the consumption of more food groups translating into significant improvement in health of children in both short and long run. This reduces health care expenditure resulting from malnutrition incidences and the budget that could be used in purchasing medication could be used to purchase more food varieties and further improve health [\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e].\u003c/p\u003e \u003cp\u003e[\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e] studied the effects of household crop diversification on child health using anthropometric measures of health including weight \u0026ndash;for-age Z-score (WAZ), and height-for age Z-score (HAZ). Based on the three waves of Tanzania national panel survey, the study findings showed that crop diversification positively affected child health with a marginal effect for HAZ measure even though it resulted into increased dietary diversity among households who diversified their crops cultivation. [\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e] examined the effects of participation in oil palm cultivation on household living standard measured in terms of health expenditure as a proxy for health status, education, nutrition, social connectedness and living condition status. The results from this study showed that households cultivating oil palm had higher dietary diversity (6.94) than that of non-cultivators (6.59). Due to higher income generated from oil palm production, cultivating households had 40% more health expenditure than non-cultivating households which implies that participating in oil palm cultivation improves household health.\u003c/p\u003e \u003cp\u003e[\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e] conducted a study on the effects of household participation in irrigation on an anthropometric measure of health called weight-for-height z-score (WHZ), women\u0026rsquo;s and household\u0026rsquo;s dietary diversity in Ethiopia and Tanzania. Through the use of panel fixed effect model, the study findings showed that, participating in irrigation agriculture resulted into higher dietary diversity and was reflected in the improved child health where children of households participating in irrigation had a WHZ of 0.87 standard deviations higher than those who did not participate. This signifies that participating in the irrigation agriculture improved household economic access to food, improved nutrition and health and thus reduces health expenditure on malnutrition related diseases.\u003c/p\u003e \u003cp\u003eHowever, on the negative side, low farm productivity which is linked to food insecurity results into malnutrition and hence high health expenditure. For example, a study by [\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e] examined the effect of food insecurity on health care expenditure in the United States using zero inflated negative binomial regression. The results showed that, food insecure households had significantly higher average annual health expenditure of \u003cspan\u003e$\u003c/span\u003e 6072 than those who were food secure (\u003cspan\u003e$\u003c/span\u003e 4208). The paper implied that food insecurity is associated with increased health care expenditure. The results from this study are in conformity with those of [\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e, \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e] in the United States of America which found that an increase in food insecurity was associated with an increase in health care costs. Food insecurity is linked to a variety of chronic health conditions like diabetes, depression, heart attack and hence leading to increased health care expenditure in curing these diseases [\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eSimilarly, agricultural production negatively affect human health through exposure to harmful agro-chemicals and injuries associated with agriculture. Through a review of literature on the pesticide formulants and their distribution pathways on human health, [\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e] showed that occupational exposure, local/imported foods, as well as occurrence of accidental spills of agrochemicals mutates the human genetic make-up and metabolism which in turn affects endocrine and reproductive development. [\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e, \u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e] examined the environmental and health hazards resulting from agricultural pesticide use in Ethiopia. The study findings showed that household members exposure to pesticides results into negative health impacts including skin irritation, nausea, headache, discomfort and dizziness. These in turn could lead to increased health care expenditure. Similarly, a study by [\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e] in China showed that, a one percent increase in pesticide application by rice producers would result in a 0.213\u0026nbsp;million US \u003cspan\u003e$\u003c/span\u003e increase in medication costs in China.\u003c/p\u003e \u003cp\u003eOur current study contributes and deviates from existing literature in the following ways. First, our study examines the effect of household\u0026rsquo;s agricultural participation on health expenditure mediated by food poverty and dietary diversity in a developing country perspective unlike previous studies [\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e]. Second, we use country representative data on aggregate crops produced to provide a holistic picture on the relationship between agricultural participation and household\u0026rsquo;s health expenditure with evidence from Tanzania. The paper is organized as follows. Section \u003cspan refid=\"Sec2\" class=\"InternalRef\"\u003e2\u003c/span\u003e presents the methodology explaining the source of data and description of key and control variables used as well as an analytical framework. Section \u003cspan refid=\"Sec5\" class=\"InternalRef\"\u003e3\u003c/span\u003e presents the empirical results while the conclusion and policy implications are presented in section \u003cspan refid=\"Sec11\" class=\"InternalRef\"\u003e4\u003c/span\u003e.\u003c/p\u003e"},{"header":"2. Methodology","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003e\u003cb\u003e2.1\u003c/b\u003e \u003cem\u003eData and definition of variables\u003c/em\u003e\u003c/h2\u003e \u003cp\u003eThis study used the nationally representative Tanzania household budget survey collected by the national bureau of statistics in 2017/18 covering a total of 26 regions found in Tanzania mainland. The survey involved a two stage cluster design in selecting a sample size. First, a total of 796 primary sampling units (PSUs) drawn from the Tanzania population and housing census 2012 frame were selected. Then, a systematic sampling of households followed where a sample size of 9552 was selected. However, the response rate was 99% reducing a sample to 9465. Furthermore, due to missing data for some variables of interest in this study, the sample was further reduced to 9374.\u003c/p\u003e \u003cp\u003eThe main variables of interest were household\u0026rsquo;s agricultural participation, food poverty, dietary diversity and health expenditure. Household\u0026rsquo;s agricultural participation is defined in this study as the involvement of at least one household member in own farm production during at least one seasons in the last 12 months. It was measured as a binary variable taking the value of one if at least one household member participated and a value of 0 if not. The measure is suitable since it reduces the ambiguities inherent in the use of extent of participation or share of farm income as a measure of participation [\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e, \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e]. Similarly, food poverty was defined as the existence of insufficient economic access to adequate quality and quantity of food to maintain a nutritionally satisfactory and socially accepted diet [\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e]. It was measured as a binary variable whereby the proportion of households whose food consumption expenditure is below the food poverty line took the value of 1 and if not otherwise [\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e, \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eFurthermore, we measured dietary diversity as the count of average number of food groups consumed at the household level over a given reference period which further measures the household\u0026rsquo;s economic access to variety of foods following [\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e]. In this study, a total of 9 healthy food groups were included to measure the household dietary diversity score (HDDS) as used by [\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e]. The score ranged from 0 signifying that a typical household did not consume a given food group to 9 if a household consumed all the food groups during the last seven days before the survey. The food groups included in this study consisted of grains/cereals; tubers and roots; legumes, nuts and seeds; vegetables; meat; fish and other sea foods; fruits; milk and milk products; and oils and fat.