Predicting Future Service Use in Dutch Mental Healthcare: A Machine Learning Approach.

A mental healthcare system in which the scarce resources are equitably and efficiently allocated, benefits from a predictive model about expected service use. The skewness in service use is a challenge for such models. In this study, we applied a machine learning approach to forecast expected service use, as a starting point for agreements between financiers and suppliers of mental healthcare. This study used administrative data from a large mental healthcare organization in the Netherlands. A training set was selected using records from 2017 (N = 10,911), and a test set was selected using records from 2018 (N = 10,201). A baseline model and three random forest models were created from different types of input data to predict (the remainder of) numeric individual treatment hours. A visual analysis was performed on the individual predictions. Patients consumed 62 h of mental healthcare on average in 2018. The model that best predicted service use had a mean error of 21 min at the insurance group level and an average absolute error of 28 h at the patient level. There was a systematic under prediction of service use for high service use patients. The application of machine learning techniques on mental healthcare data is useful for predicting expected service on group level. The results indicate that these models could support financiers and suppliers of healthcare in the planning and allocation of resources. Nevertheless, uncertainty in the prediction of high-cost patients remains a challenge.

Is change over time in psychotic symptoms related to social functioning?

OBJECTIVE: In psychosis, treatment often focuses on symptom reduction whereas social functioning is also essential. In this study, we investigate positive psychotic symptoms and medication use in relation to social functioning over a 3-year time-period in 531 patients diagnosed with psychosis. Furthermore, relations of positive symptoms with needs for care and quality of life were also investigated. METHOD: Using repeated measures analysis, changes were measured over time. Hereafter, mixed model analyses were performed to determine the associations of social functioning, needs for care, and quality of life with psychotic symptoms and patient characteristics. Finally, we assessed differences in symptoms and medication dose between those with an increase and those with a decrease in social functioning. RESULTS: Patients significantly improved in social functioning, while psychotic symptoms increased. Improvement in social functioning was associated with younger age, higher IQ, and lower social functioning at T1, but not with positive symptoms. Also, improvement in social functioning was found to be related to a decrease in the dose of clozapine. Improvement in social functioning occurs despite worsening of positive symptoms. CONCLUSIONS: The findings suggest the need to further explore the relation between symptomatology, social functioning, and medication use. In the treatment of psychotic disorders, one should reconsider the strong focus on reducing psychotic symptoms. The current focus needs to shift much more toward improving functional outcome, especially when the patient expresses a desire for change in this respect.

Using machine learning to predict mental healthcare consumption in non-affective psychosis.

OBJECTIVE: The main goal of the study was to predict individual patients' future mental healthcare consumption, and thereby enhancing the design of an efficient demand-oriented mental healthcare system by focusing on a patient population associated with intensive mental healthcare consumption. Factors that affect the mental healthcare consumption of service users with non-affective psychosis were identified, and subsequently used in a prognostic model to predict future healthcare consumption. METHOD: This study was a secondary analysis of an existing dataset from the GROUP study. Based on mental healthcare consumption, patients with non-affective psychosis were divided into two groups: low (N = 579) and high (N = 488) intensive mental healthcare consumers. Three different techniques from the field of machine learning were applied on crosssectional data to identify risk factors: logistic regression, classification tree and a random forest. Subsequently, the same techniques were applied longitudinally in order to predict future healthcare consumption. RESULTS: Identified variables that affected healthcare consumption were the number of psychotic episodes, paid employment, engagement in social activities, previous healthcare consumption, and met needs. Analyses showed that the random forest method is best suited to model risk factors, and that these relations predict future healthcare consumption (AUC 0.71, PPV 0.65). CONCLUSIONS: Machine learning techniques provide valuable information for identifying risk factors in psychosis. They may thus help clinicians optimize allocation of mental healthcare resources by predicting future healthcare consumption.

Symptomatic and Functional Remission in Young Adults with a Psychotic Disorder in a Rehabilitation Focused Team.

The aim of this study is to assess symptomatic remission (SR) and functional remission (FR) in a rehabilitation focused program for young adults with a psychotic disorder in the Netherlands, and to investigate which individual and mental health care factors are associated with SR and/or FR, by using Routine Outcome Monitoring data and data on met needs and unmet needs for care. Data of 287 young adults were collected. Almost 40% achieved or maintained SR, 34% FR, and 26% achieved or maintained both. In addition to sociodemographic factors, living independently, paid employment, higher levels of compliance with treatment, and better fulfillment of unmet needs for care in relation to psychological distress, company and daytime activities were associated with better outcomes on SR and/or FR. Our findings underscore that to successfully improve and sustain remission in young adults with a psychotic disorder, it is needed to conduct specific research into the relationship between SR and FR.