Development Of A Relapse Risk Assessment Tool For Methadone Maintenance Treatment Patients

BackgroundIt is still a challenge around the world that methadone maintenance treatment(MMT)patients occurred relapse during treatment(also known as stealing or continuing to use drugs).The proportion of relapse among MMT patients is also relatively high in China.If there are appropriate risk assessment tools to identify the high-risk groups of relapse and intervene in time before the relapse behavior of drug abusers occurs,it will be of positive significance to prevent relapse and improve the effect of MMT.However,there is still a lack of simple quantitative risk assessment tools for staff to assess the risk of relapse in MMT patients at home and abroad.ObjectivesTo develop a relapse risk assessment tool for MMT outpatients in China.Methods1.By systematically searching Embase,PubMed,China National Knowledge Network(CNKI),and Wanfang databases through the established search strategy,and combined with the study inclusion and exclusion criteria,the original studies published at home and abroad on the risk factors of drug addicts' relapse were screened,and the study-level data and candidate risk factor data were collected.Finally,the pooled relative risk(RR)and 95%CIs for each candidate risk factor across studies were calculated using Meta-analysis,and the pooled RR was used to evaluate the correlation between the risk factors and relapse,so as to screen the candidate items for relapse risk assessment tool.2.Using the retrospective open cohort study design,the patients who were newly enrolled in MMT clinics in Yunnan Province from 2013 to 2017 were selected as the study objects.Multivariable logistic regression model based on least absolute shrinkage and selection operator(lasso)regression variable selection method were constructed to explore the variables significantly related to relapse from the baseline characteristic variables of participants.3.Combined the results of systematic review and meta-analysis with the results of lasso-logistic model,and using the simple risk scoring system developed by Sullivan et al,so as to develop a relapse risk assessment tool.Taking the patients who received treatment in three MMT clinics in Taiyuan City,Shanxi Province from October to November 2020 as the validation cohort population,a baseline survey was conducted and prospectively followed up for 3 months.The relapse risk assessment tool was evaluated from three aspects:discrimination,calibration and clinical effectiveness.Results1.Systematic review and Meta-analysis:After rigorous literature search and screening,23 original literatures meeting the criteria were finally included,including 139,953 participants.A total of 36 candidate risk factors were collected from the 23 studies.After Meta-analysis and candidate risk factors screening process,9 risk factors were considered as the items of relapse risk assessment tool,which included age,gender,employment status,recent smoking status,maintenance treatment compliance,co-occurring psychiatric disorders,family history of drug abuse,relationship with family,and frequency of seeing drug-using friends in past 1 month.2.Retrospective cohort data analysis:Of 7,899 eligible patients,4,125(52.22%)cases occurred relapse in the first 6 months following enrollment.Six variables were screened as risk factors for MMT patients by Lasso regression variable selection method.they were age,employment status,living situation,repeated entry MMT clinics in the past,MMT compliance,and relationship with family.The multivariable logistic regression analysis results showed that the risk of relapse decreased by 2%(OR=0.98,95%CI:0.98-0.99)with 1 year increase in age,the risk of relapse for people who have a job currently was 1.66 times higher than that of no job people(OR=1.66,95%CI:1.50-1.85),and the risk of relapse of subjects living alone was 3.06 times higher than that of not living alone(OR=3.06,95%CI:2.64-3.55).Compared with those who had never participated in MMT,the risk of relapse increased by 32%in those who participated in MMT once(OR=1.32,95%CI:1.16-1.51),127%in those who participated in MMT twice(OR=2.27,95%CI:1.67-3.08),and 232%in those who participated in MMT at least 3 times(OR=3.32,95%CI:1.85-5.98).Compared with those with good MMT compliance within 6 months after enrollment,the risk of relapse increased by 105%(OR=2.05,95%CI:1.82-2.31)for patients with moderate MMT compliance and 374%(OR=4.74,95%CI:4.20-5.36)for those with poor MMTcompliance.Compared with subjects with good family relationships,subjects with normal family relationships had a 55%increased risk of relapse,and subjects with poor family relationships had a 146%increase in risk of relapse,their OR and 95%CI were 1.55(1.38-1.74)and 2.46(2.19-2.77),respectively.3.Relapse risk assessment tool development:The final developed relapse risk assessment tool included 11 items.i.e.age,gender,living situation,employment status,recent smoking status,repeated entry MMT clinics in the past,maintenance treatment compliance,co-occurring psychiatric disorders,family history of drug abuse,relationship with family,and frequency of seeing drug-using friends in past 1 month.By comparing with the sensitivity and specificity of relapse risk assessment tools under different cut-off values,the best cut-off value was 31 points,and the patients with a total score ?31 points were identified as high-risk groups for relapse.The index AUC and its 95%CI reflecting the distinguishing ability were 0.87(95%CI:0.82-0.91),suggesting that the model had excellent ability to distinguish relapse patients from non-relapse patients,and the Hosmer-Lemeshow test results and calibration plot reflecting the calibration ability both showed that there was a high consistency between the predicted value and the real value of the tool.The results of clinical decision curve analysis reflecting clinical effectiveness show that no matter which threshold of the relapse risk assessment tool is used to intervene patients,the decisions made can bring benefits to patients.ConclusionsBased on the risk factors of relapse among drug abusers identified in this study,we developed a relapse risk assessment tool with 11 items,namely:age,gender,living situation,employment status,recent smoking status,repeated entry MMT clinics in the past,maintenance treatment compliance,co-occurring psychiatric disorders,family history of drug abuse,relationship with family,and frequency of seeing drug-using friends in past 1 month.The total score of the tool ranges from 0 to 81 points,and the patients with a total score?31 points were identified as high-risk groups for relapse.The relapse risk assessment tool is suitable for MMT.It has been proved to have high predictive efficiency,which is helpful to find high-risk groups of relapse in time,and take comprehensive and individual intervention measures in time to delay or even avoid the occurrence of relapse.Highlights1.We developed the first relapse risk assessment tool for MMT patients in China.The tool is simple and easy to operate,with high sensitivity and specificity,and can effectively identify high-risk patients of relapse in MMT clinics.2.Relapse risk assessment tool can evaluate the risk level of relapse and take corresponding prevention and treatment strategies according to different levels,so as to improve methadone clinic intervention system,reduce relapse and improve MMT compliance,which is of important public health significance.