|Home | About | Journals | Submit | Contact Us | Français|
Background It has been argued that primary care practitioners have an important part to play in the prevention of suicide. However, levels of assessment of risk of suicide among patients treated in this setting are generally low.
Methods Cross-sectional survey of general practitioners (GPs) and people being treated in primary care who had signs of depression. The study combined open and closed questions on attitudes to screening or being screened for suicidal ideation.
Results One hundred and one of 132 patients took part in the survey and 103 of 300 GPs completed a questionnaire. A majority of both GPs and patients stated that people should be screened for suicidal ideation. However, an important minority of patients and GPs stated that asking or being asked such questions made them feel uncomfortable. Less than half of GPs had received formal training on the assessment of suicide risk. GPs told the researchers that barriers to screening included time pressures, culture and language, and concerns about the impact that screening could have on people's mental health. One-quarter of GPs and one-fifth of patients supported the notion that screening for suicidal ideation could induce a person to have thoughts of self-harm.
Conclusions GPs and family doctors should screen for suicidal risk among depressed patients and should receive training on how to do this as part of their general training in the assessment and management of mental disorders. Research should be conducted to examine what, if any, effect screening for suicidal ideation has on mental health.
Suicide is a major cause of premature mortality, which results in over one million deaths worldwide per year.1 In addition to the impact on individuals and their families, suicide also results in a substantial loss to the economy of most countries.2
Previous research showing high levels of attendance in primary care in the period prior to completed suicide has highlighted the role that primary care services might play in efforts to reduce the incidence of suicide.3–5 As depression is a significant risk factor for suicide and most people with depression are treated in primary care, the management of depression in primary care settings has also been raised as an important part of efforts to reduce the incidence of suicide. The identification and management of suicidal ideation and depression in primary care is a central component of suicide-prevention policies across Europe and North America.6,7
Despite the emphasis placed on primary care in suicide-prevention strategies, studies have found low levels of assessment of suicide risk among patients treated in this setting.8–10 A retrospective examination of general practitioner (GP) case notes of 61 men who died by suicide in the north of England found that only two documented the risk of suicide.11 Reasons for low levels of screening of suicidal ideation have rarely been examined, but include doubts about whether this provides a useful means of trying to prevent suicide and concerns about the impact on patients of being asked such questions.9
In one vignettes-based study which examined the attitude of 132 family physicians to assessing elderly depressed patients, 19% of respondents stated that they would not assess for suicide risk because they feared that asking such questions could ‘induce’ suicide.12 Support for the notion that suicidal ideation can be ‘transmitted’, comes from observational studies showing that the incidence of suicidal behaviour increases following media coverage of incidents of suicide and self-harm.13–15 While a limited amount of research has examined GP views about screening for depression, the authors are not aware of any previous attempts to examine patient responses to being screened for suicidal ideation.
This study therefore set out to survey GPs and people who use primary care services, to examine their attitudes to screening for suicidal ideation and behaviour. It specifically aimed to:
Patients who took part in the study were recruited from three general practice surgeries in north west London. Researchers attended practices on a weekly basis at the discretion of clinical staff and attempted to recruit a consecutive sample of all those presenting with non-urgent medical problems. Potential participants were given written and verbal information about the study and asked whether they were willing to take part. The researchers told potential participants that they were interested in questions that GPs may ask about health and emotional distress and how people felt about being asked such questions. Those who provided consent were screened for signs of depression using the two-item screening questionnaire developed by Arroll and colleagues.16 Those who answered yes to either of the two screening questions were asked to take part in a telephone survey which included five direct questions about suicidal ideation and behaviour that have been widely used in previous studies examining mental health in primary care and community settings, and are based on recommendations for assessing suicide risk.17–19 Participants were asked: ‘At any time in the last month:
Participants were also asked about exposure to suicidal behaviour among people close to them, as this has been highlighted as an important factor to consider when assessing for risk of suicide. Participants were asked: ‘Has any member of your family or someone else you are close to attempted to harm themselves or commit suicide in the past?’
