The Canadian Index of Wellbeing: key findings from the healthy populations domain.

Muhajarine, Nazeem ; Labonte, Ronald ; Winquist, Brandace D. 等

The Gross Domestic Product (GDP) of a country is a high-profile and often tracked measure that has emerged by default as a surrogate for wellbeing. (1) However, as the recent Commission on the Measurement of Economic Performance and Social Progress noted, "it has long been clear that GDP is an inadequate metric to gauge well-being over time particularly in its economic, environmental, and social dimensions." (2), (p.8) Reliance upon the GDP points to the need for a robust set of indicators able to measure other important dimensions of Canadians' social, economic and cultural lives. While measurement activity and indicator reporting are now commonplace across all sectors, Canada lacks a single, highly visible national instrument that monitors and publicly reports on improvements or setbacks to our collective wellbeing and its important influences. A well-designed, easily understandable, and technically sound tool that captures all of the factors contributing to our wellbeing would offer important information for all Canadians, and be useful as a guide to public policy discourse and decision-making.

The Canadian Index of Wellbeing (CIW), the result of almost eight years of preparatory work, has emerged as a tool for this purpose. Bringing together a diversity of indicators across eight principal domains--Arts, Culture and Recreation; Civic Engagement; Community Vitality; Education; Environment; Healthy Populations; Living Standards; and Time Use--a primary aim of the CIW is to monitor trends within each domain and to emphasize interconnections between them. Healthy Populations is the CIW domain that measures health outcomes and other risk (or protective) factors. This article describes the Healthy Populations framework, indicator selection process, and data sources, and a few of the key findings from the domain report.

METHODS

Healthy Populations Framework

Composed of ten "core" and six "secondary" health indicators, the CIW Healthy Populations framework covers eight subdomains (see Figure 1). Subdomains identify core aspects of health status (physical, functional, emotional, and psychological) and proximal health determinants such as lifestyle and behaviour, and health care. Since our framework is part of the larger CIW set, it does not incorporate important social determinants of health which are the foci of other CIW domains. Some key determinants of health that relate to equity (e.g., income, education, place and ethnocultural status) and the essential biological factors (notably age and sex) are considered in the Healthy Populations indicator analysis. Each subdomain is measured by at least one core indicator, and several are measured by additional secondary indicators. Age and sex differences as well as national and provincial/territorial trends are reviewed for each indicator in the full report. All core indicators were stratified by income and education to describe differences in health indicators and to identify instances of health inequity. In this paper we report the selected findings, or the "core story", of the CIW Healthy Populations domain.

[FIGURE 1 OMITTED]

Indicator selection

Indicators were selected based on the following criteria developed in this project. These criteria were set following extensive discussions at the CIW National Working Group and are consistent with the literature. (3,4)

* Quality of data: indicators are clearly defined, measurable, transparent and verifiable.

* Adequacy of data: indicators support benchmarking and monitoring over time; that is, longitudinal or repeated data are available for at least three points in time, allowing trend analyses.

* Relevance of data: indicators are commonly used, understandable, credible, meaningful, and policy relevant (that is, can change as a result of policy interventions).

* Variety of data: indicators capture both subjective and objective aspects of health outcomes, and represent both positive (desired) and negative (undesired) outcomes.

* Spatial sensitivity of data: indicators are available at national, provincial and municipal or lower levels of aggregation, allowing comparisons by geographic scale.

* Socio-demographic sensitivity of data: indicators allow for analyses by socio-economic and demographic differences; that is, data can be stratified by important socio-economic and demographic variables (sex, age group, income, cultural identity, geographic place).

Based on these criteria and a review of the international literature for comparable national indicator projects, the authors first compiled a comprehensive list of candidate indicators. A number of external experts, within Canada and internationally, were then consulted in a validation process, which involved experts indicating their recommendations and reasons for retaining or dropping a variable. The selection of the final "core" and "secondary" indicators for each subdomain rested with the authors of the domain with input from the National Working Group.

