Forecasting Long-Term Drinking Status for Individuals Treated for Alcohol Dependence: A Bayesian Approach

Forecasting Long-Term Drinking Status for Individuals Treated for Alcohol Dependence: A Bayesian Approach | 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 Forecasting Long-Term Drinking Status for Individuals Treated for Alcohol Dependence: A Bayesian Approach Durai Murukan Gunasekaran, Binu VS, Arun Kandasamy, Lakshmanan Sethuraman, and 1 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7489427/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 8 You are reading this latest preprint version Abstract Background Alcohol dependence presents a significant challenge, with treatment efforts often falling short in preventing relapse due to the nature of the condition. Ensuring sustained long-term recovery among successfully treated individuals is of paramount clinical importance. Therefore, this study aims to predict the long-term (after 12 months of successful treatment) alcohol consumption status of individuals treated for alcohol dependence, utilizing baseline characteristics and initial treatment response information. Methods A cohort study design was employed to assess both short and long-term alcohol consumption status among male participants who had undergone successful treatment for alcohol dependence at an addiction medicine outpatient department in Southern India from July 2020 to December 2021. Initially, participants were monitored for four months to observe their short-term treatment response, followed by a telephonic follow-up to assess their long-term status. This study employs the application of the Bayesian prediction model for evaluating the long-term binary outcome. Results Among four distinct fitted predictive models, the one encompassing data on initial four-month alcohol consumption, age, marital status, illness duration, and baseline clinical global impression-severity score emerged as predictors for long-term treatment outcomes. This model showed around 70% accuracy on the test dataset, with sensitivity and specificity hovering around 70%. Conclusion The findings emphasize the importance of maintaining alcohol abstinence or controlled drinking in the fourth month after successful treatment to ensure long-term success. Additionally, the results underscore the necessity of exploring variables beyond baseline information to understand factors influencing long-term outcomes comprehensively. Alcohol dependence Controlled drinking environment Bayesian India Figures Figure 1 Introduction Alcohol dependence stands as the most common form of Alcohol Use Disorder (AUD), characterized by a dysregulated pattern of alcohol consumption resulting from repeated or sustained use of the substance. As outlined in the International Classification of Diseases (ICD) -10 report, key features of alcohol dependence encompass an intense internal compulsion to consume alcohol, a prioritization of alcohol use above all other activities, and a consistent pattern of use despite encountering harm or negative repercussions ( 1 ). The World Health Organization's report underscores the global burden of alcohol dependence, indicating a prevalence of 2.6% among individuals aged 15 years and older in 2016. Notably, this prevalence exhibits gender disparities, with adult males exhibiting a markedly higher proportion of alcohol dependence (4.5%) compared to adult females (0.8%). Similarly, within the Southeast Asia region, the prevalence of alcohol dependence was reported at 2.9%, with rates of 0.3% among adult females and 5.3% among adult males. Addressing alcohol dependence necessitates a multifaceted treatment approach capable of addressing the diverse array of factors that may precipitate substance use ( 2 ). Research indicates that a significant portion of participants in alcohol treatment prefer low-risk drinking outcomes, such as controlled drinking, which is comparable to abstinence in terms of long-term functioning and can be considered successful ( 3 , 4 ). Consequently, throughout the treatment phase, individuals may transition from a state of alcohol dependence to either complete abstinence or low-risk drinking outcomes. However, this transition alone does not guarantee the enduring efficacy of treatment over a long period. Given the relapsing nature of alcohol dependence, studies have indicated a relapse rate of approximately 50% within the first year following successful treatment for alcohol dependence ( 5 ). There has been a paucity of studies focused on observing the dynamic transitions from short-term treatment outcomes and their retention rates into the long term ( 6 , 7 ). Moreover, some studies have endeavoured to identify factors influencing the retention of treatment effects over the long term, examining their interplay with short-term treatment outcomes and other baseline socio-demographic as well as clinical variables ( 8 – 12 ). Identifying high-risk individuals prone to relapse into alcohol dependence following treatment allows clinicians to provide targeted care, ensuring the enduring effectiveness of treatment and fostering long-term abstinence. Consequently, predictive models capable of identifying such high-risk populations in their initial days of treatment are instrumental in developing tailored intervention strategies to mitigate long-term relapse risk among successfully treated individuals. Therefore, the current study seeks to develop a predictive model to ascertain the long-term alcohol consumption status of male individuals with alcohol dependence who sought treatment at a tertiary care hospital in Southern India. This model incorporates a comprehensive range of factors, including socio-demographic, psychological, and environmental variables, alongside the initial treatment response pattern observed during the first four months of the treatment phase. Materials and Methods Data This is a cohort study comprising individuals seeking treatment for alcohol dependence from July 2020 to December 2021 at the Centre for Addiction Medicine (CAM) outpatient department (OPD) at the National Institute of Mental Health and Neurosciences (NIMHANS), located in Bengaluru, India. These individuals visited the CAM OPD on multiple occasions and received treatments to facilitate their transition from alcohol dependence to alcohol abstinence or controlled drinking status. The study comprised male individuals diagnosed with alcohol dependence based on ICD-10 criteria during the treatment phase. Only those who were found to be either in complete abstinence or in a controlled drinking environment at any visit within the first four months of receiving treatment from the study centre were considered to have undergone successful treatment and were eligible for inclusion. These participants were subsequently monitored for four months to understand the short-term treatment outcome, with an additional status check at the 12th month to understand the long-term treatment outcome. The primary endpoint is the status of alcohol dependence at the end of the twelve months, following complete abstinence or a controlled drinking environment achieved through treatment at NIMHANS. Exclusion criteria encompassed severe psychiatric disorders unrelated to substance dependence and significant health abnormalities. The data collected at baseline were socio-demographic and clinical information of each subject, such as their completed age in years (≤ 25, 26–40, 41–50, & >50), educational status (primary and below, middle school, high school, & graduate and above), employment status (unemployed & employed), economic status (above poverty line (APL) if annual income ≥ 20,000 INR & below poverty line (BPL) if annual income < 20,000 INR), marital status (never married & ever married), living arrangement (nuclear family/alone & joint family), first-degree family history of alcohol use (yes & no), current use of any tobacco (yes & no), age at first use of alcohol in years (25), duration of illness of alcohol dependence in years, and baseline clinical global impression-severity (CGI-s) score. The short-term treatment outcome, alcohol dependence status (abstinence/controlled drinking & alcohol dependence) at each of the first four months, were also monitored. Similarly, the same information was collected at the end of the 12th month to assess the long-term treatment outcome. Statistical Analysis A Bayesian regression model for binary response variable, long-term alcohol dependence status, was employed to forecast the likelihood of individuals transitioning into an alcohol-dependent state at the end of the 12th month following successful treatment. The dataset underwent a 90/10 split, with the training subset (90%) utilized for model development and the validation subset (10%) for model assessment. Given the limited prior knowledge regarding model parameters, a weakly informative prior proposed by Gelman was adopted ( 13 ). In this framework, the intercept was presumed to adhere to a Cauchy distribution with a mean of 0 and a standard deviation of 10, i.e., \(\:{\beta\:}_{0}\sim\:Cauchy\left(\text{0,10}\right)\) , while other parameters assumed to follow a Cauchy distribution with a mean of 0 and a standard deviation of 2.5, i.e., \(\:{\beta\:}_{\:}\sim\:Cauchy\left(\text{0,2.5}\right)\) . The variable selection procedure for fitting the final predictive model was based on the results obtained from the unadjusted individual models, focusing on one variable at a time. As a result, four different predictive models were fitted by considering various thresholds. Variables with credible intervals (CrI) of 50%, 60%, 70%, 80% and 90% that exclude the null effect in their unadjusted CrI were considered for inclusion in the final predictive model. Given that the same set of variables satisfies both 80% and 90% thresholds this study does not delve into the model based on the 90% threshold. Moreover, the Hamiltonian Monte Carlo (HMC) algorithm was employed with four Markov Chain Monte Carlo (MCMC) chains, each comprising 2000 iterations for parameter estimation to obtain the posterior distribution from both the unadjusted and adjusted models. The initial 1000 iterations were designated as a burn-in period, and the resulting samples were excluded from subsequent analysis. Even though different algorithms are available, this study employed HMC because it can theoretically generate samples from a wide range of parameter space with a high level of acceptance probability ( 14 ). The samples from the posterior obtained through HMC were summarized for the adjusted models using the posterior median, along with their corresponding 90% CrI. The list-wise deletion method was carried out to address the missingness in the independent variables while fitting and validating the models. Further, model adequacy was assessed by \(\:\widehat{R}\) statistics and effective sample size (ESS). Initially, 1000 samples were simulated from the posterior predictive distribution of all participants in the test set across the four models. These simulated samples provide the probability of being in an alcohol-dependent state at the end of the 12th month following successful treatment. Following this process, the next step involved obtaining a precise point estimate crucial for classification purposes. This was achieved by determining optimal cut-points for each model. The optimization process aimed at maximizing the combined sensitivity and specificity. Moreover, the predictive performance of these models underwent thorough evaluation and comparison. Key metrics such as the accuracy (%), area under the receiver operating characteristic curve (AUROC), sensitivity, and specificity were utilized for this purpose. All analyses were carried out using the packages “rstanarm”, “forestploter” and “grid” available under R version 4.2.3. Results During the specified timeframe, 