Do clinicians and patients think it is acceptable and useful to use an algorithm to predict risk of psychosis in primary care?

Problem

Psychosis is a serious mental illness. Early intervention in people with an at-risk mental state of developing psychosis can decrease their risk of transitioning to psychosis. However, identifying these individuals in primary care is difficult, as GPs might not be familiar with the concept of being at risk of psychosis, or feel they have the skills and knowledge to identify this group. Also, the prediction models used in secondary care for psychosis are not applicable in primary care. Our research team have developed an algorithm – called P Risk - which uses electronic primary care health record data, to help GPs identify people who may be at risk of developing psychosis. Although the P Risk is statistically accurate at identifying people at risk of psychosis, we need to assess its acceptability before implementing it in clinical practice. This study aimed to explore clinicians’ and patients’ views of the acceptability and usefulness of using P Risk in primary care.

Approach

Semi-structured interviews were held with 10 GPs, 11 patients and 6 Early Intervention (EI) team clinicians, as according to NICE guidelines, EI teams should assess and offer treatment to patients at risk of psychosis. The interviews explored participants’ views of P Risk, and were audio-recorded and analysed thematically.

Findings

Most GPs welcomed the development of the P Risk tool, as it would help them identify people who may be at risk of developing psychosis. However, some GPs raised concerns about availability of treatment for this patient group, patients’ willingness to engage with it, and the negative impact that may have on patients being told that they are at risk of psychosis. Overall, EI clinicians were positive about the use of P Risk in primary care, as the tool would remind GPs about psychosis, and assist them in discussing risk factors with patients (e.g. cannabis use), and strategies to address them. However, some EI clinicians raised concerns about EI teams’ capacity to assess a potentially high number of referrals and highlighted that not all EI teams were commissioned to work with these patients. Most patients thought P Risk would help GPs identify early symptoms of psychosis that could go unnoticed, but said revealing that one was at risk had to be worded very carefully as not to worry the patient, and only communicated if a patient’s risk was medium or high, and in the context of providing them with effective treatment.

Consequences

Whilst most clinicians and patients welcomed the development of P Risk, if the tool is to be used in primary care, there needs to be a clear pathway for assessing and offering treatment to those identified as being at risk of psychosis.