The effect of the Winter Fuel Payment on household temperature and health: a regression discontinuity design study

Background: The Winter Fuel Payment (WFP) is a non-NHS population-level policy intervention that aims to reduce cold exposure and enhance the health and well-being of older adults. Labelling this cash transfer as 'winter fuel' has been shown to lead to increased household energy expenditure, but it is not known if this expenditure produces warmer homes or health benefits. Objectives: First, the association between indoor temperature and health was established to identify the outcome measures most likely to be affected by the WFP. Then, whether or not receiving the WFP is associated with raised household temperature levels and/or improved health was assessed. Design: Random and fixed effects regression models were used to estimate the link between ambient indoor temperature and health. A regression discontinuity (RD) design analysis exploiting the sharp eligibility criteria for the WFP was employed to estimate the potential impact of the payment. Setting: The sample was drawn from the English Longitudinal Study of Ageing (ELSA), an observational study of community-dwelling individuals aged ≥ 50 years in England. Participants: Analyses examining the association between household temperature and health had a maximum sample of 12,210 adults aged 50–90 years. The RD analyses drew on a maximum of 5902 observations. Intervention: The WFP provides households with a member who is aged > 60 years (up to 2010, from which point the minimum age increased) in the qualifying week with a lump sum annual payment, typically in November or December. Main outcome measures: Differences in indoor temperature were examined, and, following an extensive literature review of relevant participant-reported health indicators and objectively recorded biomarkers likely to be affected by indoor temperature, a series of key measures were selected: blood pressure, inflammation, lung function, the presence of chest infections, subjective health and depressive symptom ratings. Data sources: The first six waves of the ELSA were drawn from, accessible through the UK Data Service (SN:5050 English Longitudinal Study of Ageing: Waves 0–7, 1998–2015). Results: Results from both random and fixed-effects multilevel regression models showed that low levels of indoor temperature were associated with raised systolic and diastolic blood pressure levels and raised fibrinogen levels. However, across the RD models, no evidence was found that the WFP was consistently associated with differences in either household temperature or the health of qualifying (vs. non-qualifying) households. Limitations: The presence of small effects cannot be ruled out, not detectable because of the sample size in the current study. Conclusions: This study capitalised on the sharp assignment rules regarding WFP eligibility to estimate the potential effect of the WFP on household temperature and health in a national sample of English adults. The RD design employed did not identify evidence linking the WFP to warmer homes or potential health and well-being effects. Future work: Further research should utilise larger samples of participants close to the WFP eligibility cut-off point examined during particularly cold weather in order to identify whether or not the WFP is linked to health benefits not detected in the current study, which may have implications for population health and the evaluation of the effectiveness of the WFP. Funding: The National Institute for Health Research Public Health Research programme