Age modification of diabetes-related hospitalization among First Nations adults in Alberta, Canada
© Campbell et al.; licensee BioMed Central Ltd. 2014
Received: 14 April 2014
Accepted: 26 September 2014
Published: 2 October 2014
We sought to determine the modifying effects of age and multimorbidity on the association between First Nations status and hospitalizations for diabetes-specific ambulatory care sensitive conditions (ACSC).
We identified 183,654 adults with diabetes from Alberta Canada, and followed them for one year for the outcome of hospitalization or emergency department (ED) visit for a diabetes-specific ACSC. We used logistic regression to determine the association between First Nations status and the outcome, assessing for effect modification by age and multimorbidity with interaction terms. In a model adjusting for age, age2, baseline A1c, duration of diabetes, and multimorbidity, First Nations people were at greater risk than non-First Nations to experience a diabetes-specific hospitalization or ED visit (unadjusted odds ratio [OR] 3.74; 95% confidence interval [CI]: 3.45-4.07). After adjustment for relevant covariates, this association varied by age (interaction: p = 0.018): adjusted OR 3.94 (95% CI: 3.11-4.99) and 5.74 (95% CI: 3.36-9.80) for First Nations compared to non-First Nations at ages 30 and 80 years, respectively.
Compared with non-First Nations, older First Nations patients with diabetes are at greater risk for diabetes-specific hospitalizations. Older First Nations patients with diabetes should be given priority access to primary care services as they are at greatest risk for requiring hospitalization for stabilization of their condition.
The Canadian Aboriginal population is a diverse one. The Canadian constitution recognizes three distinct populations of Aboriginal peoples: First Nations, Inuit and Métis . 65% of the Canadian Aboriginal population identify as First Nations . Compared to non-First Nations, First Nations people in North America have more than a two-fold increased prevalence of diabetes , and experience more diabetes-specific complications requiring hospitalization [4, 5]. Alberta’s First Nations population is heterogeneous and comprises 44 distinct First Nations with a total of over 100,000 individuals registered under the Federal Indian Act. Nearly 60% of First Nations people live on one of 134 designated reserves while the remainder live off-reserve .
It is not well known whether certain subsets of the First Nations population with diabetes are at greater risk of hospitalization; though among the non-First Nations population risk has been noted to decrease with advancing age and increases with multiple comorbid conditions . Given the increasing life expectancy and incidence of chronic conditions among First Nations, the potential modifying effect of age and multimorbidity is particularly relevant .
Using a population-based cohort we sought to determine the association between First Nations status and risk of hospitalization for diabetes-specific ambulatory care sensitive conditions (ACSCs), and to assess whether this risk varies with age and multimorbidity.
We used administrative data from Alberta Health and the Alberta Kidney Disease Network for our population-based study . For the period from Apr. 1, 1994 to Apr. 1, 2009 we obtained data on hospital admissions, physician visits, and emergency department visits from Alberta Health administrative data files. We also obtained laboratory data from a province-wide repository that captures data for all Albertans who undergo inpatient or outpatient laboratory testing. This data was de-identified prior to the researchers being granted access. A validated algorithm using ICD-9 codes from the administrative database was used to define a cohort of adults (≥18 years) with a diagnosis of diabetes between April 1, 1994 and March 31, 2008 . An individual was classified as having diabetes if they were found to have either one hospital discharge abstract with a diagnosis of diabetes or a diabetes-related code on two physician claims within 2 years. This algorithm does not distinguish between individuals with type 1 and type 2 diabetes and largely eliminates cases of gestational diabetes.
In our dataset, First Nations status reflected registration under the Federal Indian Act. In the 2006 Canadian Census, 81% of those who self-identified as First Nations were registered with the Federal Indian Act .
The outcome was hospitalization or emergency department (ED) visit for a diabetes-specific ACSC during a one-year period (April 1, 2008 to March 31, 2009), and included hypoglycemic events, diabetic ketoacidosis or hyperglycemic hyperosmolar non-ketotic states . ACSCs represent conditions which, if managed optimally in the outpatient setting, should not result in hospitalization or ED visit [13, 14]. ACSCs are commonly used as a marker of quality care for patients with chronic disease , and diabetes specifically [9, 13, 15].
Covariates and other variables of interest
Demographic data, including age and sex, were determined from the Alberta Health Registry File. The presence of comorbid conditions was generated using validated ICD-9-CM and ICD-10 coding algorithms . Specifically, we used the Charlson Comorbidity Index to identify comorbidities . Patients with a Charlson score greater than 2, in addition to diabetes, were classified as having multimorbidity. Hypertension was identified as a separate comorbidity. Baseline glycated hemoglobin (A1c) level was obtained from a provincial laboratory repository using the most recent A1c assessment prior to April 1, 2008 . Duration of diabetes was the length of time between the diabetes diagnosis date and the index date of April 1, 2008.