Participants were then asked a series of open-ended questions about their response to being screened for suicidal ideation and behaviour. Participants were asked:
The views of GPs were obtained through a cross-sectional survey. The content of the survey was based on the study aims, a review of previous literature on this topic, and input from local clinicians. GPs were asked the following questions:
In order to minimise the amount of time taken to complete the questionnaire and to assure GP colleagues that their responses would be treated in confidence, questions about demographic or other details that could be used to identify respondents were not included. Initially, questionnaires were sent out to local GPs, and then personal contacts with colleagues were used elsewhere to obtain the views of a wider range of GPs. The study set out to collect data from 96 patients and an equivalent sample of GPs as this size of sample would provide 80% power to estimate the proportion of patients that believe people should be screened for suicidal ideation, with a confidence interval of 10% and 5% level of statistical significance.
Simple descriptive statistics were used to describe the responses of those who took part in the survey, using SPSS (Version 16.0). Qualitative data from the surveys were analysed using a thematic framework.20 This was founded on the objectives of the study, developed using analytic induction from early responses and iteratively added to and amended as further data emerged. Data, uniquely coded to reflect the specific interview from which they were sourced, were pasted into the coding framework. Validation of the analytic approach was carried out on an ongoing basis through individual supervision.
One hundred and one (76.5%) of 132 patients agreed to take part in the telephone survey. One-hundred and three GPs working in London, Bristol, Sheffield, Sunderland and Durham, returned a copy of the survey, out of the approximately 300 that were sent out. Demographic data were not collected from GPs, but the mean age of patients who took part in the survey was 50 years (standard deviation (SD) = 19 years), and 68 (67.3%) were female. Twenty-eight (27.7%) patients described feeling that their life was not worth living, 19 (18.8%) stated that they had thought of ending their life at some point during the previous four weeks, and one person reported having actually tried to end their life. Twenty-eight patients knew of someone who had attempted to harm themselves or end their life in the past.
When patients were asked how they felt about being asked questions on suicidal ideation and behaviour, 39 (38.6%) stated that they did not mind being asked these questions – that they felt ‘OK’ or neutral about them. Twenty-seven patients (26.7%) stated that they were surprised at being asked such questions, and 26 (25.7%) reported feeling either upset, embarrassed or alarmed, or simply stated that they did not like being asked them.
‘I felt terrible, it upsets me, but I want to tell the truth. Talking about it makes me think of it even more.’ (P45)
‘Felt ashamed because I'm letting myself down, and I'm embarrassed because I should be stronger. The questions get to you inside, because if you think of ending your life you have failed.’ (P41)
When asked whether GPs should routinely screen people who were depressed about whether they had thoughts of trying to harm or kill themselves, 81 (80.2%) stated that they should, four (4.0%) stated that they should not, and the remaining 16 (15.8%) stated that they did not know (see Table 1). People who thought GPs should screen for suicidal ideation said they should do so because it was important for people to have a chance to talk about their feelings, for the GP to get a complete picture of the person's mental health, and to identify people at risk of suicide.
‘Because some people won't tell their friends and family but they might tell a doctor if they asked directly.’ (P14)
‘Yes, I felt when I was depressed in the past the GP didn't really have an understanding of how I felt.’ (P97)
Those who thought that GPs should not ask such questions pointed to concerns about the impact this could have on peoples' mental health.
‘If you don't have these ideas in your head it might give you ideas.’ (P12)
‘Doctors have a duty to support people, prolong life, not the other way, people might think they were suggesting that their time was up and it was time for them to go. I would be very suspicious if my doctor asked me this, why is he asking me this?’ (P91)
Finally, patients were asked about how they felt GPs should ask such questions; respondents were equally divided between those who felt they should use a direct approach and those who felt that GPs should be careful about the language that they used and avoid terms such as ‘suicide’.
