Article Text

Download PDFPDF
Does physical activity reduce seniors' need for healthcare?: a study of 24 281 Canadians
  1. John C Woolcott1,
  2. Maureen C Ashe1,
  3. William C Miller1,
  4. Peilin Shi2,
  5. Carlo A Marra1,
  6. PACC Research Team
  1. 1University of British Columbia, Vancouver, Canada
  2. 2Aging Brain Center, Institute for Aging Research, Hebrew SeniorLife, Boston, Massachusetts, USA
  1. Correspondence to Carlo Marra, 620-1081 Burrard Street, Vancouver, British Columbia, Canada V6Z 1Y6; cmarra{at}


Objectives Physical inactivity has been associated with significant increases in disease morbidity and mortality. This study assessed the association between physical activity and (1) health resource use and (2) health resource use costs.

Design and Participants The responses from 24 281 respondents >65 years to the Canadian Community Health Survey Cycle 1.1 were used to find activity levels and determine health resource use and costs. Logistic regression models were used to assess risks of hospitalisation.

Results Physical inactivity was associated with statistically significant increases to hospitalisations, lengths of stay and healthcare visits (p<0.01). Average healthcare costs (based on the 2007 value of the Canadian dollar) for the physically inactive were $C1214.15 higher than the healthcare costs of the physically active ($C2005.27 vs $C791.12, p<0.01).

Conclusion Among those >65 years, physical activity is strongly associated with reduced health resource use and costs.

Statistics from

Regular physical activity has been shown to prevent numerous chronic conditions1 and decrease mortality.2 Conversely, physical inactivity is a major health epidemic, with significant impact on health-related quality of life3 and substantial increases in all-cause mortality.4

Although it has been reported that 77% of seniors are physically inactive,3 5 few researches have assessed the association between inactivity and health resource use and costs in elderly people,6 7 and none of these researches has been completed at the population level. Given that it is estimated that the population >65 years will double over the next decade,8 there is a definite need for greater understanding of the implications of physical inactivity on health resource use and cost.

The first objective of this study was to investigate the association between inactivity and risk of hospitalisation and increased visits to health professionals. In objective 2, we assessed the healthcare costs associated with physical inactivity.


Design and participants

The data used in our analysis were obtained from the private microdata file of the Canadian Community Health Survey (CCHS) Cycle 1.1.9 The data were collected between September 2000 and November 2001 using computer-assisted interviews completed over the telephone or in person. The CCHS is a cross-sectional health survey of the Canadian population designed to provide information about chronic conditions, various health outcomes, health resource use, sociodemographics and physical activity.

For CCHS respondents >65 years, we extracted responses to questions on age, sex, physical activity, body weight, health resource use, chronic conditions, health-related quality of life and marital status.

For our primary analysis of the impact of inactivity on hospitalisations and general health visits, individuals were categorised as being active or inactive based on their reported weekly physical activity levels. The minimum level of physical activity recommended by the American Surgeon General is 1000 kcal/4187 kj per week.1 10 Respondents who expended <1000 kcal/wk were deemed inactive. For our secondary analyses, the respondents were stratified by their physical activity levels (0–499, 500–999, 1000–1499, 1500–1999 or >2000 kcal/wk).

Weekly physical activity levels were estimated from the individual response duration and frequency of leisure physical activities (such as walking, gardening and ice hockey) during the previous 3 months. The weekly energy expended was estimated from the energy score (based on metabolic equivalent of the task values including individual body weight)11 12 for each of the reported activities multiplied by the time spent each week doing the activity.

The individual's health resource use for the previous 12 months was elicited from questions on the frequency and type of visits to physicians or other health professionals and overnight stays in a healthcare facility. Visits to family physicians, general practitioners, surgeons, allergists, orthopaedists, gynaecologists, or psychiatrists, physiotherapists, optometrists, nurse practitioners, chiropractors, social workers/counsellors or psychologists were included in our analyses.

Canada's universal healthcare system typically provides coverage such that health resource use costs are paid for by a third party (such as the Ministry of Health); as such, respondents were not asked about their individual healthcare costs. To estimate health resource use costs, we used the perspective of a third party payer and incorporated unit cost estimates (see table 1) from the British Columbia Ministry of Health, deemed sufficiently representative of the Canadian healthcare system costs.13,,15

Table 1

Individual health resource use unit cost estimates

Statistical methodology

The CCHS sample was selected using a multistage stratified cluster design and random digit dialling methodologies. The selection of individual respondents was designed to over-represent the young (<19 years of age) and the elderly (>65 years of age). Frequency weights provided by Statistics Canada were applied to all point estimates in this study to adjust for the differing probabilities that individuals were selected based on the complex sampling strategy. To estimate the precision around our estimates, bootstrapping was used to estimate the variances around our point estimates. This method takes into account the unequal probabilities of selection in calculating variances.