3276 patients sought treatment at the CAM OPD. Among them, the majority, comprising 2543 individuals (77.63%), were seeking treatment for alcohol dependence, while the rest presented with various other substance abuse behaviours. From this pool of 2543 patients, only 608 (23.9%) individuals achieved either complete abstinence or a controlled drinking environment within the first four months of treatment. These 608 individuals were selected for the current study based on predetermined inclusion and exclusion criteria. The demographic analysis of the study participants unveiled a substantial 67.1%, which surpassed middle school education. Furthermore, 68.26% of the participants belonged to BPL socio-economic category. Additionally, 19.57% were unemployed, and 20.88% had never married. Moreover, a considerable segment, constituting 40.63% of the participants, disclosed a history of tobacco usage, while 19.41% reported a familial background of alcohol consumption among their first-degree family circle. Furthermore, the study found that the majority of participants, 57.89%, initiated alcohol consumption between the ages of 18 and 25 years. Approximately 75% of participants reported a duration of alcohol dependence exceeding four years. In addition, 50% of participants were assessed to be markedly ill or worse according to the CGI-S scale at the baseline. Ninety percent of the participants were randomly selected to comprise the training dataset (n = 544), while the remaining ten percent constituted the validation dataset (n = 64). Table 1 illustrates basic descriptive statistics and the effects, in terms of varying posterior estimate CrI, for the considered variables obtained from the unadjusted model using the training dataset. The results revealed that the alcohol consumption status over the past four months of attaining the abstinence/controlled drinking state played a significant role in long-term alcohol dependence status individually. Specifically, individuals identified as alcohol-dependent during the initial four months were found to be at a higher risk of continuing to exhibit alcohol dependence by the end of the 12th month. This suggests a persistent pattern of alcohol dependency over time. Additionally, variable such as age (in years) was found to be influential, as they consistently demonstrated non-null effects across various proposed thresholds. Specifically, younger age was associated with a decreased likelihood of long-term alcohol dependence, with individuals under 25 demonstrating a reduced risk across various thresholds. This underscores their importance in predicting outcomes. Furthermore, variables such as educational status, living arrangements, and age at first alcohol use were also identified as influential factors, particularly under lenient thresholds such as 50%. However, their importance diminished when stricter thresholds, like 80%, were applied. Figure 1 depicts the outcomes, showcasing 90% CrI on the posterior estimates derived from adjusted Bayesian predictive models, as per the thresholds previously mentioned. Among the four models were fitted based on the different thresholds, the alcohol dependence status during each of the first three months was not found to be significant. However, it was found to be significant in the fourth month, a finding consistent across all four models. This suggests that the consumption status in the fourth month holds greater importance in predicting long-term outcomes for individuals than in the initial three months post-treatment. Whereas, the other variables considered haven't shown any significant impact on the outcome, the recency of data on alcohol dependence status appears to play a pivotal role in forecasting an individual's long-term status over other socio-demographic and clinical variables. Further, as recommended, it was observed that model adequacy statistic \(\:\widehat{R}\) 1000 in all the fitted unadjusted and adjusted models ( 15 ) ensures convergence and less autocorrelation between the obtained samples. Table 1. Results of unadjusted Bayesian predictive model in terms of differing Credible intervals Variables Event Unadjusted Model - Posterior Estimate- ln(Odds) No (n = 317) Yes (n = 227) (P 25 ,P 75 ) (P 20 ,P 80 ) (P 15 ,P 85 ) (P 10 ,P 90 ) Status at 1st month Abstinence/Controlled drinking 255 163 REF Alcohol dependence 62 64 (0.338,0.611) (0.301,0.646) (0.264,0.683) (0.220,0.734) Status at 2nd month Abstinence/Controlled drinking 259 161 REF Alcohol dependence 58 66 (0.458,0.744) (0.419,0.780) (0.38,0.819) (0.332,0.868) Status at 3rd month Abstinence/Controlled drinking 261 162 REF Alcohol dependence 56 66 (0.471,0.758) (0.433,0.789) (0.396,0.825) (0.344,0.873) Status at 4th month Abstinence/Controlled drinking 274 154 REF Alcohol dependence 43 73 (0.947,1.248) (0.911,1.286) (0.866,1.328) (0.824,1.379) Age (in years) > 50 36 31 REF 41–50 94 52 (-0.619,-0.231) (-0.665,-0.183) (-0.726,-0.12) (-0.798,-0.052) 26–40 158 133 (-0.182-0.168) (-0.225,0.208) (-0.274,0.262) (-0.343,0.328) <=25 29 11 (-1.07,-0.522) (-1.142,-0.461) (-1.224,-0.389) (-1.339,-0.299) Educational status Primary and below 59 48 REF Middle School 40 33 (-0.184,0.227) (-0.238,0.279) (-0.303,0.336) (-0.371,0.404) Higher and Higher secondary 159 104 (-0.359,-0.057) (-0.401,-0.023) (-0.447,0.022) (-0.505,0.079) Graduate and above 59 42 (-0.314,0.061) (-0.359,0.112) (-0.411,0.158) (-0.483,0.226) Employment status Unemployed 61 45 REF Employed 256 182 (-0.184,0.117) (-0.223,0.153) (-0.269,0.194) (-0.323,0.243) Economic status APL 103 76 REF BPL 214 151 (-0.171,0.075) (-0.2,0.108) (-0.234,0.146) (-0.277,0.192) Marital status Never married 73 42 REF Ever married 244 185 (0.130,0.426) (0.095,0.458) (0.054,0.502) (-0.0003,0.558) Living arrangement Nuclear family/Alone 230 158 REF Joint Family 87 69 (0.010,0.267) (-0.020,0.3) (-0.058,0.337) (-0.113,0.387) Family History of Alcohol use No 255 187 REF Yes 62 40 (-0.288,0.018) (-0.323,0.054) (-0.368,0.102) (-0.416,0.154) Current tobacco use No 181 133 REF Yes 136 94 (-0.176,0.052) (-0.208,0.083) (-0.246,0.118) (-0.294,0.157) Age at first alcohol use (in years) > 25 63 38 REF 18–25 179 134 (0.059,0.355) (0.023,0.394) (-0.027,0.439) (-0.084,0.497) < 18 75 55 (0.0002,0.363) (-0.043,0.407) (-0.092,0.46) (-0.166,0.522) Duration of illness (in years)* 7(15 − 3) 8(16 − 4) (0.004,0.018) (0.002,0.02) (0.0003,0.022) (-0.002,0.025) Baseline CGI-s score* 4(4 − 2) 4(4 − 3) (0.076,0.162) (0.065,0.172) (0.051,0.184) (0.036,0.2) * Indicates continuous variables for which Median(Q3-Q1) has been reported; (P 25 , P 75 )-50% CrI; (P 20 , P 80 )- 60% CrI; (P 15 , P 85 )- 70% CrI; (P 10 , P 90 )- 80% CrI. Table 2 Accuracy metrics across the fitted models Model Optimal Cut-point Accuracy (%) AUROC Sensitivity Specificity Model 1 0.351 0.683 0.618 0.824 0.500 Model 2 0.355 0.683 0.619 0.824 0.500 Model 3 0.400 0.685 0.630 0.706 0.692 Model 4 0.409 0.653 0.605 0.657 0.643 Table 2 presents a comprehensive view of the optimal cut-points obtained and the resulting predictive power across all four models when applied to the test dataset. The results showed that Model 3 outperformed its counterparts based on the accuracy and AUROC criteria. Notably, the Model 3 encompassed various covariates, including All month-wise alcohol consumption status (first month to fourth month), age, marital status, duration of illness, and baseline CGI-s score. Discussion This study investigated the impact of fundamental socio-demographic factors and short-term treatment outcomes on the long-term treatment effectiveness for individuals with alcohol dependence, using the Bayesian predictive models. The age of participants has emerged as a factor in predicting their long-term outcomes in unadjusted analysis. Previous studies have consistently identified older age as a key contributor to sustained alcohol abstinence over an extended period ( 6 , 10 ). However, intriguingly, our initial unadjusted results indicated that younger age was associated with a lower likelihood of sustained alcohol dependence. Nevertheless, this association lost its significance after adjusting for other variables in the final model. This contradictory finding opens up avenues for further investigation and exploration. The presence of a recovery-oriented social support network has been acknowledged as a significant factor contributing to sustained alcohol abstinence over an extended period ( 6 ). Although this specific aspect was not directly observed in our study, marital status, often considered a proxy for social support, was found to exert a notable influence on long-term treatment outcomes. Similar to age, the significance of marital status diminished in the final fitted models. Another variable examined in this study is living arrangement, which also aimed to elucidate the impact of social support on long-term treatment outcomes. However, it was not found to be an important factor in this regard. Further, though previous research has indicated that individuals with self-sustaining income demonstrate greater treatment retention ( 10 ), the current study findings suggest that economic status does not play a significant role in treatment retention over the long-term period. This disparity highlights the need for further investigation in future studies to better understand the complex relationship between economic factors and treatment outcomes in alcohol dependence. Similarly, lower baseline alcohol consumption levels were associated with a greater likelihood of maintaining long-term abstinence ( 7 , 12 ). Given that baseline alcohol consumption serves as an indicator of the severity of alcohol consumption, this study incorporated information on duration of illness in terms of alcohol dependence and the baseline CGI-s score. Both of these variables were found to have a significant impact on achieving long-term alcohol dependence in the unadjusted model. There has been a notable lack of research focused on understanding the impact of age at first alcohol use on long-term treatment outcomes. This study endeavoured to explore this aspect, and initial unadjusted results hinted at its significance, albeit this importance diminished in the adjusted model. Nevertheless, further investigation is warranted to better understand this effect and its implications for long-term treatment efficacy. Consistent with the findings of this investigation, prior research studies have revealed that alcohol abstinence at the short-term sixth-month milestone serves as a significant predictor for sustaining abstinence throughout a one-year and five-year follow-up period ( 6 , 16 ). However, there has been a notable dearth of studies examining the impact of short-term treatment outcomes on long-term results. This lack of research has hindered our ability to compare our findings and those of previous studies. There is a discrepancy in the definitions of “long-term” and “short-term” across previous studies. However, amidst this variation, a discernible pattern has emerged, indicating that durations equal to or less than six months are often classified as short-term, while time frames exceedingly approximately 52 weeks or one year are considered long-term, aligning with our study definitions. In contrast to previous studies, this research sought to utilize the Bayesian framework for prediction, leveraging its advantages, such as incorporating prior information. Also, treating the parameter of interest as a variable rather than a constant, this approach enables us to grasp the range of potential outcomes for each individual, thereby enhancing our understanding of treatment possibilities. Even though this study has employed weakly informative priors due to the lack of evidence from previous studies, it can still provide valuable insights for future research. Specifically, this study offers a general framework for building predictive