Overqualification and risk of all-cause and cardiovascular mortality: evidence from the Canadian Census Mortality Follow-up study (1991-2001).

Smith, Brendan T. ; Smith, Peter M. ; Etches, Jacob 等

There is strong evidence demonstrating an inverse association between socio-economic position (SEP) and mortality. (1) In epidemiologic studies, SEP is typically measured as education, occupation or income. (2) These indicators attempt to measure the underlying construct of SEP, and consequently a degree of overlap exists between them. However, these indicators often reflect a specific period in the life-course and measure different dimensions of SEP, and therefore should not be used interchangeably. (2-4) Studies have attempted to understand the relationship between different SEP measures and health. (5,6) Examining the effect of multiple SEP indicators in the same study, both independently and simultaneously, can provide insight into which SEP measures are the strongest predictors of mortality.

There has been a longstanding interest in social epidemiology concerning the hypothesis that status inconsistency (i.e., having contradictory SEP measures) is associated with adverse health effects. A proposed pathway linking status inconsistency and poor health is via "goal-striving stress", where stress results from an individual's social class achievements (e.g., occupation) not meeting their aspirations (e.g., education). (7) This stress may result in neuroendocrine and immune system disruption and excess cortisol and sympathetic hormone release, which may elevate the risk of mortality, in particular from cardiovascular-related events. (8)

Overqualification--where educational attainment is higher than occupational skill requirements--is a status inconsistency measure of increasing relevance. In 2008, 28% of Canadians in non-management occupations were overqualified, with higher rates among immigrant (9) and younger workers. (10) While post-secondary educational attainment in Canada is increasing rapidly, the occupational skill level distribution of the labour market has not changed to the same extent. (10) These trends suggest that the prevalence of overqualification will likely continue to rise, underscoring the importance of understanding the relationship between overqualification and health.

There are mixed results on the association between overqualification and health, with some studies finding an association with coronary heart disease (CHD), (11) all-cause and CHD mortality, (12) declines in self-rated health (13) and adverse emotional outcomes. (14) However, a more recent study found no association between overqualification and cardiovascular disease (CVD). (15)

Comparison between studies has been hindered by the use of conflicting methodologies and multiple definitions of overqualification. This approach has led to mixed results, even within the same data source. (16,17) While methodological debate continues, (18) a clear articulation of which types of inconsistencies are being tested and how these methodologies distinguish the effects of SEP and status inconsistencies will help clarify this research area.

Recent studies examining the association between overqualification and mortality have reported mixed results. The Canadian Census Mortality Follow-up Study allows for a more meaningful measure of qualification status as occupation is classified in terms of minimum skill requirements, a measure that can be directly linked to an individual's level of education attainment. Consequently, the objective of this study was to determine whether education, occupation and overqualification increase risk of all-cause and CVD mortality.

METHODS

Study sample

The Canadian Census Mortality Follow-up Study is a nationally representative population-based cohort of non-institutionalized Canadians followed for mortality from 1991 to 2001. The cohort encompasses a 15% sample of the adult population of Canada over the age of 25 who completed the 1991 census long-form questionnaire (N=2,735,152). Record linkage methods included probabilistic linkage to income tax records. Linkage outcomes indicated that respondents who could not be matched to income tax records were more likely to be female, older than age 65, unmarried, or of lower income or educational attainment. (18) Baseline data collected in the census long-form questionnaire (1991) include data on various socio-demographic, family, household and neighbourhood characteristics of each respondent. Mortality was ascertained by linking census data to the Canadian Mortality Database, identifying over 260,000 deaths in this sample. (19) Statistics Canada estimates that 97% of all deaths to study subjects were ascertained in the Canadian Census Mortality Follow-up Study. (19)

For the purpose of this study, we restricted the original sample to respondents between 35-64 years of age at study baseline, working over 25 hours per week. These inclusion criteria are designed to account for potential bias resulting from younger individuals being more likely to be overqualified and less likely to die during the 10-year follow-up period. Similarly, it is difficult to accurately estimate occupational status in individuals over 64 years of age, and thus their exposure to overqualification as they approach and begin retirement. Therefore, it was assured that individuals had the opportunity to reach a "status-defining" occupation and are representative of labour force participants, whose exposure to overqualification was hypothesized to result in inequalities in mortality. The final sample consisted of 1,091,800 subjects (421,600 women and 670,200 men).

Exposure measurement

Educational Attainment

Education was classified into four groups using academic achievement. These included: <high school graduation, high school graduation (including trades certificate), post-secondary certificate or diploma, and university bachelor's degree or higher.