\u003c/p\u003e \u003cp\u003eThen, we defined health expenditure as the aggregated cost spent by the household for health care aspects including medication, preventive care, out of pocket expenditure, doctor consultations, and the cost of visiting health facilities expressed in Tanzania shillings (TZS) per month. Since health expenditure level of a household could be affected by other variables apart from food poverty and dietary diversity, some control variables including education level and age of household head, household size, off-farm income, location, social protection and dependency ratio informed by previous studies [\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e, \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e, \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e].\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec4\" class=\"Section2\"\u003e \u003ch2\u003e\u003cb\u003e2.2\u003c/b\u003e \u003cem\u003eStatistical Analysis\u003c/em\u003e\u003c/h2\u003e \u003cp\u003eThe empirical analysis in this study was preceded by descriptive statistics where the key variable of interest (health expenditure) and some control variables were described by agricultural participation status, food poverty and dietary diversity. A student t-test and a chi-square measure of association were applied to determine whether there was significant difference in health expenditure among the groups. The empirical analysis followed a series of step-wise regression equations. At the first stage, since participation was treated as a dummy variable, the deteminants of household\u0026rsquo;s agricultural participation were analysed using a probit model (Eq.\u0026nbsp;1) where \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\({Z}_{i}^{*}\\)\u003c/span\u003e\u003c/span\u003e is a latent variable for participation influenced by socio-economic and institutional variables (\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\({X}_{i}\\)\u003c/span\u003e\u003c/span\u003e), \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\omega\\)\u003c/span\u003e\u003c/span\u003e is a vector of instruments used while \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\({\\epsilon }_{i}\\)\u003c/span\u003e\u003c/span\u003e is the disturbance term.\u003cdiv id=\"Equa\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equa\" name=\"EquationSource\"\u003e\n$${Z}_{i}^{*}= {\\alpha }_{0}+ {X}_{i}^{{\\prime }}{\\alpha }_{i}+{\\alpha }_{1}\\omega +{\\epsilon }_{i}\\dots \\dots \\dots \\dots \\dots \\dots \\dots \\dots \\dots \\dots \\dots \\dots \\dots \\dots \\dots \\dots \\dots \\dots \\dots ..\\dots \\dots \\dots \\dots \\dots \\dots .\\dots \\dots \\dots \\dots \\dots \\left(1\\right)$$\u003c/div\u003e\u003c/div\u003e\u003cdiv id=\"Equb\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equb\" name=\"EquationSource\"\u003e\n$${Z}_{i}=\\left\\{\\begin{array}{c}1 if {Z}^{*}\u0026gt;0\\\\ 0 if {Z}^{*}\\le 0\\end{array}\\right. , \\epsilon ǀ Z\u0026sim;N\\left(0,\\left(\\begin{array}{cc}1\u0026amp; q\\\\ q\u0026amp; 1\\end{array}\\right)\\right)\\dots \\dots \\dots \\dots \\dots \\dots \\dots \\dots \\dots \\dots \\dots \\dots \\dots \\dots \\dots \\dots \\dots \\dots \\dots \\dots \\dots \\dots \\dots \\dots \\left(2\\right)$$\u003c/div\u003e\u003c/div\u003e\u003c/p\u003e \u003cp\u003eSince participation in agriculture could be influenced by both observable and unobservable factors including entrepreneurial and manageraial capability, it is subjected to endogeneity problem [\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e]. Thus, to address the problem, agricultural participation was instrumented by two instruments \u0026ndash; access to agricultural extension and ownership of agricultural capital as used by previous studies [\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e, \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e, \u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e47\u003c/span\u003e]. For an instrumental variable to be valid, it must not be correlated with the outcome variable (exclusion criteria) but must be significantly and directly related to the endogenous variable \u0026ndash; participation, relevance criteria [\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eAt the second stage, since food poverty is also treated as a dummy variable, the effect of household\u0026rsquo;s agricultural participation on food poverty was estimated by using an instrumental variable probit (Eq.\u0026nbsp;3) where \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\({Y}_{1}\\)\u003c/span\u003e\u003c/span\u003e is a binary for food poverty.\u003cdiv id=\"Equc\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equc\" name=\"EquationSource\"\u003e\n$${Y}_{1}= {\\beta }_{0}+ {X}_{i}^{{\\prime }}{\\beta }_{i}+{\\beta }_{2}Z+ {\\mu }_{i}\\dots \\dots \\dots \\dots \\dots \\dots \\dots \\dots \\dots \\dots \\dots \\dots \\dots \\dots \\dots \\dots \\dots \\dots \\dots \\dots \\dots \\dots \\dots \\dots \\dots \\dots \\dots \\dots \\dots .\\dots .. \\left(3\\right)$$\u003c/div\u003e\u003c/div\u003e\u003c/p\u003e \u003cp\u003eFurthermore, the effect of household\u0026rsquo;s agricultural participation on dietary diversity was estimated by an instrumental variable Poisson since dietary diversity (\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\({Y}_{2})\\)\u003c/span\u003e\u003c/span\u003e is a count variable. Since it has been shown by previous studies that food poverty affects household\u0026rsquo;s dietary diversity, it was included in Eq.\u0026nbsp;4 as a determinant of household\u0026rsquo;s dietary diversity. \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\({\\gamma }_{0}\\)\u003c/span\u003e\u003c/span\u003e - \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\({\\gamma }_{3}\\)\u003c/span\u003e\u003c/span\u003e are the coefficients, \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(Z\\)\u003c/span\u003e\u003c/span\u003e represents a binary variable for participation while \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\({e}_{i }\\)\u003c/span\u003e\u003c/span\u003e is a disturbance term.\u003cdiv id=\"Equd\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equd\" name=\"EquationSource\"\u003e\n$${Y}_{2}= {\\gamma }_{0}+ {X}_{i}^{{\\prime }}{\\gamma }_{i}+{\\gamma }_{2}Z+{\\gamma }_{3}{Y}_{1}+ {e}_{i }\\dots \\dots \\dots \\dots \\dots \\dots \\dots \\dots \\dots \\dots \\dots \\dots \\dots \\dots \\dots \\dots \\dots \\dots \\dots \\dots \\dots \\dots \\dots \\dots \\dots \\dots .\\dots .. \\left(4\\right)$$\u003c/div\u003e\u003c/div\u003e\u003c/p\u003e \u003cp\u003eThis was followed by estimating the effect of food poverty and dietary diversity as mediating factors of the effect of agricultural participation on household\u0026rsquo;s health expenditure (\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\({Y}_{3}\\)\u003c/span\u003e\u003c/span\u003e) in Eq.\u0026nbsp;5. In this equation, \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\({\\varOmega }_{0}\\)\u003c/span\u003e\u003c/span\u003e is the constant term, \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\({\\varOmega }_{1}\\)\u003c/span\u003e\u003c/span\u003e -\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\({\\varOmega }_{3}\\)\u003c/span\u003e\u003c/span\u003e are coefficients that were estimated, \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\({X}_{i}^{{\\prime }}\\)\u003c/span\u003e\u003c/span\u003e are socio-economic characteristics affecting health expenditure, \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\({Y}_{1}\\)\u003c/span\u003e\u003c/span\u003e and \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\({Y}_{2}\\)\u003c/span\u003e\u003c/span\u003e represent food poverty and dietary diversity respectively while \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\({\\eta }_{i }\\)\u003c/span\u003e\u003c/span\u003e is the disturbance term. Since the dependent variable is continuous, Eq.\u0026nbsp;5 was estimated by OLS and an instrumental variable generalised method of moment (IV-GMM).\u003cdiv id=\"Eque\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Eque\" name=\"EquationSource\"\u003e\n$${Y}_{3}= {\\varOmega }_{0}+ {X}_{i}^{{\\prime }}{\\varOmega }_{1}+{\\varOmega }_{2}Z+{\\varOmega }_{3}{Y}_{1}+{\\varOmega }_{4}{Y}_{2}+ {\\eta }_{i }\\dots \\dots \\dots \\dots \\dots \\dots \\dots \\dots \\dots \\dots \\dots \\dots \\dots \\dots \\dots \\dots \\dots \\dots \\dots \\dots \\dots \\dots .. \\left(5\\right)$$\u003c/div\u003e\u003c/div\u003e\u003c/p\u003e \u003cp\u003eTo test whether participation was endogenous, the Wald test of exogeneity was used while falsification tests were applied to check the strength of instruments [\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e, \u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e].