As a form of additional face validation, our final indicator selection was "mapped" against the Health Indicators Framework (HIF) used by Statistics Canada and the Canadian Institute for Health Information. (5) Our indicators overlap fully with the "health status" category and partially with other categories of the HIF, which include categories relevant to other CIW domains and so are not part of the Healthy Populations Domain. The HIF is one of the standard frameworks used by health ministries in determining their indicator use for a variety of purposes; and the independent coherence and comparability of the Healthy Populations Domain with existing health categories of public policy and data-gathering frameworks such as HIF suggests the relevance of the Domain's indicators within the larger CIW project.

Data sources

For the selected indicators, data were primarily obtained from Statistics Canada's data products, and included the Canadian Community Health Survey (CCHS) (cycles 1.1, 2.1, 3.1, 4.1) and the National Population Health Survey (1994, 1996, and 1998), the CCHS Mental Health and Wellbeing (cycle 1.2), CCHS Health Services Access Survey, Canadian Vital Statistics Database, and the Canadian Institute for Health Information, Hospital Morbidity Database.

[FIGURE 2 OMITTED]

Deriving a summary measure: Healthy Population Index

From the outset, following extensive discussions at the CIW National Working Group, it was decided that each CIW domain will have a summary measure and that this will be derived from a smaller subset of indicators within each domain. Each indicator was reviewed by members of the CIW National Working Group and voted on using a five-point rating (5=very important to 1=not at all important to the composite CIW). Voting was done individually, then again following group discussion. From this process, eight indicators ("headliners") were included in the calculation of the Healthy Population Index: percent rating their own health as excellent or very good, percent with diabetes, life expectancy at birth, percent daily or occasional smokers among adolescents (aged 12-19), percent with probable depression, percent rating health services overall as excellent or good, percent who had received immunization for influenza (age 65+) and average Health-Adjusted Life Expectancy (HALE) for those [greater than or equal to]15 years (percent of remaining years expected to be lived in good health).

All indicators were deemed equally important, therefore treated equally (not weighted) in the index (see Michalos et al. for in-depth discussion (4)). The headliner indicators included four negative indicators (i.e., increases in numerical values indicate decreases in some aspect of wellbeing) and four positive indicators (i.e., increases in numerical values indicate increases in some aspect of wellbeing). The calculation of index scores are described in detail elsewhere (see reference 4). Briefly, the principle for calculating an index score involved taking the first value for each indicator (1994 in our case, or baseline) and fixing it as a base of 100, and taking each value for subsequent years as a percentage change of this base value, and then taking the mean across all eight headliner indicators. The positive indicators were converted into a value of percent change by dividing every data value within an indicator by its first value and multiplying the result by 100. In order to standardize the index values so that increases and decreases in figures uniformly represent improvement or deterioration, respectively, the values of negative indicators were transformed into their reciprocals and the latter were turned into percentage change, similar to the manoeuvre for the positive indicators.

[FIGURE 3 OMITTED]

FINDINGS

A summary of the key findings is presented below; for a more detailed discussion, please refer to the full report. (6)

Canadians live longer, but not necessarily healthier, lives

The relatively high standard of living enjoyed by Canadians is matched by life expectancy rates that are among the best in the world. A closer look at health indicators, however, reveals a more mixed picture. Although Canadians are living longer, these additional gains in years of life are not necessarily lived in the fullest health possible. When taking into account functional limitations brought on by disease and disability, the number of years lived in full health for Canadian women is not substantially different in 2005 compared to in 1991 (data not shown); Canadian men, however, made slightly more gains during this period. As has been the case historically, women are still outliving men, but the gap is shrinking. A rich-poor gap in Canadian life expectancy was observed and can largely be explained by higher rates of death in childhood through to middle age, not in old age. Such preventable deaths early in life also negatively impact Aboriginal life expectancy, which continues to lag behind that of non-Aboriginal Canadians (data not shown).

For more than a decade, a majority of Canadians have declared that their overall health is very good or excellent (Figure 2). However, the proportion of Canadians who considered themselves as having optimal health peaked in 1998/99 at 65.2% and decreased dramatically in 2003 to 58.4%. Self-rated health began to rebound in 2005, but is still considerably lower than it was 10 years previously.

