The sex difference in self-rated health among older Turkish and Moroccan migrants in the Netherlands: an exploratory study of contributing determinants

Table 3 Sex difference in exposure to determinants of self-rated health

Variable	Beta sex *	95% CI	% Change Beta sex **
Raw model (Age)	−0.641	−1.08 ─ −0.21	Reference
Sociodemographic
Education level (high/middle vs. low)	−0.550	−1.00 ─ −0.10	−14%
Social
Living situation (together vs. alone)	−0.528	−0.98 ─ −0.08	−18%
Mosque visit (regular vs. some to never)	−0.832	−1.31 ─ −0.35	19%
Societal participation (inactive versus active)	−0.688	−1.14 ─ −0.24	−7.3%
Loneliness (0–21)	−0.684	−1.13 ─ −0.23	6.6%
Religious coping (often vs. some to never)	−0.629	−1.07 ─ −0.19	−2.0%
Social contact frequency (0–32)	−0.642	−1.08 ─ −0.20	0.2%
Lifestyle
Physical activity (Meth/w)	−0.636	−1.08 ─ −0.19	−0.8%
Health-related
Memory complaints (yes vs. no)	−0.679	−1.14 ─ −0.22	5.9%
Depression symptoms (0–60)	−0.483	−0.96 ─ −0.00	−25%
Visual difficulties (yes vs. no)	−0.585	−1.02 ─ −0.13	−0.9%
Hearing difficulties (yes vs. no)	−0.686	−1.14 ─ −0.24	7.0%
Chronic diseases (vs. none)	−0.443	−0.91 ─ 0.02	−31%
Functional limitations (vs. none)	−0.105	−0.61 ─ 0.40	−84%

* Natural exponent of the sex difference (women versus men) in odds on good self-rated health
** Percentage change = 1 – (Beta sex univariate model including the variable / Beta sex raw model)
Note: Bold is significant; p-value Beta sex < 0.05
Note: Univariate analysis; each row represents the raw model including the corresponding variable
Note: A negative percentage change means this variable causes a decrease of the sex difference when adjusted for, so partly explains it
Note: A positive percentage change means this variable causes an increase of the sex difference when adjusted for, so partly suppresses it

ISSN: 1471-2458