2 Factors associated with need for assistance and source of assistance

Warning View the most recent version.

Archived Content

Information identified as archived is provided for reference, research or recordkeeping purposes. It is not subject to the Government of Canada Web Standards and has not been altered or updated since it was archived. Please "contact us" to request a format other than those available.

This section provides an overview of the first part of the research. The objective is to identify the factors associated with the need for assistance in performing everyday activities, along with those associated with the use of informal and formal support networks. To achieve these goals, secondary data analysis of existing Statistics Canada datasets was undertaken.

2.1 Methodology

A multinomial ordered logistic regression was conducted using the 1996 National Population Health Survey (NPHS) to compute probabilities of having a disability (no, mild, moderate, or severe disability) (see text box "Definitions"). This procedure enabled us to estimate the probability, for a given individual with specific characteristics, of having a certain level of disability. Note that this regression analysis was only performed on those aged 45 and over living in private households.

A second logistic regression was then performed, this time using only the sample of those elderly with a disability, to estimate the probability of expressing a need for assistance related to their level of disability and other socio-demographic characteristics. It is important to note that a need for assistance not related to a disability is disregarded here. Data from the 1996 General Social Survey (GSS) were used for this analysis (see text box "1996 General Social Survey – Social and Community Support").5 For this specific logistic regression, the sample size was reduced to 2,290 persons aged 65 and over having been identified as has having a disability (figure 1). Only four activities are considered throughout this research: grocery shopping, everyday housework, meal preparation and personal care. Only these activities were considered since they are the ones more frequently associated with home care services. The dependent variable (outcome) is dichotomous: need or no need for assistance. Of course, we make the assumption that the more severe the disability, the greater the probability of expressing a need for assistance.

Finally, multinomial logistic regressions were conducted using the 1996 General Social Survey to compute estimated probabilities of using formal, informal or mixed assistance given a specific set of sociodemographic characteristics among those who had expressed a need for assistance. The multinomial regression was performed using only those disabled elderly having received assistance because of their disability. It is important to note that 60% of those with a disability had expressed no need for assistance to perform the activities considered in this study. Among those 977 having expressed such a need, 936 had received assistance (figure 1).

Figure 1
General Social Survey sample population

Definitions

Type of assistance: Because we are most interested in the effect of the changing socio-demographic characteristics on the demand for home care services, we examined assistance received in the areas of everyday housework, shopping for groceries, meal preparation and personal care. With the information collected in the General Social Survey, we also were able to identify the reasons behind the need for assistance: temporary or long term health or physical limitations, temporary difficult times, task sharing in the household, time constraints, etc. In this research we focused on assistance received due to a long term health problems.

Source of assistance: The "informal network" includes family, friends and neighbours. The "formal network" includes paid employees (government or non-government employees) or volunteer from a private or public agency. Volunteers are part of the formal network. We use the expression "mixed network" when someone receives assistance from both the formal and informal network.

Level of disability: Some attributes of the Health Utility Index (HUI), a composite measure of health status within the 1996/97 National Population Health Survey and the 1996 General Social Survey, were used to define disability. This composite variable is based on the Comprehensive Health Status Measurements System and takes into account both the quantitative and qualitative aspects of health. First, it provides information on the functional health of an individual using the following attributes: vision, hearing, speech, emotion, mobility, dexterity, cognition, and pain and discomfort. The Health Utility Index is a single numerical value ranging from 0 to 1 and it takes into consideration any possible combination of levels of the eight self-reported health attributes. This numerical value also embodies the views of society concerning health status based on personal preferences about various health states from a representative sample. The first three attributes (vision, hearing and speech) were dropped since they were not predictors of need for assistance. Emotion was a predictor of need for assistance but it was also dropped as the questions were not indicative of long term health needs, an area of interest for this research. For the purpose of this research, levels of disability were defined as follows:

  • No disability
  • Mild disability:
    • Mobility problem but do not need any help;
    • Dexterity problem but do not need any help from someone else (may or may not use special equipment);
    • Somewhat forgetful and a little difficulty in thinking;
    • Moderate and/or severe pain prevents performing some or few tasks.
  • Moderate disability:
    • Requires wheel chair or mechanical support to walk;
    • Dexterity problem and need help to perform some tasks;
    • Very forgetful and a lot of difficulty in thinking;
    • Severe pain prevents performing most tasks.
  • Severe disability:
    • Can not walk or need help from others to walk;
    • Dexterity problem and need help for most or all tasks;
    • Unable to remember or think.