models potentially guiding subsequent studies in developing more accurate and robust models. This study also has several limitations. Several studies investigated the influence of gender on predicting long-term treatment outcomes, consistently finding that female gender is significantly associated with achieving controlled drinking ( 6 , 7 , 12 ). However, this study was limited to male individuals due to the scarcity of information on female participants, preventing us from directly comparing gender differences in treatment outcomes. Another crucial aspect left unexplored was the motivation of individuals to quit alcohol. Previous studies have underscored that commitment to alcohol abstinence is associated with retention in maintaining non-problematic drinking behaviour over the long term ( 10 ). Additionally, various studies have highlighted the pivotal role of motivation to quit alcohol and the establishment of specific drinking goals in maintaining long-term alcohol abstinence or controlled drinking. These factors have been consistently emphasized across multiple research endeavours, reaffirming their significance in sustaining positive outcomes ( 8 , 11 ). However, this study could not address the role of motivation in predicting long-term outcomes due to data unavailability. In addition, the study lacks consideration of any time-varying follow-up information beyond alcohol consumption status, such as participant’s emotional states, levels of craving, and other pertinent factors, unlike previous studies ( 17 ). This limitation presents another drawback of the study. Expanding the exploration of the aforementioned potential factors in future research could provide even richer insights into treatment effectiveness and also potentially enhance the predictive capabilities of the model. Conclusion The integration of a Bayesian predictive model within this study has emerged as a pivotal tool for anticipating the long-term alcohol consumption status of individuals successfully treated for alcohol dependence. Through comprehensive analysis of relevant socio-demographic factors, clinical indicators, and initial alcohol consumption patterns among participants, the model has identified that the alcohol dependence status at the end of the fourth month has a significant impact on the long-term treatment outcome. Even though some variables were not found to have a significant effect individually, their combined contribution has improved the predictive nature of the model, demonstrating its efficacy in forecasting long-term effects to a considerable extent. Consequently, clinicians can glean invaluable insights to proactively address and manage potential future outcomes, thereby enhancing their capacity to deliver targeted and effective support throughout individual’s journeys towards sustained recovery. Declarations Ethics approval and consent to participate All participants provided informed consent to use their hospital record data for research purposes. This study was approved by the NIMHANS Internal Institute Ethics Committee (Basic Sciences) on 18th November 2020.We declare that this study was conducted in accordance with the Declaration of Helsinki. Consent for publication Not applicable. Availability of data and materials The data supporting this study are derived from the NIMHANS CAM department database. Due to licensing restrictions, these data are not publicly available. However, they may be obtained from the authors upon reasonable request and with appropriate permission from the NIMHANS CAM department. Conflicts of interest There are no conflicts of interest. Financial Disclosure The authors declare that there was no external funding or financial support received for the research, development, or publication of this article. Author Contributions Statement Durai Murukan Gunasekaran performed the data analysis and prepared the manuscript. Binukumar Bhaskarapillai, Binu V.S., and Arun Kandasamy conceptualised the core idea of this work. Lakshmanan Sethuraman assisted in data acquisition. Binukumar Lakshmanan provided overall supervision of the study. All authors reviewed the manuscript. Acknowledgement The authors are grateful to participants for sharing their information. Special thanks to Dr. N. Sreekumaran Nair, Professor and Head, Biostatistics, JIPMER, Puducherry, India, and Dr. K. Thennarasu, Professor and Head, Biostatistics, Bengaluru, NIMHANS, India, for generous suggestions and insightful recommendations. References Edition S. International statistical classification of diseases and related health problems 10th revision Volume 2 Instruction manual. 2019. Hammer JH, Parent MC, Spiker DA, World Health Organization. Global status report on alcohol and health 2018 [Internet]. Vol. 65, Global status report on alcohol. 2018.. Mann K, Aubin HJ, Witkiewitz K. Reduced Drinking in Alcohol Dependence Treatment, What Is the Evidence? Eur Addict Res. 2017;23(5):219–30. Witkiewitz K. Temptation to Drink as a Predictor of Drinking Outcomes Following Psychosocial Treatment for Alcohol Dependence. 2013;37(3):529–37. Sliedrecht W, de Waart R, Witkiewitz K, Roozen HG. Alcohol use disorder relapse factors: A systematic review. Psychiatry Res [Internet]. 2019;278(May):97–115. Available from: https://doi.org/10.1016/j.psychres.2019.05.038 Weisner C, Ray GT, Mertens JR, Satre DD, Moore C. Short-term alcohol and drug treatment outcomes predict long-term outcome. Drug Alcohol Depend. 2003;71(3):281–94. Moos RH, Moos BS. Long-Term Influence of Duration and Frequency of Participation in Alcoholics Anonymous on Individuals with Alcohol Use Disorders. J Consult Clin Psychol. 2004;72(1):81–90. Grau E, Kemmann D, Brieger P. Zugangsvariablen als Prädiktoren für das Therapieergebnis bei Alkohollangzeitentwöhnungsmaβnahmen. Rehabil. 2014;53(1):38–42. Gueorguieva R, Wu R, Fucito LM, O’Malley SS. Predictors of abstinence from heavy drinking during follow-up in COMBINE. J Stud Alcohol Drugs. 2015;76(6):935–41. Haug S, Schaub MP. Treatment outcome, treatment retention, and their predictors among clients of five outpatient alcohol treatment centres in Switzerland. BMC Public Health [Internet]. 2016;16(1):1–10. Available from: http://dx.doi.org/10.1186/s12889-016-3294-4 Berglund KJ, Rauwolf KK, Berggren U, Balldin J, Fahlke C. Outcome in Relation to Drinking Goals in Alcohol-Dependent Individuals: A Follow-up Study 2.5 and 5 Years after Treatment Entry. Alcohol Alcohol. 2019;54(4):439–45. Ingesson-Hammarberg S, Jayaram-Lindström N, Hammarberg A. Predictors of treatment outcome for individuals with alcohol use disorder with a goal of controlled drinking. Addict Sci Clin Pract [Internet]. 2024;19(1):1–15. Available from: https://doi.org/10.1186/s13722-024-00443-z van de Schoot R, Depaoli S, King R, Kramer B, Märtens K, Tadesse MG, et al. Bayesian statistics and modelling. Nat Rev Methods Prim. 2021;1(1). Nishio M, Arakawa A. Performance of Hamiltonian Monte Carlo and No-U-Turn Sampler for estimating genetic parameters and breeding values. Genet Sel Evol [Internet]. 2019;51(1):1–12. Available from: https://doi.org/10.1186/s12711-019-0515-1 Muth C, Oravecz Z, Gabry J. User-friendly Bayesian regression modeling: A tutorial with rstanarm and shinystan. Quant Methods Psychol. 2018;14(2):99–119. Kline-Simon AH, Falk DE, Litten RZ, Mertens JR, Fertig J, Ryan M, et al. Posttreatment Low-Risk Drinking as a Predictor of Future Drinking and Problem Outcomes Among Individuals with Alcohol Use Disorders. Alcohol Clin Exp Res. 2013;37(SUPPL.1):373–80. Witkiewitz K. “Success” Following Alcohol Treatment: Moving Beyond Abstinence. Alcohol Clin Exp Res. 2013;37(SUPPL.1):9–13. Additional Declarations No competing interests reported. Cite Share Download PDF Status: Under Review Version 1 posted Reviewers agreed at journal 25 Nov, 2025 Reviewers agreed at journal 16 Sep, 2025 Reviewers agreed at journal 11 Sep, 2025 Reviewers invited by journal 11 Sep, 2025 Editor invited by journal 04 Sep, 2025 Editor assigned by journal 02 Sep, 2025 Submission checks completed at journal 02 Sep, 2025 First submitted to journal 29 Aug, 2025 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-7489427","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":516763173,"identity":"ef7f983a-1401-49fe-8974-2b754316ea74","order_by":0,"name":"Durai Murukan Gunasekaran","email":"","orcid":"","institution":"National Institute of Mental Health and Neuro Sciences (NIMHANS)","correspondingAuthor":false,"prefix":"","firstName":"Durai","middleName":"Murukan","lastName":"Gunasekaran","suffix":""},{"id":516763174,"identity":"79d1d305-a730-40f3-945b-9bc1953628cd","order_by":1,"name":"Binu VS","email":"","orcid":"","institution":"National Institute of Mental Health and Neuro Sciences (NIMHANS)","correspondingAuthor":false,"prefix":"","firstName":"Binu","middleName":"","lastName":"VS","suffix":""},{"id":516763176,"identity":"86f39928-b0bb-4e00-83eb-5b98bd5b8764","order_by":2,"name":"Arun Kandasamy","email":"","orcid":"","institution":"National Institute of Mental Health and Neuro Sciences (NIMHANS)","correspondingAuthor":false,"prefix":"","firstName":"Arun","middleName":"","lastName":"Kandasamy","suffix":""},{"id":516763177,"identity":"49666d67-e2e4-4457-823c-f74054c1202d","order_by":3,"name":"Lakshmanan Sethuraman","email":"","orcid":"","institution":"National Institute of Mental Health and Neuro Sciences (NIMHANS)","correspondingAuthor":false,"prefix":"","firstName":"Lakshmanan","middleName":"","lastName":"Sethuraman","suffix":""},{"id":516763178,"identity":"c139b502-f774-4e2b-b6b5-9efeb46d4d19","order_by":4,"name":"Binukumar Bhaskarapillai","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA10lEQVRIiWNgGAWjYBACA2YGBsYGEEuC+QCIlCFFC1sCiOQhrIUBroUHxGYgrMWcnfeY5My2bfLms3s+v7pRY8HDwH746AZ8Wiyb+dIkN7bdNpxz5+w265xjQIfxpKXdwOuwwzxmkg/bbjPOkMjdZpzDBtQiwWNGlBb7GRI5z4xz/hGrBeiwRKAW5se5bURosWzmMbacce528gyZY2bMuX0SPGyE/GLOf8bwZk/ZbdsZ0s2PP+d8q5PjZz98DK8WZMAmASaJVQ4CzB9IUT0KRsEoGAUjBwAAnS5FgPaLlXQAAAAASUVORK5CYII=","orcid":"","institution":"National Institute of Mental Health and Neuro Sciences (NIMHANS)","correspondingAuthor":true,"prefix":"","firstName":"Binukumar","middleName":"","lastName":"Bhaskarapillai","suffix":""}],"badges":[],"createdAt":"2025-08-29 14:23:20","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7489427/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7489427/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":91834533,"identity":"e30a2820-1af4-41fb-9c40-e08f8705ad97","added_by":"auto","created_at":"2025-09-22 09:26:18","extension":"jpeg","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":398016,"visible":true,"origin":"","legend":"","description":"","filename":"Figure1.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7489427/v1/e18d662e36fd2458edc305ff.jpeg"},{"id":91834531,"identity":"81808776-f49b-4020-bb09-4912f9d68aeb","added_by":"auto","created_at":"2025-09-22 09:26:18","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":33229,"visible":true,"origin":"","legend":"","description":"","filename":"table1.docx","url":"https://assets-eu.researchsquare.com/files/rs-7489427/v1/892413ed5a6dcf91bd5707c3.docx"},{"id":91834534,"identity":"35ae9f9e-79e6-4669-acc7-9f3764160ed3","added_by":"auto","created_at":"2025-09-22 09:26:18","extension":"docx","order_by":2,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":771137,"visible":true,"origin":"","legend":"","description":"","filename":"manuscriptbmc.docx","url":"https://assets-eu.researchsquare.com/files/rs-7489427/v1/94c30eb9dcd0b3c4decf9963.docx"},{"id":91834535,"identity":"511f00f4-f46d-4c57-a913-1f8051413e11","added_by":"auto","created_at":"2025-09-22 09:26:18","extension":"docx","order_by":3,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":24743,"visible":true,"origin":"","legend":"","description":"","filename":"table2.docx","url":"https://assets-eu.researchsquare.com/files/rs-7489427/v1/8d268060f12586a7a67586ac.docx"},{"id":91836005,"identity":"5d2b9c1e-de21-4e35-aeb7-a3acbd481f12","added_by":"auto","created_at":"2025-09-22 09:34:18","extension":"json","order_by":4,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":7479,"visible":true,"origin":"","legend":"","description":"","filename":"27b5234ec6a448119f3bf34f6d13f9d1.json","url":"https://assets-eu.researchsquare.com/files/rs-7489427/v1/48ff7fb0c76c53c4e9eba184.json"},{"id":91834539,"identity":"9809e144-c3e4-496d-8316-246c7d57a858","added_by":"auto","created_at":"2025-09-22 09:26:19","extension":"xml","order_by":5,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":101458,"visible":true,"origin":"","legend":"","description":"","filename":"27b5234ec6a448119f3bf34f6d13f9d11enriched.xml","url":"https://assets-eu.researchsquare.com/files/rs-7489427/v1/9632c091afe314ddbd483281.xml"},{"id":91836006,"identity":"b4257a0f-9b67-475b-99d6-822b2c2debe4","added_by":"auto","created_at":"2025-09-22 09:34:19","extension":"jpeg","order_by":6,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":398016,"visible":true,"origin":"","legend":"","description":"","filename":"Figure1.