Occupation Skill Requirements

Occupation was categorized using the National Occupational Classification (NOC) system. (17) The NOC, developed by Human Resources and Development Canada, groups occupations based on minimum training and/or education required to work in an occupation. Occupations were classified into five categories: requiring no education (unskilled), requiring secondary school education (semi-skilled), requiring post-secondary education below bachelors level (technical/skilled/supervisory), managers, and requiring bachelors education or higher (professional).

Overqualification

Overqualification was defined as a mismatch between educational attainment and occupational achievement. To examine the health consequences of overqualification--above the health effects of occupation alone--we specified statistical interactions between education and occupation. Here, we examined whether the mortality risk associated with a given occupation differs depending on the respondent's level of education. We hypothesized that the effects of occupation on mortality would differ by level of education, being greater in overqualified respondents.

Outcome ascertainment

All-cause and CVD mortality were examined in this study. Underlying cause and date of death were ascertained from the Canadian Mortality Database. CVD mortality was classified using the World Health Organization's International Classification of Diseases, Ninth Revision (ICD-9) (20) for deaths occurring from 1991 through 1999 and Tenth Revision (ICD-10) (21) for deaths occurring in 2000 or 2001.

Confounders

Confounders included age, income adequacy, marital status, years since immigration, ethnicity, Aboriginal origins, province of residence and community size. Income adequacy, divided into quintiles, was calculated as the ratio of total family income to the Statistics Canada low income cut-off (LICO) for family and community size. (22)

Statistical analyses

Sex-specific Cox proportional hazards models were used to calculate hazard ratios (HR) and 95% confidence intervals (95% CI) to test the association between education and occupation (independently) on risk of all-cause and CVD mortality. Follow-up was calculated from study baseline (census day) until date of death or end of study (December 31, 2001), whichever comes first. All analyses were adjusted for age, with a subsequent model adjusting for education or occupation (when not the exposure of interest) and a limited set of possible confounders (age, income adequacy, marital status, years since immigration, ethnicity, Aboriginal origins, province of residence and community size). To examine the effect of overqualification on all-cause and CVD mortality, we evaluated two models: 1) an interaction term for education and occupation in models that simultaneously adjusted for both variables, and 2) a categorical variable specifying a respondent as overqualified, qualified or underqualified. Analyses were conducted using SAS version 9.1 (SAS Institute, Cary, NC).

The Canadian Census Mortality Follow-up Study was approved by the Statistics Canada Policy Committee after consultations with the Statistics Canada Confidentiality and Legislation Committee, the Data Access and Control Services Division, and the Federal Privacy Commissioner. The protocol for this research was reviewed and approved by the Health Sciences Research Ethics Board of the University of Toronto.

RESULTS

Table 1 describes the sex-specific distribution of baseline covariates by qualification status. In both men and women, younger workers were more likely to be overqualified, while older workers were more likely to be underqualified for their occupation. Further, men and women with no visible minority status were less likely to be overqualified and more likely to be underqualified compared to visible minority groups.

Age-adjusted Cox proportional hazards models showed a graded inverse association between both education and occupation and all-cause mortality in men (education: HR=1.94, 95% CI: 1.87-2.01 for <high school graduation vs. university degree; occupation: HR=1.86, 95% CI: 1.78-1.95 for unskilled vs. professional occupation) (Table 2). In women, education and occupation were inversely associated with all-cause mortality in age-adjusted models (education; HR=1.55, 95% CI: 1.45-1.66 for <high school graduation vs. university degree; occupation: HR=1.42, 95% CI: 1.32-1.53 for unskilled vs. professional occupation). In both men and women, these associations were attenuated and remained statistically significant in fully-adjusted models.

Age-adjusted Cox proportional hazards models demonstrated inverse associations between both education and occupation with CVD mortality in men (education; HR=2.01, 95% CI: 1.87-2.15 for <high school graduation vs. university degree; occupation: HR=1.91, 95% CI: 1.75-2.07 for unskilled vs. professional occupation) (Table 2). In women, similar inverse associations between education and occupation and CVD mortality were observed (education; HR=2.45, 95% CI: 2.02-2.35 for <high school graduation vs. university degree, and occupation: HR=1.96, 95% CI: 1.642.35 for unskilled vs. professional occupation). In both men and women, associations were attenuated and remained statistically significant in fully-adjusted models.

Limited evidence was found to support the hypothesized association between overqualification and all-cause or CVD mortality. In the analytic models estimating an interaction between education and occupation, the interaction terms for both men and women were not statistically significant on the multiplicative scale. In analytic models estimating over- or underqualification via a categorical variable, there was a weak association for men between overqualification and all-cause mortality in the fully-adjusted model (HR: 1.06, 95% CI: 1.01-1.12 compared to the qualified group) (Table 3).

