Table 4. Results from logistic regression analysis of the outcome chronic pain at the one-year follow-up.

		Unadjusted			Adjusteda
		OR	CI	p	OR	CI	p
Model 1
n = 1523, n = 1523
Stress	Yes (ref. No)	1.7	1.1–2.6	0.010	1.6	1.0–2.4	0.029
Non-restorative sleep	Yes (ref. No)	1.5	1.0–2.3	0.036	1.5	1.0–2.3	0.037
Inactivity	Yes (ref. No)	1.7	1.1–2.7	0.029	1.8	1.1–3.0	0.016
Model 2				<0.001			<0.001
n = 1523, n = 1523
Interaction variable
Interactions (ref. no stress, no non-restorative sleep, no inactivity)	Only stress	2.3	1.3–4.1	0.007	2.2	1.2–4.0	0.010
	Only non-rest. sleep	1.5	0.7–3.2	0.298	1.6	0.7–3.5	0.228
	Only inactivity	1.7	0.6–5.1	0.360	1.7	0.6–5.3	0.345
	Stress+inactivity	1.4	0.4–5.0	0.575	1.5	0.4–5.2	0.535
	Stress+non-rest. sleep	2.6	1.4–4.7	0.002	2.4	1.3–4.4	0.006
	Inact.+non-rest. sleep	6.2	2.3–17.0	0.001	6.9	2.5–19.2	<0.001
	All three	5.1	2.4–11.1	<0.001	5.2	2.4–11.5	<0.001
Model 3
n = 1523, n = 1523
Stress	Yes (ref. No)	1.7	1.1–2.5	0.016	1.6	1.0–2.4	0.042
Inactivity	Yes (ref. No)	1.7	1.0–2.7	0.039	1.8	1.1–2.9	0.024
Sleep disturbanceb	Yes (ref. No)	0.9	0.5–1.6	0.671	0.8	0.5–1.5	0.518
Tirednessc	Yes (ref. No)	1.9	1.2–2.9	0.004	1.9	1.2–2.9	0.004

^a Included confounders in all models (Model 1-Model 3): gender, education

^b Yes = presence of at least one of “frequent awakenings” and “difficulty falling asleep”.

^c Yes = presence of at least one of “non-restorative sleep” and “tired during the daytime”.

OR = odds ratio, CI = confidence interval. The sample size in the unadjusted and adjusted model is given.