Exploring the relationships between provision of welfare benefits advice and the health of elderly people: a longitudinal observational study and discussion of methodological issues

Provision of welfare benefits advice to maximise financial benefit uptake is a shared goal for social and health policy in the UK. The study was designed to explore the wider impact on elderly people provided with specialist welfare benefits advice, in terms of their health and quality of life. This paper reports on a longitudinal postal survey of community dwellers aged 60 and over referred for specialist welfare benefits advice within social services and who were followed up at 5 months (over 2003-2004). Outcome measures included the Short Form-36 (SF-36), the General Health Questionnaire-12 (GHQ-12) and the Barthel Index (postal version), along with questions relating to chronic illness. We also report on the methodological issues that arise from conducting research of this nature. Out of 233 elderly people sent questionnaires, 77 (33%) returned a completed baseline questionnaire. Of the 156 who did not return questionnaires, 35 (22%) gave reasons of being 'too unwell' to participate. Between baseline and follow-up, 8 (10%) of the 77 respondents died; 52 of the 69 (75%) remaining participants completed follow-up questionnaires. Although low, these response rates are comparable with other similar questionnaire surveys examining the outcome of providing benefits advice to individuals. Elderly people (mean age = 80.3, SD = 8.6 years) receiving welfare benefits advice usually reported the presence of a longstanding illness or disability, and the use of healthcare services. Baseline SF-36 scores were extremely low and remained largely unchanged at follow-up; however, there were significant improvements in GHQ-12 scores (mean difference = -1.45, 95% CI = -2.63 to -0.27, P = 0.017). Significant increases in benefit income were identified in 65% of respondents with complete financial data sets (mean increase = pound14.73 per week; 95% CI = 5.27-24.18, d.f. = 39, P = 0.003). Participants were very vulnerable in their health status (compared with normative data for elderly people), and this may have contributed to the difficulty in engaging them in the research. There are methodological issues around establishing cause and effect in this type of study, which cannot be readily designed out on account of ethical issues. Extraction and analysis of financial status and benefit eligibility with a view to determining absolute changes in the material well-being of vulnerable individuals over time is a complex and challenging task. Use of suitable measures is essential. Innovative strategies are necessary to maximise survey response rates amongst the vulnerable elderly population.