\u003c/p\u003e \u003c/div\u003e"},{"header":"3. Results","content":"\u003cdiv id=\"Sec6\" class=\"Section2\"\u003e \u003ch2\u003e\u003cb\u003e3.1\u003c/b\u003e \u003cem\u003eDescriptive results\u003c/em\u003e\u003c/h2\u003e \u003cp\u003eThe descriptive statistics are presented in Tables\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e and \u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e. From Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e, out of the 9374 households included in the sample, 71.6% participated in agriculture while the remaining percent did not participate. About 6% the households were food poor with an average household size of 4.9 people. A large proportion of households (53.1%) and 41.1% consumed at least seven food groups and between four to six food groups respectively while a small proportion (5.8%) consumed at most three food groups. This implies that, most households had higher and medium dietary diversity. The everage age of typical household head was 47.1 years with household\u0026rsquo;s average monthly health expenditure of 7954.8 TZS and an everage earning from off-farm income activities of 946893.4 Tanzania shillings. Large percent of households sampled in this study resides in rural area and over half of them had no formal education.\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\u003eDescriptive statistics\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=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\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\u003ePercent (%)\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAgricultural participation (yes\u0026thinsp;=\u0026thinsp;0, 0\u0026thinsp;=\u0026thinsp;no)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e71.6\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eFood poverty (1\u0026thinsp;=\u0026thinsp;yes, 0\u0026thinsp;=\u0026thinsp;no)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e5.9\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHousehold\u0026rsquo;s Dietary diversity score (HDDS)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLower HDDS\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e5.8\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMedium HDDS\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e41.1\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHigher HDDS\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e53.1\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHealth expenditure (monthly TZS)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e7954.8\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAge of household head (mean)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e47.1\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eOff-farm income (TZS/annum)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e946893.4\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHousehold size (mean)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e4.9\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eEducation of household head (1\u0026thinsp;=\u0026thinsp;no formal education)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e50.3\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLocation of household (1\u0026thinsp;=\u0026thinsp;rural, 0\u0026thinsp;=\u0026thinsp;urban)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e70.8\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003ctfoot\u003e \u003ctr\u003e\u003ctd colspan=\"2\"\u003e\u003cem\u003eSource\u003c/em\u003e: Tanzania household budget survey 2017/18\u003c/td\u003e\u003c/tr\u003e \u003c/tfoot\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003e \u003cdiv class=\"BlockQuote\"\u003e \u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e presents household\u0026rsquo;s health expenditure disaggregated by agricultural participation status, food poverty incidence and dietary diversity. Results show that there was no significant difference in health expenditure by agricultural participation status. However, the difference was observed for the case of mediating variables -food poverty incidence and dietary diversity. Food poor households had significantly lower health expenditure than non-food poor households. Similarly, low food diverse households had higher health expenditure than the high and medium diverse households whereby the medium diverse households had the lowest health expenditure.\u003c/p\u003e \u003c/div\u003e \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\u003eHealth expenditure by agricultural participation food poverty and dietary diversity\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"4\"\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=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\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\u003eCategory\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eHealth expenditure (Monthly TZS)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003et/F\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAgricultural participation\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e8259.2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e-0.99\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eNo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e7188.2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eFood poverty\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e1863.2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e3.14**\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eNo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e8333.1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHousehold dietary diversity (HDDS)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eLow\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e10276.8\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e4.56*\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMedium\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e6221.8\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eHigh\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e9034.1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eEducation of household head\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eNo formal\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e7286.4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.39\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eFormal\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e8629.8\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLocation\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eRural\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e6967.1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e3.18**\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eUrban\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e10349.9\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eStandard errors in parentheses, * p\u0026thinsp;\u0026lt;\u0026thinsp;0.05, ** p\u0026thinsp;\u0026lt;\u0026thinsp;0.01, *** p\u0026thinsp;\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003cp\u003e \u003cstrong\u003eSource\u003c/strong\u003e \u003cp\u003eAuthor computations from the 2017/18 HBS\u003c/p\u003e \u003c/p\u003e \u003cp\u003eThe extent of health expenditure also significantly differed across rural-urban localities. Rural based households had relatively lower health expenditure than their urban counterpart. This is a reflection of the difference in access to health infrastructure, resources and health information flow between rural and urban areas. Health expenditure also exhbits a curve-linear relationship with age of household head and household off-farm income (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). At first, health expenditure increases with age of household head, reaches maximum at about age 60 years while above this age health expenditure declines. Similarly, initially, health expenditure increases at a decreasing rate with an increase in household\u0026rsquo;s off-farm income level while at some point it starts declining. Health expenditure is also positively related to household size where at first as the household increases, health expenditure also increases but at decreasing rate until it reaches a maximu of around a household size of 35 members.