DISCUSSION

Overall, strong socio-economic gradients in all-cause and CVD mortality were observed. These associations remained in fully-adjusted models. Adjusting for education and occupation simultaneously attenuated the mortality risk associated with each SEP measure. However, as in previous studies, (23) level of education still displayed a stepwise gradient for both all-cause and CVD mortality, while the occupational gradient became less apparent. The lack of interaction between education and occupation demonstrates a consistent effect of occupation on mortality across educational strata. A weak association was observed between overqualification and all-cause mortality in men. However, overall these findings suggest that overqualification does not confer additional all-cause or CVD mortality risk over and above that associated with individual SEP measures.

Prior literature

Previous studies have demonstrated strong social gradients for both education and occupation in all-cause and CVD mortality, (24) a finding confirmed in this report. In addition, the overqualification hypothesis was not confirmed in this study. These findings add to the mixed results on the association between overqualification and health. Two recent studies from Germany reported an association between status inconsistencies and CVD, (11,15) although only one found an association between overqualification and CVD. (11) However, these studies compared status-inconsistent with status-consistent individuals, an approach that does not differentiate between SEP and status-inconsistency measures. In addition, the use of ordinal--sample-dependent--rankings of SEP measures lacks a theoretical basis for quantifying both the level (as it groups all inconsistencies together) and type of inconsistency. The present study introduces a new measure of overqualification, making it difficult to compare with previous studies. (11,12,15) However, the classification of occupation by minimum skill level required allows a more accurate classification of overqualification. One exception is management occupations, which were included as an independent category as they reflect a range of potential education attainments. A sensitivity analysis individually classifying management occupations into occupation skill requirement categories, defined in a previous study using the NOC classification system, (13) did not change the study results.

Strengths and limitations

Our results should be interpreted taking into account the following limitations. The duration of exposure to overqualification as well as transitions between occupations are not captured in this study, as exposure measures were only assessed at one time point. Therefore some non-differential exposure misclassification may have occurred, which would have biased estimates toward the null. Future studies are necessary to better understand the time-dependent effect of overqualification on mortality. There may be misclassification across underqualified and qualified categories in our main independent variable as on-the-job training and experience may be considered in place of education by some employers. Unfortunately our data source did not contain information on years of relevant work experience, so we were unable to take job experience into account. In addition, immigrant populations are more likely to be overqualified (9) and have better-than-average health given the Canadian immigration system. Therefore, we controlled for years since immigration. Sensitivity analyses excluding all immigrants did not significantly change the results from this study (data not shown). Further, information on important mediators (e.g., health behaviours) of the relationship between education, occupation, overqualification and mortality were not available in this study. Mediators, such as time-dependent health behaviours, have been shown to explain a portion of social inequalities in mortality. (25) Finally, given the limited information on health status available in this cohort, it was not possible to assess whether the association between overqualification and mortality was potentially confounded by this factor (e.g., individuals in poor health seek occupations below their educational attainment).

A major strength of this study was the large sample size of a nationally representative cohort of Canadians, which enabled us to test the association between overqualification and all-cause and CVD mortality. Moreover, we present a more direct measure of overqualification and a statistical methodology that clearly distinguishes between the effects of SEP and overqualification.

CONCLUSION

Education and occupation gradients in all-cause and CVD mortality were observed in this study. However, there was little evidence that overqualification resulted in additional all-cause or CVD mortality, suggesting that the effects of occupation are consistent across educational strata.

Received: January 20, 2012

Accepted: May 21, 2012

REFERENCES

(1.) Kaplan GA, Keil JE. Socioeconomic factors and cardiovascular disease: A review of the literature. Circulation 1993;88:1973-98.

(2.) Galobardes B, Shaw M, Lawlor DA, Lynch JW, Davey Smith G. Indicators of socioeconomic position (part 1). J Epidemiol Community Health 2006;60(1):7-12.

(3.) Kuh D, Ben-Shlomo Y, Lynch J, Hallqvist J, Power C. Life course epidemiology. J Epidemiol Community Health 2003;57(10):778-83.

(4.) Krieger N, Williams DR, Moss NE. Measuring social class in US public health research: Concepts, methodologies, and guidelines. Annu Rev Public Health 1997;18:341-78.

(5.) Davey Smith G, Hart C, Hole D, MacKinnon P, Gillis C, Watt G, et al. Education and occupational social class: Which is the more important indicator of mortality risk? J Epidemiol Community Health 1998;52(3):153-60.