Our statistical analyses provided a comparison of both the health resource use and their associated costs between those individuals deemed inactive and active. Using the physically active as a reference group, logistic regression models were used to estimate adjusted odds ratios (OR) for predictors of hospitalisation.

Our secondary analysis of the relationship between health resource use costs and physical activity levels was completed by estimating costs of individuals in each of the five strata of activity. The mean cost of visits to a health professional and the total health resource use were separately estimated and compared across groups to determine whether health resource use differed by physical activity strata. All the data were analysed using the SAS (V.9.1) software.16


The analyses were completed on individual level data of 24 281 respondents. Of the respondents, 18 258 (75.19%) were inactive. The mean age of respondents was 73.82 years, with 15.36% (n=3976) of respondents reporting an overnight hospitalisation in the previous year, with a mean length of stay in hospital of 2.60 days (95% CI 2.22 to 2.98) (table 2).

Table 2

Characteristics of elderly respondents of the CCHS Cycle 1.1

The mean cost of visits to physicians or health professionals for individuals >65 years was estimated to be $C422.14, whereas the average total cost of all health resource use was $C1705.41. Extrapolating the results of the respondents to the Canadian population showed health resource use costs to exceed $C7.03 billion (based on the 2007 value of the Canadian dollar).

As shown in table 2, comparing inactive to active seniors showed that the inactive individuals had higher likelihood of being hospitalised (17.37% vs 9.22%, p<0.01) with longer lengths of stay (3.18 vs 0.82 days, p<0.01). In addition, physically inactive respondents reported significantly more visits to physicians and health professionals than the physically active (p<0.01).

The logistic regression analysis showed physically inactive individuals to have elevated risk of hospitalisation (adjusted OR 1.84; 95% CI 1.61 to 2.11) when compared with active respondents. There was an increased risk of hospitalisation among those >80 years (adjusted OR 1.62; 1.38 to 1.91) and those with >3 chronic conditions (adjusted OR 3.97; 2.87 to 5.50).

Health resource use costs were significantly higher for inactive individuals compared with those for individuals who were active (visit cost/total health resource use cost: $C331.22/$C2005.27 vs $C451.92/$C791.12, p<0.01). As shown in table 3, when stratifying costs by physical activity, health resource use costs decreased as individuals increased their activity. There was a non-statistically significant health resource use cost increase (p>0.10) when activity levels exceeded 2000 kcal/wk (compared with the costs for those with activity levels between 1500 and 1999 kcal/wk).

Table 3

Total health resource use cost by activity level


Our investigation showed that physically inactive seniors have significantly higher health resource use and associated costs than active individuals. When compared with an active individual, the risk of hospitalisation for inactive individuals is almost twice as great, with longer lengths of stay. Furthermore, the health resource use costs of an inactive senior were higher than 2.5 times greater than the health resource use costs of an active senior, and extrapolation of health resource use costs show the Canadian population of inactive seniors to have estimated health resource use costs >$C5.60billion (based on the 2007 value of the Canadian dollar).

The inactive respondents were more likely to be older, unmarried, have more chronic conditions and report lower health-related quality of life than the active respondents. The individuals with the lowest physical activity (<499 kcal/wk) had the highest health resource use costs.

Limitations to our study include the CCHS physical activity data only reflecting leisure physical activity rather than all activities. In addition, the severity of the chronic conditions is not reported, restricting our ability to assess the impact of disease severity on physical activity levels. Finally, the CCHS Cycle 1.1 data are cross-sectional, which does not allow inference of causality.


In those >65 years, physical activity has a strong association with health resource use and health resource use costs. Given the growth of the population >65 years, assessment, introduction and continuation of programmes that promote physical activity in this group are needed to decrease health resource use and their related costs while improving the general level of physical fitness and health.

What is already known on this topic

Physical inactivity is associated with increased chronic conditions and increased risk of mortality.

What this study adds

Our study identifies that among seniors, physical inactivity is associated with significantly increased costs due to health resource use including hospitalisations and health professional visits.


The research and analysis were completed with data from the private-use microdata file of the CCHS Cycle 1.1, accessed through Statistics Canada. The opinions expressed do not represent the views of Statistics Canada.



  • Competing interest None.

  • Funding This research was supported (in part) by the Canadian Institutes of Health Research Team Development Grant (M.C.A., W.C.M. and C.A.M.), the Canadian Institutes of Health Research (J.C.W., W.C.M. and C.A.M.), the Michael Smith Foundation for Health Services Research (J.C.W., M.C.A. and C.A.M.), the Canadian Arthritis Network (C.A.M.) and the Canada Research Chair in Pharmaceutical Outcomes (C.A.M.).

  • Ethics approval This study was conducted with ethics approval obtained from the University of British Columbia.

  • Provenance and peer review Not commissioned; externally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.