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7489427/v1/979f8985274549e06261727f.jpeg"},{"id":91834538,"identity":"8c3467f4-661d-44a1-b3fc-f12939be5fc6","added_by":"auto","created_at":"2025-09-22 09:26:19","extension":"jpeg","order_by":7,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1196180,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage1.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7489427/v1/2ed5a68ca5376e0330b3e342.jpeg"},{"id":91834541,"identity":"1d6c4c42-eeec-4ccc-84e3-fabf2abfd843","added_by":"auto","created_at":"2025-09-22 09:26:19","extension":"png","order_by":8,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":129110,"visible":true,"origin":"","legend":"","description":"","filename":"OnlineFigure1.png","url":"https://assets-eu.researchsquare.com/files/rs-7489427/v1/eb5cd5d9e6c8d9a83fb531fc.png"},{"id":91834536,"identity":"1b15d624-a669-4a9b-a12f-5425add2c2cb","added_by":"auto","created_at":"2025-09-22 09:26:18","extension":"png","order_by":9,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":270234,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-7489427/v1/765115100e89de2e968758de.png"},{"id":91834537,"identity":"2bc231e8-84b2-4a59-9d72-9112abeff277","added_by":"auto","created_at":"2025-09-22 09:26:19","extension":"xml","order_by":10,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":100185,"visible":true,"origin":"","legend":"","description":"","filename":"27b5234ec6a448119f3bf34f6d13f9d11structuring.xml","url":"https://assets-eu.researchsquare.com/files/rs-7489427/v1/3f7103e2e22f538087c3e02b.xml"},{"id":91836007,"identity":"8350f4f2-eace-4358-91d0-152594b46634","added_by":"auto","created_at":"2025-09-22 09:34:19","extension":"html","order_by":11,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":108140,"visible":true,"origin":"","legend":"","description":"","filename":"earlyproof.html","url":"https://assets-eu.researchsquare.com/files/rs-7489427/v1/0cf652730b718361476f2cd7.html"},{"id":91834530,"identity":"c9a9b09e-0512-41ae-9f5c-6740b1a5001a","added_by":"auto","created_at":"2025-09-22 09:26:18","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":274591,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cem\u003e\u003cstrong\u003eResults of adjusted Bayesian predictive models based on differing thresholds on unadjusted posterior Credible intervals\u003c/strong\u003e\u003c/em\u003e\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-7489427/v1/29ff268cadbe7dab88b3ee76.png"},{"id":91838064,"identity":"f2247a70-3bfd-42f8-9e82-121c49555462","added_by":"auto","created_at":"2025-09-22 09:42:19","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1398397,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7489427/v1/c4110b6b-e461-4285-98b6-08018d322d68.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Forecasting Long-Term Drinking Status for Individuals Treated for Alcohol Dependence: A Bayesian Approach","fulltext":[{"header":"Introduction","content":"\u003cp\u003eAlcohol dependence stands as the most common form of Alcohol Use Disorder (AUD), characterized by a dysregulated pattern of alcohol consumption resulting from repeated or sustained use of the substance. As outlined in the International Classification of Diseases (ICD) -10 report, key features of alcohol dependence encompass an intense internal compulsion to consume alcohol, a prioritization of alcohol use above all other activities, and a consistent pattern of use despite encountering harm or negative repercussions (\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eThe World Health Organization's report underscores the global burden of alcohol dependence, indicating a prevalence of 2.6% among individuals aged 15 years and older in 2016. Notably, this prevalence exhibits gender disparities, with adult males exhibiting a markedly higher proportion of alcohol dependence (4.5%) compared to adult females (0.8%). Similarly, within the Southeast Asia region, the prevalence of alcohol dependence was reported at 2.9%, with rates of 0.3% among adult females and 5.3% among adult males. Addressing alcohol dependence necessitates a multifaceted treatment approach capable of addressing the diverse array of factors that may precipitate substance use (\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eResearch indicates that a significant portion of participants in alcohol treatment prefer low-risk drinking outcomes, such as controlled drinking, which is comparable to abstinence in terms of long-term functioning and can be considered successful (\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e, \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e). Consequently, throughout the treatment phase, individuals may transition from a state of alcohol dependence to either complete abstinence or low-risk drinking outcomes. However, this transition alone does not guarantee the enduring efficacy of treatment over a long period. Given the relapsing nature of alcohol dependence, studies have indicated a relapse rate of approximately 50% within the first year following successful treatment for alcohol dependence (\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e). There has been a paucity of studies focused on observing the dynamic transitions from short-term treatment outcomes and their retention rates into the long term (\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e, \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e). Moreover, some studies have endeavoured to identify factors influencing the retention of treatment effects over the long term, examining their interplay with short-term treatment outcomes and other baseline socio-demographic as well as clinical variables (\u003cspan additionalcitationids=\"CR9 CR10 CR11\" citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eIdentifying high-risk individuals prone to relapse into alcohol dependence following treatment allows clinicians to provide targeted care, ensuring the enduring effectiveness of treatment and fostering long-term abstinence. Consequently, predictive models capable of identifying such high-risk populations in their initial days of treatment are instrumental in developing tailored intervention strategies to mitigate long-term relapse risk among successfully treated individuals. Therefore, the current study seeks to develop a predictive model to ascertain the long-term alcohol consumption status of male individuals with alcohol dependence who sought treatment at a tertiary care hospital in Southern India. This model incorporates a comprehensive range of factors, including socio-demographic, psychological, and environmental variables, alongside the initial treatment response pattern observed during the first four months of the treatment phase.\u003c/p\u003e"},{"header":"Materials and Methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e\u003ch2\u003eData\u003c/h2\u003e\u003cp\u003eThis is a cohort study comprising individuals seeking treatment for alcohol dependence from July 2020 to December 2021 at the Centre for Addiction Medicine (CAM) outpatient department (OPD) at the National Institute of Mental Health and Neurosciences (NIMHANS), located in Bengaluru, India. These individuals visited the CAM OPD on multiple occasions and received treatments to facilitate their transition from alcohol dependence to alcohol abstinence or controlled drinking status. The study comprised male individuals diagnosed with alcohol dependence based on ICD-10 criteria during the treatment phase. Only those who were found to be either in complete abstinence or in a controlled drinking environment at any visit within the first four months of receiving treatment from the study centre were considered to have undergone successful treatment and were eligible for inclusion. These participants were subsequently monitored for four months to understand the short-term treatment outcome, with an additional status check at the 12th month to understand the long-term treatment outcome. The primary endpoint is the status of alcohol dependence at the end of the twelve months, following complete abstinence or a controlled drinking environment achieved through treatment at NIMHANS. Exclusion criteria encompassed severe psychiatric disorders unrelated to substance dependence and significant health abnormalities.\u003c/p\u003e\u003cp\u003eThe data collected at baseline were socio-demographic and clinical information of each subject, such as their completed age in years (\u0026le;\u0026thinsp;25, 26\u0026ndash;40, 41\u0026ndash;50, \u0026amp; \u0026gt;50), educational status (primary and below, middle school, high school, \u0026amp; graduate and above), employment status (unemployed \u0026amp; employed), economic status (above poverty line (APL) if annual income\u0026thinsp;\u0026ge;\u0026thinsp;20,000 INR \u0026amp; below poverty line (BPL) if annual income\u0026thinsp;\u0026lt;\u0026thinsp;20,000 INR), marital status (never married \u0026amp; ever married), living arrangement (nuclear family/alone \u0026amp; joint family), first-degree family history of alcohol use (yes \u0026amp; no), current use of any tobacco (yes \u0026amp; no), age at first use of alcohol in years (\u0026lt;\u0026thinsp;18, 18\u0026ndash;25, \u0026amp; \u0026gt;25), duration of illness of alcohol dependence in years, and baseline clinical global impression-severity (CGI-s) score. The short-term treatment outcome, alcohol dependence status (abstinence/controlled drinking \u0026amp; alcohol dependence) at each of the first four months, were also monitored. Similarly, the same information was collected at the end of the 12th month to assess the long-term treatment outcome.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec4\" class=\"Section2\"\u003e\u003ch2\u003eStatistical Analysis\u003c/h2\u003e\u003cp\u003eA Bayesian regression model for binary response variable, long-term alcohol dependence status, was employed to forecast the likelihood of individuals transitioning into an alcohol-dependent state at the end of the 12th month following successful treatment. The dataset underwent a 90/10 split, with the training subset (90%) utilized for model development and the validation subset (10%) for model assessment. Given the limited prior knowledge regarding model parameters, a weakly informative prior proposed by Gelman was adopted (\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e). In this framework, the intercept was presumed to adhere to a Cauchy distribution with a mean of 0 and a standard deviation of 10, i.e., \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{\\beta\\:}_{0}\\sim\\:Cauchy\\left(\\text{0,10}\\right)\\)\u003c/span\u003e\u003c/span\u003e, while other parameters assumed to follow a Cauchy distribution with a mean of 0 and a standard deviation of 2.5, i.e., \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{\\beta\\:}_{\\:}\\sim\\:Cauchy\\left(\\text{0,2.5}\\right)\\)\u003c/span\u003e\u003c/span\u003e. The variable selection procedure for fitting the final predictive model was based on the results obtained from the unadjusted individual models, focusing on one variable at a time. As a result, four different predictive models were fitted by considering various thresholds. Variables with credible intervals (CrI) of 50%, 60%, 70%, 80% and 90% that exclude the null effect in their unadjusted CrI were considered for inclusion in the final predictive model. Given that the same set of variables satisfies both 80% and 90% thresholds this study does not delve into the model based on the 90% threshold.