(6.) Singh-Manoux A, Clarke P, Marmot M. Multiple measures of socio-economic position and psychosocial health: Proximal and distal measures. Int J Epidemiol 2002;31(6):1192-99.

(7.) Dressler WW. Social consistency and psychological distress. J Health Soc Behav 1988;29(1):79-91.

(8.) McEwen BS. Protective and damaging effects of stress mediators. N Engl J Med 1998;338(3):171-79.

(9.) Gilmore J. The 2008 Canadian Immigrant Labour Market: Analysis of Quality of Employment. The Immigrant Labour Force Analysis Series. Ottawa, ON: Statistics Canada, 2009. Report No.:71-606-X, no. 5. Available at: http://www.statcan.gc.ca/bsolc/olc-cel/olc-cel?catno=71-606- X2009001&lang=eng (Accessed June 1, 2011).

(10.) Li C, Gervais G, Duval A. The Dynamics of Overqualification: Canada's Underemployed University Graduates. Ottawa: Statistics Canada, 2006. Report No.: 11-621-MIE - No. 039. Available at: http://www.statcan.gc.ca/bsolc/olc-cel/olc-cel?catno=11-621- M2006039&lang=eng (Accessed June 1, 2011).

(11.) Peter R, Gassler H, Geyer S. Socioeconomic status, status inconsistency and risk of ischaemic heart disease: A prospective study among members of a statutory health insurance company. J Epidemiol Community Health 2007;61(7):605-11.

(12.) Faresjo T, Svardsudd K, Tibblin G. The concept of status incongruence revisited: A 22-year follow-up of mortality for middle-aged men. Scand J Soc Med 1997;25(1):28-32.

(13.) Smith P, Frank J. When aspirations and achievements don't meet. A longitudinal examination of the differential effect of education and occupational attainment on declines in self-rated health among Canadian labour force participants. Int J Epidemiol 2005;34(4):827-34.

(14.) Lundberg J, Kristenson M, Starrin B. Status incongruence revisited: Associations with shame and mental wellbeing. Sociol Health Illn 2009;31(4):478-93.

(15.) Braig S, Peter R, Nagel G, Hermann S, Rohrmann S, Linseisen J. The impact of social status inconsistency on cardiovascular risk factors, myocardial infarction and stroke in the EPIC-Heidelberg cohort. BMC Public Health 2011;11:104.

(16.) Horan PM, Gray BH. Status inconsistency, mobility and coronary heart disease. J Health Soc Behav 1974;15(4):300-10.

(17.) Statistics Canada. National Occupational Classification for Statistics 2006. Ottawa: Statistics Canada, 2007.

(18.) Zhang X. Status inconsistency revisited: An improved statistical model. Eur Sociol Rev 2008;24(2):155-68.

(19.) Wilkins R, Tjepkema M, Mustard C, Choiniere R. The Canadian census mortality follow-up study, 1991 through 2001. Health Rep 2008;19(3):25-43.

(20.) World Health Organization. Manual of the International Statistical Classification of Diseases, Injuries and Causes of Death. Ninth Revision. Geneva, Switzerland: WHO, 1977.

(21.) World Health Organization. International Statistical Classification of Diseases and Related Health Problems, Tenth Revision. Geneva: WHO, 1992.

(22.) Statistics Canada. 1991 Census Dictionary (Catalogue 92-301E) Ottawa: Supply and Services Canada, 1992.

(23.) Winkleby MA, Jatulis DE, Frank E, Fortmann SP. Socioeconomic status and health: How education, income, and occupation contribute to risk factors for cardiovascular disease. Am J Public Health 1992;82(6):816-20.

(24.) Mackenbach JP, Bos V, Andersen O, Cardano M, Costa G, Harding S, et al. Widening socioeconomic inequalities in mortality in six Western European countries. Int J Epidemiol 2003;32(5):830-37.

(25.) Stringhini S, Sabia S, Shipley M, Brunner E, Nabi H, Kivimaki M, et al. Association of socioeconomic position with health behaviors and mortality. JAMA 2010;303(12):1159-66.

Brendan T. Smith, MSc, [1,2] Peter M. Smith, PhD, MPH, [1-3] Jacob Etches, PhD, [2] Cameron A. Mustard, ScD [1,2]

Author Affiliations

[1.] Dalla Lana School of Public Health, University of Toronto, Toronto, ON

[2.] Institute for Work & Health, Toronto, ON

[3.] School of Public Health and Preventive Medicine, Monash University, Melbourne, Australia

Correspondence: Brendan T. Smith, Institute for Work & Health, 481 University Ave, Toronto, ON M5G 2E9, Tel: 416-927-2027, ext. 2189, Fax: 416-927-4167, E-mail: bsmith@iwh.on.ca

Acknowledgement: The authors acknowledge the leadership of Dr. Michael Wolfson in the creation of the Canadian Census Mortality Follow-up Study.

