Profiles of physician follow-up care, correlates and outcomes among patients affected by an incident mental disorder

Table 4 Characteristics and comparisons of adverse outcomes by profiles (N = 170,957)

Adverse outcomes (measured 12 months after the end of the mental health [MH] follow-up care period)	Profile 1: Follow-up care by usual psychiatrist	Profile 2: Low MH follow-up care but high prior consultations for physical reasons	Profile 3: Follow-up care by general practitioners (GP) and psychiatrists	Profile 4: High follow-up care by usual GP and prior consultations for physical reasons	Profile 5: Low MH follow-up care and prior consultations for physical reasons	Total
Group size: N (%)	2,243 (1.3)	85,391 (49.9)	19,155 (11.2)	39,574 (23.2)	24,594 (14.4)	170,957 (100.0)
	%	%	%	%	%	n	%
Frequent emergency department (ED) use (at least 4 visits/year) for any reason ^a	6.06^2,3,4,5	2.18^1,3,4,5	3.86^1,2,4,5	2.89^1,2,3,5	2.91^1,2,3,4	4,594	2.69
Hospitalization for MH reasons	18.55^2,3,4,5	3.13^1,3,4,5	8.43^1,2,4,5	4.16^1,2,3,5	5.62^1,2,3,4	7,732	4.52
Death by any cause	0.88^2,3,5	3.96^1,4,5	2.46^1,4,5	1.72^2,3,5	3.03^1,2,3,4	5,489	3.21
Death by cause: ^b Accidental or intentional	25.00	7.43	10.56	8.66	11.44	466	8.49
Other causes	75.00	92.57	89.44	91.34	88.56	5,023	91.51

Superscript numbers (e.g., 1, 2) indicate significant differences between profiles at p < 0.05 from logistic regression adjusted by sex and age. The percentages found in this table (Table 4) are not adjusted by sex and age – for adjusted results, see Fig. 1. In the article, the adverse outcomes adjusted by age and sex only are presented in the Results section, and further explored in the Discussion section
^a Because of the high number of missing diagnoses justifying ED use in the database, all ED visits were considered
^b The percentage of death by cause are of those who died. Data were too small to consider the differences between profiles of follow-up care and proportion of death related to specific causes
For definitions of the variables included in the study, see the Methods section. For the related health databases linked with each variable, see Table 1. Details on diagnostic codes are presented in Appendix 1

ISSN: 2731-4553