Background: There are regional differences in the Netherlands in the numbers of emergency compulsory admissions (Inbewaringstelling: IBS). We looked at three 24/7 facilities to investigate the relationships between patient and consultation characteristics on the one hand, and numbers of emergency compulsory admissions on the other, against the background of the level of urbanisation.
Method: We compared emergency consultations in 18-64 year olds in Apeldoorn, Amsterdam and Rotterdam between 2012 and 2016 in terms of socio-demographic, procedural and clinical characteristics, and in terms of outcome. We used the Severity of Psychiatric Illness Scale (SPI) to determine disorder severity.
Results: Apeldoorn had as many consultations per 100,000 inhabitants as the highly urbanised city of Rotterdam. GPs there referred 68% of patients, compared with 25% in Amsterdam and 50% in Rotterdam. In Apeldoorn, 17% of the patients were psychotic, compared with 35% in the other regions. In addition, 66% of the patients there had a low SPI score, compared with 40% in the large cities. Amsterdam and Rotterdam had 3.5 times higher risk of emergency compulsory admissions as Apeldoorn. After adjustment for socio-demographic, procedural and clinical characteristics, this difference with Apeldoorn was 1.5 for Amsterdam and 2.6 for Rotterdam. SPI score and psychotic disorder were found to be the most important predictors of IBS admission.
Conclusion: Differences in consultation numbers, referral patterns and the location of consultations indicate that there are regional differences in the position of the 24/7 facility in the mental health care system. The numbers of emergency compulsory emissions were related in part to the level of urbanisation and the associated epidemiological differences but probably also to differences in the position of the crisis facility in the mental health care system. Differences in admission numbers were primarily linked to differences in diagnostic characteristics and disorder severity and, to a lesser extent, to referral patterns and socio-demographic characteristics. However, these variables did not explain all the observed inter-regional differences.