Association between Insight, Appointment Attendance and Secondary Mental Health Service Use a Naturalistic Cohort Study using Electronic Health Records

BackgroundWe aimed to determine whether insight at discharge from Crisis treatment and home Resolution Teams (CRTs) was associated with appointment attendance and further use of CRTs or inpatient admission.MethodsWe extracted routine electronic health data from Camden & Islington(C&I) NHS Foundation Trust including individuals with a primary diagnosis of non-affective psychosis (ICD-10, F20-29) who presented at least once to C&I CRTs between January 1, 2008 and December 31, 2017. We defined insight as a composite construct, including Illness Recognition, Treatment Compliance and Symptom Relabelling, and tested whether insight could be coded using data extracted from progress notes. A total of 317 records were analysed. Multivariable Poisson, Negative Binomial regressions were used to estimate differences in attendance rates; Cox modelling was used to estimate differences in risk of service use. We analysed the models at 12-months from first CRTs discharge, and at the end of cohort (all-time).ResultsMost of the descriptions of insight did not map onto our conceptualisation of insight, only 87(27.4%) had no missing data in all 3 components. People with partial insight had a lower attendance rate than people with lack of insight at 12 months (incident-rate-ratio = 0.945,95% CI 0.896-0.997). People with full insight had a decreased rate of service use compared to people with lack of insight at all-time (hazard-ratio = 0.638, 95% CI 0.424-0.960). ConclusionsOverall, better insight was not associated with higher face-to-face appointment attendance, but was associated with lower rates of service use after first discharge from CRTs.