The decline in those who report their health in superlative terms needs some further explanation. Given the consistency of findings between this subjective measure and the objective and derived HALE measure, it could safely be said that the decline in self-reported health is likely not an artefact. Self-rated health is one of the most intensely researched measures of overall subjective health, and over the years studies have found a wide array of correlates of self-rated health. The overall Canadian population's age structure is likely not a solid explanation as our period of study is relatively short (15 years) and we observe downward trends every two years of data availability. The likely explanation would tend to be a complex mix of factors, including but not exclusively, changing prevalence of chronic diseases, mental health, and other psychosocial and social factors.

[FIGURE 4 OMITTED]

The decline in the share of the population who consider themselves in excellent or very good health has been most marked among Canadian teenagers, with a drop of 12.5 percentage points from its 1998/99 peak. This is matched by a steadily increasing share of teenagers who report problems with everyday functions (an increase of 8.8 percentage points from 1994/95 to 2005), and increasing obesity rates--a trend that augurs poorly for the health of this generation as they advance in age.

After peaking at the start of the new millennium, depression rates gradually declined. However, the 2007/08 estimate represented a slight increase. Throughout this period, the prevalence of depression has been consistently higher in women than in men. At their peak in 2000/2001, depression rates in women were nearly double those of men (9.4% and 5%, respectively). While this difference has diminished slightly in recent years, more women report depression than men. A notable income gradient in depression was also observed for both sexes, with males being most impacted by lower income levels.

Healthy living trends (some) catching on

Two encouraging findings were that the number of Canadians who smoke dropped markedly from 1996/97 to 2009, particularly among youth (Figure 3), and that there is a greater proportion of Canadians now compared to a decade ago reporting that they are at least "moderately active". A steep household income gradient was observed for both findings. In 2007/08, smoking rates in the lowest income households were more than double those of the highest.

Obesity continues to rise, despite more physical activity

Despite the overall and sustained rise in physical activity, the number of Canadians who are obese continues to increase (from 11.9% in 1997 to 17.9% in 2009). At the same time, more Canadians are now living with chronic diseases such as diabetes (which includes treated and untreated diabetes), the prevalence of which has doubled over 15 years (3% in 1994/95 to 6% in 2009). Diabetes rates are highest in the oldest age group, with nearly one in five people aged 65 and over affected in 2009. Income and education, even at relatively low levels, have a marked protective effect for diabetes. While this is true for both sexes, the effect is greater for women (data not shown).

[FIGURE 5 OMITTED]

Higher income adds both quantity and quality to life

Higher incomes and higher levels of education are associated with longer life expectancy and better self-reported health. Interestingly, the positive impact of income and education on self-rated health is most marked among women. At higher levels of income, women are more likely than men to consider themselves in very good or excellent health. Income and education effects were observed, with varying patterns, for most of the indicators analyzed. Income increases in the lower income brackets have the greatest impact in reducing the prevalence of diabetes, depression, and teen smoking.

Place matters

On a number of indicators, there were interesting differences between physical and mental health at the provincial and territorial levels. For instance, Newfoundlanders have lower life expectancies and generally higher rates of diabetes and other adverse health conditions than people in other Canadian provinces and territories; however, they have among the lowest levels of depression and are most likely to consider themselves as having excellent or very good health. On the other hand, British Columbians and Albertans enjoy among the longest life expectancies and lowest levels of obesity and diabetes; but they are also more likely than Newfoundlanders to report high levels of depression and less likely to say they are satisfied with the quality of their health services. These intriguing geographic differences may point towards cultural differences in self-evaluations, and should be the subject of further inquiry.

