Effectiveness of integrated management on hypertension and mortality in rural China: A CHHRS study

Summary A study was conducted to investigate whether an integrated management (IM) model led by public healthcare providers is effective in reducing cardiovascular disease (CVD)-specific and all-cause mortality rates in low-income rural populations with hypertension. The study recruited 14,234 patients with hypertension aged 18 years or older and allocated them to either an IM group or a usual care (UC) group. During a median follow-up of 48.0 months, the incidences of CVD-specific and all-cause deaths were lower in the IM group than in the UC group. The hazard ratios for CVD-specific mortality and all-cause mortality among patients in the IM group were 0.60 and 0.62, respectively. The results showed that the IM model led by public health providers resulted in clinically significant reductions in CVD-specific and all-cause mortality rates in low-income rural populations with hypertension.


Specific prescriptions
The antihypertensive effect of follow-up   Shown is the analysis using all the original data.b Shown is the analysis using multiple imputed data.
c Shown is the analysis using the Fine-Gray competing-risk regression to estimate the subdistribution HR (SHR), considering CVD mortality and non-CVD mortality as competing events.d Shown is a random selection of data from 10 townships for internal validation.
e Shown is after excluding participants died within the first six months of follow-up.f Shown is the hazard ratio from the multivariable Cox proportional-hazards model, with additional adjustment for age, sex, BMI, WC, Current smoking, Current alcohol drinking, SBP, DBP, HR, diabetes mellitus, stroke, CHD, AF, HCY, TCHO, TG, LDL-C, HDL-C, UA, eGFR, AST, ALT, GGT, Creatinine, Hypoglycemic drugs, Lipid-lowering drugs, Antiplatelet drugs, Antihypertensive drugs, and laboratory tests on presentation.The analysis included all 14232 patients.g Shown is the primary analysis with a hazard ratio from the multivariable Cox proportional-hazards model with the same covariates with inverse probability weighting according to the propensity score.The analysis included 28504 patients (14232 who received Integration management and 14272 who did not).h Shown is the hazard ratio from a multivariable Cox proportional-hazards model with the same covariates with matching according to the propensity score.The analysis included 9172 patients (4586 who received Integration management and 4586 who did not).i Shown is the hazard ratio from a multivariable Cox proportional-hazards model with the same strata and covariates, with additional adjustment for the propensity score.The analysis included 9172 patients (4586 who received Integration management and 4586 who did not).

Table S4. Hazard ratios (95% CIs) for the composite endpoint for all variables included as covariates in the Cox multivariable model with inverse probability weighting by the propensity score*, related to
Figure S1.Flow chart of the study participants, related to STAR Methods * The high rate of follow-up in this study (as of August 2022) can be attributed to the recent establishment of a complete COVID-19 virus nucleic acid and Vaccine Information Registration System in China.As a result, the accuracy and completeness of the resident information registration rate reached almost 100%, making it easier to track and monitor participants throughout the study.Abbreviations: CDC, Centers for Disease Control and Prevention.

Figure S2 .
Figure S2.Management protocol of the study , related to STAR Methods

Figure S5 .
Figure S5.Scatter distribution of the estimated propensity score for receiving integration management, among patients who did and did not actually receive the treatment, related to STAR Methods

Table 2 .
This Cox multivariable model, which was inverse probability weighted by the propensity score, was designed to control for potential confounding-by-indication of the exposure (receiving integration management) and composite outcome.It was additional stratified on sex, chronic lung disease and BMI, for which parameter estimates are not generated or shown.The results in this table are provided for the reader's information but should be not be interpreted to provide information on predictors or causes of the composite outcome *