Is the rise in long-term sickness on the UK Labour Force Survey due to proxy responses or survey mode?

There has been a considerable increase in the number of people in the UK reporting that they are economically inactive due to long-term sickness or disability since the coronavirus pandemic, which has attracted policy and research interest. The key data source for these trends is the UK Labour Force Survey (LFS), which has seen a sharp decline in response rates and issues associated with population weighting over recent years. It therefore remains unclear whether the rise in reported rates of sickness is a genuine phenomenon or a statistical artefact. We examine the role of proxy responses and survey mode on the LFS to test the robustness of the trends in sickness. In so doing, we shed light on the important role of proxy responses on the LFS, particularly in the domains of young people and health status. We find considerable selection effects associated with proxy response, and very high rates amongst young people. The increase in incidence of reported long-term sickness may be slightly smaller when accounting for changing survey response patterns, although other findings also suggest the reverse; either way, the increase is not simply a methodological artefact.