Recovery from musculoskeletal injury: the role of social support following a transport accident

Table 3 Multiple logistic regression models of the relation between social support and persistent pain

	Persistent pain^a
	OR	95 % CI
Models
1. Marital status^b
Married or in de facto relationship	1.28	0.97–1.69
Widowed/separated/divorced	1.62*	1.11–2.37
Never married (ref)
Hosmer-Lemeshow fit test	X ²(8) = 11.36 p = 0.18
2. Children^b
Yes	0.96	0.77–1.20
No (ref)
Hosmer-Lemeshow fit test	X ²(8) = 3.12 p = 0.92
3. Family composition^b
Married or in de facto relationship with children	1.14	0.78–1.68
Married or in de facto with no children	1.13	0.76–1.68
Widowed/separated/divorced with children	1.49	0.87–2.55
Widowed/separated/divorced with no children	1.41	0.85–2.33
Never married with children	0.84	0.55–1.26
Never married with no children (ref)
Hosmer-Lemeshow fit test	X ²(8) = 10.27 p = 0.25
4. Family^c
Definitely	0.40*	0.24–0.68
Yes, sometimes	0.68	0.39–1.19
No, not at all	0.58	0.31–1.07
Not often (ref)
Hosmer-Lemeshow fit test	X ²(8) = 9.95 p = 0.26
5. Friends^d
Definitely	0.29*	0.17–0.47
Yes, sometimes	0.50*	0.30–0.89
No, not at all	1.14	0.58–2.25
Not often (ref)
Hosmer-Lemeshow fit test	X ²(8) = 12.23 p = 0.14
6. Neighbours^c
Definitely	0.59*	0.41–0.84
Yes, sometimes	0.78	0.54–1.15
No, not at all	1.27	0.89–1.81
Not often (ref)
Hosmer-Lemeshow fit test	X ²(8) = 5.39 p = 0.72

^aRestricted to those who were at least 3 months post-injury
^bAll models adjusted for gender, age, education, country of birth, residential location, injury types, prior health, days post-injury and hospitalisation
^cAll models adjusted for family composition, gender, age, education, country of birth, residential location, injury types, prior health, days post-injury and hospitalisation
^dDue to problem with model fit, non-significant variables were removed. The final model was adjusted for age, education and days post-injury
*p < .05

ISSN: 1477-7525