Suicide is a global public health challenge, around 700 000 people die from suicide every year. A large proportion was in contact with healthcare close in time before death, suggesting healthcare to be an important resource in the work with prevention of suicide.

The overall aim of this thesis was to increase the knowledge and understanding of suicide as an incident of patient harm, and to find possibilities of changes in the approach to suicide investigations which could contribute to increased learning and improve suicide prevention in healthcare.

Four studies were performed: in the first two studies we reviewed investigations of healthcare performed of suicide cases reported to the supervisory authority as patient harm. Study III was a scoping narrative literature review of the problems with the current approaches to investigations of suicide as patient harm and possible changes for improvement. Study IV was an interview study in which I explored the requirements for valuable investigations of suicide from the views of persons with lived experience of suicidality and professionals. All studies were performed in a Swedish context.

The majority of suicides reported as incidents of patient harm were reported by a psychiatry healthcare provider. Most suicides occurred shortly after the last contact with healthcare and during outpatient care. Demographically, these cases were representative compared to the suicide cases in the entire population.

As incidents of patient harm, suicides differ from most other kinds of reported patient harm in some ways. Only a small proportion occurs in hospitals, most occur in the home of the patient without any witnesses or staff around. Suicide is an act performed by the patient himself/herself and is usually the final outcome of the complex interplay of several different variables with different impacts in different contexts, varying over time and between individuals.

It was found that the adaptation of the investigations to the requirements of the supervisory authority contributed to the fact that the learning from the healthcare’s investigations of suicide has levelled off, the same shortcomings and actions were reported over time. The investigations were performed with a strict healthcare provider perspective, with focus on the last contact with the patient, routines, and what went wrong. This resulted in suggested measures for improvement at an organizational micro level without organizational sustainability over time and with a risk to not address organizational system deficiencies.

The investigations of suicide as potential patient harm should integrate current knowledge in suicidology and patient safety to enable learning and insights valuable for healthcare improvement. This include a holistic perspective of the patient’s situation, analysis of a longer time period and factors of importance for suicidality, suicide prevention, and patient safety, professionalization of the investigations, analyses across organizational boundaries, and focus on learning. A framework to guide this analysis is suggested in this thesis.

The development of knowledge in the science fields of patient safety and suicidology imply the need for a cultural shift in the understanding of suicide as an incident of patient harm. Instead of making a difficult and often to some extent speculative assessment if a suicide had been prevented if other actions had been performed in the contacts with healthcare, and therefore should be investigated and reported as a severe patient harm, or not, the focus in the analyses should be on risk management over time. I propose a framework with factors of importance for a safe healthcare at suicidality to guide this analysis.