Comparing inequalities in oral and general health: findings of the Canadian Health Measures Survey.
Ravaghi, Vahid ; Quinonez, Carlos ; Allison, Paul J. 等
Oral health constitutes an important part of general health. The US
Surgeon General's report on oral health asserts that oral health is
a critical component of health and must be included in the provision of
health care. (1) Despite this, there are differences in provision of
general and oral health care in many developed countries, including
Canada. While Canada has universal health insurance providing for
hospital and physician care, this excludes oral health care. (2) There
are also social justice challenges in Canada's oral health care
system, in that the most socially and economically vulnerable people
have the highest level of oral health problems but also the greatest
difficulty in accessing oral health care. Two nationally representative
surveys of Canadians have reported that a large proportion of the
population faces financial barriers to accessing oral health care, and
that these limitations are more pronounced among lower income groups.
(3,4) Moreover, the limited access to oral health care among the poor is
likely to influence the gap in oral health between the poor and the
rich. The poor, who avoid seeing dental professionals because of issues
such as the affordability and acceptability of care, receive less
preventive treatments and postpone curative treatments; therefore, it is
arguable that they develop more severe oral health problems and have
more untreated disease.
Although the inequality in use of health services is not uncommon
in health sectors, the size of inequalities can differ based on the
structure of health systems. For example, in Canada, Allin et al.
demonstrated that the inequality in oral health care was the largest
when compared to other parts of the health sector (e.g., physician and
hospital care). (5) These differences in provision of oral and general
health care may accordingly translate into variation in the size of
inequalities for oral and general health.
In Canada, and internationally, there has been no comparative study
on the magnitude of income-related inequalities in oral and general
health. The aim of this study is thus to estimate and compare the
magnitude of inequality in oral health, here indicated by the number of
decayed and missing teeth, and general health, indicated by obesity and
high blood pressure, using a nationally representative sample of
Canadian adults.
METHODS
Sample
Data for this study are from the 2007/09 Canadian Health Measures
Survey (CHMS). This survey is the most recent national health survey of
the Canadian population, and collected data on oral health indicators.
The CHMS is a probabilistic multistage stratified survey of households,
excluding institutionalized populations. The CHMS collected data from
5,604 Canadians aged 6-79 years. For the purpose of this study, we
analyzed data for Canadian adults aged 18-79 years. The CHMS aimed to
provide national estimates by collecting data from several sites
covering 97% of the population. The methodology and sampling framework
of the CHMS have been described by Statistics Canada elsewhere. (6)
Because we investigated the number of decayed teeth in this study, we
excluded edentulous adults with no teeth in the upper and lower jaws
(n=302). Subjects with missing data were also excluded from analyses
(n=351), and of those, the majority were missing data on income (n=326).
There was no significant difference with regard to oral and general
health between the respondents with missing values on income and those
who reported income.
Health outcomes
We examined two oral health outcomes: the number of decayed teeth
and the number of missing teeth; and two general health outcomes:
obesity and high blood pressure. The latter were chosen as, like oral
disease, they represent diseases strongly associated with behaviour
(e.g., oral hygiene and diet). In the CHMS data, blood pressure was
measured by a clinician. Obesity and high blood pressure are not only
major public health problems, they are also the leading global risks for
mortality in the world according to the World Health Organization. (7)
With regard to blood pressure, Statistics Canada has classified
participants into six groups (within acceptable range, at high end of
acceptable range, above acceptable range, moderately high, high, and
very high). For the purpose of this study, we collapsed participants
into two groups: No high blood pressure (within acceptable range) versus
high blood pressure (at high end of acceptable range, above acceptable
range, moderately high, and high). Obesity was ascertained using Body
Mass Index (BMI). BMI was calculated as weight (kg)/height ([m.sup.2]).
