A protective personal factor against disability and dependence in the elderly: an ordinal regression analysis with nine geographically-defined samples from Spain

Table 4 Summary of ordinal logistic regression analyses for disability and dependence

	WHODAS 2.0 (n = 377)			Extended Katz index (n = 377)
	OR	95% CI	p	OR	95% CI	p
Target predictors
Comprehensibility (Low)
Mid	0.40	0.22–0.70	0.002	0.82	0.47–1.44	ns
High	0.39	0.21–0.74	0.004	0.79	0.43–1.47	ns
Manageability (Low)
Mid	1.30	0.72–2.36	ns	1.95	1.09–3.51	0.025
High	0.76	0.40–1.44	ns	1.31	0.70–2.46	ns
Meaningfulness (Low)
Mid	0.63	0.36–1.10	ns	0.88	0.50–1.53	ns
High	0.33	0.19–0.58	0.000	0.50	0.29–0.87	0.014
Social factors
Availability of confidant (No)
Yes	0.56	0.31–1.01	0.054	0.87	0.49–1.53	ns
Social contacts (Once a month or less)
Biweekly	1.12	0.54–2.31	ns	1.09	0.54–2.19	ns
Weekly	0.99	0.53–1.88	ns	0.78	0.42–1.45	ns
Everyday	0.70	0.36–1.36	ns	0.70	0.37–1.33	ns
Environmental factors
Social resources accessibility (Very difficult)
Difficult	1.44	0.30–6.91	ns	0.23	0.06–0.86	0.029
Not difficult nor easy	0.18	0.04–0.84	0.029	0.14	0.04–0.56	0.006
Easy	0.06	0.02–0.20	0.000	0.07	0.02–0.22	0.000
Very easy	0.04	0.01–0.13	0.000	0.03	0.01–0.99	0.000
Municipality size (1–10,000 hab.)
> 10,000 habitants	0.50	0.30–0.84	0.003	1.84	1.11–3.05	0.018

Notes. Models adjusted for sex, age, and morbidity. Reference category in parenthesis. Data were split by tertiles in order to obtain low, mid, and high levels in each of the SOC domains. WHO-DAS-II outcome levels: no problem, mild, moderate or above. Extended Katz Index output levels: independent, instrumental activities dependent, basic activities dependent. No significant departures from the proportional odds assumption were detected for any outcome (p > 0.2). ns = p > 0.05

ISSN: 1471-2318