Abstract
Background: Dementia, diabetes, chronic kidney disease and cardiovascular disease (CVD) share common pathophysiology, possibly through metabolic stress. Multifactorial individual-level interventions including lifestyle modification have been reported to reduce risk of diabetes and CVD, mainly in high risk population and many done in clinic settings. Attempts to prevent dementia / Alzheimer’s disease (drugs and supplements) in individuals have not been successful. Little evidence exists to describe the effect of community-based lifestyle interventions on risk of dementia and such NCDs in intermediate risk middle-age and elderly populations. The objective of the project in Year 3 was to investigate the impact of community-based lifestyle modification program on health behaviors, physiologic and biochemical variables and the risk of diabetes, chronic kidney disease and CVD over 2 years of follow-up. This study also aimed to examine the economic impact of such the intervention. Methods: In this cluster randomized controlled trial, 60 villages were selected using multistage stratified random sampling from 25 districts of Ubon Ratchathani. Men and women aged 45-75 years without the above preexisting diseases were invited to participate. A sample size of 4,000 individuals is required to attend screening at baseline to assure approximately 3,600 individuals enrolled in the study (1,800 individuals in the Intervention and Control groups). Main interventions to enhance physical activity and healthy diet were developed and tested before enrollment of participants. After baseline assessment, multi-module multi-level Interventions to address four health behaviors (physical activity, diet, smoking and alcohol consumption) at four different levels (individual, household, group/ knowledge management and community levels) were given in the Intervention group. Site monitoring visits were performed to assure the quality of data collection and intervention. Primary outcome was incident dementia at 10 years. In Year 3 (Follow-up Year 2), we investigated the impact of the above interventions on health behaviors, physiologic and biochemical variables and the risk of diabetes, chronic kidney disease and CVD, reporting relative risk (95%confidence interval) and relative risk reduction. Intention-to-treat data analysis was performed. Incremental cost-effectiveness ratio (ICER) were calculated. Results: Based on multi-stage stratified random sampling (described in Year 1), a total of 60 villages were selected for this trial and randomly divided into two groups: Intervention group which received the above-mentioned 4x4 lifestyle modification intervention and Control group which received healthcare in normal practice. In Year 3 refresher course for key research assistant team from 60 villages, which included 60 nurses and health care officers, 120 health volunteers and 60 community leaders, were undertaken. Enrolment started on 1st March 2016 and, up to July 2017, a total of 3,983 individuals aged 45-75 years were screened, with 3,505 participants meeting the inclusion/ exclusion criteria. There were 1,767 and 1,738 participants in the Intervention and Control groups respectively. Participants in both groups were comparable according to their socio-demographic characteristics, personal and family history, and key laboratory results (e.g. fasting plasma glucose, HbA1c, and LDL cholesterol). Due to unexpected changes in central laboratories and severe flooding, Year 3 data collection were delayed. A total of 3,314 participants (94.5%) attended Year 3 follow-up, including 1,650 and 1,664 participants in the Intervention and Control groups. Over 2 years of follow-up, participants in the Intervention group reported better health behaviours (physical activity, diet and smoking) than the control group. Although there was no difference in anthropometric indices and physiologic variables, i.e. blood pressure, between the two groups, those in the intervention group had lower fasting plasma glucose and HbA1c than those in the control group (Fasting plasma glucose 84.4±13.0 vs 85.6±14.3 mg%, p=0.015 and HbA1c 5.58 (0.57)% vs 5.63 (0.58)%, p=0.021 respectively) They had lower total cholesterol and triglyceride than those in control group (Total cholesterol 186.6±37.1 vs 189.6±37.1 mg%, p=0.028 and Triglyceride 147.0±80.6 vs 160.0±102.7 mg%, p<0.001) Overall the Intervention and control groups had comparable risk of diabetes. However, in the very high risk defined as having family history of diabetes and pre-diabetes blood sugar levels, participants in the intervention groups had a 37% reduced risk compare to the control groups (2-year incidence of 14% and 19%, relative risk 0.63 (95%CI 0.35-1.09), one-tailed p-value 0.041). In all participants, there was no difference in risk of worsening kidney function (defined as >25% reduction of eGFR) between two groups. In a subgroup analysis in female participants, the intervention group had 59% reduced risk of worsening kidney function (2-year incidence of 1.1% vs 2.6%, Relative risk 0.41 (95%CI 0.20-0.87)). The risk of mild cognitive impairment was lower in the intervention than control groups, with 2-year incidence of relative risk of 2.1% และ 4.0% respectively (relative risk 0.53 (95%CI 0.36-0.78), p=0.003). There was no difference in CVD risk over 2 years. In base case scenario of cost-effectiveness analyses, community-based 4x4 lifestyle modification program resulted in an ICER of 21,196 Baht per one case of diabetes prevented compared to normal practice. The ICER was 145,721 Baht per one case of worsening renal function and 129,530 Baht per one MCI case prevented. Further, a qualitative study suggested that key success factors of the project included community leaders, healthcare officers and common interest and engagement of community members. Participants strongly suggested the project to continue. Discussion: Albeit some delay, key milestones in Year 3 data collection, provision of the designed interventions and outcome ascertainment were achieved. In this short follow-up, With changes in key health behaviours and metabolic variables, this community-based multi-module and multi-level lifestyle modification was effective in prevention diabetes, worsening kidney function and MCI in high risk and some other particular groups and this might be cost-effective. Lessons learned from this community-based disease prevention project may be extended and adapted for other communities in Thailand. Crucially, the sustainability of this kind of project may be enhanced by more community engagement and continuing financial and non-financial supports from local administrative organizations.
