This study examines the characteristics of 'care networks' - groups of individuals comprised of family, friends and neighbours caring for frail seniors. Care networks vary according to their size, relationship, gender, age and proximity. Variations in care network characteristics explain differences in the types and amount of care received. Care network characteristics that might place seniors at risk of receiving inadequate care are small size and higher proportions of non-kin, male, and geographically distant members. Existing health policy in Canada does not account for these risk factors. Background: The transfer of responsibility for care of frail seniors from the public sphere to individual family members and friends is justified in part, by the belief they comprise an appropriate, available, and cost-effective resource for addressing the challenges of caring for the elderly. This study tests the conventional assumption that such informal care is sufficient to sustain care to frail seniors in the longterm. Past research has focused on individual caregivers and the economic, social, and health costs incurred by them in providing care. This study focuses on 'care networks', or the group of individuals who together provide unpaid care to seniors. It addressed how caregiving responsibilities are distributed across a group of caregivers, and how the nature of that group affects the adequacy of care.

Methods: Data from the 1996 Statistics Canada General Social Survey (GSS) were used to examine the characteristics of the care networks of frail seniors. The study utilized the subset of 1,104 respondents aged 65 years or older who received assistance during the previous year from at least one family member, friend, or neighbour because of the respondent's long-term health condition. Data about the age, gender, relationship to respondent, and geographical proximity of each of these individuals were analysed. Analysis was performed to predict the relationship between these characteristics and the amount of care received, the number of care tasks provided, and the types of care tasks performed.

Findings: Care networks vary considerably in size, relationship composition, gender composition, age composition, and proximity. Although care networks ranged in size from 1 to 8 members, the average size was 1.5 members. Half were comprised entirely of women, 22% included both men and women, and 28% included only men. Seventy eight percent included just family members, but 15% included only non-family members. The majority of network members was under the age of 65 and lived within a halfday drive, with 33% of network members co-residing with recipients. Care recipients, who lived with care network members, had larger care networks, and those whom women cared for received more hours of care than others did. Those with larger care networks, with networks comprised of kin and who coresided with care providers also received assistance with more tasks. Men were significantly more likely to receive assistance than women. Having a higher proportion of female caregivers increased the likelihood of receiving housekeeping and personal care assistance, but decreased the likelihood of receiving assistance with household maintenance and repairs.

Conclusions: Having large care networks is the strongest predictor of receiving more assistance with more tasks. Yet, in general, care networks are small in size and are dominated by women and family members. Network characteristics that might place seniors at risk of receiving inadequate care include small size and higher proportions of non-kin, male, and geographically distant members. These risk factors are poorly reflected in existing home and community care policy. Given that the majority of networks are small, the ability of Canadians to absorb increasing responsibility for the care of home care recipients is uncertain.

Reference: Fast J, Keating N, Otfinowski P, Derksen L, "Characteristics of Family/Friend Care Networks of Frail Seniors", Canadian Journal on Aging, 2004; 23(1), 5-19.