BJPsych Bulletin
Assaults by patients on psychiatric trainees: frequency and training issues
Guido Pieters, Ewoud Speybrouck, Véronique De Gucht, Staf Joos



To conduct a survey on the frequency and severity of assaults by patients on psychiatric trainees, and the availability of training on managing violent patients; a self-report questionnaire developed in the USA was adapted to the Flemish situation. Data were collected from 99 psychiatric trainees from the Dutch speaking part of Belgium, representing a 60% response rate.


As many as 56% of the respondents had been confronted with at least one physical assault by a patient during their residency, whereas 72% had already been threatened by a patient. Only a small minority had received any training related to patient violence.


Some formal teaching or training in managing (potentially) violent patients should be incorporated in psychiatric training in Flanders.

The risk of being confronted with patient violence is more than twice as high in psychiatric residents than in residents from other medical disciplines (39% v. 16%) (Milstein, 1987). Overall, the reported prevalence rates of physical assaults by patients on psychiatric trainees vary between 36% and 56% (Black et al, 1994; Schwartz & Park, 1999). Verbal threats occur even more frequently, with prevalence figures ranging from 48% (Fink et al, 1991) to 96% (Black et al, 1994). The existing studies have been conducted either in the USA (Ruben et al, 1980; Milstein, 1987; Gray, 1989; Fink et al, 1991; Black et al, 1994; Schwartz & Park, 1999) or in Canada (Chaimowitz & Moscovitch, 1991). No single study has looked at frequency data on assaults by patients on psychiatric residents in Europe. Data on the availability of training courses on risk assessment and management of (potentially) violent patients are scarce. In an American study, 72% of the respondents had received training on management of violent patients, and only 33% of them perceived this training to be adequate (Schwartz & Park, 1999). In a survey of Canadian residents, 51.5% of the respondents had received practical training in dealing with violent patients; 24% considered this training to be adequate (Chaimowitz & Moscovitch, 1991).

The objective of the present study was to conduct a survey on the frequency and severity of physical assaults as well as verbal threats by patients on psychiatric trainees, and the availability of training on managing violent patients, in the Dutch speaking part of Belgium.



In March 2000, a survey was mailed to all (164—65 male, 99 female) psychiatric trainees from the Dutch speaking part of Belgium, who had been identified through the mailing list of the Flemish Training Committee for Psychiatry. For anonymity, a random number was assigned to each of the addresses on the mailing list. This number was noted on the questionnaire before sending it to the appropriate person. After 3 months, the questionnaire was addressed a second time to non-respondents. All subjects received a brief cover letter, a survey and a return envelope.

Study measures

The short survey developed by Schwartz & Park (1999) was slightly changed and translated into Dutch. It is a multiple choice questionnaire, using examples to rate the severity of physical harm caused by assaults. Some questions were omitted, such as length and weight of the respondent, and questions about specific body parts the physical assaults were directed at. In addition, at some points a distinction was made between assaults during the whole training and assaults in the course of the previous year.

Statistical analyses

All statistics were performed with the Statistical Package for the Social Sciences, version 9.0 for Windows. For the most part descriptive statistics were conducted. To study the relationship between different categorical variables, χ2 analysis was used. The level of significance was set at 0.05 (two-tailed).


Response rate

Ninety-nine psychiatric trainees (50 male, 49 female) completed the questionnaire, representing a response rate of 60%. The response rate was 77% for male trainees and 49% for female trainees. Nineteen per cent of respondents were in their first year of training (n=19), 23% in their second year (n=23), 13% in their third year (n=13), 24% in their fourth year (n=24), and 20% were in their fifth year of training (n=20). The median length of psychiatric training was 3 years.

Physical assaults

Table 1 describes the frequency, severity and nature of physical assaults. No significant differences were found between men and women (χ2=0.02, d.f.=1, P=0.89), or between more or less experienced trainees (χ2=7.3, d.f.=4, P=0.12) with respect to the incidence of physical assault in the course of residency.

View this table:
Table 1.

