Public Health Agency of Canada
www.publichealth.gc.ca
National Diabetes Surveillance System (NDSS) Methods Report, Abridged Version
National Diabetes Surveillance System (NDSS)
- In each province and territory, the health insurance registry database is linked to the physician billing and hospitalization databases, using a unique person-specific identifier assigned through the provincial and territorial health insurance systems.
- This resulting database is used to determine individuals who have received a diagnosis of diabetes, based on the NDSS validated definition.
- To protect an individual's privacy, data were aggregated by age group and sex for the years 1995-1996 to 2004-2005. Each province and territory sent the resulting aggregate data set to the Public Health Agency of Canada (PHAC). These data form the National Diabetes Surveillance System (NDSS) database and are used for analysis and to describe the burden of diagnosed diabetes among Canadians. In order to adjust for age distribution differences between the provinces and territories over time, the aggregate rates were age-standardized, using the 1991 Canadian Census as the reference population. In addition, for selected statistics, the 95% confidence interval was presented, showing the likelihood of including the true statistic, 19 times out of 20.
Abridged Methods for the Provincial, Territorial, and Canadian NDSS Data
NDSS Case Definition: Individuals were considered an NDSS diabetes case1, and included in the NDSS summary data set, when they had at least one hospitalization with a diagnosis of diabetes (Table 1) or had at least two physician visits with a diagnosis of diabetes within a two-year period. The case date was determined by the dates of the hospitalization or the most recent of the two physician visits. Prevalence and incidence rates would be artificially affected (lower and higher rates, respectively) by capturing cases that were first diagnosed prior to the start of the capture period (April 1, 1995-1996 to March 31 2004-2005). Therefore, the NDSS reporting time period began with the 1997-1998 fiscal year (prevalence rates), and 1998-1999 (incidence rates). The NDSS data for this report includes summary information on individuals aged 1 year and older, excluding cases of gestational diabetes, from all provinces and territories for the fiscal years 1997-1998 to 2004-2005 (prevalence) and 1998-1999 to 2004-2005 (incidence).
Prevalence: The NDSS defines diabetes prevalence as the percent of cases2 among individuals with valid health insurance coverage during the selected capture period (April 1, 1995-1996 – March 31 of the index year).
| Prevalence = | Total number of individuals with a diabetes case date during the capture period (prevalent cases) | X 100 |
| _______________________________ | ||
| Total number of individuals with a valid health insurance number^ during the capture period |
Incidence: The NDSS defines diabetes incidence as the rate of new cases2 occurring, among individuals, who were not already a prevalent case, with valid health insurance coverage during the fiscal year (April 1 – March 31 of the selected year).
| Incidence = | Total number of individuals with a diabetes case date in a selected fiscal year | X 1000 | |
| ______________________________ | |||
|
^Quebec and Newfoundland use census-based population estimates to report these numbers.
Mortality
Death rates: NDSS mortality analyses were based on deaths from any cause, from provincial and territorial health insurance registries, and calculated separately for individuals with and without diabetes, during the capture period. The denominator includes individuals in the NDSS summary data set with or without diabetes. The ratio between the two rates reflects the significance of diabetes mortality among Canadians.
| Death rate (with diabetes) = |
Total # of deaths in the 2004-2005 fiscal year among individuals with diabetes |
| _________________________________ | |
| Total number of individuals with diabetes |
| Death rate (without diabetes) = |
Total # of deaths in the 2004-2005 fiscal year among individuals without diabetes |
| _________________________________ | |
| Total number of individuals without diabetes |
| Rate ratio of death rates = |
Death rate among individuals with diabetes |
| ________________________________ | |
| Death rate among individuals without diabetes |
For Quebec, the number of deaths among individuals without diabetes was calculated by subtracting the number of deaths among individuals with diabetes, as determined by the NDSS, from the count of all deaths from their vital statistics department. The denominator for the rate among persons without diabetes is the difference between census-based population estimates and the number of persons with diabetes.
Life Expectancy: Life Expectancy was computed by using an abridged life table; applying the method of Keyfitz3 to 2002-2003 to 2004-2005 age-specific NDSS death rates. For a particular age group the corresponding life expectancy value may be interpreted as the average years of life remaining from that age until death. Life expectancy was calculated by age group, sex, and diabetes status. Because diabetes status was not available for people younger than one year of age; the 2002 to 2004 sex-specific death rates for the Canadian population were used to model the mortality experience of individuals with and without diabetes for this age group.
Health Services Utilization
General and Specialist Practitioner Rates and Rate Ratios: The number of general and specialist practitioner visits,4 for individuals with and without diabetes, for the 2004-2005 fiscal year, were aggregated and age-standardized rates and rate ratios were computed using similar methods as described for the death rates and rate ratios of death rates.
