Table 5.

Logistic regression showing associations between GP referral to secondary care (model 1) and GP cooperating with another municipal social or care service (model 2) in the year after index consultation, and the initial diagnosis, parental education, and GP characteristics

	Model 1 Referrals		Model 2 Collaborative contacts

	OR	95% CI	OR	95% CI
Female (ref: male)	1.270^a	1.11 to 1.45	1.183^b	1.02 to 1.37

Diagnosis at first consultation
Sleep disturbance	1	1.00 to 1.00	1	1.00 to 1.00
Internalising problems	3.784^a	2.82 to 5.08	1.275	0.96 to 1.69
Externalising problems	3.857^a	2.79 to 5.33	1.725^a	1.26 to 2.36
Attention problems	2.732^a	1.99 to 3.76	1.412^b	1.04 to 1.92
Psychological symptom, other, P29	2.907^a	2.09 to 4.05	1.536^c	1.11 to 2.13
Miscellaneous	2.922^a	2.03 to 4.21	1.978^a	1.38 to 2.83

GP consultations 1 year prior to index episode
0	1	1.00 to 1.00	1	1.00 to 1.00
1–2	0.981	0.85 to 1.13	1.102	0.94 to 1.29
≥3	0.984	0.83 to 1.17	1.523^a	1.26 to 1.84

Length of parental education, years
Mother
≤10	1	1.00 to 1.00	1	1.00 to 1.00
11–13	1.003	0.87 to 1.16	0.860	0.74 to 1.01
>13	1.119	0.95 to 1.32	0.853	0.71 to 1.03

Father
≤10	1	1.00 to 1.00	1	1.00 to 1.00
11–13	1.035	0.90 to 1.19	0.682^a	0.59 to 0.79
>13	1.015	0.83 to 1.24	0.646^a	0.52 to 0.81

GP characteristics
Age	0.984^a	0.98 to 0.99	0.995	0.99 to 1.00
Male GP (ref: female)	0.998	0.86 to 1.16	0.971	0.83 to 1.14
GP approved specialist	1.069	0.91 to 1.26	0.811^b	0.68 to 0.96
List size, per 100	1.031^c	1.01 to 1.05	0.957^a	0.94 to 0.98

GP practice variables
Consultations with P-diagnosis (psychological diagnosis), %
<8.7	1	1.00 to 1.00	1	1.00 to 1.00
8.7–11.3	1.164	0.92 to 1.47	1.159	0.91 to 1.48
>11.3–14.4	1.382^c	1.10 to 1.73	1.124	0.89 to 1.42
>14.4	1.313^b	1.05 to 1.64	1.006	0.79 to 1.28

Percentage of long consultations	0.999	0.99 to 1.00	1.002	1.00 to 1.01

↵a P<0.001.
↵b P<0.05.
↵c P<0.01.
OR = odds ratio.