Sources of funding: Funding for this research was provided by the Canadian Population Health Initiative of the Canadian Institute for Health Information, the Canadian Institutes of Health Research, and the Research Advisory Council of the Ontario Workplace Safety and Insurance Board. Brendan Smith is supported through a Frederick Banting and Charles Best Canada Graduate Scholarships Doctoral Award, Canadian Institutes of Health Research. Peter Smith is supported by a New Investigator Award from the Canadian Institutes of Health Research.

Conflict of Interest: None to declare.

Table 1. Baseline Characteristics According to Measure of
Status Consistency of Non-institutionalized Men and Women
Aged 35-64, Canadian Census Mortality Follow-up Study,
Canada (1991-2001)

                                            Men

                                  n       Deaths    Qualified
                                                       (%)

Total Sample                   670,200    34,750        37
Education
University degree              118,400      3347        71
Postsecondary diploma           87,700      3113        50
Secondary graduation           260,900    12,686        32
<secondary graduation          203,300    15,604        17

Occupation skill
    requirements
Professional                    91,800      3029        69
Management                     113,100      4705        37
Technical,
    skilled, supervisory       235,600    12,225        16
Semi-skilled                   169,500    10,147        42
Unskilled                       60,200      4644        58

Age group
35-39                          169,000      2702        39
40-44                          156,500      3833        40
45-49                          122,400      4982        38
50-54                           95,300      6482        34
55-59                           76,400      8229        32
60-64                           50,600      8522        31

Income adequacy
    quintile
5: Highest                     198,800      9649        42
4                              172,800      8620        36
3                              144,400      7450        35
2                              100,400      5561        33
1: Lowest                       53,800      3470        32

Marital status
Married                        545,400    26,410        37
Common-law                      38,400      1831        37
Widowed                           4400       622        33
Separated                       13,700       917        37
Divorced                        26,300      2048        37
Never married                   42,100      2922        42

Years since immigration
Non-immigrant                  512,400    27,635        36
Immigrated at age <14           24,300       731        40
0-5 years                       13,700       296        40
6-10 years                      10,500       271        42
11-20 years                     41,200      1265        41
21+ years                       65,600      4507        36
Non-permanent resident            2500        45        40

Ethnicity
Not a visible minority         618,100    33,139        36
Black                             9000       305        42
East Asian                      16,800       531        41
Latin American                    1900        39        38
South Asian                     12,200       376        43
South-east Asian/Pacific          5400       150        39
South-west Asian or Arab          6000       177        44
Multiple visible minority          800        33        41

Aboriginal origins
No Aboriginal origins          650,400    33,576        37
Not registered
    Indian/band member          11,200       568        34
Registered Indian/
    band member                   8600       606        32

Province
Alberta                         62,600      2924        36
British Columbia                80,900      3717        37
Manitoba                        27,500      1603        33
New Brunswick                   16,700       857        35
Newfoundland                    13,300       675        34
Northwest Territories             4700       216        36
Nova Scotia                     20,700      1092        35
Ontario                        247,800    13,027        38
Prince Edward Island              3100       180        33
Quebec                         167,400      9138        38
Saskatchewan                    23,900      1262        30
Yukon                             1500        59        37

Community size
[greater than or equal
    to]1,000,000               207,400    10,469        39
500,000-999,999                109,200      5366        41
100,000-499,999                 99,900      5057        38
10,000-99,999                   93,300      4943        36
<10,000 with
    metropolitan
    influence                  151,200      8317        31
<10,000 no
    metropolitan
    influence                     9200       598        28

                                        Men

                                 Over-      Under-
                               qualified   qualified
                                    (%)         (%)

Total Sample                        11          52
Education
University degree                   29       N/A *
Postsecondary diploma               24          26
Secondary graduation                 8          60
<secondary graduation            N/A *          83

Occupation skill
    requirements
Professional                     N/A *          31
Management                          13          50
Technical,
    skilled, supervisory             6          78
Semi-skilled                        13          46
Unskilled                           42       N/A *

Age group
35-39                               14          46
40-44                               13          47
45-49                               11          52
50-54                                9          57
55-59                                8          60
60-64                                8          61

Income adequacy
    quintile
5: Highest                          10          48
4                                   11          53
3                                   12          53
2                                   13          53
1: Lowest                           14          54