We used logistic regression to determine the odds ratios (ORs) of a hospitalization or ED visit for a diabetes-specific ACSC for First Nations compared with non-First Nations, adjusting for demographic and clinical characteristics. We used interaction terms to assess for effect modification by age (age*First Nations status) and by multimorbidity status (multimorbidity*First Nations status), as well as joint modification by both age and multimorbidity (age*multimorbidity*First Nations status). Based on the literature [14, 19, 20], we considered the following potential confounders: duration of diabetes, baseline A1c, hypertension, sex, multimorbidity status, age, and a quadratic age term (age*age). Age, age2 and A1c were treated as continuous variables in analyses. We used a backwards stepwise elimination technique to obtain the most parsimonious model that was still hierarchically well formulated. Ethics approval was obtained from the University of Calgary Research Ethics Board.
N = 8007
N = 175,647
Persons having a hospitalization for diabetes-related ACSC N (%)
Age, years Mean (SD)
Gender, women N (%)
Charlson Comorbidity Indexǂ
0–1, N (%)
2+, N (%)
Hypertension N (%)
Baseline A1c, % Mean (SD)
Duration of diabetes, years
Adjusted age-specific odds ratio of diabetes-specific ACSC hospitalization
Odds ratio (95% CI)
First Nations Status, including age interaction
1.33 (1.31-1.35) per % increase in A1c
Duration of diabetes
1.11 (1.10-1.11) per year of exposure
Multimorbidity did not modify the relationship between First Nations status and hospitalization or ED visit for a diabetes-specific ACSC (interaction: p = 0.197).
In this population-based cohort of patients with diabetes the risk of hospitalization or ED visit for a diabetes-specific ACSC was consistently higher for First Nations compared with non-First Nations – and the magnitude of the excess risk increased with age. By age 80 the risk was almost six times higher for First Nations compared to non-First Nations, even after adjusting for multimorbidity and other factors. Given the aging First Nations population and the increasing prevalence of diabetes, these results are particularly relevant and worthy of further exploration.
Our study supports previous findings suggesting that First Nations are at higher risk of hospitalizations for diabetes-specific ACSCs than non-First Nations [4, 15, 21]. However, this is the first study, to our knowledge, to further assess this risk by age and multimorbidity status. These findings have important implications for provision of diabetes care. Providers should be aware that among First Nations patients, those with advanced age may have a higher excess risk and thus warrant closer follow-up.
A strong association between diabetes incidence and age is known to exist, where rates of diabetes increase concordantly with age [22, 23]. However, the risk of diabetes-specific ACSC hospitalizations among the non-First Nations population with diabetes has been shown to decrease with age and multimorbidity . In contrast, by adjusting for age in a non-linear fashion (age2), we were able to identify that the risk of hospitalization decreases until age 60 and thereafter increases with advancing age. The reason for this pattern may be improved disease management through adulthood which is later complicated by incident comorbid conditions with advancing age.
Furthermore, we found that the excess risk of diabetes-specific ACSC hospitalization among First Nations adults compared to non-First Nations adults increased with advancing age, and did not vary by the presence of multimorbidity. The reasons for these excess risks in First Nations patients, as compared to non-First Nations patients are complex and have their roots in social and historical inequities . These inequities may lead to the accumulation of a variety of factors over time, which may explain the excess risk in older First Nations adults. These factors include: poor glycemic control, inconsistent medication adherence, inadequate specialized care for diabetes-specific problems, and, consequently, an increased risk of developing long-term complications and adverse health outcomes [4, 25]. While life expectancy in general is increasing, the First Nations population in particular has experienced an increase in life expectancy . As they age this population is at a particularly high risk of adverse outcomes related to their diabetes in the future. We feel that the results of our study are likely to apply to other Canadian First Nations populations outside of Alberta.
