Skip to main content
. 2022 Jun 28;62(6):4192–4208. doi: 10.1007/s10943-022-01603-8

Table 3.

Baseline religiosity and odds ratios (95% CI) for incident Parkinson’s disease

Study, analysis Religion very important Religion somewhat important Religion not very important Religion not at all important P value for trend
Pooled
PD cases [n (%)] 11 (0.5) 16 (0.7) 25 (1.0) 22 (0.9)
N 2313 2290 2628 2565
Model 1 1 [reference] 1.62 (0.75, 3.53) 2.44 (1.16, 5.13)* 2.34 (1.07, 5.11)* 0.021*
Model 2 1 [reference] 2.90 (1.20, 7.00)* 7.82 (2.85, 21.47)*** 9.99 (3.28, 30.36)***  < 0.001***
ELSA
PD cases [n (%)] 5 (0.4) 6 (0.4) 21 (1.0) 20 (0.9)
N 1219 1432 2140 2333
Model 1 1 [reference] 1.08 (0.33, 3.57) 2.60 (0.96, 6.99) 2.23 (0.82, 6.08) 0.047*
Model 2 1 [reference] 1.29 (0.34, 4.91) 5.83 (1.37, 24.83)* 6.89 (1.48, 32.01)* 0.006**
MIDUS
PD cases [n (%)] 6 (0.5) 10 (1.2) 4 (0.8) 2 (0.9)
N 1094 858 488 232
Model 1 1 [reference] 2.30 (0.81, 6.51) 1.68 (0.45, 6.21) 1.74 (0.33, 9.10) 0.406
Model 2 1 [reference] 5.53 (1.67, 18.32)** 8.42 (1.68, 42.10)** 13.69 (1.63, 114.97)* 0.003**

ELSA, English Longitudinal Study of Aging; MIDUS, Midlife in the United States study; PD, Parkinson’s disease

Model 1: age, sex and either cohort (pooled), region (ELSA), or sample (MIDUS)

Model 2: Model 1 + ethnicity, education, marital status, smoking status, alcohol consumption, physical activity levels, self-rated health, diabetes, hypertension, mental health conditions, frequency of private religious practices, and frequency of religious service attendance

*P < 0.05

**P < 0.01

***P < 0.001