We used the definition of obesity recommended by WHO, in which
individuals with BMI >30 were considered obese. (8)
Income
We used "total annual household income" as a measure of
socioeconomic status for a number of reasons. Alternative indicators of
socio-economic status such as educational attainment and occupational
status tend to be stable or provide little variation among adults, and
therefore would mask substantial socio-economic variation in health
outcomes. More importantly, the calculation of a relative concentration
index requires a ranking socio-economic measure. In the CHMS, total
annual household income was originally reported as a 12-level ordinal
variable. Participants reported their total annual household income
range from zero to more than $100,000. Using this variable, we
calculated the "equivalized household income" employing a
"modified-OECD scale" approach. (9) This approach takes into
account the number of people in the household and their ages. We
initially merged the three lower income groups to permit sufficient
numbers of observations. The upper boundary for the last category of
income variable (more than $100,000) was set at $120,000 to be
consistent with the income range in the previous categories of income.
This decision was made to calculate the midpoint of income range for
those in this income category. We used the formula below to calculate
the equivalized household income using the median point of the ranges of
the ordinal variable. This method originally defines those aged 0-14 as
children and those 15 and older as adults. However, for calculating the
equivalized household income, we defined children as those aged 0-15 due
to availability of information in the CHMS dataset.
Equivalized household income =
Midpoint of income range/1+0.5*(No. of adults-1)+0.3*(No. of
children)
Concentration index
Several standard approaches exist for measuring inequalities in
health by income level. (10) We used the "concentration index"
to quantify income-related health inequalities. This approach was first
developed by Wagstaff et al. (11) and now has become a common
measurement tool in the epidemiological and health economics literature
to investigate the magnitude of inequality in health. The concentration
index is derived from the concentration curve (Figure 1). Values of the
concentration index range from -1 to +1 with 0 indicating no inequality,
negative values indicating concentration of the outcome among lower
income groups, and positive values indicating concentration of the
outcome in higher income groups. The greater the absolute value of the
concentration index, the greater the degree of concentration in a
negative or positive direction. Wagstaff (12) pointed out that when the
health outcome is binary, such as obesity and hypertension, and for
large samples, the bounds of the concentration index are equal to p-1
and 1-p, respectively, where p is the proportion of outcome variable. In
order to permit comparison of the concentration indices of binary
outcomes for this study (obesity and high blood pressure) with those of
other outcomes (decayed teeth and missing teeth), the concentration
index needs to be normalized, because otherwise the bounds are not -1
and +1. For binary outcomes of the study, the normalized concentration
index was calculated according to the method by Wagstaff:12 [C.sub.n] =
C/(1-p).
Statistical analyses
We first examined the distribution of oral and general health
outcomes across the quintiles of equivalized household income. We then
calculated the concentration index of oral and general health outcomes
for both sexes and for the entire population. All data analyses were
adjusted for age. Sampling weights suggested by Statistics Canada were
used for all stages of the data analyses to adjust for sample design
effects in order to produce nationally representative estimates.
Data analyses were performed using STATA 11.1 and ADePT (4.0). The
ADePT is a statistical program developed by the World Bank. In this
study, we used the ADePT's Health Module for analyzing health
inequality and for estimating the variance of the linear model.
RESULTS
We analyzed the data for 3,413 Canadian adults (1,601 men and 1,812
women). The prevalence of obesity and high blood pressure was 23.4% and
14.2%, respectively. The mean number of decayed teeth and the number of
missing teeth was 0.58 (SD=0.007) and 2.07 (SD=0.013), respectively.
Figure 2a shows the weighted proportions of obesity and high blood
pressure according to equivalized household income for the entire
population and for both sexes. Figure 2b shows the weighted mean number
of decayed and missing teeth across the quintiles of equivalized
household income for the entire population and for both sexes. Those
from higher socio-economic status had a lower prevalence of obesity and
high blood pressure and lower mean numbers of decayed and missing teeth.
However, the graded shape of the relationship between income and oral
health is more consistent than the shape of the relationship between
income and general health outcomes.
Table 1 presents the concentration indices for both general and
oral health outcomes. This table suggests three things for the entire
population: 1) the concentration indices for all general and oral health
outcomes were negative, which indicates a higher concentration of
general health outcomes and oral health outcomes among the poor; 2) the
concentration indices for oral health outcomes suggested statistically
significant deviation from equality whereas the concentration indices of
general health outcomes did not; and 3) the absolute values for the
concentration indices for oral health outcomes were greater than those
for the general health outcomes. We performed additional analyses which
confirmed that the differences between the CIs for oral and general
health are statistically significant (details of analyses not shown;
available upon request from the corresponding author). Sex-stratified
calculations of the concentration indices replicated the above findings
for both sexes, with the exception of obesity among females, which was
statistically significant.