A strength of our study was the population-based nature of our data, which allowed us to define a large and representative cohort of patients with diabetes. However, there are also inherent limitations based on the data sources, including potential misclassification of outcomes or comorbidities. However, such misclassification is unlikely to have differed by First Nations status, and therefore (if present) would bias the results towards the null. We did not have data on measures of socioeconomic status (income, social environment, education, etc.…), therefore we could not fully adjust our models for potential confounding variables which may partially contribute to differences seen in our analysis. We were also unable to assess whether the risk of hospitalization varied by diabetes type (Type 1 vs. Type 2), this would be problematic if type 2 diabetes were known to be associated with higher rates of hospitalizations as First Nations patients are much more likely to have type 2 diabetes, compared with the general population  and this may result in differential misclassification. However, the reason for hospitalization and length of stay is known to vary by type of diabetes, but the number of hospitalizations is not particularly variable, and in fact, those with Type 1 diabetes seem to be at higher risk for hospitalization . Therefore, the inability to distinguish between Type 1 and Type 2 diabetes is unlikely to bias the results of our study.
Our study highlights a substantial difference in risk of hospitalization or ED visit for diabetes-specific ACSC by age for First Nations people, as compared to non-First Nations people. Given the aging of the First Nations population, this risk is particularly concerning and warrants further study.
The authors would like to thank Jianguo Zhang for formatting and preparing the dataset. This research was funded by an Alberta Innovates-Health Solutions Team Grant.
- Government of Canada: Constitution Act: Rights of the Aboriginal Peoples of Canada. Book Constitution Act: Rights of the Aboriginal Peoples of Canada. Section 35. 1982, Ottawa, ONGoogle Scholar
- Statistics Canada: Aboriginal Peoples in Canada: First Nations People, Métis and Inuit [article online]. 2011, Available from http://www12.statcan.gc.ca/nhs-enm/2011/as-sa/99-011-x/99-011-x2011001-eng.pdfGoogle Scholar
- Oster RT, Johnson JA, Hemmelgarn BR, King M, Balko SU, Svenson LW, Crowshoe L, Toth EL: Recent epidemiologic trends of diabetes mellitus among status aboriginal adults. CMAJ. 2011, 183 (12): E803-8. 10.1503/cmaj.101882.PubMed CentralView ArticlePubMedGoogle Scholar
- Gao S, Manns BJ, Culleton BF, Tonelli M, Quan H, Crowshoe L, Ghali WA, Svenson LW, Ahmed S, Hemmelgarn BR, for the Alberta Kidney Disease Network: Access to health care among status aboriginal people with chronic kidney disease. Can Med Assoc J. 2008, 179 (10): 1007-1012. 10.1503/cmaj.080063.View ArticleGoogle Scholar
- Deved V, Jette N, Quan H, Tonelli M, Manns B, Soo A, Barnabe C, Hemmelgarn BR, for the Alberta Kidney Disease Network: Quality of care for first nations and non-first nations people with diabetes. Clin J Am Soc Nephrol. 2013, 8 (7): 1188-1194. 10.2215/CJN.10461012.PubMed CentralView ArticlePubMedGoogle Scholar
- Health Determinants for First Nations in Alberta [article online]. 2010, Available from http://report.hcom.ca/people/
- Ahern MM, Hendryx M: Avoidable hospitalizations for diabetes: comorbidity risks. Dis Manag. 2007, 10 (6): 347-355. 10.1089/dis.2007.106709.View ArticlePubMedGoogle Scholar
- Wilson K, Rosenberg MW, Abonyi S, Lovelace R: Aging and health: an examination of differences between older aboriginal and non-aboriginal people. Can J Aging. 2010, 29 (3): 369-382. 10.1017/S0714980810000309.View ArticlePubMedGoogle Scholar
- Manns BJ, Tonelli M, Zhang J, Campbell DJ, Sargious P, Ayyalasomayajula B, Clement F, Johnson JA, Laupacis A, Lewanczuk R, McBrien K, Hemmelgarn BR: Enrolment in primary care networks: impact on outcomes and processes of care for patients with diabetes. CMAJ. 2012, 184 (2): E144-52. 10.1503/cmaj.110755.PubMed CentralView ArticlePubMedGoogle Scholar
- Hux JE, Ivis F, Flintoft V, Bica A: Diabetes in Ontario: determination of prevalence and incidence using a validated administrative data algorithm. Diabetes Care. 2002, 25 (3): 512-516. 10.2337/diacare.25.3.512.View ArticlePubMedGoogle Scholar
