Economic assessment of the impact of telecare on the use of social care resources using a zero-inflated, hierarchical linear statistical model.

INTRODUCTION: There is an ever-increasing demand for social care in the UK, with expenditure predicted to double to £56 billion by 2038/39. Many councils are under budget restrictions putting pressure on the number of services provided and their quality. Telecare complements social care and involves the implementation of technology to keep individuals more independent. METHODS: This study utilised a retrospective time-series analysis of data provided by Lancashire County Council between the period January-2013 to March-2018. A generalised linear mixed model (GLMM) was used to control for potential confounders. Two groups were identified: those using telecare (telecare group, n = 699) and those who did not (control group, n = 839). RESULTS: The fixed effects data showed that telecare group start £75 per week lower in cost and as time progressed this reduced further by 9p per service user per week. In contrast, control group costs rose 5p per week per user. This effect was independent of age but was affected by measure of dependency. Analysis was then utilised to make predictions based on weighted averages. The scenario showed a total difference of £4,949 per service user over the whole year. A second scenario pro-rata'd costs for the full year showed a difference of £6,214, where telecare would avoid costs of £17 million per year. DISCUSSION: This analysis demonstrates that there is evident potential for the use of telecare to reduce social care resource use and costs. This study also highlights the use of a GLMM as a novel method of analysing observed data by controlling confounders.

NICE, in Confidence: An Assessment of Redaction to Obscure Confidential Information in Single Technology Appraisals by the National Institute for Health and Care Excellence.

INTRODUCTION: Health technology assessment (HTA) aims to provide a transparent framework within which normative judgements can be applied for decision making. Such transparency enables the public to understand the rationale for decision making, but conflicts with companies being able to offer commercially sensitive discounts. We investigated how to balance these conflicting ideals. METHODS: National Institute for Health and Care Excellence (NICE) submissions were reviewed for products with an approved, simple Patient Access Scheme (PAS) discount. The approach to censoring was noted (e.g. total cost and clinical outcomes redacted). Submissions were then assessed for transparency (i.e. whether the decision appeared justifiable given the available information) and confidentiality (i.e. whether the PAS discount could be 'back calculated'). RESULTS: One hundred and eighteen products have an approved commercial arrangement, of which 110 have simple PAS discounts considered within the NICE Single Technology Appraisal programme. A definitive incremental cost-effectiveness ratio was presented within final NICE guidance in only 20 appraisals. Documentation for seven appraisals allowed for the straightforward 'back calculation' of PAS discounts. Furthermore, a large amount of information was censored as academic-in-confidence and remains so many years later. CONCLUSION: Appropriate redaction ensures discounts remain confidential, yet maintains the transparency of the HTA decisions made. Complete redaction does not allow for transparent, justifiable decision making. However, redacting 'enough' information to preclude direct estimation of discounts provides a means of maintaining both transparency and confidentiality. This study demonstrates a lack of consensus regarding presentation of results, and the importance of appropriate redaction.