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eWe also implemented a correlation analysisto measure the degree of association between agricultural participation, food poverty, dietary diversity and health expenditure. Results in Table\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e show a significantly weak and positive association between food poverty and agricultural participation while it was weak and negative for the agricultural participation and dietary diversity. The results further show that there is no direct link between agricultural participation and health expenditure given insignificant correlation coefficient. This study also found significantly weak and negativa correlation between food poverty and health expenditure and household\u0026rsquo;s dietary diversity. However, household\u0026rsquo;s dietary diversity score exhibited a positive correlation with health expenditure implying that improved dietary diversity is directly correlated with an increase in household\u0026rsquo;s health expenditure.\u003c/p\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\u003eCorrelation analysis\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\u003eVariables\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003e(1)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003e(2)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003e(3)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003e(4)\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e(1) Agricultural participation\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1.000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e(2) Food poverty\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.033**\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1.000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e(3) Health expenditure\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.010\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e-0.032**\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e1.000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e(4) Household dietary diversity score\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e-0.056**\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e-0.165**\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.021**\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e1.000\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"5\" nameend=\"c5\" namest=\"c1\"\u003e \u003cp\u003e** p\u0026thinsp;\u0026lt;\u0026thinsp;0.05\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=\"Sec7\" class=\"Section2\"\u003e \u003ch2\u003e3.2 Empirical results and Discussion\u003c/h2\u003e \u003cdiv id=\"Sec8\" class=\"Section3\"\u003e \u003ch2\u003e3.2.1 Determinants of household\u0026rsquo;s agricultural participation\u003c/h2\u003e \u003cp\u003eWe first present the effect of household\u0026rsquo;s agricultural participation on intermediate variables (food poverty and dietary diversity) where agriculture is treated as an endogenous variable in Table\u0026nbsp;\u003cspan refid=\"Tab4\" class=\"InternalRef\"\u003e4\u003c/span\u003e while access to extension services and agricultural capital were used as instrumental variables. From the Wald test of exogeneity, the results confirm that household\u0026rsquo;s agricultural participation was potentially endogenous (Wald χ\u003csup\u003e2\u003c/sup\u003e\u0026thinsp;=\u0026thinsp;571.5, p\u0026thinsp;=\u0026thinsp;0.0000) which implies that the use of instruments to instrument participation was crucial. Similarly, results from the falsification test on the selected instruments shows that the instruments are jointly significant in the agricultural participation equation but not in the food poverty and dietary diversity equations. This signifies that the instruments used met both the exclusion and relevance restriction criteria.\u003c/p\u003e \u003cp\u003eResults from Table\u0026nbsp;\u003cspan refid=\"Tab4\" class=\"InternalRef\"\u003e4\u003c/span\u003e show that, household\u0026rsquo;s agricultural participation is significantly influenced by the age of household head, household size, off-farm income, access to mobile extension services and agricultural capital access. The coefficient of age of household head was positive and significant implying that the likelihood of a household to participate in agriculture increases with an increase in the age of household head.\u003c/p\u003e \u003cp\u003eThe coefficient of household size was positive and significant suggesting that an increase in household size is also a motivating factor for household\u0026rsquo;s agricultural participation in the study area. This is explained by the reason that a large household size may imply high supply of family labour to support agricultral activities if and only if the large proportion of family members are in their working age as opposed to many dependents.\u003c/p\u003e \u003cp\u003eOff-farm income earned by a household from off-farm activities significantly and negatively affects the likelihood of household\u0026rsquo;s agricultural participation. Off-farm income demotevates agricultural participation since it acts as a shock absorber during agricultural crop failure, increases purchasing power of a household and makes a household to have choice on investment decisions based on amount of return and risky nature of an activity.\u003c/p\u003e \u003cp\u003eEducation coefficient was positive and significant implying that households headed by heads with no formal education had higher likelihood of participation in agriculture than those headed by heads with formal education. This suggests that, education creates more employment opportunities, helps a household in accessing information on low and high return activities and thus are more able to make informed decisions, grab more opportunities unlike those with no formal education which are faced with limited income generating activities and information space.\u003c/p\u003e \u003cp\u003eThe coefficient of the instrumental variables were also significant though with unexpected signs. Access to extension services provides agronomic and technological information that facilitate adoption of newly improved technologies that further enahnce participation in agriculture. Agricultural capital is also an important input in agricultural production through increasing the efficiency of other factors including labour and land. The negative signs might in these instrumental variables could be explained by the low extent nature of agricultural extension and agricultural capital access in Tanzania.\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\u003eEffect of agricultural participation on food poverty and dietary diversity\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\" colspan=\"2\" nameend=\"c3\" namest=\"c2\"\u003e \u003cp\u003eIV-Probit\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colspan=\"2\" nameend=\"c5\" namest=\"c4\"\u003e \u003cp\u003eIV-Poisson\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eVariable\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eParticipation\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eFood poverty\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eParticipation\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eHDDS\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eParticipation (1\u0026thinsp;=\u0026thinsp;yes)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e2.251***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e-4.167*\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e(0.0170)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e(1.773)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAge of household head\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.000803**\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e-0.0019**\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.000766*\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.00204\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e(0.000302)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e(0.000684)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e(0.000298)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e(0.00196)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHousehold size\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.00414*\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e-0.00183\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.00317\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.0172\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e(0.00161)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e(0.00372)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e(0.00174)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e(0.00956)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLN (off-farm Income)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e-0.0756***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.165***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e-0.0761***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.284*\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e(0.00638)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e(0.0145)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e(0.00659)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e(0.139)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eEducation (1\u0026thinsp;=\u0026thinsp;no formal)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.0762***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e-0.165***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.0738***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.262\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e(0.00962)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e(0.0218)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e(0.00966)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e(0.140)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMobile exten access(1\u0026thinsp;=\u0026thinsp;yes)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e-0.0141***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e-0.0203\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e(0.00243)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e(0.0115)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAg-capital access (1\u0026thinsp;=\u0026thinsp;yes)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e-0.0128***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.0257\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e(0.00271)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e(0.0151)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eConstant\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1.645***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e-3.744***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e1.660***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e8.343**\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e(0.0892)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e(0.204)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e(0.0912)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e(2.950)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAthrho\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e-3.527***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e(0.148)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLnsigma\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e-0.811***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e(0.00744)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eWald test of exogeneity\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c3\" namest=\"c2\"\u003e \u003cp\u003eχ\u003csup\u003e2\u003c/sup\u003e\u0026thinsp;=\u0026thinsp;571.5, p\u0026thinsp;=\u0026thinsp;0.0000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eWald χ\u003csup\u003e2\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c3\" namest=\"c2\"\u003e \u003cp\u003e17563.3, p\u0026thinsp;\u0026gt;\u0026thinsp;chi2\u0026thinsp;=\u0026thinsp;0.0000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eN\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e9374\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e9374\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\u003eStandard errors in parentheses, * p\u0026thinsp;\u0026lt;\u0026thinsp;0.05, ** p\u0026thinsp;\u0026lt;\u0026thinsp;0.01, *** p\u0026thinsp;\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003cp\u003e \u003cstrong\u003eSource\u003c/strong\u003e \u003cp\u003eAuthor computations from the 2017/18 HBS\u003c/p\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec9\" class=\"Section3\"\u003e \u003ch2\u003e3.2.2 Effect of agricultural participation on mediating variables (Food poverty and HDDS)\u003c/h2\u003e \u003cp\u003eOn the first case, results from the instrumental variable probit in Table\u0026nbsp;\u003cspan refid=\"Tab4\" class=\"InternalRef\"\u003e4\u003c/span\u003e show that households participationg in agriculture had higher likelihood of being food poor than non-participating households. This can be explained by the reason that due to scarcity of resources including time allocated in the agricultural sector which limits households participation in off-farm activities, depending on own farm production with low productivity and farm gate price and under developed markets leads households participating entirely in the agriculture sector as their main source of income being food poor.\u003c/p\u003e \u003cp\u003eAmong the control variables, age of the household head, off-farm income and education level of household head had significant effect on household\u0026rsquo;s poverty incidence. Food poverty decreases with an increase in age of household head. This could be explained by the reason that aged household heads have accumulated life time wealth, have established long run channels of income generating activities, market access information as well as access to remittances and social protection that helps to reduce food poverty incidence. Unexpectedly, the coefficients for off-farm income and education level of household head showed negative signs signifying that an increase in household\u0026rsquo;s off-farm income led to an increase in food poverty incidence while having non-formal education translates into lower food poverty.\u003c/p\u003e \u003cp\u003eOn the second case, results further show that agricultural participation significantly and negatively affect household\u0026rsquo;s dietary diversity. The result suggests that households participating in agriculture were less food diverse than non-participating households. This could be explained by the reason that, most of the agricultural participating households reside in rural areas with underdeveloped markets and information infrastructure leading to information asymmetry, over-dependency on a single source of income (agriculture) which is prone to climate variability, low and unstable prices which translates into low income and purchasing power particularly on nutrient dense foods (meat, milk and variety of fruits) they do not produce in their own farms [\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e, \u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e, \u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eOff-farm income was also found to positively and significantly affecting household\u0026rsquo;s dietary diversity. This suggests that dietary diversity increases with an increase in household\u0026rsquo;s off-farm income. Off farm income increases the purchasing power of a household by supplementing the own produced food with market purchased nutrient-dense food items and thus leading to an increase the number of food groups consumed. However in this stance, off farm income would effectively work out to increase the level of household\u0026rsquo;s dietary diversity given that there is market access facilitated by developed infrastructure and availability of strong institutions.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec10\" class=\"Section3\"\u003e \u003ch2\u003e3.2.3 Effects of food poverty and dietary diversity on household\u0026rsquo;s health expenditure\u003c/h2\u003e \u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab5\" class=\"InternalRef\"\u003e5\u003c/span\u003e presents the effect of food poverty and dietary diversity on household\u0026rsquo;s health expenditure as mediating factors between agricultural participation and health expenditure. Both, the F and Wald chi-square tests were significant (p\u0026thinsp;\u0026lt;\u0026thinsp;0.01) signifying that both models predicted well the data. Results from both OLS and IV-GMM show that, participating in agriculture does not directly affect household\u0026rsquo;s health expenditure but its effect can be indirectly observed through food poverty incidence and dietary diversity as mediating factors. The coefficient for food poverty and dietary diversity were significant and with similar signs for both the OLS and the IV-GMM although there was a slight difference in magnitude.\u003c/p\u003e \u003cp\u003eFood poverty coefficient had significant and negative effect on household\u0026rsquo;s health expenditure suggesting that food poor households had relatively lower health expenditure than food secure households. Health expenditure was 54.1\u0026ndash;55.7% lower among food poor households than food secure households. This is attributed by the reason that being food poor, there is a trade-off between income directed to food purchase and medication/health care investment. Given that, over 70% of household\u0026rsquo;s income in Tanzania as in other developing countries is alloacted in food purchase, being food poor forces a household to allocate most of its income earned to purchase more food rather than investing in health and thus reduce health expenditure. Food poverty strongly predict poor mental, physical health and chronic diseases particularly to adults and children under five years [\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e].