No abstract

IntroductionSchizophrenia is a severe, chronic and disabling mental illness. Non-adherence to medication and relapse may lead to poorer patient function. This randomised controlled study, under the acronym LEAN (Lay health supporter, e-platform, award, and iNtegration), is designed to improve medication adherence and high relapse among people with schizophrenia in resource poor settings.Methods/analysisThe community-based LEAN has four parts: (1) Lay health supporters (LHSs), mostly family members who will help supervise patient medication, monitor relapse and side effects, and facilitate access to care, (2) an E-platform to support two-way mobile text and voice messaging to remind patients to take medication; and alert LHSs when patients are non-adherent, (3) an Award system to motivate patients and strengthen LHS support, and (4) iNtegration of the efforts of patients and LHSs with those of village doctors, township mental health administrators and psychiatrists via the e-platform. A random sample of 258 villagers with schizophrenia will be drawn from the schizophrenic ‘686’ Program registry for the 9 Xiang dialect towns of the Liuyang municipality in China. The sample will be further randomised into a control group and a treatment group of equal sizes, and each group will be followed for 6 months after launch of the intervention. The primary outcome will be medication adherence as measured by pill counts and supplemented by pharmacy records. Other outcomes include symptoms and level of function. Outcomes will be assessed primarily when patients present for medication refill visits scheduled every 2 months over the 6-month follow-up period. Data from the study will be analysed using analysis of covariance for the programme effect and an intent-to-treat approach.Ethics and disseminationUniversity of Washington: 49464 G; Central South University: CTXY-150002-6. Results will be published in peer-reviewed journals with deidentified data made available on FigShare.Trial registration numberChiCTR-ICR-15006053; Pre-results.

No abstract

Background Postpartum depression (PPD) is a serious public health problem. Building a predictive model for PPD using data during pregnancy can facilitate earlier identification and intervention. Objective The aims of this study are to compare the effects of four different machine learning models using data during pregnancy to predict PPD and explore which factors in the model are the most important for PPD prediction. Methods Information on the pregnancy period from a cohort of 508 women, including demographics, social environmental factors, and mental health, was used as predictors in the models. The Edinburgh Postnatal Depression Scale score within 42 days after delivery was used as the outcome indicator. Using two feature selection methods (expert consultation and random forest-based filter feature selection [FFS-RF]) and two algorithms (support vector machine [SVM] and random forest [RF]), we developed four different machine learning PPD prediction models and compared their prediction effects. Results There was no significant difference in the effectiveness of the two feature selection methods in terms of model prediction performance, but 10 fewer factors were selected with the FFS-RF than with the expert consultation method. The model based on SVM and FFS-RF had the best prediction effects (sensitivity=0.69, area under the curve=0.78). In the feature importance ranking output by the RF algorithm, psychological elasticity, depression during the third trimester, and income level were the most important predictors. Conclusions In contrast to the expert consultation method, FFS-RF was important in dimension reduction. When the sample size is small, the SVM algorithm is suitable for predicting PPD. In the prevention of PPD, more attention should be paid to the psychological resilience of mothers.