[Tobacco smoking and psychiatric intensive care unit: Impact of the strict smoking ban on the risk of violence].

BACKGROUND

Tobacco smoking is the main cause of death among mentally ill persons. Since February 2007, smoking has been strictly forbidden in French covered and closed psychiatric wards. The fear of an increased violence risk induced by tobacco withdrawal is one of the most frequent arguments invoked against this tobacco ban. According to the literature, it seems that the implementation of this ban does not imply such a risk. All these studies compared inpatients' violence risk before and after the tobacco ban in a same psychiatric ward.

AIM

We aimed to analyse the strict tobacco withdrawal consequences on the violence risk in a retrospective study including patients hospitalised in a psychiatric intensive care unit of the university hospital of Caen during the same period.

METHODS

We compared clinical and demographic data and the violence risk between the smoker group (strict tobacco withdrawal with proposed tobacco substitution) and the non-smoker group (control group). In order to evaluate the violence risk, we used three indicators: a standardised scale (the Bröset Violence Checklist) and two assessments specific to the psychiatric intensive care setting ("the preventing risk protocol" and the "seclusion time"). The clinical and demographic data were compared using the Khi2 test, Fisher test and Mann-Whitney test, and the three violence risk indicators were compared with the Mann-Whitney test. Firstly, comparisons were conducted in the total population, and secondly (in order to eliminate a bias of tobacco substitution) in the subgroup directly hospitalised in the psychiatric intensive care setting. Finally, we analysed in the smoker group the statistical correlation between tobacco smoking intensity and violence risk intensity using a regression test.

RESULTS

A population of 72 patients (50 male) was included; 45 were smokers (62.5%) and 27 non-smokers. No statistically significant differences were found in clinical and demographic data between smoker and non-smoker groups in the whole population, as well as in the subgroup directly hospitalised in the psychiatric intensive care setting. Whatever the violence risk indicators, no statistically significant difference was found between the smoker group and the non-smoker group in the total population, as well as the subgroup directly hospitalised in the psychiatric intensive care setting. Moreover, no correlation was found between the tobacco smoking intensity and the violence risk intensity in the smoker group.

CONCLUSION

Strict tobacco withdrawal does not appear to constitute a violence risk factor in psychiatric intensive care unit inpatients. However, further studies are needed to confirm these results. They should be prospective and they should take into account larger samples including patients hospitalised in non-intensive care psychiatric wards.