‘I think you shouldn't beat around the bush, just come out with it “have you thought about suicide?”’ (P22)
‘Better to ask it gently, like “would you ever think of taking your own life?” I don't like the word “suicide” and I don't think people should say the word when they are asking about it.’ (P37)
‘To say the word suicide in my Hindu religion is very shocking, it would be a very sinful thing to think about … Be aware of religion before asking, be culturally sensitive.’ (P66)
Nearly all the GPs that took part in the study stated that they sometimes screened suicidal ideation (n = 96, 93.2%) (see Table 1). Most GPs stated that they screened those who had signs or symptoms of depression (n = 89, 86.4%), psychosis (n = 22, 21.4%), or chronic physical health problems (n = 22, 21.4%); very few mentioned screening those with substance misuse problems and those with a past history of deliberate self-harm.
GPs were asked why they screened for suicidal ideation. The most commonly stated reason was to assess and manage risk (n = 89, 86.4%). Thirty (29.1%) stated that screening for suicidal ideation was part of the effort they could make to prevent suicide; 21 (20.4%) stated that it helped to open up a dialogue about a sensitive matter that they wanted patients to feel able to talk to them about, and a small minority (n = 7, 6.8%) stated that they screened for ‘medico-legal reasons’.
GPs were asked if they received any formal training for screening for suicidal ideation and behaviour. The majority of practitioners (n = 62, 60.2%) stated they had not, while 40 (38.8%) stated that that they had. Approximately half of those that had received training did so while undertaking specialist training posts in psychiatry.
The two most commonly stated barriers to screening for suicidal ideation were time and issues related to screening among people from ethnic minority communities. Seventeen doctors (16.5%) stated that they did not have enough time to screen for suicidal ideation; one GP stated that asking such questions was ‘opening a bag of worms’ (GP36). Culture issues were mentioned by 15 (14.6%) GPs and were felt to be barriers to screening because of problems asking sensitive questions via an interpreter, and because GPs were unsure about cultural attitudes to suicide among some of the patients they saw who came from ethnic minority communities.
A small minority of GPs stated that they were embarrassed to ask about suicidal ideation, or were concerned that asking such questions could ‘make things worse by suggestion’ (e.g. GP52). Conversely, GPs stated that an established diagnosis of depression, a past history of deliberate self-harm, or knowing the patient well, all made it easier to screen for suicidal ideation.
When asked directly whether exposure to questions or information about suicidal behaviour could induce a person to have thoughts of self-harm, 37 (35.9%) stated that it could, 62 (60.2%) stated that it did not, and 3 (2.9%) stated that they did not know. Nineteen (18.4%) GPs mentioned the role of the media (including television and the internet) in transmitting suicidal behaviour, and nine (8.7%) stated that particular groups such as young people and those with a history of mental health problems may be particularly susceptible to social transmission of suicidal behaviour.
‘I think there are some vulnerable groups who might be at risk of individual or mass hysterical or attention-seeking reactions such as seen in schools or evangelical groups. They might get “ideas” from exposure to such information.’ (GP62)
Finally, when participants were asked if they thought that screening for suicidal ideation in primary care could trigger suicidal behaviour in vulnerable patients, the majority (n = 70, 68%) stated that it did not, 26 (25.2%) said that it could and seven (6.8%) stated that they were uncertain. Twenty-six GPs provided information about the types of patients who may be vulnerable to social transmission through screening for suicidal ideation. The most commonly identified categories of vulnerable patients were individuals with personality disorders and drug misuse problems.
This study explored attitudes to screening for suicidal ideation among an equivalent sample of GPs and patients attending primary care, as determined by a power calculation. The authors believe that the approach used to select the study sample increases the validity of the findings. All the patients who participated in the survey had some evidence of depression and would therefore have been considered appropriate for screening for suicidal ideation according to the majority of GPs that took part in the study. All participating patients were asked about their reaction to being asked screening questions rather than being asked to comment on a hypothetical scenario. A limitation of the study was that all the patient interviews were conducted over the telephone and it is not known whether this had an impact on how they felt about being asked such questions compared to their being asked them in a face-to-face consultation. The poor response rate to the GP survey raises questions about the generalisability of the study findings. While the authors believe that the barriers to screening that they raised are likely to be experienced by other GPs, it is not known whether the proportion who identified these barriers would be the same among other groups of GPs.