Marital status
Married                             11          53
Common-law                          11          52
Widowed                              9          58
Separated                           12          51
Divorced                            13          50
Never married                       17          41

Years since immigration
Non-immigrant                       10          53
Immigrated at age <14               13          47
0-5 years                           28          32
6-10 years                          22          36
11-20 years                         18          41
21+ years                            9          54
Non-permanent resident              25          35

Ethnicity
Not a visible minority              10          53
Black                               17          41
East Asian                          20          39
Latin American                      28          34
South Asian                         24          32
South-east Asian/Pacific            34          28
South-west Asian or Arab            24          33
Multiple visible minority           19          39

Aboriginal origins
No Aboriginal origins               11          52
Not registered
    Indian/band member              11          56
Registered Indian/
    band member                     10          58

Province
Alberta                             11          52
British Columbia                    13          50
Manitoba                            10          57
New Brunswick                        9          56
Newfoundland                         7          59
Northwest Territories               11          53
Nova Scotia                         11          54
Ontario                             12          50
Prince Edward Island                 9          57
Quebec                              11          51
Saskatchewan                         9          61
Yukon                               14          49

Community size
[greater than or equal
    to]1,000,000                    14          47
500,000-999,999                     12          47
100,000-499,999                     11          51
10,000-99,999                       10          54
<10,000 with
    metropolitan
    influence                        8          61
<10,000 no
    metropolitan
    influence                        8          64

                                           Women

                                  n       Deaths    Qualified
                                                       (%)

Total Sample                   421,600    11,168        41
Education
University degree               62,200      1039        71
Postsecondary diploma           86,600      1939        35
Secondary graduation           157,200      3954        46
<secondary graduation          115,600      4236        22

Occupation skill
    requirements
Professional                    73,800      1458        52
Management                      38,400       937        35
Technical,
    skilled, supervisory       121,300      3226        23
Semi-skilled                   148,900      4164        44
Unskilled                       39,300      1383        63

Age group
35-39                          115,500      1152        43
40-44                          109,000      1735        43
45-49                           81,300      1999        40
50-54                           56,200      2139        38
55-59                           38,400      2285        36
60-64                           21,300      1858        36

Income adequacy
    quintile
5: Highest                     130,700      3160        44
4                              107,500      2660        40
3                               85,300      2275        39
2                               60,900      1830        38
1: Lowest                       37,200      1243        37

Marital status
Married                        289,900      6793        40
Common-law                      26,500       569        41
Widowed                         14,000       835        38
Separated                       14,900       426        41
Divorced                        38,700      1321        41
Never married                   37,700      1224        46

Years since immigration
Non-immigrant                  322,300      8955        41
Immigrated at age <14           15,700       304        42
0-5 years                         8900       120        40
6-10 years                        7200       118        41
11-20 years                     29,100       447        42
21+ years                       37,000      1197        40
Non-permanent resident            1500        27        40

Ethnicity
Not a visible minority         382,900    10,506        41
Black                             8400       185        42
East Asian                      12,200       222        41
Latin American                    1400        15        42
South Asian                       7300        99        41
South-east Asian/Pacific          6200        88        38
South-west Asian or Arab          2700        41        43
Multiple visible minority          700        12        40

Aboriginal origins
No Aboriginal origins          407,100    10,680        41
Not registered
    Indian/band member            7600       221        38
Registered Indian/
    band member                   6900       267        35

Province
Alberta                         39,100      1027        41
British Columbia                48,600      1284        41
Manitoba                        17,200       532        39
New Brunswick                   10,000       231        42
Newfoundland                      8100       199        40
Northwest Territories             3100        73        40
Nova Scotia                     12,200       337        41
Ontario                        162,600      4526        41
Prince Edward Island              2100        56        40
Quebec                         103,100      2475        41
Saskatchewan                    14,500       397        39
Yukon                             1000        31        39

Community size
[greater than or equal
    to]1,000,000               143,600      3681        42
500,000-999,999                 72,400      1892        43
100,000-499,999                 61,800      1701        42
10,000-99,999                   55,500      1470        40
<10,000 with
    metropolitan
    influence                   83,300      2254        37
<10,000 no
    metropolitan
    influence                     4900       170        33

                                        Women

                                 Over-      Under-
                               qualified   qualified
                                    (%)         (%)

Total Sample                        14          46
Education
University degree                   29       N/A *
Postsecondary diploma               32          33
Secondary graduation                 7          47
<secondary graduation            N/A *          78

