Fig 1.
A flow chart of data preprocessing and analysis using the 2019 Behavioral Risk Factor Surveillance System (BRFSS) dataset including the 2019 cognitive decline module.
Table 1.
a: Socio-demographic and behavioral characteristics of male adults aged 45 and older as a function of subjective cognitive decline status (SCD); Behavioral Risk Factor Surveillance System Data 2019.
b: Health-related and comorbid characteristics of male adults aged 45 and older as a function of subjective cognitive decline status (SCD); Behavioral Risk Factor Surveillance System Data 2019.
Fig 2.
(a) The propensity score distribution for all racial/ethnicity groups and (b) The propensity score distribution for the Black or African-American groups. The left density plot shows before the PSM with the full sample, and the right density plot shows after PSM with the matched sample of veterans and non-veterans. The blue-color and peach-color density plots represent veterans and non-veterans, respectively.
Fig 3.
Distribution of propensity scores between matched and unmatched treated (veteran) and control (non-veteran) units.
(a) All (White, Black or African-American, and Hispanic or Latin-American) males and (b) Black or African-American males. No unmatched treated units are displayed due to unbalanced data.
Table 2.
Standardized mean differences between the treated (veterans) and control (non-veterans) groups for before and after propensity score matching using the 1:1 nearest neighbor method of matching.
Table 3.
Odds ratio (OR) of the binary logistic regression model fitted before and after propensity score matching for the all (White, Black or African-American, and Hispanic or Latin-American) participants who responded yes to the subjective cognitive decline module in BRFSS 2019 data.
Table 4.
Odds ratio (OR) of the binary logistic regression model fitted before and after propensity score matching for Black or African American male participants who responded yes to the subjective cognitive decline module in BRFSS 2019 data.