In the present study we found that, patients with misuse problems received lowest commitment, empathy and generated more irritation than patients with other diagnoses. These findings have not to our knowledge been presented before. The psychiatrists and GPs had somewhat more positive attitudes to suicidal patients than the internists, and female physicians were more positive than males. This study shows that all three groups of physicians had positive attitudes towards suicidal patients and were understanding and willing to help them. The levels of competence and commitment tended to align with the physicians’ areas of specialization. The finding that the physicians were most irritated by and least committed to treat substance misuse patients than patients with other diagnoses is concerning. Substance misuse patients often have large and complex problems, and increased morbidity and mortality rates . In the review of Saunders et al. , attitudes among health care professionals were largely negative. Compared to these findings, however,the present study show that physicians report more positive attitudes,and willingness to help suicidal patients. The results are consistent with another study where more positive attitudes have been found among health-care professionals working in psychiatric wards than those working in other fields . This is also in line with a study of professionals in mental health-care in Norwegian outpatient clinics .
In our study, physicians showed a higher degree of understanding than acceptance of suicide among people with an incurable illness. In Norway, as in many other countries, euthanasia is illegal. The distinction between suicide among patients with incurable illnesses and other serious somatic and psychiatric diagnoses is delicate. It is important that this issue is addressed because it raises ethical questions and should be considered when educational initiatives are raised. During a medical career, there is a probability that a patient may, by implication or directly, ask for help to end his or her life. It is therefore important that physicians have reflected on their own personal and professional attitudes. Even though all groups stated that patients who ask for help to commit suicide should not receive it, they were able to understand why patients with an incurable disease may want to end their life. This reflects an understanding attitude for patients’ decisions, but maintains the premise that the health-care professional’s role is not to help patients end their lives.
The levels of competence and commitment tended to correspond with the physicians’ areas of specialization. Training can improve clinicians’ attitudes toward suicide, confidence in working with clients at risk of suicide, and, most importantly, their skills in clinical practice . A survey in London showed that 25% of all people who completed suicide had seen their GP during the previous month . Consistent with the findings of our study, GPs are interested in further training . It has been shown that directed efforts of GPs can reduce the number of suicides . In study by Poma et al., sixty-one percent of doctors admitted difficulties in exploring suicidal ideation, but tended to ascribe this to a reluctant attitude of patients . The study underscores the finding that GPs’ need assistance in the difficult task of recognizing suicidal patients. Apparently, there is a lack of instruments available for GPS to discover suicidal risk at an early stage .
Fifteen percent of the internists had participated in courses in assessment and treatment of suicidal patients during the last five years. This is probably somewhat less than optimal. However, it should be emphasized that the need for such training is most important to the staff in accident and emergency departments, because they deal with suicide attempt patients more frequently. In a review of self-harming patients’ attitudes to clinical services, the main points for improved patient satisfaction were increased knowledge and communication skills among staff . In our study, the internists found it more difficult than GPs and psychiatrists to talk to patients after a suicide attempt. All three professional groups, however, frequently encounter patients with depression, substance misuse and personality disorders, and should have the competence to detect suicidal risk.
The generalizability of the findings is limited by the low response rate. However, the relative differences between the groups’ specialties and gender are probably still reliable because of the robust sample size and consistent figures. We used a previously validated scale to measure health-care professionals’ attitudes and the reliability of the scale was found satisfactory.
A review study of scales used to measure attitudes to suicidal behaviour  concluded that there are other, more reliable, scales than the USP. The USP was developed specifically to measure health-care professionals’ attitudes to patients after a suicide attempt, and was therefore found to be most suitable for the specific research questions in our study.
In a study by Creed and Pfeffer, the results showed more positive attitudes among physicians towards somatic diagnoses than self harm . There is no available validated scale for this purpose, so in the present study a new self rating scale was used in order to determine whether this was still present and compare the current attitudes to somatic diagnoses and substance misuse with these findings. The disorders were selected because they are common, and all clinical physicians have treated these patients.
Because the subject of this study is sensitive, a completely anonymous survey may have increased the response rate. To increase the level of discretion and decrease concern among the participants that stating their exact age and information of their profession might make them recognizable, we therefore used categories of age and experience rather than the exact values. The use of electronic questionnaires did not facilitate participation; no participants chose to answer in this way. It should also be considered that those responding might be more dedicated and interested in the current topic and therefore represent a response bias. Unfortunately we did not have the opportunity to perform drop out analyses in this study.
Clinical interpretations and relevance
The relevance of attitudes in clinical practice has previously been emphasized, because attitudes may influence care decisions. If the commitment in clinical practice is consistent with the findings in this study, it appears that patients with substance misuse problems receive the least attention.
We know that working with suicidal patients can be stressful and demanding, especially because of the continuous threat that these patients may eventually take their own lives. Colson et al.  underlined the importance of staff members’ perceptions that some patient groups are difficult to treat, and suggested that this could influence the patients and the treatment process, with implications for progress and prognosis. They found that patients with suicidal depressed behaviour were one of the groups most commonly associated with staff members’ perceptions of being difficult to treat. They suggest that continuous supervision is important to prevent negative attitudes and anxiety of failure, especially among physicians with responsibility for treatment.
To prevent these factors, and to increase confidence in treating patients in a suicidal crisis, continuously updated knowledge is needed, especially with regard to assessment of suicidal risk and treatment options. This is highlighted in the National Institutes for Clinical Excellence guidelines on the care of people who self-harm . The findings in this study underline the importance of implementing appropriate training in basic medical education, regular internal training and continued education. Such educational initiatives should probably be modified for each group of physicians, so they are perceived as relevant and meaningful to their specific situation. Half of the participants reported participation in some form of suicidology education during the previous five years. The next step is to determine how such education can be made more readily available to as many physicians as possible. A successful example was the STORM project, in which training of professionals in primary and mental health care and accident and emergency departments showed improved skills in the assessment and management of suicide risk, and high levels of satisfaction with the training .
Because the results might be influenced by social desirability, an indirect way of measuring general attitudes in the wards could have been to ask the physicians what they thought of their colleagues’ attitudes. This technique was used by Album and Westin,  who asked physicians how they thought other physicians would prioritize different diagnoses, and found that patients with depression and anxiety received the lowest scores. Further, as Taylor et al. suggest,  more insight could be obtained from patients’ own experiences with different parts of the health-care system using standardized interview schedules. Finally, to make educational initiatives more interesting, further research should also investigate what kind of training and education professionals might find useful in their clinical work.