OBJECTIVES: This analysis pragmatically reviews prior health technology assessment (HTA) manufacturer submissions for treatments indicated for mental health, behavioral and neurodevelopmental conditions. It systematically appraises their methodological approaches, establishing key modelling challenges and potential solutions.

METHODS: All submissions were identified from NICE’s website. A data extraction table was developed to identify key features from the submissions, including model structures and data sources for clinical, resource use and health-related quality of life (HRQoL) data. Key challenges and solutions were identified and evaluated.

RESULTS: Eleven HTA submissions were identified from NICE’s website. Two were terminated, leaving nine for evaluation. HRQoL and resource use data were infrequently collected in the submissions’ underlying clinical trials, commonly addressed by utilizing alternative values from literature. External assessment groups (EAGs) typically acknowledged the necessity of this approach, but criticized where alternative sources represented a different population to that being modelled, or greatly predating the submissions. Five submissions modelled disease pathways beyond the acute phase of treatment, but did not source long-term clinical trial data. Where manufacturers extrapolated data from the acute phase of treatment to generate long-term data, this was well received by EAGs. Lastly, societal costs and caregiver burden were not included in the manufacturers’ base case analyses owing to insufficient data.

CONCLUSIONS: The methodological challenges identified within this review include infrequent collection of HRQoL and resource use data and exclusion of societal costs. Data gaps for resource use and HRQoL are generally addressed with literature-derived values. Some models were restricted to shorter time horizons due to a lack of long-term clinical data, compromising their representation of the long-term nature of mental health disorders. Future research should focus on improving data collection practices and better integration of these solutions to strengthen the evidence base for treatments in this area.