\u003c/p\u003e\u003cp\u003eMoreover, the Hamiltonian Monte Carlo (HMC) algorithm was employed with four Markov Chain Monte Carlo (MCMC) chains, each comprising 2000 iterations for parameter estimation to obtain the posterior distribution from both the unadjusted and adjusted models. The initial 1000 iterations were designated as a burn-in period, and the resulting samples were excluded from subsequent analysis. Even though different algorithms are available, this study employed HMC because it can theoretically generate samples from a wide range of parameter space with a high level of acceptance probability (\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e). The samples from the posterior obtained through HMC were summarized for the adjusted models using the posterior median, along with their corresponding 90% CrI. The list-wise deletion method was carried out to address the missingness in the independent variables while fitting and validating the models. Further, model adequacy was assessed by \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\widehat{R}\\)\u003c/span\u003e\u003c/span\u003e statistics and effective sample size (ESS).\u003c/p\u003e\u003cp\u003eInitially, 1000 samples were simulated from the posterior predictive distribution of all participants in the test set across the four models. These simulated samples provide the probability of being in an alcohol-dependent state at the end of the 12th month following successful treatment. Following this process, the next step involved obtaining a precise point estimate crucial for classification purposes. This was achieved by determining optimal cut-points for each model. The optimization process aimed at maximizing the combined sensitivity and specificity. Moreover, the predictive performance of these models underwent thorough evaluation and comparison. Key metrics such as the accuracy (%), area under the receiver operating characteristic curve (AUROC), sensitivity, and specificity were utilized for this purpose. All analyses were carried out using the packages \u0026ldquo;rstanarm\u0026rdquo;, \u0026ldquo;forestploter\u0026rdquo; and \u0026ldquo;grid\u0026rdquo; available under R version 4.2.3.\u003c/p\u003e\u003c/div\u003e"},{"header":"Results","content":"\u003cp\u003eDuring the specified timeframe, 3276 patients sought treatment at the CAM OPD. Among them, the majority, comprising 2543 individuals (77.63%), were seeking treatment for alcohol dependence, while the rest presented with various other substance abuse behaviours. From this pool of 2543 patients, only 608 (23.9%) individuals achieved either complete abstinence or a controlled drinking environment within the first four months of treatment. These 608 individuals were selected for the current study based on predetermined inclusion and exclusion criteria.\u003c/p\u003e\n\u003cp\u003eThe demographic analysis of the study participants unveiled a substantial 67.1%, which surpassed middle school education. Furthermore, 68.26% of the participants belonged to BPL socio-economic category. Additionally, 19.57% were unemployed, and 20.88% had never married. Moreover, a considerable segment, constituting 40.63% of the participants, disclosed a history of tobacco usage, while 19.41% reported a familial background of alcohol consumption among their first-degree family circle. Furthermore, the study found that the majority of participants, 57.89%, initiated alcohol consumption between the ages of 18 and 25 years. Approximately 75% of participants reported a duration of alcohol dependence exceeding four years. In addition, 50% of participants were assessed to be markedly ill or worse according to the CGI-S scale at the baseline.\u003c/p\u003e\n\u003cp\u003eNinety percent of the participants were randomly selected to comprise the training dataset (n\u0026thinsp;=\u0026thinsp;544), while the remaining ten percent constituted the validation dataset (n\u0026thinsp;=\u0026thinsp;64). Table\u0026nbsp;1 illustrates basic descriptive statistics and the effects, in terms of varying posterior estimate CrI, for the considered variables obtained from the unadjusted model using the training dataset. The results revealed that the alcohol consumption status over the past four months of attaining the abstinence/controlled drinking state played a significant role in long-term alcohol dependence status individually. Specifically, individuals identified as alcohol-dependent during the initial four months were found to be at a higher risk of continuing to exhibit alcohol dependence by the end of the 12th month. This suggests a persistent pattern of alcohol dependency over time.\u003c/p\u003e\n\u003cp\u003eAdditionally, variable such as age (in years) was found to be influential, as they consistently demonstrated non-null effects across various proposed thresholds. Specifically, younger age was associated with a decreased likelihood of long-term alcohol dependence, with individuals under 25 demonstrating a reduced risk across various thresholds. This underscores their importance in predicting outcomes. Furthermore, variables such as educational status, living arrangements, and age at first alcohol use were also identified as influential factors, particularly under lenient thresholds such as 50%. However, their importance diminished when stricter thresholds, like 80%, were applied.\u003c/p\u003e\n\u003cp\u003eFigure \u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003e depicts the outcomes, showcasing 90% CrI on the posterior estimates derived from adjusted Bayesian predictive models, as per the thresholds previously mentioned. Among the four models were fitted based on the different thresholds, the alcohol dependence status during each of the first three months was not found to be significant. However, it was found to be significant in the fourth month, a finding consistent across all four models. This suggests that the consumption status in the fourth month holds greater importance in predicting long-term outcomes for individuals than in the initial three months post-treatment. Whereas, the other variables considered haven\u0026apos;t shown any significant impact on the outcome, the recency of data on alcohol dependence status appears to play a pivotal role in forecasting an individual\u0026apos;s long-term status over other socio-demographic and clinical variables.\u003c/p\u003e\n\u003cp\u003eFurther, as recommended, it was observed that model adequacy statistic \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\widehat{R}\\)\u003c/span\u003e\u003c/span\u003e \u0026lt; 1.1 and ESS \u0026gt;1000 in all the fitted unadjusted and adjusted models (\u003cspan class=\"CitationRef\"\u003e15\u003c/span\u003e) ensures convergence and less autocorrelation between the obtained samples.\u003c/p\u003e\n\u003cdiv class=\"gridtable\"\u003e\n \u003cdiv align=\"left\" class=\"colspec\"\u003e\u003cbr\u003e\u003c/div\u003e\n \u003cdiv align=\"left\" class=\"colspec\"\u003e\u003cstrong\u003e\u003cem\u003eTable 1. Results of unadjusted Bayesian predictive model in terms of differing Credible intervals \u0026nbsp;\u003c/em\u003e\u003c/strong\u003e\u003c/div\u003e\n \u003cdiv align=\"left\" class=\"colspec\"\u003e\u003cbr\u003e\u003c/div\u003e\u003cbr\u003e\n \u003ctable id=\"Taba\" border=\"1\"\u003e\n \u003cthead\u003e\n \u003ctr\u003e\n \u003cth align=\"left\" colspan=\"2\" rowspan=\"2\"\u003e\n \u003cp\u003eVariables\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\" colspan=\"2\"\u003e\n \u003cp\u003eEvent\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\" colspan=\"4\"\u003e\n \u003cp\u003eUnadjusted Model - Posterior Estimate- ln(Odds)\u003c/p\u003e\n \u003c/th\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eNo (n\u0026thinsp;=\u0026thinsp;317)\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eYes (n\u0026thinsp;=\u0026thinsp;227)\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003e(P\u003csub\u003e25\u003c/sub\u003e,P\u003csub\u003e75\u003c/sub\u003e)\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003e(P\u003csub\u003e20\u003c/sub\u003e,P\u003csub\u003e80\u003c/sub\u003e)\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003e(P\u003csub\u003e15\u003c/sub\u003e,P\u003csub\u003e85\u003c/sub\u003e)\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003e(P\u003csub\u003e10\u003c/sub\u003e,P\u003csub\u003e90\u003c/sub\u003e)\u003c/p\u003e\n \u003c/th\u003e\n \u003c/tr\u003e\n \u003c/thead\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" rowspan=\"2\"\u003e\n \u003cp\u003e\u003cstrong\u003eStatus at 1st month\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003eAbstinence/Controlled drinking\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e255\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e163\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colspan=\"4\"\u003e\n \u003cp\u003eREF\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003eAlcohol dependence\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e62\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e64\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e(0.338,0.611)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e(0.301,0.646)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e(0.264,0.683)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e(0.220,0.734)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" rowspan=\"2\"\u003e\n \u003cp\u003e\u003cstrong\u003eStatus at 2nd month\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003eAbstinence/Controlled drinking\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e259\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e161\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colspan=\"4\"\u003e\n \u003cp\u003eREF\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003eAlcohol dependence\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e58\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e66\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e(0.458,0.744)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e(0.419,0.780)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e(0.38,0.819)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e(0.332,0.868)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" rowspan=\"2\"\u003e\n \u003cp\u003e\u003cstrong\u003eStatus at 3rd month\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003eAbstinence/Controlled drinking\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e261\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e162\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colspan=\"4\"\u003e\n \u003cp\u003eREF\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003eAlcohol dependence\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e56\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e66\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e(0.471,0.758)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e(0.433,0.789)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e(0.396,0.825)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e(0.344,0.873)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" rowspan=\"2\"\u003e\n \u003cp\u003e\u003cstrong\u003eStatus at 4th month\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003eAbstinence/Controlled drinking\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e274\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e154\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colspan=\"4\"\u003e\n \u003cp\u003eREF\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003eAlcohol dependence\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e43\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e73\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e(0.947,1.248)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e(0.911,1.286)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e(0.866,1.328)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e(0.824,1.379)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" rowspan=\"4\"\u003e\n \u003cp\u003e\u003cstrong\u003eAge (in years)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026gt;\u0026thinsp;50\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e36\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e31\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colspan=\"4\"\u003e\n \u003cp\u003eREF\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003e41\u0026ndash;50\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e94\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e52\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e(-0.619,-0.231)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e(-0.665,-0.183)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e(-0.726,-0.12)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e(-0.798,-0.052)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003e26\u0026ndash;40\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e158\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e133\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e(-0.182-0.168)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e(-0.225,0.208)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e(-0.274,0.262)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e(-0.343,0.328)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026lt;=25\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e29\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e11\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e(-1.07,-0.522)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e(-1.142,-0.461)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e(-1.224,-0.389)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e(-1.339,-0.299)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" rowspan=\"4\"\u003e\n \u003cp\u003e\u003cstrong\u003eEducational status\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003ePrimary and below\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e59\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e48\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colspan=\"4\"\u003e\n \u003cp\u003eREF\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003eMiddle School\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e40\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e33\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e(-0.184,0.227)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e(-0.238,0.279)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e(-0.303,0.336)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e(-0.371,0.404)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003eHigher and Higher secondary\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e159\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e104\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e(-0.359,-0.057)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e(-0.401,-0.023)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e(-0.447,0.022)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e(-0.505,0.079)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003eGraduate and above\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e59\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e42\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e(-0.314,0.061)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e(-0.359,0.112)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e(-0.411,0.158)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e(-0.483,0.226)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" rowspan=\"2\"\u003e\n \u003cp\u003e\u003cstrong\u003eEmployment status\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003eUnemployed\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e61\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e45\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colspan=\"4\"\u003e\n \u003cp\u003eREF\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003eEmployed\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e256\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e182\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e(-0.184,0.117)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e(-0.223,0.153)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e(-0.269,0.194)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e(-0.323,0.243)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" rowspan=\"2\"\u003e\n \u003cp\u003e\u003cstrong\u003eEconomic status\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003eAPL\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e103\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e76\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colspan=\"4\"\u003e\n \u003cp\u003eREF\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003eBPL\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e214\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e151\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e(-0.171,0.075)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e(-0.2,0.108)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e(-0.234,0.146)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e(-0.277,0.192)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" rowspan=\"2\"\u003e\n \u003cp\u003e\u003cstrong\u003eMarital status\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003eNever married\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e73\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e42\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colspan=\"4\"\u003e\n \u003cp\u003eREF\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003eEver married\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e244\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e185\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e(0.130,0.426)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e(0.095,0.458)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e(0.054,0.502)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e(-0.0003,0.558)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" rowspan=\"2\"\u003e\n \u003cp\u003e\u003cstrong\u003eLiving arrangement\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003eNuclear family/Alone\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e230\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e158\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colspan=\"4\"\u003e\n \u003cp\u003eREF\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003eJoint Family\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e87\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e69\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e(0.010,0.267)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e(-0.020,0.3)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e(-0.058,0.337)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e(-0.113,0.387)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" rowspan=\"2\"\u003e\n \u003cp\u003e\u003cstrong\u003eFamily History of Alcohol use\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003eNo\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e255\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e187\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colspan=\"4\"\u003e\n \u003cp\u003eREF\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003eYes\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e62\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e40\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e(-0.288,0.018)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e(-0.323,0.054)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e(-0.368,0.102)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e(-0.416,0.154)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" rowspan=\"2\"\u003e\n \u003cp\u003e\u003cstrong\u003eCurrent tobacco use\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003eNo\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e181\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e133\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colspan=\"4\"\u003e\n \u003cp\u003eREF\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003eYes\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e136\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e94\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e(-0.176,0.052)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e(-0.208,0.083)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e(-0.246,0.118)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e(-0.294,0.157)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" rowspan=\"3\"\u003e\n \u003cp\u003e\u003cstrong\u003eAge at first alcohol use (in years)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026gt;\u0026thinsp;25\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e63\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e38\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colspan=\"4\"\u003e\n \u003cp\u003eREF\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003e18\u0026ndash;25\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e179\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e134\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e(0.059,0.355)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e(0.023,0.394)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e(-0.027,0.439)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e(-0.084,0.497)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026lt;\u0026thinsp;18\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e75\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e55\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e(0.0002,0.363)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e(-0.043,0.407)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e(-0.092,0.46)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e(-0.166,0.522)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" colspan=\"2\"\u003e\n \u003cp\u003e\u003cstrong\u003eDuration of illness (in years)*\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e7(15\u0026thinsp;\u0026minus;\u0026thinsp;3)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e8(16\u0026thinsp;\u0026minus;\u0026thinsp;4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e(0.004,0.018)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e(0.002,0.02)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e(0.0003,0.022)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e(-0.002,0.025)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" colspan=\"2\"\u003e\n \u003cp\u003e\u003cstrong\u003eBaseline CGI-s score*\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e4(4\u0026thinsp;\u0026minus;\u0026thinsp;2)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e4(4\u0026thinsp;\u0026minus;\u0026thinsp;3)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e(0.076,0.162)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e(0.065,0.172)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e(0.051,0.184)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e(0.036,0.2)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" colspan=\"8\"\u003e\n \u003cp\u003e\u003cstrong\u003e* Indicates continuous variables for which Median(Q3-Q1) has been reported; (P\u003c/strong\u003e\u003csub\u003e\u003cstrong\u003e25\u003c/strong\u003e\u003c/sub\u003e,\u003cstrong\u003eP\u003c/strong\u003e\u003csub\u003e\u003cstrong\u003e75\u003c/strong\u003e\u003c/sub\u003e\u003cstrong\u003e)-50% CrI; (P\u003c/strong\u003e\u003csub\u003e\u003cstrong\u003e20\u003c/strong\u003e\u003c/sub\u003e,\u003cstrong\u003eP\u003c/strong\u003e\u003csub\u003e\u003cstrong\u003e80\u003c/strong\u003e\u003c/sub\u003e\u003cstrong\u003e)- 60% CrI; (P\u003c/strong\u003e\u003csub\u003e\u003cstrong\u003e15\u003c/strong\u003e\u003c/sub\u003e,\u003cstrong\u003eP\u003c/strong\u003e\u003csub\u003e\u003cstrong\u003e85\u003c/strong\u003e\u003c/sub\u003e\u003cstrong\u003e)- 70% CrI; (P\u003c/strong\u003e\u003csub\u003e\u003cstrong\u003e10\u003c/strong\u003e\u003c/sub\u003e,\u003cstrong\u003eP\u003c/strong\u003e\u003csub\u003e\u003cstrong\u003e90\u003c/strong\u003e\u003c/sub\u003e\u003cstrong\u003e)- 80% CrI.