Pervasive health gaps remain for Aboriginal populations

Considerable progress has been made in terms of Aboriginal health and wellbeing in Canada, but challenges remain. (7) In recent years, Aboriginal people have experienced longer life expectancies and reduced mortality rates, although the gap in these rates between Aboriginal and non-Aboriginal Canadians remains unacceptably high. Life expectancy at birth in 2000 for the Aboriginal population was 68.9 years for males and 76.6 for females--a full 7.4 and 5.2 years lower than for Canadian men and women, respectively. The infant mortality rate for Aboriginal populations has fallen dramatically, from 27.6 deaths per 1,000 live births in 1979 to 6.4 in 2000. Still, it was 16% higher than the comparable Canadian rate for that same year. First Nations living on-reserve also rate their health in poorer terms. In 2002-03, 79.7% said their health was "excellent" or "very good" compared to 88% of the general Canadian population. (8) One potential explanation for this discrepancy could be the much higher rate of diabetes on-reserve (19.7% vs. 5.2%), a condition with which one in five adults had been diagnosed. The differences between diabetes rates on- and off-reserve were highest for young and middle-aged adults, with implications for future generations of seniors. Unintentional injury and suicide remain key challenges for Aboriginal people, as does infectious disease. Potential years of life lost from injury were almost 3.5 times higher than the national rate, with motor vehicle accidents being a primary cause of death for all age groups. The Healthy Populations report highlights a small portion of the long-known health disparities between Aboriginal and non-Aboriginal Canadians, a health gap that in recent years has narrowed somewhat but still remains unacceptably large.

[FIGURE 6 OMITTED]

Equity: The driving value

Equity is the core value that drives the CIW project. Health equity similarly informs the work undertaken in the Healthy Populations Domain. Health equity refers to the absence of unfair and avoidable or remediable differences in health among populations or groups defined socially, economically, demographically or geographically. (9) Various analyses of health indicators and comparisons to other nations suggest the potential for our collective capacity to improve the health of all Canadians. Income and other socioeconomic conditions that influence health are shaped by both private economic practices ("the market") and public policies (regulation, taxes, transfers and social protection spending). Income-related negative health effects are caused by both material conditions (inadequate access to resources for health) and psychosocial dynamics. (10,11) These conditions can be mitigated by government regulation, programs and services.

Despite the availability of universal health care services, with which a large majority of Canadians are satisfied when they encounter it as patients, the persistence of significant health gaps suggests both the need for health interventions tailored to socially excluded groups and the potential health benefits of initiatives outside the health field. This latter point was emphasized most strongly in the August 2008 Report of the World Health Organization Commission on Social Determinants of Health, Closing the gap in a generation: Health equity through action on the social determinants of health, to which many Canadian public health researchers and policy analysts made substantial contributions. (12)

Pulling it together: The Healthy Populations Index

A key deliverable for the CIW project is the calculation of a single, overall measure, or index, of Canadians' wellbeing. Such indices were also calculated for each individual CIW domain. * Figure 6 presents the Healthy Populations Domain index along with the overall Canadian Index of Wellbeing and the GDP from 1994 to 2010. As shown, there is an overall increase or positive trend in the health of populations that began in 2001 (corresponding to a 6.6% increase in the index), and while this increase tracks closely with the overall CIW index, it is far below the rapidly climbing increase in GDP. The gains in GDP do not necessarily translate directly to gains in overall wellbeing or health of populations. Decreasing teen smoking rates and increasing influenza immunization rates in seniors were two of the key drivers behind the recent improvement in the Healthy Population index, while increasing prevalence of diabetes, depression, and stagnant self-rated health in recent years suppressed further gains in the overall score.

CONCLUSION

The Canadian Index of Wellbeing is a multifaceted measurement and monitoring tool developed to engage Canadians in conversations about their health and wellbeing that go beyond health care or the economy, and about acting on changes that matter in their lives. The Healthy Population domain of the CIW stands as one important component in measuring wellbeing; however, a greater value of the domain is to understand the contributions it makes to and receives from other domains of the CIW that, collectively, contribute to overall wellbeing. Understanding the interactions between the indicators comprising CIW domains, currently and over the years past, is the next challenge facing CIW researchers.