[FIGURE 1 OMITTED]
DISCUSSION
This study compared income-related inequalities in Canada between
oral and general health outcomes. We found that oral and general health
outcomes were concentrated among lower income groups; however, only for
oral health outcomes was there a significant deviation from equality. We
also found that the magnitude of inequalities was greater for oral
health outcomes compared to general health outcomes. Our study not only
adds to the existing evidence for oral health inequalities in Canada,
(13) but also suggests that inequality in oral health may be a greater
problem.
The greater magnitude of inequalities in oral health outcomes
compared to general health is in line with dental care being the most
unequal aspect of health care in Canada. (5) Access to oral health care
is a great challenge for many Canadians, particularly low income groups,
(3,4) and financial barriers, in turn, are linked to poorer oral health.
(3) We suggest that the variation in the provision of medical and dental
care in Canada is among the factors that are responsible for the greater
magnitude of oral health inequality. While Canada has nationalized
hospital and physician care through its universal health insurance
program, oral health care is almost wholly privately financed, with
approximately 60% of dental care paid through employment-based
insurance, and 35% through out-of-pocket expenditures. (14) This
strongly links the utilization of and access to dental care with the
ability to pay.
[FIGURE 2a OMITTED]
[FIGURE 2b OMITTED]
More equitable access to health care plays an important role in
reducing inequalities in health. (15,16) The differences in the
provision of medical care in Canada (universal health insurance) and the
US (mostly privately funded) have provided a platform for comparing the
effects of health care on health inequalities. A number of studies
suggest that health inequalities are more pronounced in the US than in
Canada (17-19) due to cross-national differences in social policies,
particularly in universal health insurance.
In the dental literature, there has been recent interest in
evaluating the role of oral health care in reducing oral health
inequalities, and the existing evidence has yielded inconsistent
findings. (20-23) Most notably, Wamala et al. found that approximately
60% of inequalities in oral health are due to "refraining from
seeking dental treatment". (20) Similarly, it is suggested that
dental attendance can partly explain oral health inequalities with
regard to the number of sound teeth. (22) In contrast, other studies did
not suggest a major effect on oral health care of reducing health
inequalities. (21,23) These studies looked at different characteristics
of oral health care (dental visits, patterns of dental service use,
etc.) in various settings, which might explain the contradictory
findings. Our study, however, did not examine to what extent social
policies and the characteristics of health care are responsible for
inequalities in oral and general health. One great advantage of the
concentration index approach, which we used in this study, is that the
concentration index is a "decomposable" index. (24) This
enables quantifying the extent to which potential factors contribute to
health inequalities. We advise that future studies employ the analysis
of "decomposition of the concentration index" to evaluate what
factors are responsible--and to what extent--for inequalities in oral
and general health and how these factors differ between oral and general
health inequalities.
For high blood pressure and obesity, there appear to be no
statistically significant income-related inequalities. The absence of
significant income-related inequalities for these two general health
outcomes in Canada is partially in keeping with the existing literature.
A study of elderly Canadians did not find income-related inequalities
for high blood pressure, (25) and other Canadian studies have documented
the absence of significant inequalities for major health outcomes such
as mortality (17) and quality of life. (18) Yet with regard to obesity,
studies of Canadian populations have yielded mixed findings concerning
the relationship between socioeconomic status and obesity. (26-28) There
are a number of explanations for the inconsistency between our findings
and those of previous studies in Canada. First, the relationship between
socio-economic status and obesity is complex and varies extensively
according to sex and the measurement of socio-economic status. (29)
Second, previous studies looked at obesity by analyzing BMI scores
whereas our study classified individuals into two groups, i.e., obese
and non-obese. Third, it is possible that the status of inequalities has
varied over time.