Occupation skill
    requirements
Professional                     N/A *          48
Management                          13          52
Technical,
    skilled, supervisory             7          69
Semi-skilled                        20          36
Unskilled                           36       N/A *

Age group
35-39                               16          40
40-44                               14          43
45-49                               13          47
50-54                               11          51
55-59                               11          53
60-64                               11          54

Income adequacy
    quintile
5: Highest                          12          44
4                                   13          47
3                                   14          46
2                                   16          46
1: Lowest                           17          45

Marital status
Married                             13          47
Common-law                          13          46
Widowed                             11          51
Separated                           14          45
Divorced                            14          44
Never married                       17          37

Years since immigration
Non-immigrant                       12          47
Immigrated at age <14               12          45
0-5 years                           30          30
6-10 years                          26          33
11-20 years                         22          36
21+ years                           13          47
Non-permanent resident              30          30

Ethnicity
Not a visible minority              13          47
Black                               18          41
East Asian                          21          38
Latin American                      29          29
South Asian                         27          32
South-east Asian/Pacific            36          26
South-west Asian or Arab            23          34
Multiple visible minority           25          35

Aboriginal origins
No Aboriginal origins               14          45
Not registered
    Indian/band member              13          49
Registered Indian/
    band member                     12          53

Province
Alberta                             15          45
British Columbia                    17          42
Manitoba                            13          48
New Brunswick                       13          45
Newfoundland                        10          50
Northwest Territories               12          48
Nova Scotia                         14          45
Ontario                             14          45
Prince Edward Island                15          44
Quebec                              12          48
Saskatchewan                        13          49
Yukon                               22          39

Community size
[greater than or equal
    to]1,000,000                    15          43
500,000-999,999                     14          43
100,000-499,999                     14          44
10,000-99,999                       13          47
<10,000 with
    metropolitan
    influence                       12          51
<10,000 no
    metropolitan
    influence                       10          57

* N/A or not applicable refers to situations where the
qualification status consistency category is not possible.
For example, it is impossible for individuals with the
highest level of education to be underqualified for their
occupation, or individuals with the lowest level of
education to be overqualified for their occupation.

Table 2. Cox Proportional Hazard Models Evaluating the
Association Between Educational Attainment and Occupational
Status With All-cause and CVD Mortality in Non-
institutionalized Men and Women Aged 35-64, Canadian Census
Mortality Follow-up Study, Canada (1991-2001)

                                                     All-cause
                                                     Mortality

                                                     Model 1 *

Men                            n       Events     HR      95% CI
  Education
University degree           118,400      3347    1.00
Postsecondary diploma        87,700      3113    1.30    1.24-1.36
Secondary graduation        260,900    12,686    1.62    1.56-1.68
<secondary graduation       203,300    15,604    1.94    1.87-2.01

Occupation skill
    requirements
Professional                 91,800      3029
Management                  113,100      4705
Technical, skilled,
      supervisory           235,600    12,225
Semi-skilled                169,500    10,147
Unskilled                    60,200      4644

Women
Education
University degree            62,200      1039    1.00
Postsecondary diploma        86,600      1939    1.22    1.13-1.32
Secondary graduation        157,200      3954    1.37    1.28-1.46
<secondary graduation       115,600      4236    1.55    1.45-1.66

Occupation skill
      requirements
Professional                 73,800      1458
Management                   38,400       937
Technical, skilled,
      supervisory           121,300      3226
Semi-skilled                148,900      4164
Unskilled                    39,300      1383

                                         All-cause Mortality

                                 Model 2 *            Model 3
                                                    ([dagger])
                                                 ([double dagger])

Men                           HR       95% CI     HR       95% CI
  Education
University degree                                1.00
Postsecondary diploma                            1.19    1.13-1.26
Secondary graduation                             1.40    1.33-1.46
<secondary graduation                            1.53    1.46-1.61

Occupation skill
    requirements
Professional                1.00                 1.00
Management                  1.18    1.13-1.24    1.02    0.97-1.07
Technical, skilled,
      supervisory           1.44    1.39-1.50    1.09    1.04-1.14
Semi-skilled                1.69    1.62-1.76    1.16    1.11-1.22
Unskilled                   1.86    1.78-1.95    1.21    1.15-1.28

Women
Education
University degree                                1.00
Postsecondary diploma                            1.15    1.06-1.25
Secondary graduation                             1.28    1.19-1.39
<secondary graduation                            1.41    1.30-1.54

Occupation skill
      requirements
Professional                1.00                 1.00
Management                  1.17    1.08-1.27    1.06    0.97-1.15
Technical, skilled,
      supervisory           1.24    1.16-1.32    1.07    0.99-1.14
Semi-skilled                1.27    1.20-1.35    1.06    0.99-1.14
Unskilled                   1.42    1.32-1.53    1.14    1.05-1.24