\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\u003eEffect of food poverty and dietary diversity on health expenditure\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\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003eVariable\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colspan=\"2\" nameend=\"c3\" namest=\"c2\"\u003e \u003cp\u003eLN (Health expenditure)\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eOLS\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eIV-GMM\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eParticipation (1\u0026thinsp;=\u0026thinsp;yes, 0\u0026thinsp;=\u0026thinsp;no)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e-0.0209\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e-0.748\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e(0.0453)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e(2.711)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAge of household head\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.00476***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.00478*\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e(0.00133)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e(0.00212)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHousehold size\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.0592***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.0589***\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e(0.00679)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e(0.00880)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLN (off-farm Income)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.127***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.0748\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e(0.0279)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e(0.210)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eEducation lev (1\u0026thinsp;=\u0026thinsp;no formal, 0\u0026thinsp;=\u0026thinsp;formal)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e-0.0835*\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e-0.0472\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e(0.0417)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e(0.204)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eFood poverty (1\u0026thinsp;=\u0026thinsp;yes, 0\u0026thinsp;=\u0026thinsp;no)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e-0.814***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e-0.779***\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e(0.0976)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e(0.122)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHousehold dietary diversity score (HDDS)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.116***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.113***\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e(0.0126)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e(0.0206)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eConstant\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e5.324***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e6.543\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e(0.396)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e(4.649)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eWald χ\u003csup\u003e2\u003c/sup\u003e (7)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eF\u0026thinsp;=\u0026thinsp;41.95\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e278.73\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eProb\u0026thinsp;\u0026gt;\u0026thinsp;χ\u003csup\u003e2\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eProb\u0026thinsp;\u0026gt;\u0026thinsp;F\u0026thinsp;=\u0026thinsp;0.0000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.0000\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eR-squared\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.0497\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.0056\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\u003eStandard errors in parentheses, * p\u0026thinsp;\u0026lt;\u0026thinsp;0.05, ** p\u0026thinsp;\u0026lt;\u0026thinsp;0.01, *** p\u0026thinsp;\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003cp\u003e \u003cstrong\u003eSource\u003c/strong\u003e \u003cp\u003eAuthor computations from the 2017/18 HBS\u003c/p\u003e \u003c/p\u003e \u003cp\u003eHowever, our result deviates from those of previous studies including those of [\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e, \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e, \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e, \u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e]. These studies altogether concluded that food poor households had reatively higher health care expenditure and utilization than food secure households. The reasons put forward by these studies was that food poor households are prone to outbreak of malnutrition related diseases to children under five years, women of reproductive age and cadio-vascular diseases particularly to elders leading to an increase in visitation to health care facilities and services which ultimately increases household\u0026rsquo;s health expenditure. The difference emanates in the complex conceptualisation of the health expenditure and health status of a person. Higher health expenditure due to an increase in investment in health care services could be a proxy for improved health as used in our current study and supported by previous studies [\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e, \u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e]. In contrary, higher health expenditure might also reflect presence of diseases, injuries [\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eHousehold\u0026rsquo;s dietary diversity score had positive and significant effect on household\u0026rsquo;s health expenditure. This suggests that, households consuming diverse food groups had higher health expenditure than less food diverse households. As the household consumes one additional food group, health expenditure increases by about 11.3\u0026ndash;11.6% ceteris paribus. Higher dietary diversity translates into improved nutrition and more economic access to food varieties which in turn protects the household against outbreak of diseases due to malnutrition and ensures food security. With absence or reduced malnutrition and being food secure, household income that could be allocated in purchasing nutrient dense foods is instead allocated in health care improvement and thus leading to an increase in household\u0026rsquo;s health expenditure.\u003c/p\u003e \u003cp\u003eApart from food poverty and dietary diversity as variables of interest, other control variables included in the model including age and education of household head, off-farm income and household size had also significant effect on household\u0026rsquo;s health expenditure. As the age of household head increase by one year, household\u0026rsquo;s health expenditure increases marginally by about 0.5%. Aged people are prone to age-related diseases including cardiovascular diseases caused by a decline in health capital over years which in turn needs continuous health care investment resulting in an increase in health expenditure.\u003c/p\u003e \u003cp\u003eThe coefficient of household size was positive and significant implying that household\u0026rsquo;s health expenditure level increases as a household size increases. As one member increases in the household, household\u0026rsquo;s health expenditure increases by about 6% holding other factors affecting health expenditure constant. This suggests that, an increase in members of a household translates in more demand for health care services and thus leading to an increase in household\u0026rsquo;s health expenditure level.\u003c/p\u003e \u003cp\u003eAs expected, the coefficient of household\u0026rsquo;s off farm income was positive and significant implying that household\u0026rsquo;s health expenditure level increases as off-farm income increases. With an increase in off farm income by one Tanzania shilling, household\u0026rsquo;s health expenditure increases by 12.7%. Income facilitates access and utilization of health care services and acts as a motivator to increase more investment in improving health of household members such as regular visit to health facilities for body check-up particularly to highly ranked and expensive health facilities [\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e]. An increase in income may also translate in more investment in health insurance services which altogether leads to an increase in household\u0026rsquo;s health expenditure.\u003c/p\u003e \u003cp\u003eSimilarly, having no formal education translates in relatively lower health expenditure than having formal education. Households headed by heads with no formal education had 8.01% lower health expenditure than households headed by heads with formal education. The plausible reason for this results is that, with no formal education, households might have limited access to health information and services, low investment in health due to having no health education, lower access to more income generating activities and thus low income that is invested in health care improvement.