Aims: The aim of the multivariable analysis was to examine factors associated with survey responses to the questions about respondents’ attitudes to providing care if in the future their parent requires it.
Methods: Survey respondents were asked about six different tasks which for the purpose of this analysis were grouped into two categories:
- Low level needs: help getting to social activities, supervision/emotional support
- High level needs: help with bathing 3-4 times per week, help with feeding, help with toileting, help with supporting treatments
The response categories for each task comprised: the respondent, the parent’s spouse/partner, a family member other than spouse/partner in the household, a family member outside the household, a paid carer, live-in paid carer, parent moves to live with a family member, family member moves to live with parent, admission to a care home, other, none of these, don’t know. Responses of non-spouse coresident family member and extra-resident family member were combined to form a binary ‘other family member’ category, which was coded as 1 if either type of family member was indicated a likely future support source. The categories of parent figure moving in with a relative and relative moving in with the parent figure were similarly combined.
Separate regressions were run for each of the two levels of need for each of these response categories (except ‘other’ and ‘none of these’). For example, one regression considered factors associated with the respondent indicating they would provide care for low-level needs and another considered factors associated with the respondent indicating that paid carers would likely provide care for high-level needs.
The dependent variable in each case was a binary indicator of whether or not the respondent indicated that the particular support source being considered would likely provide care for one or more of the low-level needs or (separately) high-level needs.
The independent variables explored comprised: age, gender, ethnicity, housing tenure and education of the respondent, whether the respondent was a current or past carer (or neither), whether they lived with a spouse/partner, whether they lived with a child, whether they (if employed full-time) expected to be retired or work only part-time within five years, their relationship to their parent figure, their parent’s age, whether they co-resided with their parent (and travel time to visit their parent if not), and (when considering the respondent as a likely source of support) whether another family member is likely to provide help.
This note covers specifically the first of these regressions, considering factors associated with the respondent indicating they will likely provide support if in the future a parent/parent figure develops a low-level care need.
Results: Women are more likely to indicate themselves as a likely support source for low-level care needs in the future (OR=1.24, p=0.02) holding other variables constant. Those who will likely be retired are similarly more likely to indicate themselves as a future support source for low-level needs (OR=1.5, p=0.007). Those with care experience are also far more likely to indicate themselves as a future support source if they have cared in the past (OR=1.5, p<0.001) or are currently caring (OR=2.15, p<0.001). Respondents are less likely to indicate themselves as a carer for a father than a mother (OR=0.57, p<0.001) and considerably less likely to indicate themselves as a likely support source for a mother-in-law (OR=0.25, p<0.001), father-in-law (OR=0.26, p<0.001) or uncle/aunt (OR=0.09, p<0.002). Distance is also a significant factor, with co-resident respondents being more likely to indicate themselves as a likely support source (OR=2.21, p<0.001) compared to those living between 1 and 30 mins away from the parent/parent figure, whereas those who live 30-60 mins away are less likely to indicate themselves a likely support source (OR=0.66, p=0.001) and those living over 1 hour away are less likely still to do so (OR=0.24, p<0.001). Finally, those who indicate other family members will likely provide care are considerably more likely to indicate that they themselves will likely be a future support source (OR=4.2, p<0.001).