This study benefitted from the concentration index approach, which
is an innovative method in health inequality research. (24) Using the
concentration index gives us a more comprehensive picture of the health
inequalities across the population. Traditionally, regression analyses
have been used to investigate the association between socio-economic
status and health outcomes. Accordingly, odds ratios or beta
coefficients are reported to indicate the magnitude and direction of
association. This approach, despite being common, is somewhat limited in
that regression-based analyses do not allow measuring inequalities
across the whole range of the socio-economic hierarchy. Comparing
inequality across studies or over time using traditional regression
analysis is also difficult because studies employ different categories
of socio-economic status. The concentration index is limited in that it
can only be applied if a strict ranking socio-economic variable is
present. In this survey, a large portion of the sample had missing
values for the actual values of income. Thus, to minimize the bias due
to missing values, we opted to use ordinal rather than actual values of
income.
Comparative analysis is an important tool for understanding health
inequalities. Although much is known about inequalities in both oral and
general health, no comparative research has investigated differences in
the magnitude of these inequalities. Previous comparative study of
inequalities in oral and general health are limited in that they only
investigated the absence or presence of inequalities. (30) Our study is
the first, to our knowledge, to compare the magnitude of these
inequalities. One limitation associated with comparing general and oral
health outcomes is that the investigated outcomes do not represent all
domains of general and oral health. Therefore, caution should be taken
with regard to extrapolating the findings of this study to other general
and oral health outcomes. This study also had some other limitations.
Due to the cross-sectional nature of the survey, we are unable to
establish causal relationships between income and health. Further, the
CHMS excluded those Canadians living in institutions, on crown land or
Indian reserves, or in remote regions, as well as full-time members of
the Canadian Forces. Sample size was designed to obtain estimates of the
prevalence of health conditions in the Canadian population as a whole
and so when analyses are performed using variables with multiple
categories, the power of the analyses is reduced. We also collapsed the
blood pressure and obesity variables to dichotomous, which may have
reduced the power of the study. Excluding the edentulous individuals may
have influenced the calculation of equivalized income.
It is no longer sufficient to look at the presence or absence of
inequalities. Attention should be paid to the magnitude of inequalities
in order to identify priority areas for intervention when tackling
health inequalities. Inequalities in oral health have decreased in
Canada over the past decades. (31) Despite this, our findings suggest
that inequalities in oral health may represent a greater challenge in
comparison with inequalities in general health. The variation in the
funding of oral health care and general health care, in this regard, is
of interest. The funding of oral health care in Canada has received
attention in recent years with dental professionals demanding more
strategic involvement from governments, (32) and the public supporting
its incorporation into Canada's national system of health insurance
yet concurrently not ranking dental care as a first priority for
government funding. Given this, and that implementing universal health
care is suggested as an effective approach to reduce inequalities in
oral health, (33) policy makers may need to consider shifting towards
some level of universal oral health care.
Acknowledgements: This research was funded in part by the
Association of Canadian Faculties of Dentistry, the Canadian Association
for Dental Research, the Institute for Musculoskeletal Health and
Arthritis, the Nova Scotia Health Research Foundation, the Order of
Dentists of Quebec and the Network for Oral and Bone Health Research.
Access to data from the 2007-2009 Canadian Health Measures Survey was
obtained through Statistics Canada at McGill University.
Conflict of Interest: None to declare.
REFERENCES
(1.) US Department of Health and Human Services. Oral Health in
America: A Report of the Surgeon General. Rockville, MD: US Department
of Health and Human Services, National Institutes of Health, National
Institute of Dental and Craniofacial Research, 2000. NIH publication
00-4713.
(2.) Stamm J, Waller M, Lewis D, Stoddart G. Dental Care Programs
in Canada: Historical Development, Current Status, and Future
Directions--A Report Prepared on Contract for the Department of National
Health and Welfare, Canada. Ottawa, ON: Canadian Government Publishing
Centre, 1986.
(3.) Locker D, Maggirias J, Quinonez C. Income, dental insurance
coverage, and financial barriers to dental care among Canadian adults. J
Public Health Dentistry 2011;71(4):327-34.