The gains made in health trends in recent years, however, are surprisingly modest and hide vast discrepancies in the health of Canadians. Without proper policy intervention, Canada will continue to see a health divide along income lines. Inaction will lead to increased pressures on the health care system in the long run. With the current rates of growth in obesity and diabetes, we might experience diminishing life expectancy, particularly among the younger generation of Canadians. The relationship between income and health calls for a multipronged approach that does not simply address health policies, but also social policies in order to advance our collective wellbeing. By implementing greater intergovernmental efforts in benchmarking, streamlining, and coordinating health programming and targets, we will enhance equity and effectiveness across regions and populations.

Appendix A. Indicators included in the Healthy Population
Index, Canada

Year    Self-rated      Diabetes       Life
          Health       Prevalence   Expectancy
       (% excellent     (% yes)      at Birth
       or very good)                 (years)

1994       100.0         100.0        100.0
1995       100.2          96.8        100.3
1996       100.5          93.8        100.5
1997       101.9          89.6        100.8
1998       103.3          85.7        101.0
1999       101.3          81.1        101.3
2000        99.3          76.9        101.5
2001        97.3          73.2        101.8
2002        94.9          69.0        102.0
2003        92.6          65.2        102.2
2004        93.9          63.2        102.5
2005        95.2          61.2        102.8
2006        94.8          56.1        103.3
2007        94.5          51.7        103.3
2008        93.3          50.8        103.3

Year   Teen Smoking   Depression      Patient
           Rate         (% with     Satisfaction
       (aged 12-19,    probable     With Overall
        % daily or    depression)      Health
        occasional                    Services
         smokers)                    (% rating
                                    services as
                                     excellent
                                      or good)

1994      100.0          100.0         100.0
1995       98.4          112.8         100.0
1996       96.8          129.3         100.0
1997      102.0          123.3         100.0
1998      107.7          117.8         100.0
1999      109.0           98.1         100.0
2000      110.4           84.1         100.0
2001      111.8           73.6         100.0
2002      124.4           80.9         101.4
2003      140.3           89.8         102.8
2004      154.8           95.5         101.9
2005      172.7          101.9         100.9
2006      173.4           94.6         101.9
2007      174.2           88.3         102.8
2008      183.3          88.3          102.8

Year    Influenza      Average HALE     Average
       Immunization   for 15+ (% of    of Healthy
           Rate         remaining      Population
        (age 65+)     years expected   Indicators
         (% yes)      to be lived in
                       good health)

1994      100.0           100.0          100.0
1995      100.0           101.1          101.2
1996      100.0           102.2          102.9
1997      106.3            99.3          102.9
1998      112.6            96.5          103.1
1999      118.9            96.3          100.8
2000      125.2            96.0           99.2
2001      131.5            95.8           98.1
2002      130.9            97.1          100.1
2003      130.3            98.4          102.7
2004      134.6            97.2          105.4
2005      138.8            96.1          108.7
2006      136.5            96.1          107.1
2007      134.2            96.1          105.7
2008      134.2            96.1          106.6

Acknowledgements: This paper is an abbreviated version of a full-length report published by the Canadian Index of Wellbeing foundation that is available at: https://uwaterloo.ca/canadian-index-wellbeing/sites/ca.canadian-index-wellbeing/files/ uploads/files/CIW2012-HowAreCanadiansReallyDoing-23Oct2012_0.pdf. The authors acknowledge the contributions of Jacqueline Quail, Elisabeth Bartlett, Matthew Sanger, Toby Sanger and Fleur Macqueen Smith to the full report. We also thank colleagues in the CIW project and from Statistics Canada for their feedback and assistance; external anonymous reviewers and a number of experts who commented on the full report and on the selection of indicators; and the Atkinson Foundation for their initial leadership and support of the CIW project.

Conflict of Interest: None to declare.

REFERENCES

(1.) Stiglitz JE, Sen A, Fitoussi J. Report by the Commission on the Measurement of Economic Performance and Social Progress. Available at: http://www.stiglitz-sen-fitoussi.fr/documents/rapport_anglais.pdf (Accessed September 13, 2011).