\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e"},{"header":"4. Conclusion and Policy implications","content":"\u003cp\u003eNutrition sensitive and food pro-poor agriculture have received attention by policy makers and nutritionists particularly in developing countries due due its linkage with person\u0026rsquo;s health. However, the relationship between agricultural production and health expenditure has received little attention by scholars while the existing literature presents mixed results. This study aimed at examining the effect of household\u0026rsquo;s agricultural participation on health expenditure mediated by food poverty and dietary diversity. Using the 2017/18 nationally representative Tanzania household budget survey and an instrumental variable Probit, IV-Poisson and IV-GMM, this study found the following results. Agricultural participating househlds were more likely to be food poor and less food diverse than non \u0026ndash;participating households. There was no difference in household\u0026rsquo;s health ependiture between participating and non-participating households in agricultural production. The difference was observed by food poverty incidence and dietary diversity level which were used as mediating variables. Food poor households had about 54.1\u0026ndash;55.7% lower health expenditure than non-food poor households. Similarly, an increase in household\u0026rsquo;s dietary diversity by one more food group leads to an increase in household\u0026rsquo;s health expenditure by 11.3\u0026ndash;11.6%.\u003c/p\u003e \u003cp\u003eOur study has the following policy implications. First, agricultural policies in Tanzania and other developing countries should go beyond fostering increased agricushould participation. The focus should be on promoting and investing in poverty reducing and nutrition-sensitive agricultural production which have direct effect on food security and dietary diversity and thereby affecting household\u0026rsquo;s health status and health expenditure. Second, to more improve investment in health, family planning to take into account the household size, promotion of off-farm income generating activities as well as investment in education particularly health education is inevitable.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eEthics approval and consent to participate\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe utilized secondary data that is nationally representative, obtained from the Tanzania National Bureau of Statistics website, where it is openly accessible for research purposes. Since the data was acquired from the Tanzania National Bureau of Statistics specifically for research, ethical approval was not necessary for this study.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAvailability of data and materials\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eData used in the analysis may be available upon reasonable requests.\u003c/p\u003e\n\u003cp\u003e\u0026nbsp;Please contact Mr. Furaha Rashid \u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u0026nbsp; [email protected]\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting interests\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors declare that they have no confict of interest\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis study did not receive any form of financial support\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthors\u0026apos; contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eFR was responsible for designing the study and authoring the manuscript. RL carried out the data analysis and its interpretation. All authors reviewed and gave their approval to the final version of the manuscript.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eAddai KN, Ng\u0026rsquo;ombe JN, Temoso O. (2022). Food Poverty, Vulnerability, and Food Consumption Inequality among Smallholder Households in Ghana: A Gender-Based Perspective. Soc Indic Res, 1\u0026ndash;29.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAssefa T, Abide EB. Determinants of food insecurity in rural households: A case of lemo district, southern Ethiopia. Heliyon. 2023;9(1):1\u0026ndash;7.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAthukorala W, Wilson C, Robinson T. Determinants of health costs due to farmers\u0026rsquo; exposure to pesticides: an empirical analysis. J Agric Econ. 2012;63(1):158\u0026ndash;74.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBalaji SJ, Jhajhria A, Kumar S, Immanuelraj TK, Kar A. Agriculture\u0026ndash;nutrition linkages: A preliminary investigation for rural India. Outlook on Agriculture. 2017;46(4):302\u0026ndash;8.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBenfica RS, Kilic T. (2015). The Effects of smallholder agricultural involvement on household food consumption and dietary diversity: Evidence from Malawi. In \u003cem\u003eInternational Conference of Agricultural Economists, Milan Italy\u003c/em\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBerkowitz SA, Basu S, Meigs JB, Seligman HK. Food insecurity and health care expenditures in the United States, 2011\u0026ndash;2013. Health Serv Res. 2018;53(3):1600.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBernzen A, Mangnus E, Sohns F. Diversify, produce or buy? An analysis of factors contributing to household dietary diversity among shrimp and non-shrimp farmers in coastal Bangladesh. Food Secur. 2022;14(3):741\u0026ndash;61.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBui LK, Hoang H. Non-farm employment, food poverty and vulnerability in rural Vietnam. Environ Dev Sustain. 2021;23(5):7326\u0026ndash;57.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChegere MJ, Stage J. Agricultural production diversity, dietary diversity and nutritional status: Panel data evidence from Tanzania. World Dev. 2020;129:104856.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChrisendo D, Siregar H, Qaim M. Oil palm cultivation improves living standards and human capital formation in smallholder farm households. World Dev. 2022;159:106034.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDean EB, French MT, Mortensen K. Food insecurity, health care utilization, and health care expenditures. Health Serv Res. 2020;55:883\u0026ndash;93.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDillon A, McGee K, Oseni G. Agricultural production, dietary diversity and climate variability. J Dev Stud. 2015;51(8):976\u0026ndash;95.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eEcker O. Agricultural transformation and food and nutrition security in Ghana: Does farm production diversity (still) matter for household dietary diversity? Food Policy. 2018;79:271\u0026ndash;82.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eUNICEF, WHO and WFP. The state of food security and nutrition in the World 2018. Building climate resilience for food security and nutrition. Rome: Food and agriculture organisation of the United Nations; 2018.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWFP and WHO. brief to the state of food security and nutrition in the world 2021. Transforming food systems for food security, improved nutrition and affordable healthy diets for all. Rome: Food and agriculture organisation of the United Nations; 2021.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFraval S, Yameogo V, Ayantunde A, Hammond J, De Boer IJ, Oosting SJ, Van Wijk MT. Food security in rural Burkina Faso: the importance of consumption of own-farm sourced food versus purchased food. Agric Food Secur. 2020;9(1):1\u0026ndash;17.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGaillard C, Verger EO, Dury S, Dop MC, Ati E, Medina Study Group. J., and (2022). Farm production diversity and women\u0026rsquo;s dietary diversity: Evidence from central Tunisia. PLoS ONE, 17(2), e0263276.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGarcia SP, Haddix A, Barnett K. Peer reviewed: incremental health care costs associated with food insecurity and chronic conditions among older adults. Prev Chronic Dis. 2018;15:E108.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGrossman M. On the concept of health capital and the demand for health. J Polit Econ. 1972;80(2):223\u0026ndash;55.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHawkes C, Ruel MT. (2006). Overview: Understanding the links between agriculture and health. 2020 Vis briefs, (13 (1)).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJohnston MP. Secondary data analysis: A method of which the time has come. Qualitative and quantitative methods in libraries. 2017;3(3):619\u0026ndash;26.