(4.) Report on the Findings of the Oral Health Component of the
Canadian Health Measures Survey 2007-2009. Available at:
http://www.hc-sc.gc.ca/
ahc-asc/branch-dirgen/fnihb-dgspni/ocdo-bdc/project-eng.php (Accessed
January 1, 2014).
(5.) Allin S. Does equity in healthcare use vary across Canadian
provinces? Healthc Policy 2008;3(4):83-99.
(6.) Statistics Canada. Canadian Health Measures Survey (CHMS).
Available at: http://www.statcan.gc.ca/cgi-bin/imdb/p2SV.pl?Function=getSurvey&SDDS= 5071&lang=en&db= imdb&adm=8&dis=2
(Accessed January 1, 2014).
(7.) Mathers C, Stevens G, Mascarenhas M. Global Health Risks:
Mortality and Burden of Disease Attributable to Selected Major Risks.
Geneva, Switzerland: World Health Organization, 2009.
(8.) World Health Organization. Global Database on Body Mass Index:
An Interactive Surveillance Tool for Monitoring Nutrition Transition.
Available at: http://apps.who.int/bmi/ (Accessed January 1, 2014).
(9.) Organisation for Economic Co-operation and Development. What
Are Equivalence Scales? 2010. Available at:
http://www.oecd.org/eco/growth/ OECD-Note-EquivalenceScales.pdf
(Accessed January 1, 2014).
(10.) Mackenbach JP, Kunst AE. Measuring the magnitude of
socio-economic inequalities in health: An overview of available measures
illustrated with two examples from Europe. Soc Sci Med
1997;44(6):757-71.
(11.) Wagstaff A, Paci P, Vandoorslaer E. On the measurement of
inequalities in health. Soc Sci Med 1991;33(5):545-57.
(12.) Wagstaff A. The bounds of the concentration index when the
variable of interest is binary, with an application to immunization
inequality. Health Econ 2005;14(4):429-32.
(13.) Ravaghi V, Quinonez C, Allison JP. The magnitude of oral
health inequalities in Canada: Findings of the Canadian Health Measures
Survey. Community Dent Oral Epidemiol 2013;41(6):490-98.
(14.) Chaplin R, Earl L. Household spending on health care. Health
Rep 2000;12(1):57-65 (Eng); 61-70 (Fre).
(15.) Mackenbach JP. An analysis of the role of health care in
reducing socioeconomic inequalities in health: The case of the
Netherlands. Int J Health Serv 2003;33(3):523-41.
(16.) Gelormino E, Bambra C, Spadea T, Bellini S, Costa G. The
effects of health care reforms on health inequalities: A review and
analysis of the European evidence base. Int J Health Serv
2011;41(2):209-30.
(17.) Ross NA, Wolfson MC, Dunn JR, Berthelot JM, Kaplan GA, Lynch
JM. Relation between income inequality and mortality in Canada and in
the United States: Cross sectional assessment using census data and
vital statistics. Br Med J 2000;320(7239):898-902.
(18.) Huguet N, Kaplan MS, Feeny D. Socioeconomic status and
health-related quality of life among elderly people: Results from the
Joint Canada/United States Survey of Health. Soc Sci Med
2008;66(4):803-10.
(19.) McGrail KM, van Doorslaer E, Ross NA, Sanmartin C.
Income-related health inequalities in Canada and the United States: A
decomposition analysis. Am J Public Health 2009;99(10):1856-63.
(20.) Wamala S, Merlo J, Bostrom G. Inequity in access to dental
care services explains current socioeconomic disparities in oral health:
The Swedish National Surveys of Public Health 2004-2005. J Epidemiol
Community Health 2006;60(12):1027-33.
(21.) Somkotra T, Detsomboonrat P. Is there equity in oral
healthcare utilization: Experience after achieving Universal Coverage.
Community Dent Oral Epidemiol 2009;37(1):85-96.
(22.) Donaldson AN, Everitt B, Newton T, Steele J, Sherriff M,
Bower E. The effects of social class and dental attendance on oral
health. J Dent Res 2008;87(1):60-64.
(23.) Ravaghi V, Underwood M, Marinho V, Eldridge S. Socioeconomic
status and self-reported oral health in Iranian adolescents: The role of
selected oral health behaviors and psychological factors. J Public
Health Dent 2012;72(3):198-207.
