Gintare Malisauskaite, Olena Nizalova, Katerina Gousia, Hansel Teo, Julien Forder
How often people drink alcohol might affect their need for (long-term) care in old age. The existing research on this issue is somewhat out of date, not focused on the UK, and does not consider unpaid care from family and friends – sometimes called informal care. This study should help with predicting the amount of the various types of care and the corresponding public spending that might be needed in the future.
We theorise that there are two main ways through which alcohol may influence long-term care use: a) alcohol negatively impacts health and this increases the need for care in older age; and b) alcohol negatively impacts relationships, ability for self-care, responsiveness and this reduces the amount of care offered or used.
We used a statistical analysis that compared data on how often people over 65 said they consumed alcohol during a week with data on how much formal and informal care they used. The aim was to determine whether drinking causes a need for long-term care. To do this we need to allow for people’s decisions, personal habits, behaviour and preferences, factors that affect both care and alcohol use but in separate ways. People born with certain genes are more likely to drink alcohol and these genes should not directly affect people’s need for care in later life. Accordingly, we predict drinking behaviour based on this genetic information and compare this with the person’s use of care, with a reduced risk that any relationship we see is caused not by alcohol use but these other factors.
We compared the findings between our analyses with and without using the genetic information to isolate the alcohol effect from other factors. We considered men and women together and separately. We used the English Longitudinal Study of Ageing (ELSA) (2002-2017) dataset to conduct our analysis at an individual level. Additionally, we accounted for people’s personal characteristics (age, background etc.), lifestyle choices, and existing health conditions.
We found that frequent drinking increases formal care use, although this effect was not especially strong. We did not find any relationship with informal care use. Furthermore, we did no find that alcohol consumption affects future formal or informal care use when looking at genders separately.
We would expect family (informal) carers would help with long-term care needs, but our findings suggest this is not the case here. This may mean that alcohol use damages relationships with family and friends and then the long-term care burden likely falls on formal care.
There are some limitations with the analysis. One potential problem is that higher alcohol use may cause respondent’s earlier death, so the data only includes survivors. We cannot predict the overall direction of the bias, as the named influences may have opposite effects towards long-term care use.
Overall, our analysis weakly supports the claim that frequent alcohol use does impact on health and so increases the need for care (provided by professionals). This result provides further justification for policies that seek to reduce frequent drinking (five days a week to every day). This relationship is somewhat tentative because it is possible that people who drink more may also be less likely to seek care (even if they may benefit from care). Similarly, drinking may be perceived negatively and limit the care offered – particularly by families. We found some evidence of such an effect. Taken together these factors may partly explain why the headline effect is relatively small. If this is the case, our findings are likely to be underestimating the beneficial effects (on long-term care use) of policies to support problem drinking.