(2.) Osberg L, Sharpe A. Centre for the Study of Living Standards. Beyond GDP: Measuring Economic Well-being in Canada and the Provinces, 1981-2010. Available at: http://www.csls.ca/reports/csls2011-11.pdf (Accessed September 13, 2011).

(3.) Hagerty MR. Quality of life indexes for national policy: Review and agenda for research. SocIndicators Res 2001;55(1):1-96.

(4.) Mihalos AC, Smale B, Labonte R, Muhajarine N, Scott K, Moore K, et al. The Canadian Index of Wellbeing: Composite Index Report. Technical Report 1.0. Waterloo, ON: Canadian Index of Wellbeing and University of Waterloo, 2011. Available at: http://ciw.ca/reports/en/Reports%20and%20FAQs/Cana dian_Index_of_Wellbeing-TechnicalPaper-FINAL.pdf (Accessed July 6, 2012).

(5.) Statistics Canada. Health Indicator Framework. Available at: http://www.statcan.gc.ca/pub/82-221-x/2011002/hifw-eng.htm (Accessed September 13, 2011).

(6.) Labonte R, Muhajarine N, Winquist B, Quail J. Healthy Populations: A Report of the Canadian Index of Wellbeing. Available at: http://ciw.ca/en/CanadianIndexOfWellbeing/DomainsOfWellbeing/HealthyPopulations.html (Accessed October 18, 2011).

(7.) First Nations & Inuit Health Branch, Health Canada. A Statistical Profile on the Health of First Nations in Canada for the Year 2000. 2005. Available at: http://www.hc-sc.gc.ca/fniah-spnia/pubs/aborig-autoch/2009-statsprofil/index-eng.php (Accessed September 13, 2011).

(8.) Health Canada. A Statistical Profile on the Health of First Nations in Canada: Self-rated Health and Selected Conditions, 2002 to 2005. Available at: http://www.hc-sc.gc.ca/fniah-spnia/pubs/aborig-autoch/index-eng.php (Accessed October 17, 2011).

(9.) Solar O, Irwin A. Towards a Conceptual Framework for Analysis and Action on the Social Determinants of Health: Discussion Paper for the Commission on Social Determinants of Health. Geneva, Switzerland: World Health Organization, 2007. Available at: http://www.who.int/social_determinants/ resources/csdh_framework_action_05_07 (Accessed September 13, 2011).

(10.) Lynch JW, Smith GD, Kaplan GA, House JS. Income inequality and mortality: Importance to health of individual income, psychosocial environment, or material conditions. BMJ 2000;320:1220-24.

(11.) Marmot M, Wilkinson RG. Psychosocial and material pathways in the relation between income and health: A response to Lynch et al. BMJ 2001;322:1233-36.

(12.) Commission on Social Determinants of Health. Closing the Gap in a Generation: Health Equity Through Action on the Social Determinants of Health. Final Report of the Commission on Social Determinants of Health. Geneva: WHO, 2008. Available at: http://www.who.int/ social_determinants/ final_report/csdh_finalreport_2008.pdf (Accessed September 13, 2011).

* Refer to the Composite Index Report (4) for a full discussion on the methodology and individual results for each Healthy Populations domain indicator and the other CIW domains.

Received: October 27, 2011

Accepted: July 18, 2012

Nazeem Muhajarine, PhD, [1,2] Ronald Labonte, PhD, FCAHS, [3] Brandace D. Winquist, MSc [1]

Author Affiliations

[1.] Department of Community Health and Epidemiology, College of Medicine, University of Saskatchewan, Saskatoon, SK

[2.] Saskatchewan Population Health and Evaluation Research Unit, University of Saskatchewan, Saskatoon, SK

[3.] Department of Epidemiology and Community Medicine, Faculty of Medicine, and Institute of Population Health, University of Ottawa, Ottawa, ON

Correspondence: Dr. Nazeem Muhajarine, Health Sciences Building, University of Saskatchewan, 107 Wiggins Road, Saskatoon, SK S7N 5E5, Tel: 306-966-7940, E-mail: [email protected]