\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n \u003c/table\u003e\n\u003c/div\u003e\n\u003cdiv class=\"gridtable\"\u003e\n \u003ctable id=\"Tab1\" border=\"1\"\u003e\n \u003ccaption language=\"En\"\u003e\n \u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e\n \u003cdiv class=\"CaptionContent\"\u003e\n \u003cp\u003eAccuracy metrics across the fitted models\u003c/p\u003e\n \u003c/div\u003e\n \u003c/caption\u003e\n \u003cthead\u003e\n \u003ctr\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eModel\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eOptimal\u003c/p\u003e\n \u003cp\u003eCut-point\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eAccuracy\u003c/p\u003e\n \u003cp\u003e(%)\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eAUROC\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eSensitivity\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eSpecificity\u003c/p\u003e\n \u003c/th\u003e\n \u003c/tr\u003e\n \u003c/thead\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eModel 1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.351\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.683\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.618\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.824\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.500\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eModel 2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.355\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.683\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.619\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.824\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.500\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eModel 3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.400\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.685\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.630\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.706\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.692\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eModel 4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.409\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.653\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.605\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.657\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.643\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n \u003c/table\u003e\n\u003c/div\u003e\n\u003cp\u003eTable\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003e presents a comprehensive view of the optimal cut-points obtained and the resulting predictive power across all four models when applied to the test dataset. The results showed that Model 3 outperformed its counterparts based on the accuracy and AUROC criteria. Notably, the \u003cem\u003eModel 3\u003c/em\u003e encompassed various covariates, including All month-wise alcohol consumption status (first month to fourth month), age, marital status, duration of illness, and baseline CGI-s score.\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eThis study investigated the impact of fundamental socio-demographic factors and short-term treatment outcomes on the long-term treatment effectiveness for individuals with alcohol dependence, using the Bayesian predictive models.\u003c/p\u003e\u003cp\u003eThe age of participants has emerged as a factor in predicting their long-term outcomes in unadjusted analysis. Previous studies have consistently identified older age as a key contributor to sustained alcohol abstinence over an extended period (\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e, \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e). However, intriguingly, our initial unadjusted results indicated that younger age was associated with a lower likelihood of sustained alcohol dependence. Nevertheless, this association lost its significance after adjusting for other variables in the final model. This contradictory finding opens up avenues for further investigation and exploration. The presence of a recovery-oriented social support network has been acknowledged as a significant factor contributing to sustained alcohol abstinence over an extended period (\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e). Although this specific aspect was not directly observed in our study, marital status, often considered a proxy for social support, was found to exert a notable influence on long-term treatment outcomes. Similar to age, the significance of marital status diminished in the final fitted models. Another variable examined in this study is living arrangement, which also aimed to elucidate the impact of social support on long-term treatment outcomes. However, it was not found to be an important factor in this regard. Further, though previous research has indicated that individuals with self-sustaining income demonstrate greater treatment retention (\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e), the current study findings suggest that economic status does not play a significant role in treatment retention over the long-term period. This disparity highlights the need for further investigation in future studies to better understand the complex relationship between economic factors and treatment outcomes in alcohol dependence.\u003c/p\u003e\u003cp\u003eSimilarly, lower baseline alcohol consumption levels were associated with a greater likelihood of maintaining long-term abstinence (\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e, \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e). Given that baseline alcohol consumption serves as an indicator of the severity of alcohol consumption, this study incorporated information on duration of illness in terms of alcohol dependence and the baseline CGI-s score. Both of these variables were found to have a significant impact on achieving long-term alcohol dependence in the unadjusted model. There has been a notable lack of research focused on understanding the impact of age at first alcohol use on long-term treatment outcomes. This study endeavoured to explore this aspect, and initial unadjusted results hinted at its significance, albeit this importance diminished in the adjusted model. Nevertheless, further investigation is warranted to better understand this effect and its implications for long-term treatment efficacy.\u003c/p\u003e\u003cp\u003eConsistent with the findings of this investigation, prior research studies have revealed that alcohol abstinence at the short-term sixth-month milestone serves as a significant predictor for sustaining abstinence throughout a one-year and five-year follow-up period (\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e, \u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e). However, there has been a notable dearth of studies examining the impact of short-term treatment outcomes on long-term results. This lack of research has hindered our ability to compare our findings and those of previous studies.\u003c/p\u003e\u003cp\u003eThere is a discrepancy in the definitions of \u0026ldquo;long-term\u0026rdquo; and \u0026ldquo;short-term\u0026rdquo; across previous studies. However, amidst this variation, a discernible pattern has emerged, indicating that durations equal to or less than six months are often classified as short-term, while time frames exceedingly approximately 52 weeks or one year are considered long-term, aligning with our study definitions. In contrast to previous studies, this research sought to utilize the Bayesian framework for prediction, leveraging its advantages, such as incorporating prior information. Also, treating the parameter of interest as a variable rather than a constant, this approach enables us to grasp the range of potential outcomes for each individual, thereby enhancing our understanding of treatment possibilities.\u003c/p\u003e\u003cp\u003eEven though this study has employed weakly informative priors due to the lack of evidence from previous studies, it can still provide valuable insights for future research. Specifically, this study offers a general framework for building predictive models potentially guiding subsequent studies in developing more accurate and robust models. This study also has several limitations. Several studies investigated the influence of gender on predicting long-term treatment outcomes, consistently finding that female gender is significantly associated with achieving controlled drinking (\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e, \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e, \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e). However, this study was limited to male individuals due to the scarcity of information on female participants, preventing us from directly comparing gender differences in treatment outcomes. Another crucial aspect left unexplored was the motivation of individuals to quit alcohol. Previous studies have underscored that commitment to alcohol abstinence is associated with retention in maintaining non-problematic drinking behaviour over the long term (\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e). Additionally, various studies have highlighted the pivotal role of motivation to quit alcohol and the establishment of specific drinking goals in maintaining long-term alcohol abstinence or controlled drinking. These factors have been consistently emphasized across multiple research endeavours, reaffirming their significance in sustaining positive outcomes (\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e, \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e). However, this study could not address the role of motivation in predicting long-term outcomes due to data unavailability. In addition, the study lacks consideration of any time-varying follow-up information beyond alcohol consumption status, such as participant\u0026rsquo;s emotional states, levels of craving, and other pertinent factors, unlike previous studies (\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e). This limitation presents another drawback of the study. Expanding the exploration of the aforementioned potential factors in future research could provide even richer insights into treatment effectiveness and also potentially enhance the predictive capabilities of the model.