Comparing baseline characteristics among respondents (study participants) versus non-respondents (including those with missing data of covariates or those without any follow-up data) was performed using Student’s t-test for continuous variables and chi-square test for categorical variables.

Baseline characteristics across weight change categories are illustrated as mean ± standard deviation (SD) for continuous variables and number (%) for categorical variables.

The multivariable Cox proportional regression analysis was applied to evaluate the association of weight change categories with incident CVD/CHD by reporting Hazard ratios (HRs) with 95% confidence intervals (CIs) in two models: Model 1: adjusted for age and sex; Model 2: further adjusted for BMI, educational level, current smoking (at first follow-up), GLDs use (at baseline or first follow-up), family history of premature CVD, hypertension, hypercholesterolemia, CKD, and FPG.

To address the low power and possibility of bias caused by missing data, as a sensitivity analysis, we examined the impact of weight change on incident CVD/CHD with imputed baseline and first follow-up missing data for covariates using stochastic single imputation with predictive mean matching (PMM) [24, 25].

Interactions of weight change categories with age groups (≥ 60 years versus < 60 years), sex (men versus women), BMI groups (≥ 30 kg/m2 versus < 30 kg/m2), and GLDs use (yes versus no) were checked by the log–likelihood ratio test in Model 2 and HRs with 95% CIs calculated for each subgroup.

Time to event was considered as the time of censoring or the outcome (incident CVD/CHD) occurring, whichever came first. We censored subjects in the case of leaving the district, lost to follow-up, or being without any event in the study until 20 March 2018.

Using the Schoenfeld residual test, the proportionality in the Cox model was evaluated; all proportionality assumptions were appropriate. STATA version 14 (StataCorp LP, College Station, Texas) was employed for statistical analyses. A P-value of < 0.05 and P-value for interaction of < 0.2 was considered to be statistically significant.

Note: The content above has been extracted from a research article, so it may not display correctly.

Please log in to submit your questions online.
Your question will be posted on the Bio-101 website. We will send your questions to the authors of this protocol and Bio-protocol community members who are experienced with this method. you will be informed using the email address associated with your Bio-protocol account.

We use cookies on this site to enhance your user experience. By using our website, you are agreeing to allow the storage of cookies on your computer.