Appendix B and C

Appendix B

2001 Census

Census information is collected either with a short questionnaire (2A) or with a long questionnaire (2B), which is administered to a random sample of one in five (20%) households. The long 2B form collects additional information including Aboriginal ancestry, Aboriginal identity, Band/First Nation membership and Registered Indian status. The 2B data are weighted to provide estimates for the entire population.

A different form (2D) is used for the Northern and Reserves Questionnaire. It is a long questionnaire similar to 2B and is used to enumerate the Yukon, the Northwest Territories (except Whitehorse and Yellowknife), Nunavut, Indian reserves, Indian settlements, Indian government districts and terres réservées. More information about the 2001 Census is available at: http://www12.statcan.gc.ca/english/census01/home/Index.cfm.

Dissemination area (DA) characteristics in this report were derived from the 2001 Community Profiles and are available at:  /bsolc/olc-cel/olc-cel?catno=95F0495XCB2001002&lang=eng. Public use 2001 Community Profiles data can also be accessed through the Data Liberation Initiative (DLI) .

Hospital Morbidity Database

The Hospital Morbidity Database (HMDB) is part of the larger hospital Discharge Abstract Database (DAD). The DAD contains a census of acute-care hospital separations (discharges and in-hospital deaths) for the April to March fiscal year. It is submitted to the Canadian Institute for Health Information (CIHI), in addition to other information provided by hospitals in provinces that do not submit to the DAD (Quebec, parts of Manitoba and Alberta, depending on the fiscal year). Additional information about the annual HMDB and detailed information about data quality are available at: www.cihi.ca.

For the HMDB, 2001/2002 was a transition year, as Canadian hospitals were completing implementation of the 10th revision of the International Disease Classification coding system. Depending on the province or territory, discharge information was reported according to one or more of three classification systems: International Classification of Diseases, 10th revision, Canadian Adaptation (ICD10-CA); International Classification of Diseases, 9th revision (ICD 9); or International Classification of Diseases, 9th revision - Clinical Modification (ICD 9-CM). For some analysis, coding conversion tables with information about the conversion grade quality produced by CIHI were used to convert codes to the ICD-9 system. For broad disease category classification (Figure 6), validated internationally accepted disease constructs that are comparable across classification systems and listed in the “International Shortlist for Hospital Morbidity Tabulation” (ISHMT) ⁴⁶ were used. Background information about the rationale, history and process of development for the ISHMT is available at: http://www.who.int/classifications/icd/implementation/morbidity/ishmt/en/index.html.

Appendix C

Supplementary methods information

Dissemination areas (DAs), the smallest geographic areas for which all census data are available, were used for this analysis. DA populations range from 400 to 700 persons and can represent a single city block in heavily populated urban areas or a much larger area in remote regions. In the 2001 Census, there were 52,993 dissemination areas.

In 2001, population information was missing for 2,689 DAs, which could not be classified as either a high- or low-percentage Aboriginal (0.05% of all 2001 DAs). Excluding Quebec, the number of unclassifiable DAs totaled 2,406, with an estimated population of 27,820 (0.1% of the total population of Canada excluding Quebec).

For census population counts, no adjustment was made for undercounting below the Census Division (CD) level. For information about under-enumerated areas for the 2001 Census and estimated populations in Indian Reserves, see Appendix A, Table E, and http://www12.statcan.ca/english/census01/Meta/indres.cfm.

Based on census information, a percentage Aboriginal was calculated for each DA.¹² The percentage Aboriginal is the number of people in the DA claiming Aboriginal identity, divided by the DA’s total population.

The PCCF+19 was used to assign a DA code to each postal code in the hospital discharge database. In some parts of Canada, postal codes do not correspond precisely to a single DA. In these cases, the PCCF+ completes an unbiased assignment to one DA after taking weighted population counts for each possible DA into consideration. Assignment of a percentage Aboriginal to each DA accounts for the fact that DA code assignment for postal codes has been determined probabilistically from (mostly rural) postal codes that span multiple DAs. The percentage Aboriginal is a weighted average of the percentage Aboriginal in all of the DAs that could have been selected.

Quebec was excluded from these analyses because the data submitted by that province include only Forward Sortation Area (FSA) codes rather than 6-digit postal codes.

More information about the PCCF+ and the reference files used in assigning of DA codes to the neighbourhood income quintile value for each DA using postal codes is available at: \\geodepot2\FTP\Geographie_2006_Geography\Geo_Data_Products-Produits_de_données_Géo\PCCFplus_version5E_Mar09, or from Russell Wilkins (613-951-5305; russell.wilkins@statcan.gc.ca), Health Analysis Division, Statistics Canada.