(24.) Harper S, Lynch J. Commentary: Using innovative inequality
measures in epidemiology. Int J Epidemiol 2007;36(4):926-28.
(25.) Kaplan MS, Huguet N, Feeny DH, McFarland BH. Self-reported
hypertension prevalence and income among older adults in Canada and the
United States. Soc Sci Med 2010;70(6):844-49.
(26.) Godley J, McLaren L. Socioeconomic status and body mass index
in Canada: Exploring measures and mechanisms. Can Rev Sociol
2010;47(4):381-403.
(27.) Shields M, Tjepkema M. Trends in adult obesity. Health Rep
2006;17(3):53 59.
(28.) McLaren L, Auld MC, Godley J, Still D, Gauvin L. Examining
the association between socioeconomic position and body mass index in
1978 and 2005 among Canadian working-age women and men. Int J Public
Health 2010;55(3):193-200.
(29.) McLaren L. Socioeconomic status and obesity. Epidemiologic
Rev 2007;29:29-48.
(30.) Sabbah W, Tsakos G, Chandola T, Sheiham A, Watt RG. Social
gradients in oral and general health. J Dent Res 2007;86(10):992-96.
(31.) Elani HW, Harper S, Allison PJ, Bedos C, Kaufman JS.
Socio-economic inequalities and oral health in Canada and the United
States. J Dent Res 2012;91(9):865-70.
(32.) Quinonez CR, Figueiredo R, Locker D. Canadian dentists'
opinions on publicly financed dental care. J Public Health Dentistry
2009;69(2):64-73.
(33.) Sehgal AR. Universal health care as a health disparity
intervention. Ann Intern Med 2009;150(8):561-62.
Received: April 12, 2013
Accepted: October 17, 2013
Vahid Ravaghi, DDS, MSc, PhD, [1] Carlos Quinonez, DMD, MSc, PhD,
FRCDC, [2] Paul J. Allison, BDS, FDS RCS (Eng), PhD [3]
Author Affiliations
[1.] Oral Health & Society Research Unit, Faculty of Dentistry,
McGill University, Montreal, QC
[2.] Assistant Professor and Program Director, Dental Public
Health, Faculty of Dentistry, University of Toronto, Toronto, ON
[3.] Professor, Faculty of Dentistry, McGill University, Montreal,
QC
Correspondence: Vahid Ravaghi, Lecturer, The School of Dentistry,
College of Medical and Dental Sciences, University of Birmingham, St
Chad's Queensway, Birmingham, B4 6NN, United Kingdom, Tel:
+44(0)1214665377, E-mail: v.ravaghi@bham.ac.uk
Table 1. Concentration Index for General Health Outcomes and Oral
Health Outcomes by Sex
Health Outcome Total Population
Concentration SE (95% CI)
Index
General Obesityt -0.05 0.027 (-0.1, 0)
health High blood -0.04 0.039 (-0.11, 0.04)
pressure
([dagger])
Oral Decayed teeth -0.25* 0.047 (-0.34, -0.16)
health Missing teeth -0.15* 0.021 (-0.19, -0.11)
Health Outcome Men
Concentration SE (95% CI)
Index
General Obesityt -0.02 0.041 (-0.1, 0.06)
health High blood -0.05 0.056 (-0.16, 0.06)
pressure
([dagger])
Oral Decayed teeth -0.20 * 0.064 (-0.33, -0.08)
health Missing teeth -0.08 * 0.032 (-0.14, -0.02)
Health Outcome Women
Concentration SE (95% CI)
Index
General Obesityt -0.08 * 0.036 (-0.15, -0.01)
health High blood -0.04 0.053 (-0.15, 0.06)
pressure
([dagger])
Oral Decayed teeth -0.34 * 0.053 (-0.45, -0.24)
health Missing teeth -0.20 * 0.026 (-0.25, -0.15)
* Statistically significant; p<0.05.
([dagger]) Concentration indices for binary outcomes
(obesity and high blood pressure) are normalized according
to Wagstaff, 2005. (12)