- Aboriginal Affairs and Northern Development Canada [article online]. 2013, Available from http://www.aadnc-aandc.gc.ca/eng/1100100013791/1100100013795
- Canadian Institute for Health Information: Technical Note: Ambulatory Care Sensitive Conditions (ACSC) 2010. 2011, Ottawa, OntarioGoogle Scholar
- Booth GL, Hux JE: Relationship between avoidable hospitalizations for diabetes mellitus and income level. Arch Intern Med. 2003, 163 (1): 101-106. 10.1001/archinte.163.1.101.View ArticlePubMedGoogle Scholar
- Stratton IM, Adler AI, Neil HA, Matthews DR, Manley SE, Cull CA, Hadden D, Turner RC, Holman RR: Association of glycaemia with macrovascular and microvascular complications of type 2 diabetes (UKPDS 35): Prospective observational study. BMJ. 2000, 321 (7258): 405-412. 10.1136/bmj.321.7258.405.PubMed CentralView ArticlePubMedGoogle Scholar
- Campbell D, Ronksley P, Hemmelgarn B, Zhang J, Barnabe C, Tonelli M, Manns B: Association of enrolment in primary care networks with diabetes care and outcomes among first nations and low-income Albertans. Open Med. 2012, 6 (4): e155-PubMed CentralPubMedGoogle Scholar
- Quan H, Sundararajan V, Halfon P, Fong A, Burnand B, Luthi JC, Saunders LD, Beck CA, Feasby TE, Ghali WA: Coding algorithms for defining comorbidities in ICD-9-CM and ICD-10 administrative data. Med Care. 2005, 43 (11): 1130-1139. 10.1097/01.mlr.0000182534.19832.83.View ArticlePubMedGoogle Scholar
- Charlson ME, Pompei P, Ales KL, MacKenzie CR: A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis. 1987, 40 (5): 373-383. 10.1016/0021-9681(87)90171-8.View ArticlePubMedGoogle Scholar
- Hemmelgarn BR, Clement F, Manns BJ, Klarenbach S, James MT, Ravani P, Pannu N, Ahmed SB, MacRae J, Scott-Douglas N, Jindal K, Quinn R, Culleton BF, Wiebe N, Krause R, Thorlacius L, Tonelli M: Overview of the alberta kidney disease network. BMC Nephrol. 2009, 10: 30-2369-10-30-View ArticleGoogle Scholar
- Lee JM, Davis MM, Gebremariam A, Kim C: Age and sex differences in hospitalizations associated with diabetes. J Womens Health (Larchmt). 2010, 19 (11): 2033-2042. 10.1089/jwh.2010.2029.View ArticleGoogle Scholar
- Rosenthal MJ, Fajardo M, Gilmore S, Morley JE, Naliboff BD: Hospitalization and mortality of diabetes in older adults. A 3-year prospective study. Diabetes Care. 1998, 21 (2): 231-235. 10.2337/diacare.21.2.231.View ArticlePubMedGoogle Scholar
- Green C, Blanchard JF, Young TK, Griffith J: The epidemiology of diabetes in the Manitoba-registered first nation population: current patterns and comparative trends. Diabetes Care. 2003, 26 (7): 1993-1998. 10.2337/diacare.26.7.1993.View ArticlePubMedGoogle Scholar
- Public Health Agency of Canada: Diabetes In Canada: Facts And Figures From A Public Perspective. 2011, Ottawa, OntarioGoogle Scholar
- Centers for Disease Control and Prevention: National Diabetes Fact Sheet: National Estimates And General Information On Diabetes And Prediabetes In The United States, 2011. 2011, Atlanta, GeorgiaGoogle Scholar
- Loppie C, Wien F: Health Inequalities And Social Determinants Of Aboriginal Peoples’ Health. 2009, Prince George, British Columbia: National Collaborating Centre for Aboriginal HealthGoogle Scholar
- Gao S, Manns BJ, Culleton BF, Tonelli M, Quan H, Crowshoe L, Ghali WA, Svenson LW, for the Alberta Kidney Disease Network: Prevalence of chronic kidney disease and survival among aboriginal people. J Am Soc Nephrol. 2007, 18 (11): 2953-2959. 10.1681/ASN.2007030360.View ArticlePubMedGoogle Scholar
- Dyck R, Osgood N, Lin TH, Gao A, Stang MR: Epidemiology of diabetes mellitus among First Nations and non-First Nations adults. CMAJ. 2010, 182 (3): 249-256. 10.1503/cmaj.090846.PubMed CentralView ArticlePubMedGoogle Scholar
- Donnan PT, Leese GP, Morris AD, Diabetes Audit and Research in Tayside, Scotland/Medicine Monitoring Unit Collaboration: Hospitalizations for people with type 1 and type 2 diabetes compared with the nondiabetic population of Tayside, Scotland: a retrospective cohort study of resource use. Diabetes Care. 2000, 23 (12): 1774-9. 10.2337/diacare.23.12.1774.View ArticlePubMedGoogle Scholar
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly credited. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.