                                           CVD Mortality

                                 Model 1 *           Model 2 *

Men                           HR       95% CI     HR      95% CI
  Education
University degree           1.00
Postsecondary diploma       1.34    1.23-1.47
Secondary graduation        1.64    1.53-1.76
<secondary graduation       2.01    1.87-2.15

Occupation skill
    requirements
Professional                                     1.00
Management                                       1.20    1.11-1.31
Technical, skilled,
      supervisory                                1.48    1.37-1.59
Semi-skilled                                     1.77    1.64-1.91
Unskilled                                        1.91    1.75-2.07

Women
Education
University degree           1.00
Postsecondary diploma       1.54    1.25-1.90
Secondary graduation        1.80    1.49-2.19
<secondary graduation       2.45    2.02-2.96

Occupation skill
      requirements
Professional                                     1.00
Management                                       1.31    1.06-1.62
Technical, skilled,
      supervisory                                1.38    1.18-1.63
Semi-skilled                                     1.62    1.39-1.89
Unskilled                                        1.96    1.64-2.35

                              CVD Mortality

                                Model 3
                              ([dagger])
                              ([section])

Men                          HR      95% CI
  Education
University degree           1.00
Postsecondary diploma       1.22    1.11-1.34
Secondary graduation        1.38    1.27-1.50
<secondary graduation       1.52    1.40-1.66

Occupation skill
    requirements
Professional                1.00
Management                  1.04    0.95-1.13
Technical, skilled,
      supervisory           1.09    1.00-1.19
Semi-skilled                1.20    1.09-1.31
Unskilled                   1.22    1.11-1.35

Women
Education
University degree           1.00
Postsecondary diploma       1.40    1.12-1.74
Secondary graduation        1.61    1.29-2.00
<secondary graduation       2.01    1.61-2.51

Occupation skill
      requirements
Professional                1.00
Management                  1.06    0.85-1.33
Technical, skilled,
      supervisory           1.02    0.85-1.23
Semi-skilled                1.11    0.92-1.32
Unskilled                   1.22    0.99-1.51

* Adjusted for age.

([dagger]) Adjusted for age, education, occupation, income
adequacy, marital status, years since immigration,
ethnicity, Aboriginal origins, province of residence and
community size.

([double dagger]) When added to Model 3, an interaction term
between education and occupation for all-cause mortality was
not statistically significant (men: p=0.22; women: p=0.74).

([section]) When added to Model 3, an interaction term
between education and occupation for CVD mortality was not
statistically significant (men: p=0.25; women: p=0.16).

Table 3. Cox Proportional Hazard Models Evaluating the
Association Between Status Inconsistency and All-cause and
CVD Mortality in Non-institutionalized Men and Women Aged
35-64, Canadian Census Mortality Follow-up Study, Canada
(1991-2001)

                               n    Events
Men
  Status Inconsistency
    Qualified             247,100   11,051
    Overqualified         76,000      3100
    Underqualified        347,100   20,599
Women
  Status Inconsistency
    Qualified             171,500     4141
    Overqualified         57,600      1295
    Underqualified        192,400     5732

                                 All-cause Mortality

                              Model 1 *          Model 2
                                                 ([dagger])

                           HR      95% CI     HR      95% CI
Men
  Status Inconsistency
    Qualified             1.00               1.00
    Overqualified         1.00   0.96-1.04   1.06   1.01-1.12
    Underqualified        1.15   1.12-1.17   1.00   0.96-1.05
Women
  Status Inconsistency
    Qualified             1.00               1.00
    Overqualified         0.98   0.92-1.04   1.04   0.96-1.13
    Underqualified        1.10   1.06-1.15   1.02   0.95-1.10

                                          CVD Mortality

                              Model 1*           Model 2
                                                 ([dagger])

                           HR      95% CI     HR      95% CI
Men
  Status Inconsistency
    Qualified             1.00               1.00
    Overqualified         0.97   0.91-1.05   1.05   0.96-1.15
    Underqualified        1.12   1.08-1.17   0.96   0.89-1.03
Women
  Status Inconsistency
    Qualified             1.00               1.00
    Overqualified         0.97   0.83-1.13   1.19   0.98-1.46
    Underqualified        1.13   1.03-1.24   0.90   0.76-1.07

* Adjusted for age.

([dagger]) Adjusted for age, education, occupation, income
adequacy, marital status, years since immigration,
ethnicity, Aboriginal origins, province of residence and
community size.