Use of the least intrusive coercion at Danish psychiatric wards: A register-based cohort study of 131,632 first and subsequent coercive episodes within 35,812 admissions

OBJECTIVE

Psychiatric legislation in Denmark implies a principle of using the least intrusive types of coercion first. The intrusiveness is not universally agreed upon. We examined the order in which coercive measures during admission were used, implying that the first used should be less intrusive than the following types.

METHODS

For coercive episodes reported to the national administrative register for the period 2011-16, the order of 12 legal coercive interventions during each admission was examined. Comparing with mechanical restraint, the odds ratio (OR) and confidence interval (95%CI) of being first or subsequent used types were estimated using conditioned (96,611 episodes) and unconditioned (131,632 episodes) logistic regression models, stratified on sex.

RESULTS

Totally 17,796 patients aged 18+ were subjected to at least one coercive episode. The median time between admission and the first episode was 4 days in men and 6 for women. For females, involuntary detention, forced feeding, coercive treatment of somatic disorder, locking of doors and close observations in females were used before mechanical restraint, and forced follow-up, involuntary electro convulsive therapy (ECT), forced treatment, use of gloves and straps, physical restraint and forced intramuscular medication was used later. In men, only involuntary detention was used before mechanical restraint, while involuntary ECT, close observations, administration of drugs, use of gloves and straps, physical restraint and forced intramuscular medication was used after mechanical restraint.

CONCLUSION

The order of used coercive measures is not consistent with the international ranking of the least intrusive types, especially in men and in younger adults.

Representativeness of personality and involvement preferences in a web-based survey on healthcare decision-making

Background

Obtaining a sample that is representative of the group of interest is of utmost importance in questionnaire studies. In a survey using a state authorized web-portal for citizen communication with authorities, we wanted to investigate the view of adult men on patient involvement in health care decision-making regarding Prostate-Specific Antigen test for prostatic cancer. In this paper, we report on sample characteristics and representativeness of our sample in terms of personality and baseline involvement preferences.

Methods

We compared personality profiles (BFI-10) and baseline healthcare decision-making preferences (CPS) in our sample (n = 6756) to internationally available datasets. Pooled data from a) US, UK, Canada, Australia, and New Zealand (n = 1512), b) Germany, Netherlands, Switzerland, and Belgium (n = 1136), and c) Norway, Sweden, Finland, and Denmark (n = 1313) were used for BFI-10 comparisons. Regarding CPS, we compared our sample with three previous datasets relating to decision-making in cancer (n = 425, 387, and 199).

Results

Although statistically significant differences particularly appeared in large dataset comparisons, sample BFI-10 and CPS profiles mostly were within the range of those previously reported. Similarity was greatest in BFI-10 comparisons with group a) where no statistically significant difference could be established in factors ‘agreeableness’ and ‘neuroticism’ (p = .095 and .578, respectively).

Conclusion

Despite some variation, our sample displays personality and baseline preference profiles that are generally similar to those described in previous international studies. For example, this was the case with the BFI-10 ‘agreeableness’ measure (incl. trust and fault-finding items), an important factor in healthcare decision-making.