Defensive practice among psychiatrists: a questionnaire survey
- 1Stockton Learning Disability Service, Stockton
- 2Medicine, Health Policy and Practice, University of East Anglia, Norwich
- Correspondence to: Dr Kirsty Passmore, Stockton Learning Disability Service, 163 Durham Road, Stockton TS19 0EA; UK;
- Received 4 March 2002
- Accepted 12 August 2002
Objective: There has been little research on the prevalence of defensive practice within hospital settings. The aim of this report was to examine the extent of defensiveness among psychiatrists and to examine the relationship between defensiveness and seniority, as well as the effect of previous experiences on the level of defensiveness.
Design: A postal questionnaire survey on defensive practice.
Setting: Northern Region of England.
Subjects: 154 psychiatrists in the region.
Results: 96 responses were received from 48 equivalent consultants, 18 specialist registrars, and 23 equivalent senior house officers. Overall, 75% of those who replied had taken defensive actions within the past month. In particular, 21% had admitted patients overcautiously and 29% had placed patients on higher levels of observations. Junior psychiatrists were particularly prone to practise defensively. Important contributing factors included previous experience of complaints (against colleague or self), critical incidents, and legal claims.
Conclusion: Almost three quarters of the psychiatrists who responded had practised defensively within the past month. The higher propensity of junior trainees to practise defensively may be attributable to their lack of confidence and experience. Experience of complaints (colleague or self) and critical incidents were important factors for defensive practice. Better and more structured training might reduce the high level of defensive practice and the way complaints and investigations are handled should be improved to maintain a truly “no blame” environment conducive to learning from past experience.
A “no blame” and learning culture is essential for the delivery of quality health care,1,2 but the close examination of practitioners’ practice required for a learning culture might result in defensive practice. One definition of defensive practice is the “ordering of treatments, tests and procedures for the purpose of protecting the doctor from criticism rather than diagnosing or treating the patient”.3
It has been proposed that there are both positive and negative aspects of defensive practice.4 Examples of positive aspects might include improvements in the quality of services with more detailed explanations being given to patients and increased patient satisfaction. Examples of negative aspects might include the prescription of unnecessary treatments, increased observation levels of inpatients, and increased rates of follow up.
Although defensive practice has been examined in the USA and Canada,5–8 few such studies had been carried out in the UK until the 1990s. In high risk specialties such as obstetrics, the majority of obstetricians and midwives believe that litigation has caused a rise in defensiveness.9 However, defensive practice is common even among general practitioners, who are commonly regarded as being in a low risk specialty, with 98% claiming to have made some practice changes as a result of the possibility of a patient complaining.4 A survey of all consultants in the Oxford Regional Health Authority in 2000 concluded that “extensive evidence of defensive medicine was not found”,10 but this survey focused only on consultants’ responses to hospital complaints. No surveys have been carried out on defensive practice among doctors in specific low risk hospital specialties in the UK.
This report examines the extent of defensiveness among practitioners in one such low risk specialty, psychiatry. It also aims to examine the relationship between defensiveness and seniority, as well as the effect of previous experiences on the level of defensiveness.
We sent questionnaires to doctors working in the field of psychiatry within the Northern Region of England including trainees, non-consultant grades, and consultant psychiatrists. The questionnaire contained questions relating to mental health and the law including a section on defensive practice. The relevant section of the questionnaire is shown in fig 1. Non-respondents were sent a reminder letter one month later but the responses were analysed anonymously.
In the section on defensive practice, respondents were asked if they had taken any of four specified actions within the past month because of worries about possible consequences such as complaints, disciplinary action by managers, legal action, or publicity in the media. The specified actions were: admitting patients to hospital when the patient’s condition could be managed as an outpatient, placing patients on a higher level of observation than warranted by the patient’s condition, writing in patients’ records specific remarks such as “not suicidal” and dictating letters more than necessary for managing the patient’s illness. Respondents who had taken one of the above actions were also asked whether the following factors were important considerations for their actions: previous complaints or legal claims against themselves, previous complaints or legal claims against their colleagues, previous critical incidents, and concerns about media interest. The data were analysed using SPSS 8.0.
Out of 154 questionnaires sent, 96 were returned (response rate 62%) from 48 equivalent consultant grades, 18 specialist registrars, 23 senior house officers (SHOs), and seven non-consultant career grades but one did not contain valid responses.