Data for Quebec and Nunavut were excluded from the health services utilization analyses.
Preliminary Disease Codes for Comorbid Hospitalizations
Comorbid Hospitalizations: The NDSS comorbid hospitalizations analyses were based on the number of individuals with or without diabetes, during the capture period, and hospitalizations for select comorbidities and lower limb amputation procedures5, associated with diabetes in the 2004-2005 fiscal year. The age-standardized comorbid hospitalization rates and rate ratios were calculated separately for persons with and without diabetes as follows:
| Rate of comorbid hospitalizations (with diabetes) = |
Total number of individuals hospitalized in the 2004-2005 fiscal year among individuals with diabetes |
| _________________________________ | |
| Total number of individuals with a valid health insurance number and diabetes during the 2004-2005 fiscal year |
| Rate of comorbid hospitalizations (without diabetes) = |
Total number of individuals hospitalized in the 2004-2005 fiscal year among individuals without diabetes |
| ______________________________ | |
| Total number of individuals with a valid health insurance number without diabetes during the 2004-2005 fiscal year |
| Rate of comorbid hospitalizations = |
Comorbid hospitalization rate among individuals with diabetes |
| _________________________________ | |
| Comorbid hospitalization rate among individuals without diabetes |
Quebec was excluded from the analyses. Lower limb amputations data for Yukon and Nunavut were excluded from the analyses.
International Classification of Disease Codes (ICD) Codes Used in NDSS Reporting and Analyses
| Table 1. Diabetes Diagnosis ICD Codes | |
| ICD-9 | 250 |
| ICD-10-CA | E10-E14 |
| Gestational Diabetes Exclusion ICD Codes | |
| ICD-9 | 650-669 |
| ICD-10-CA | O265, O290-O296, O298-O299, O300-O302, O308-O309, O318, O320-O326, O328-O329, O330-O339, O340-O349, O350-O359, O360-0369, O40, O410-O411, O418-O419, O420-O422, O429, O430-O431, O438-O439, O60, O610-O611, O618-O619, O620-O624, O628-O629, O630-O632, O639, O640-O645, O648-O649 O650-O655, O658-O659, O660-O665, O668-O669, O680-O683, O688-O689, O690-O695, O698-O699, O700-O703, O709, O710-O719, O720-O723 , O730-O731, O740-O749, O750-O759, O800-O801, O808-O809, O810-O815, O820-O822, O828-O829, O830-O834, O838-O839, O840-O842, O848-O849, O890-O896, O898-O899 , O904, O908, O95-O97, Z354-Z356 |
| Table 2. Comorbid Hospitalization ICD Codes | |
| Cardiovascular Disease | ICD-10-CA: I00-I78 |
| Ischaemic Heart Disease | ICD-10-CA: I20-I25 |
| Hypertensive Disease | ICD-10-CA: I10-I13, I15 |
| Acute Myocardial Infarction | ICD-10-CA: I21-I22 |
| Heart Failure | ICD-10-CA: I50 |
| Stroke | ICD-10-CA: I60-I69 |
| Renal Disease | ICD-10-CA: N18-N19 |
| Lower Limb Amputations CCP, and CCI Codes6 | |
| Canadian Classification of Diagnostic, Therapeutic and Surgical Procedures (CCP) | 96.11-96.15 |
| Canadian Classification of Health Interventions (CCI) | 1SQ93, 1VA93, 1VC93, 1VG93, IVQ93, 1WA93, 1WE93, 1WJ93, 1WL93, 1WM93 |
- From this point forward in the report, "diabetes" refers to a diabetes case, as defined by the NDSS case definition.
- Coded using the International Classification of Disease (ICD) 9th Revision, code 250 or ICD-10-CA (E10 to E14)
- Keyfitz N. A life table that agrees with the data. Part I. Journal of the American Statistical Association 1966; 61 (314): 305-12.
- Specialists did not include parahealth professionals, such as, chiropractors, naturopaths, dentists, or optometrists.
- Select comorbid hospitalizations were for the following diseases: cardiovascular, ischaemic heart, hypertension, renal and conditions: acute myocardial infarction, heart failure, and stroke. They were coded using ICD-9, ICD-10-CA, and the Canadian Classification of Health Interventions (CCI), Table 2.
Lower limb amputations were coded using The Canadian Classification of Diagnostic, Therapeutic, and Surgical Procedures (CCP), Table 2. - Disease Exclusion ICD Codes: ICD-10-CA: C40, C41, C461, C47, C49, C80, C962, D160, E830, G901, H473, L590, M431, M432, M898, N079, N818, P293, P960, Q, R294, S, T
To share this page just click on the social network icon of your choice.