Addi van Bergen

Supplementary material chapter 2 171 Confounding, match-ing & stratification** etc. Results per indicator*** Correlations and effect estimates per indicator $ Combined result $$ Methodological limitations # Study quality ## Confounder: gender, age. age 2 , ethnicity, migrant, marital status, job status, educational level, social class, region and (transitions in) SAH and LLTI. c) Mediator/moderator: rural vs urban; car access, mobile phone ownership, internet use Effect on SP Transition from: -Low to high: + -High to low : ns -Stable high : + Effect MH1 on SP (wave 3) MH1 ß (95% CI) -Stable low is reference -Low=>high 0.28 (0.01-0.54) ^ -Stable high 0.91 (0.64-1.18) ^^ + 2 3 5 + ^ p<.05; ^^ p<.01 + SP Effect SP on MH2 (wave 4) SP: OR adj =1.07 (1.02-1.13) Other factors in model: physical health, environment, social relations. Not in model: P1 C. + S P2 ns E P1 C MH ⇦ S: ß=-0.77 MH ⇨ P2: ß=-0.58 (Model fit criteria: RMSEA<0.05; GFI>0.90 & CFI>0.90) +/0 4 7” + + S E1 E2 P1 P2 P3 P4 P5 C1 ns C2 (p<.001) Kendall tau_b S:0.23; E1:0.20; E2:0.18; P1:0.05; P2:0.09; P3:0.16; P4:0.05; P5:0.11; C1:0.09 + 2 4 5 6 7 + Stratified by gender. Covariate: D=disability other than mental health + S E g) ♀ D*S: p < .05; D*E: p < .05; S*E: p < .05; D*S* E: p < .05 + 1 2 3 4 5” 6 + + S E g) ♂ D*S: p < .05; D*E: p < .05; S*E: p < .05; D*S* E: p < .05 + 1 2 3 4 5” 6 Adj for gender, age, migrant status and education + S ns P S*P S: ß=-0.339 p<.001 + 1 3 4 5” 7” + + S P S: ß=-0.154 p<.05 P: ß=-0.197 p<.01