Self-reported health and subsequent mortality: an analysis of 767 deaths from a large Thai cohort study

BMC Public Health

Table 4 Longitudinal outcomes (survival and mortality 2009–2012) by change in self-reported health (SRH) ^† in the Thai cohort study, 2005-2009

Change in self-reported health 2005–09 (ΔSRH)	2012 Survival status (%) by ΔSRH^††		Hazard ratios (95% CI) for mortality from 2009 to 2012 by ΔSRH
Change in self-reported health 2005–09 (ΔSRH)	Survived (n = 52859)	Died (n = 247)	Total (n = 247)	Males (n = 164)	Females (n = 83)
All good	91.1	87.6	Ref	Ref	Ref
Better	3.5	3.2	0.86 [0.42-1.77]	0.33 [0.08-1.34]	1.94 [0.83-4.50]
Worse	4.4	7.2	1.57 [0.96-2.56]	0.96 [0.44-2.06]	2.70 [1.41-5.20]***
All poor	1.0	2.0	1.50 [0.60-3.70]	1.07 [0.26-4.46]	2.18 [0.63-7.62]

^†SRH = self-reported health; six ordinal categories reduced to two ─ “good” and “poor”. “Good” included excellent, very good, good, and fair; “poor” included poor and very poor.
^††Column percentage.
p < 0.05, p < 0.01, ***p < 0.005; hazard ratios compare mortality over the follow-up period by change in SRH over the previous period. All three models include age category (5 groups), household income in 2009 (3 categories), health insurance status in 2005, urban or rural residence in 2009, smoking (2009), drinking (2009), physical activities (2009), ischemic heart disease (2009), cancer (2009), hypertension (2009), and diabetes (2009); for total population, models include sex as well.

ISSN: 1471-2458