Frequency and characteristics of physical assaults, verbal threats and of training received on topics of aggression in Flemish psychiatric trainees


Table 1 describes the frequency and nature of threats that trainees received during residency. All 71 of those receiving threats had received verbal threats face-to-face. No differences were found between male and female respondents (χ2=0.69, d.f.=1, P=0.41) or between more or less experienced trainees (χ2=2.6, d.f.=4, P=0.62). There was a significant positive association between physical assaults and threats (χ2=11.17, d.f.=1, P<0.001).

Preceding events

Both in the cases of physical assaults and verbal threats, ‘setting a limit’ appeared to be the most frequently occurring preceding event, immediately followed by ‘refusing to meet a patient’s request’.

Training issues

Table 1 describes the frequency and content of training in managing violent patients during residency. Only 1 respondent out of 7 (14%, n=3) felt that this training was adequate. Of those respondents who had not received training during residency, 9% (n=7) had sought training outside of their training scheme. A total of 4 respondents out of 99 (4%) reported seeking self-defence training to be better prepared to deal with violent patients. As far as safety procedures were concerned, none of the respondents indicated the presence of metal detectors, whereas 67% (n=66) indicated that most doors to consulting rooms opened inward.

A large majority of the respondents considered a seminar, developed especially for psychiatric trainees, to be (very) beneficial. This seminar should include basic techniques to avoid serious harm, information about violence in psychiatry, and interviewing techniques for violence-prone patients. On average, respondents would be willing to spend 7 h in such a seminar.


The results of this study indicate that more than half of the respondents were physically assaulted by a patient and almost 3 out of 4 respondents were threatened by a patient during their residency. These figures should be interpreted with caution, because 40% of trainees did not return the survey. Even if we assume that all non-respondents were not confronted with aggression, at least 33% of Flemish psychiatric trainees were physically assaulted and 43% were threatened during their training. Interestingly, no differences were found between trainees in their earlier years and those in their later years of psychiatric training. This seems in accordance with findings of a study that demonstrated that experienced psychiatrists were as likely to be victims of assault as were inexperienced psychiatrists (Dubin et al, 1988). The fact that we found no difference in confrontation with violent incidents between men and women should be interpreted in the light of the different response rates between sexes. If we suppose that confrontation with violence would make one more inclined to return the questionnaire, then the higher response rate of male trainees could point to the fact that they are more often confronted with violence. If we assume that non-respondents were not involved with violent incidents, then 27% of women and 44% of men were confronted with aggression.

When comparing these results with those from previous studies conducted in the USA and Canada, it becomes clear that our respondents score in the upper range as far as the occurrence of both physical assaults and verbal threats are concerned. These differences might be accounted for both by methodological differences between studies and by differences in organisation of mental health systems and work environments for trainees between Flanders and North America. Psychiatric training in Flanders is to a large extent confined to in-patient settings. The lack of information on the patients involved in aggressive incidents is a weakness of this study.

With respect to the perceived reasons for being assaulted, most of the respondents attribute the assaults to setting limits on the patient, or refusing to satisfy a patient’s request, which is very similar to the findings in other studies (Ruben et al, 1980; Powell et al, 1994; Schwartz & Park, 1999). Only 21% of the respondents in the present study report some formal teaching or training in managing violent patients, a percentage that is considerably lower than those reported elsewhere (Chaimowitz & Moscovitch, 1991; Schwartz & Park, 1999). On the basis of the existing studies, training should comprise the following aspects:

  • risk assessment,

  • practical skills for managing violent patients,

  • the (psycho)dynamics of aggression.

The importance of training subjects in self-defence should not be underestimated, as two studies have shown that this kind of training results in subjects being involved in significantly fewer aggressive incidents (Infantino & Musingo, 1985; Phillips & Rudestam, 1995). The training course proposed by Schwartz & Park (1999), involving both lectures on patient violence and practical exercises, is an excellent example of how this kind of training could be organised. In addition to the fact that it taps into a wide range of topics related to patient violence, it has the advantage of being limited in time, necessitating a total of 8–11h, which is congruent with the amount of time the respondents of the present study are willing to invest in a seminar developed especially for psychiatric trainees.

Future research on this topic should use prospective designs to exclude recall bias and should incorporate valid measures of the impact of aggression on trainees. In addition, researchers should make an effort to collaborate on international surveys, thus allowing comparisons of different psychiatric training systems as well as different healthcare systems.


View Abstract