\u003c/p\u003e"},{"header":"Conclusion","content":"\u003cp\u003eThe integration of a Bayesian predictive model within this study has emerged as a pivotal tool for anticipating the long-term alcohol consumption status of individuals successfully treated for alcohol dependence. Through comprehensive analysis of relevant socio-demographic factors, clinical indicators, and initial alcohol consumption patterns among participants, the model has identified that the alcohol dependence status at the end of the fourth month has a significant impact on the long-term treatment outcome. Even though some variables were not found to have a significant effect individually, their combined contribution has improved the predictive nature of the model, demonstrating its efficacy in forecasting long-term effects to a considerable extent. Consequently, clinicians can glean invaluable insights to proactively address and manage potential future outcomes, thereby enhancing their capacity to deliver targeted and effective support throughout individual\u0026rsquo;s journeys towards sustained recovery.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003e\u003cem\u003eEthics approval and consent to participate\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAll participants provided informed consent to use their hospital record data for research purposes. This study was approved by the NIMHANS Internal Institute Ethics Committee (Basic Sciences) on 18th November 2020.We declare that this study was conducted in accordance with the Declaration of Helsinki.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eConsent for publication\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot applicable.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eAvailability of data and materials\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe data supporting this study are derived from the NIMHANS CAM department database. Due to licensing restrictions, these data are not publicly available. However, they may be obtained from the authors upon reasonable request and with appropriate permission from the NIMHANS CAM department.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eConflicts of interest\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThere are no conflicts of interest.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eFinancial Disclosure\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors declare that there was no external funding or financial support received for the research, development, or publication of this article.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eAuthor Contributions Statement\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eDurai Murukan Gunasekaran performed the data analysis and prepared the manuscript. Binukumar Bhaskarapillai, Binu V.S., and Arun Kandasamy conceptualised the core idea of this work. Lakshmanan Sethuraman assisted in data acquisition. Binukumar Lakshmanan provided overall supervision of the study. All authors reviewed the manuscript.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eAcknowledgement\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors are grateful to participants for sharing their information. Special thanks to Dr. N. Sreekumaran Nair, Professor and Head, Biostatistics, JIPMER, Puducherry, India, and Dr. K. Thennarasu, Professor and Head, Biostatistics, Bengaluru, NIMHANS, India, for generous suggestions and insightful recommendations.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eEdition S. International statistical classification of diseases and related health problems 10th revision Volume 2 Instruction manual. 2019. \u003c/li\u003e\n\u003cli\u003eHammer JH, Parent MC, Spiker DA, World Health Organization. Global status report on alcohol and health 2018 [Internet]. Vol. 65, Global status report on alcohol. 2018.. \u003c/li\u003e\n\u003cli\u003eMann K, Aubin HJ, Witkiewitz K. Reduced Drinking in Alcohol Dependence Treatment, What Is the Evidence? Eur Addict Res. 2017;23(5):219\u0026ndash;30. \u003c/li\u003e\n\u003cli\u003eWitkiewitz K. Temptation to Drink as a Predictor of Drinking Outcomes Following Psychosocial Treatment for Alcohol Dependence. 2013;37(3):529\u0026ndash;37. \u003c/li\u003e\n\u003cli\u003eSliedrecht W, de Waart R, Witkiewitz K, Roozen HG. Alcohol use disorder relapse factors: A systematic review. Psychiatry Res [Internet]. 2019;278(May):97\u0026ndash;115. Available from: https://doi.org/10.1016/j.psychres.2019.05.038\u003c/li\u003e\n\u003cli\u003eWeisner C, Ray GT, Mertens JR, Satre DD, Moore C. Short-term alcohol and drug treatment outcomes predict long-term outcome. Drug Alcohol Depend. 2003;71(3):281\u0026ndash;94. \u003c/li\u003e\n\u003cli\u003eMoos RH, Moos BS. Long-Term Influence of Duration and Frequency of Participation in Alcoholics Anonymous on Individuals with Alcohol Use Disorders. J Consult Clin Psychol. 2004;72(1):81\u0026ndash;90. \u003c/li\u003e\n\u003cli\u003eGrau E, Kemmann D, Brieger P. Zugangsvariablen als Pr\u0026auml;diktoren f\u0026uuml;r das Therapieergebnis bei Alkohollangzeitentw\u0026ouml;hnungsma\u0026beta;nahmen. Rehabil. 2014;53(1):38\u0026ndash;42. \u003c/li\u003e\n\u003cli\u003eGueorguieva R, Wu R, Fucito LM, O\u0026rsquo;Malley SS. Predictors of abstinence from heavy drinking during follow-up in COMBINE. J Stud Alcohol Drugs. 2015;76(6):935\u0026ndash;41. \u003c/li\u003e\n\u003cli\u003eHaug S, Schaub MP. Treatment outcome, treatment retention, and their predictors among clients of five outpatient alcohol treatment centres in Switzerland. BMC Public Health [Internet]. 2016;16(1):1\u0026ndash;10. Available from: http://dx.doi.org/10.1186/s12889-016-3294-4\u003c/li\u003e\n\u003cli\u003eBerglund KJ, Rauwolf KK, Berggren U, Balldin J, Fahlke C. Outcome in Relation to Drinking Goals in Alcohol-Dependent Individuals: A Follow-up Study 2.5 and 5 Years after Treatment Entry. Alcohol Alcohol. 2019;54(4):439\u0026ndash;45. \u003c/li\u003e\n\u003cli\u003eIngesson-Hammarberg S, Jayaram-Lindstr\u0026ouml;m N, Hammarberg A. Predictors of treatment outcome for individuals with alcohol use disorder with a goal of controlled drinking. Addict Sci Clin Pract [Internet]. 2024;19(1):1\u0026ndash;15. Available from: https://doi.org/10.1186/s13722-024-00443-z\u003c/li\u003e\n\u003cli\u003evan de Schoot R, Depaoli S, King R, Kramer B, M\u0026auml;rtens K, Tadesse MG, et al. Bayesian statistics and modelling. Nat Rev Methods Prim. 2021;1(1). \u003c/li\u003e\n\u003cli\u003eNishio M, Arakawa A. Performance of Hamiltonian Monte Carlo and No-U-Turn Sampler for estimating genetic parameters and breeding values. Genet Sel Evol [Internet]. 2019;51(1):1\u0026ndash;12. Available from: https://doi.org/10.1186/s12711-019-0515-1\u003c/li\u003e\n\u003cli\u003eMuth C, Oravecz Z, Gabry J. User-friendly Bayesian regression modeling: A tutorial with rstanarm and shinystan. Quant Methods Psychol. 2018;14(2):99\u0026ndash;119. \u003c/li\u003e\n\u003cli\u003eKline-Simon AH, Falk DE, Litten RZ, Mertens JR, Fertig J, Ryan M, et al. Posttreatment Low-Risk Drinking as a Predictor of Future Drinking and Problem Outcomes Among Individuals with Alcohol Use Disorders. Alcohol Clin Exp Res. 2013;37(SUPPL.1):373\u0026ndash;80. \u003c/li\u003e\n\u003cli\u003eWitkiewitz K. \u0026ldquo;Success\u0026rdquo; Following Alcohol Treatment: Moving Beyond Abstinence. Alcohol Clin Exp Res. 2013;37(SUPPL.1):9\u0026ndash;13. \u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"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":"bmc-public-health","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"pubh","sideBox":"Learn more about [BMC Public Health](http://bmcpublichealth.biomedcentral.com/)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/pubh/default.aspx","title":"BMC Public Health","twitterHandle":"@BMC_series","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"em","reportingPortfolio":"BMC Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"Alcohol dependence, Controlled drinking environment, Bayesian, India","lastPublishedDoi":"10.21203/rs.3.rs-7489427/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7489427/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003ch2\u003eBackground\u003c/h2\u003e\u003cp\u003eAlcohol dependence presents a significant challenge, with treatment efforts often falling short in preventing relapse due to the nature of the condition. Ensuring sustained long-term recovery among successfully treated individuals is of paramount clinical importance. Therefore, this study aims to predict the long-term (after 12 months of successful treatment) alcohol consumption status of individuals treated for alcohol dependence, utilizing baseline characteristics and initial treatment response information.\u003c/p\u003e\u003ch2\u003eMethods\u003c/h2\u003e\u003cp\u003eA cohort study design was employed to assess both short and long-term alcohol consumption status among male participants who had undergone successful treatment for alcohol dependence at an addiction medicine outpatient department in Southern India from July 2020 to December 2021. Initially, participants were monitored for four months to observe their short-term treatment response, followed by a telephonic follow-up to assess their long-term status. This study employs the application of the Bayesian prediction model for evaluating the long-term binary outcome.\u003c/p\u003e\u003ch2\u003eResults\u003c/h2\u003e\u003cp\u003eAmong four distinct fitted predictive models, the one encompassing data on initial four-month alcohol consumption, age, marital status, illness duration, and baseline clinical global impression-severity score emerged as predictors for long-term treatment outcomes. This model showed around 70% accuracy on the test dataset, with sensitivity and specificity hovering around 70%.\u003c/p\u003e\u003ch2\u003eConclusion\u003c/h2\u003e\u003cp\u003eThe findings emphasize the importance of maintaining alcohol abstinence or controlled drinking in the fourth month after successful treatment to ensure long-term success. Additionally, the results underscore the necessity of exploring variables beyond baseline information to understand factors influencing long-term outcomes comprehensively.\u003c/p\u003e","manuscriptTitle":"Forecasting Long-Term Drinking Status for Individuals Treated for Alcohol Dependence: A Bayesian Approach","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-09-22 09:26:13","doi":"10.21203/rs.3.rs-7489427/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"reviewerAgreed","content":"13994605281114894085954416038568506085","date":"2025-11-25T11:24:08+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"117456862304837413219282855998110770518","date":"2025-09-16T09:13:12+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"14209136059088833467771570110196347407","date":"2025-09-12T03:05:32+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-09-11T05:54:15+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2025-09-04T04:57:51+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-09-02T14:19:13+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-09-02T14:18:35+00:00","index":"","fulltext":""},{"type":"submitted","content":"BMC Public Health","date":"2025-08-29T14:15:53+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"bmc-public-health","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"pubh","sideBox":"Learn more about [BMC Public Health](http://bmcpublichealth.biomedcentral.com/)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/pubh/default.aspx","title":"BMC Public Health","twitterHandle":"@BMC_series","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"em","reportingPortfolio":"BMC Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"74a90dc3-de6a-47ae-bdc1-1d20685d923d","owner":[],"postedDate":"September 22nd, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[],"tags":[],"updatedAt":"2025-09-22T09:26:14+00:00","versionOfRecord":[],"versionCreatedAt":"2025-09-22 09:26:13","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7489427","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7489427","identity":"rs-7489427","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}