Analyzing musculoskeletal neck pain, measured as present pain and periods of pain, with three different regression models: a cohort study

Table 4 Marginalmodels – analyses of pain at present.

Explanatory variables from baseline	Exposed cases	Simple model		Adjusted model			Proportion ratio
		Odds ratio	P-value	Odds ratio	P-value	95% CI
Women/men	160	2.9	< 0.001	2.6	< 0.001	2.02; 3.41	2.4
Overweight	22	1.1	0.463				1.1
Breakfast regularly	168	0.64	0.001	0.77	0.058	0.593; 0.995	0.71
Snuff use	5	-	-				-
Smoking	23	2.2	< 0.001	2.0	0.003	1.35; 3.00	1.8
Physical activity		0.98	0.157	0.99	0.318	0.967; 1.01	0.99
High work/study demands (ref: not too high)	104		< 0.001		0.004
Not affecting home life	71	1.4		1.3		1.06; 1.61	1.3
Affecting home life	45	1.8		1.5		1.16; 1.99	1.6
High home life demands	12	1.5	0.020	1.2	0.243	0.868; 1.77	1.4
Work/study time		1.0	0.266				1.0
Good relationship with superiors	206	0.78	0.139	0.89	0.507	0.624; 1.26	0.82
Good relationship with colleagues	207	0.76	0.175	0.92	0.700	0.614; 1.39	0.81
Computer use pattern (ref: 0)	120		< 0.001		0.012
One 4 h period without a break	24	1.0		1.0		0.748; 1.34	1.0
At least two 4 h periods without a break	76	1.6		1.4		1.11; 1.71	1.4
Asthma	23	1.2	0.360				1.1
Perceived stress	163	1.8	< 0.001	1.6	< 0.001	1.34; 2.01	1.6

All odds ratios are adjusted for gender (using men as the reference category). For the adjusted odds ratios both p-values and 95% confidence intervals (CI) are presented. The total number of respondents varied between N = 737 – 885, because of incomplete data. Estimates were not calculated for explanatory variables with five or fewer exposed cases.

ISSN: 1471-2474