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKafle K, Paliwal N, Benfica R. (2018). \u003cem\u003eWho works in agriculture? Exploring the dynamics of youth involvement in Tanzania\u0026rsquo;s and Malawi\u0026rsquo;s Agri-food system\u003c/em\u003e. A selected paper presented at the 2018 Agricultural and Applied Economics Association Annual meeting, Washington, D.C., August 5\u0026ndash;7.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKalyabina VP, Esimbekova EN, Kopylova KV, Kratasyuk VA. Pesticides: formulants, distribution pathways and effects on human health\u0026ndash;a review. Toxicol Rep. 2021;8:1179\u0026ndash;92.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKe XU, Saksena P, Holly A. The determinants of health expenditure: a country-level panel data analysis. Geneva: World Health Organization. 2011;26:1\u0026ndash;28.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKennedy G, Ballard T, Dop MC. Guidelines for measuring household and individual dietary diversity. Nutrition and Consumer Protection Division, Food and Agriculture Organization of the United Nations; 2013.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKumar N, Harris J, Rawat R. If they grow it, will they eat and grow? Evidence from Zambia on agricultural diversity and child undernutrition. J Dev Stud. 2015;51(8):1060\u0026ndash;77.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLai W. Pesticide use and health outcomes: Evidence from agricultural water pollution in China. J Environ Econ Manag. 2017;86:93\u0026ndash;120.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLovo S, Veronesi M. Crop diversification and child health: empirical evidence from Tanzania. Ecol Econ. 2019;158:168\u0026ndash;79.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMa W, Zhou X, Renwick A. Impact of off-farm income on household energy expenditures in China: Implications for rural energy transition. Energy Policy. 2019;127:248\u0026ndash;58.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMuthini D, Nzuma J, Qaim M. Subsistence production, markets, and dietary diversity in the Kenyan small farm sector. Food Policy. 2020;97:101956.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNBS. (2019). Tanzania mainland household budget survey 2017-18. Key Indicators report. Dodoma, Tanzania. Retrieved on September 1, 2022 from World Wide Web at \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.nbs.go.tz/index.php/en/census-surveys/poverty-indicators\u003c/span\u003e\u003cspan address=\"https://www.nbs.go.tz/index.php/en/census-surveys/poverty-indicators\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eOchieng J, Afari-Sefa V, Lukumay PJ, Dubois T. (2017). Determinants of dietary diversity and the potential role of men in improving household nutrition in Tanzania. PLoS ONE, 12(12), e0189022.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eO'Connor N, Farag K, Baines R. What is food poverty? A conceptual framework. Br Food J. 2016;118(2):429\u0026ndash;49.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eOgutu SO, G\u0026ouml;decke T, Qaim M. Agricultural commercialisation and nutrition in smallholder farm households. J Agric Econ. 2020;71(2):534\u0026ndash;55.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eOlabisi M, Obekpa HO, Liverpool-Tasie LSO. Is growing your own food necessary for dietary diversity? Evidence from Nigeria. Food Policy. 2021;104:102144.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePassarelli S, Mekonnen D, Bryan E, Ringler C. Evaluating the pathways from small-scale irrigation to dietary diversity: evidence from Ethiopia and Tanzania. Food Secur. 2018;10(4):981\u0026ndash;97.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePeltz A, Garg A. Food insecurity and health care use. Pediatrics. 2019;144(4):e20190347.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSchneider MT, Chang AY, Chapin A, Chen CS, Crosby SW, Harle AC, Dieleman JL. (2021). Health expenditures by services and providers for 195 countries, 2000\u0026ndash;2017. BMJ Global Health, 6(7), e005799.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eShively G, Sununtnasuk C. Agricultural diversity and child stunting in Nepal. J Dev Stud. 2015;51(8):1078\u0026ndash;96.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSibhatu KT, Qaim M. Meta-analysis of the association between production diversity, diets, and nutrition in smallholder farm households. Food Policy. 2018;77:1\u0026ndash;18.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSignorelli S, Haile B, Kotu BH. (2017). Exploring the agriculture-nutrition linkage in northern Ghana. IFPRI discussion paper 1697. Washington, D.C: Intenational food policy research institute.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eURT (United Republic of Tanzania). (2021). Integrated Labour force survey 2020/21. Retrieved on September 15, 2022 from World Wide Web \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.nbs.go.tz/index.php/en/census-surveys/labour-statistics/688-integrated-labour-force-survey-2020-21\u003c/span\u003e\u003cspan address=\"https://www.nbs.go.tz/index.php/en/census-surveys/labour-statistics/688-integrated-labour-force-survey-2020-21\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWagner N, Tasciotti L. Urban agriculture, dietary diversity and child health in a sample of Tanzanian town folk. Int J Shape Model. 2018;39(2):234\u0026ndash;51.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWHO (World Health Organization). (2021). Global expenditure on health: public spending on the rise? Retrieved on September 16, 2022 from World Wide Web at \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.who.int/publications/i/item/9789240041219\u003c/span\u003e\u003cspan address=\"https://www.who.int/publications/i/item/9789240041219\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWondimu M, Gezahegn A, Gedefaw Y. (2022). Health and Environmental Hazards Associated with Agricultural Pesticides in Ethiopia. Asian Res J Agric, 14\u0026ndash;26.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWorld Bank. (2022). Current health expenditure as a percentage of GDP. Retrieved on September 5, 2022 from World Wide Web at \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://data.worldbank.org/indicator/SH.XPD.CHEX.GD.ZS\u003c/span\u003e\u003cspan address=\"https://data.worldbank.org/indicator/SH.XPD.CHEX.GD.ZS\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZanello G, Shankar B, Poole N. Buy or make? Agricultural production diversity, markets and dietary diversity in Afghanistan. Food Policy. 2019;87:101731.\u003c/span\u003e\u003c/li\u003e\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":"Agricultural participation, health expenditure, food poverty, dietary diversity","lastPublishedDoi":"10.21203/rs.3.rs-3728355/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-3728355/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eAgriculture and health are linked through natural environment, food production, nutrition and production of medicinal plants. The existing literature on the effect of food poverty and dietary diversity as mediating factors for agriculture on household\u0026rsquo;s health expenditure presents mixed results and are country specific. This study aimed at examining the link between agriculture and household\u0026rsquo;s health expenditure mediated by food poverty and dietary diversity using the nationally representative Tanzania household budget survey data 2017/18. We employed an instrumental variable generalized method of moment (IV-GMM) as a method of analysis. Results show that participation in agriculture solely has no effect on health expenditure but the effect stems from food poverty incidence and dietary diversity level as mediating factors. Food poor households had 54.1\u0026ndash;55.7% lower health expenditure than food secure households. An additional food group consumed by a household leads to 11.3\u0026ndash;11.6% increase in household health expenditure. Thus, policies aiming at improving health should go beyond merely fostering agricultural participation and instead place more emphasis on pro-poor and nutrition sensitive agriculture.\u003c/p\u003e","manuscriptTitle":"Grow and eat healthy: Mediating effect of food poverty and dietary diversity with evidence from Tanzania","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-01-08 13:46:31","doi":"10.21203/rs.3.rs-3728355/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":"db7bec59-6e07-431a-bb17-d9a0c773deaf","owner":[],"postedDate":"January 8th, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2024-12-12T16:23:41+00:00","versionOfRecord":[],"versionCreatedAt":"2024-01-08 13:46:31","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-3728355","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-3728355","identity":"rs-3728355","version":["v1"]},"buildId":"qtupq5eGEP_6zYnWcrvyt","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}