Most of the GPs that took part in the survey stated that they screened for suicidal ideation, and most patients thought that they should. GPs stated that they selectively screened those with depression. Few raised the issue of screening among other high-risk groups such as those with a history of deliberate self-harm and those with substance misuse problems. A small but significant proportion of GPs stated that they felt uncomfortable screening for suicidal ideation and one-quarter of patients stated that they did not like being asked these questions. One in four GPs and one in five patients supported the notion that screening for suicidal ideation could increase the likelihood of someone thinking about or actually harming themselves. Most GPs had not received any formal training in how to assess suicide risk.
Previous research indicates that screening of suicidal ideation in primary care is generally low.8–10 This study has helped establish possible reasons for this. These include time pressures, cultural and language barriers, and concerns about the impact of asking some people questions about suicidal ideation. While most patients thought that it was important to screen for suicidal ideation, one-quarter felt uncomfortable about being asked such questions. A small number of respondents highlighted stigmatisation of suicide within their religion. Attitudes to suicide differ greatly depending on cultural and religious context,21 and these findings suggest that awareness of these cultural factors is important when screening for suicidal ideation and behaviour.
Time constraints are a frequently cited obstacle to screening and other initiatives aimed at prevention.22,23 As there is a consensus that people who are contemplating suicide should be referred to secondary care services for further assessment and treatment,24 systems need to be in place to support the work of primary care practitioners so that they are confident that they can respond to those found to have thoughts of ending their life.
The other significant barrier found to screening for suicidal ideation was the concern that asking such questions could have an adverse effect on a person's mental health. It is noteworthy that more GPs than patients among those who took part in the survey supported the notion that screening for suicidal ideation could induce someone to have thoughts of self-harm. The belief that suicidal ideation can be ‘induced’ by screening has been described as a myth by commentators.25 The impact of screening for suicidal ideation among high school students in the United States was tested in a randomised trial.26 Students who completed a questionnaire which asked about suicidal thoughts were no more likely to have such thoughts two days later, than those who did not complete baseline screening. Experimental research examining the impact of screening for suicidal ideation in primary care settings has not been conducted. Studies examining the impact of screening for health-related problems in primary care have been criticised for failing to consider the harm they can do.27 The authors believe that experimental research could examine the impact of screening patients in primary care for suicidal ideation, and that the study finding that one in five GPs is concerned about the impact of screening on mental health provides a strong case for conducting such a study.
GPs and patients treated in primary care believe that people should be screened for suicidal ideation. However, an important minority of patients and GPs stated that asking or being asked such questions makes them feel uncomfortable. Research should be conducted to examine what effect, if any, screening for suicidal ideation has on mental health. In the absence of any clear evidence that screening is harmful, GPs and family doctors should assess the risk of suicide among vulnerable patients by finding out whether those with depression and other mental disorders have thoughts or plans to end their life. GPs should be trained to assess suicide risk as part of their general training in the assessment and treatment of mental disorders.
Priya Bajaj, Specialist Registrar in Psychiatry, Central and North West London NHS Foundation Trust, UK.
Elena Borreani, Research Associate, Department of Psychological Medicine, Imperial College London, UK.
Pradip Ghosh, Specialist Registrar in Psychiatry, Central and North West London NHS Foundation Trust, UK.
Caroline Methuen, Specialist Registrar in Psychiatry, West London Mental Health NHS Trust.
Melissa Patel, Undergraduate Medical Student.
Michael Joseph, Reader in Mental Health Services Research, Department of Psychological Medicine, Imperial College London, UK.
This project was part funded through an educational grant from Imperial College London.
Research ethics committee approval was obtained prior to the start of data collection.