Table 2.

Health and voting in the 1958 and 1970 cohorts: main effects.

		Model 1 Separately		Model 2 Together		Model 3 + Birth		Model 4 + Ages 23/30		Model 5 + Time-varying
		OR	95%CI	OR	95%CI	OR	95%CI	OR	95%CI	OR	95%CI
NCDS 1958	Self-rated health
	Fair	0.73	(0.68–0.80)	0.74	(0.68–0.81)	0.76	(0.70–0.83)	0.82	(0.75–0.91)	0.85	(0.77–0.94)
	Poor or worse	0.61	(0.52–0.70)	0.62	(0.53–0.72)	0.67	(0.57–0.78)	0.75	(0.63–0.89)	0.83	(0.69–1.00)
	(Good or better = ref.)
	Limitations in everyday activities
	Limited (No = ref.)	0.81	(0.74–0.90)	0.97	(0.88–1.08)	0.96	(0.87–1.07)	1.02	(0.91–1.15)	1.11	(0.98–1.25)
BCS 1970	Self-rated health
	Fair	0.67	(0.60–0.76)	0.68	(0.60–0.77)	0.68	(0.59–0.78)	0.77	(0.67–0.88)	0.82	(0.72–0.95)
	Poor or worse
	(Good or better = ref.)	0.51	(0.41–0.63)	0.52	(0.41–0.66)	0.54	(0.42–0.69)	0.65	(0.50–0.84)	0.68	(0.52–0.90)
	Limitations in everyday activities
	Limited (No = ref.)	0.79	(0.67–0.92)	0.95	(0.80–1.13)	0.95	(0.78–1.15)	1.02	(0.84–1.24)	1.07	(0.87–1.31)

Estimates represent odds ratios (OR) from weighted random-effects logistic models. Bolded estimates are statistically significant at the p < .05 level. Models 1–5 controlled for age. Model 3 included: gender, region, mother's age, mother's smoking, mother's weight, father's social class, birth weight. Model 4 included: intention to vote in the next election, educational attainment. Model 5 included: social class, employment status, parenthood, marital status, housing tenure.