 

1996 General Social Survey – Social and community support

Most of the data used for this study came from Statistics Canada's 1996 General Social Survey, Cycle 11: Social and Community Support. The target population for the survey was all Canadians 15 years of age or over living in private households. Full-time residents of institutions as well as residents of the Yukon and Northwest Territories were excluded. Data were collected using Computer Assisted Telephone Interviewing (CATI), systematically excluding households without telephones. Statistics Canada estimates that less than 2% of the target population resides in this type of household and that their characteristics are not different enough from those of the rest of the target population to have an impact on the estimates. Survey estimates were adjusted (weighted) to account for persons without telephones. In total, the sample consisted of 12,756 respondents. The response rate was 85.3%.

Two of the survey's objectives were to learn about the types of assistance Canadians provide or receive, as well as to gain a better understanding of the dynamics that link a person's social network and the assistance this person gives and/or receives. To this end, the questionnaire was designed to collect detailed information on the type of assistance provided or received for the following activities: meal preparation, house cleaning, laundry and sewing, house maintenance and outside work, grocery shopping, transportation, banking and bill paying, personal care (bathing, toileting, care of toenails/fingernails, brushing teeth, shampooing and hair care or dressing) as well as moral or emotional support. Since we were interested in the effect of changing socio-demographic characteristics on the use of home care services, we concentrated on four activities that are more commonly associated with those services: everyday housework, shopping for groceries, meal preparation and personal care.

2.2 Results: the need for assistance in performing everyday activities

The results of this second analysis show the expected association between age and disability, age being positively associated with disability. Being a woman increases the probability of having higher levels of disability. Also, marital status has the expected effect, as past research has shown the protective effect of marriage. Married persons have lower probabilities of having higher levels of disability. As far as level of schooling, the association is very strong; those having no high school diploma, compared to those who do, have a much greater probability of having higher levels of disability. There are also some regional differences as elderly persons living in the province of Quebec have the lowest probabilities.

The results show that those with a severe level of disability were three times (odds ratio of 2.9) as likely to express a need for assistance compared to those with a mild disability. They are also twice as likely as those with a moderate level of disability to express a need for assistance. Also, disabled men were about half as likely to express a need for assistance compared to disabled women. This result could very well reflect the division of domestic labour, especially among these older generations. Disabled men, for example, may not associate the meal preparation done by their spouse as a need for assistance related to their disability, however disabled women will likely make this association if in the same situation. Finally, it is in the Atlantic provinces that the need for assistance is the greatest.

2.3 Results: factors associated with the use of informal and formal support networks

Results suggest that among disabled elderly males, those living with a spouse aged less then 75 years old are more likely to receive assistance from their informal network only or a mix of formal and informal. The number of surviving children has no significant effect on the source of assistance that disabled elderly males receive. However, a higher level of schooling increases quite significantly the probability of receiving assistance from the formal network, either in conjunction with the informal network or not. As can be expected, a higher level of disability is associated with greater odds of receiving a mix of formal and informal assistance. As for disabled elderly females, all things being equal, the absence of surviving children increases significantly the odds of receiving assistance from the formal network only. Also, the presence of a spouse is not nearly as significant as it is for disabled elderly males when looking at the odds of receiving informal versus formal assistance. If for males the presence of a spouse, especially if this spouse is less than 75 years old, increases the odds of receiving assistance from the informal network, this relationship is not significant among disabled females. As in the case of older males, a severe disability was associated with a much greater probability of receiving a mix of formal and informal assistance.

From the results shown in this section, it seems clear that demographic trends will have an impact on the use of formal home care services in the future. A decline in fertility rates affects the extent and nature of our immediate social environment. When baby boomers reach old age, they will have fewer children to provide them with assistance when needed. Results using the 1996 General Social Survey showed that for females, the presence of surviving children lower the odds of using the formal network. If these patterns of utilization are predictors of future patterns, it would indicate that the absolute and relative number of disabled females relying on the formal network will increase. Moreover, trends in divorce rates show that a greater proportion of those cohorts will enter old age as divorced individuals and may be living alone. This will in turn increase the need for formal home care services.

The second part of the research was designed to project the number of older persons with different levels of disability using the results presented above. Based on the results of the second logistic regressions and taking into account the changing nature and extent of the family network, we will then look, using projections done using microsimulation, at the number of disabled elderly who could be using their formal, informal or mixed network in the future.

 

5 . The 1996 General Social Survey was used instead of the 1996 National Population Health Survey because this latter survey does not provide information on the source of the assistance being provided.