Table 1 shows the number (%) of respondents who had practised defensively. Overall, 71 respondents (75%) had taken at least one of the four actions within the past month. For actions that did not directly affect patient care, 66% of respondents had written in patient’s records and 50% had dictated for defensive purposes. For actions that directly affect patient care, 21% of respondents had admitted patients overcautiously and 29% of respondents had placed patients on a higher level of observations than necessary.
SHOs were significantly more likely to take the two actions which directly affect patient care, followed by specialist registrars and consultants. (Overcautious admission: SHO 43%, specialist registrar 31%, consultant equivalents 9%; χ2 by trend = 12.2, df=1, p=0.002. Higher level of observations: SHO 65%, specialist registrar 18%, consultant equivalent 16%; χ2 by trend = 19.5, df=1, p<0.0005.) However, there was no significant relationship between the psychiatrist’s grade and the likelihood of taking the two actions that did not directly affect patient care. (Writing in patients’ record specific remarks: SHO 78%, specialist registrar 65%, consultant equivalent 57.1%; χ2 by trend p>0.05. Dictating letters for defensive purposes: SHO 52%, specialist registrar 65%, consultant equivalent 43%; χ2 by trend p>0.05.)
Among the 71 respondents who had taken any of the four actions within the past month, the number (%) who regarded the following factors as important were: a colleague’s experience of complaints, 40 (56%); concerns about media interest, 27 (38%); previous critical incident, 24 (34%); previous legal claim against colleague, 17 (24%); previous experience of complaints, 16 (23%); and previous legal claim against self, four (6%).
This study found that almost three quarters of the psychiatrists who responded had practised defensively within the past month. As psychiatry is regarded as a low risk specialty by the UK medical defence organisations, these results might indicate an even higher level of defensive medical practice among other hospital specialties.
It might be argued that writing in patient’s records or dictating more than perceived necessary to manage the patient’s illness by the clinicians may improve record keeping and communication, and may be considered as a positive aspect of defensive practice. However, unnecessary hospital admissions and close observations can have adverse effects on patients’ independence and autonomy. Furthermore, these activities result in inefficient use of resources in the NHS. These two actions represent negative aspects of defensive practice.
The higher propensity of junior trainees to admit patients to hospital and to place patients on higher levels of observations than necessary may be attributable to their lack of confidence and experience. A US study found that resident psychiatrists trained in consultation-liaison psychiatry ordered “constant observation” less frequently than psychiatrists without such training.11 Furthermore, resident psychiatrists ordered “constant observation” less frequently when experienced members of staff were available for supervision compared with after hours. Another US study of walk-in psychiatric patients found that less experienced staff (first or second year residents) admitted twice as many patients than more experienced staff (third year residents and attending physicians).12 However, a more structured training programme for the second year residents resulted in a rapid reduction in their rates of admission. Therefore, better and more structured training might reduce the high level of defensive practice among SHOs in our study.
Experience of complaints (colleague or self) and critical incidents were important factors for defensive practice. Complaints are on the increase. In a survey of consultants in the Oxford Regional Health Authority,10 56% of all consultants had received at least one complaint. These complaints have an important effect on the consultants at an emotional level and consultants rely almost exclusively on medical networks (rather than managers) for support. Taken together, our findings indicate that the way complaints and investigations are handled should be improved to maintain a truly “no blame” environment conducive to learning from past experience. Such an environment is necessary for minimising detrimental effects on patient care.
Although valuable lessons can be learnt from investigations after critical incidents,13 over a third of the psychiatrists surveyed who had practised defensively attributed their behaviour to such previous incidents. Critical incidents, such as suicides, homicides, and deaths while detained under the Mental Health Act, are often investigated by both the trust and the coroner’s inquest. Negligence claims from the relatives may follow. These investigations may provide the source of incentives to act defensively. Following recent high profile investigations such as the Griffiths inquiry and the Bristol and Alder Hey inquiries, another source of stress to the doctors involved in inquiries is the perception that some inquiries are themselves subjected to bias.14,15
There is a dilemma between the creation of a “no blame culture”1 and the need to learn from the past.13 Our results demonstrate that external pressure such as complaints and investigations into critical incidents led to increased defensive practice. Following the Bristol and Alder Hey affairs, there is a perception among doctors that inquiries are used to scapegoat clinicians for systems failure.16 These highlight the difficulties of creating truly “no blame cultures”.
A limitation in our study is our rather low response rate of 62%. This may be attributable to the sensitive nature of the topic. However, it is already slightly higher than other similar surveys—the response rates in a survey among general practitioners4 and consultants10 were 60% and 52% respectively.