4. Indicators based on Statistics Canada surveys

4.1 National Population Health Survey

For detailed information on this survey, including methodology, go to the NPHS – Household component or NPHS – North component.

4.2 Canadian Community Health Survey

For detailed information on this survey, including methodology, go to Canadian Community Health Survey (CCHS).

Different period estimates available

Since 2007, data for the Canadian Community Health Survey (CCHS) are collected yearly instead of every two years. While a sample of approximately 130,000 respondents has been interviewed during the reference periods of 2003 and 2005, it has been changed to 65,000 respondents each year starting in 2007. Two indicator profiles based on CCHS core content are available, featuring the same variables, geography and age breakdowns, but different reference periods. The CANSIM table 105–0501 includes data from 2008, 2007, 2005 and 2003. These estimates present the most up–to–date population health characteristics and will be updated yearly. The CANSIM table 105–0502 includes estimates from 2007/2008 combined, 2005 and 2003. The two–year combined data are less current than annual estimates, but have higher precision (less variability). Users should refer to the CANSIM annual data table 105–0501 as the primary source for most current estimates from the survey. However, where data quality flags indicate suppression (F) or higher variability (E), the CANSIM two–year data table 105–0502 should be used.

"Not stated" category
Rates calculated from Canadian Community Health Survey (CCHS) data for Health Indicators no longer include the 'not stated' categories in the denominator. 'Not stated' refers to non–response which covers the 'Refusal', 'Don't Know', and 'Not Stated' response categories. This approach ensures a coherent method across all reported CCHS indicators, as well as with rates published in Statistics Canada's publication Health Reports, which generally exclude the 'not stated' responses from the denominator.

Regular medical doctor indicator
In 2005 and 2003, the indicator in French only included "médecin de famille". Starting in 2007, this concept was widened to "médecin régulier", which includes "médecin de famille".

Household Food Insecurity indicator
This variable is based on a set of 18 questions and indicates whether households both with and without children were able to afford the food they needed in the previous 12 months. The levels of food security are defined as: 1– Food secure: No, or one, indication of difficulty with income–related food access; 2– Moderately food insecure: Indication of compromise in quality and/or quantity of food consumed; 3– Severely food insecure: Indication of reduced food intake and disrupted eating patterns.

This variable is based on the CCHS Food Security module, a set of 18 questions, and indicates whether households both with and without children were able to afford the food they needed in the previous 12 months. A full Household measure of food security was based on all 18 questions, with separate Adult and Child measures being derived from 10 questions about adult food security and 8 about child food security, respectively. Note that percentages for the Household and Adult measures are calculated using the whole sample as the denominator, whereas percentages for the Child measure have only households with children under 18 years as the denominator.

The respondents were aged 12 or older, but questions refer to the food insecurity status of all members of the household. Household weights were used to investigate household characteristics (e.g. living arrangement, presence of children in the household), and person–level weights were used to investigate the characteristics of the respondents (e.g. age, sex), relative to food insecurity status.

Household food security exists in a household when there is sufficient income (financial resources) to provide access by all people at all times to enough food for an active, healthy life. Conversely, household income–related food insecurity results when lower household incomes lead to compromises in food quality or quantity.

Breasfeeding practices indicator – exclusive breastfeeding

Starting in 2010, data for this indicator have been updated for all years included in the table. The numerator includes mothers who have exclusively breast fed for at least 6 months and who may or may not be still breastfeeding. The denominator includes all mothers who had a baby in the past five years but excludes mothers who were still breastfeeding and who had not introduced any other liquids or solid foods to the baby's feeds.

Previously, this indicator included in the numerator, only mothers who had stopped breastfeeding and for whom we knew they had introduced other liquids or solid foods to the baby's feeds when the baby was 6 months or more. The denominator included mothers who had stopped breastfeeding and for whom we knew when they had introduced other liquids or solid foods to the baby's feeds.

This modification will produce lower rates of six months exclusive breastfeeding as mothers who have had a baby in the past five years and have not breastfed are now included in the denominator. This change was implemented to produce more comparable rates over time and is more consistent with methods used in calculating other indicators.

4.3 Bootstrap resampling method

To ensure high data quality for estimates from the NPHS, the CCHS and NLSCY , a weighted bootstrap resampling procedure (and for the NPHS–North, a modified bootstrap procedure) was used to calculate coefficients of variation (CVs) for totals and rates. If the CV was greater than 33.3% or the sample size was less than 10, the data were suppressed and an "F" symbol appears in the data cell. If the CV is greater than 16.5% and no greater than 33.3%, the data should be interpreted with caution and an "E" symbol appears in the same cell as the data. Data with CVs of 16.5% or less are presented without restrictions.

Sampling theory dictates that sample survey results of exactly 100% or 0% must have a coefficient of variation of exactly 0. In reality it is possible that in rare circumstances the true estimate may be lower than 100% or conversely greater than 0% and results should be interpreted as such.

Date modified: