1. 
The table shows body mass index (BMI) changes
of the 1994/1995 household population aged 18 to 56,
in cycle 1 and in another cycle of the National Population Health
Survey. Population, age and sex are based on the first survey cycle (cycle 1)
in 1994/1995. 
2. 
Body mass index (BMI) is calculated by dividing
the respondent's body weight (in kilograms) by their height (in metres) squared.
The index is: under 18.5 (underweight); 18.5 to 24.9 (normal
weight); 25.0 to 29.9 (overweight); 30.0 or
higher (obese). 
3. 
The table excludes persons who died or who
were institutionalized after cycle 1 survey interview in 1994/1995.
Estimated number of persons who have died: 1,279 persons after cycle 5 and 1,640 persons
after cycle 6. These data are not all confirmed yet with the Canadian
Vital Statistics Database. Persons living in an institution are excluded because
they are not asked certain questions. Estimated number of persons who have
been institutionalized: 161 persons at cycle 5 and 144 persons
at cycle 6. 
4. 
The table also excludes nonresponse (missing
values), that is, "not applicable", "don't know", "refusal" and "not stated".

5. 
No estimates for body mass index are provided
for 1994/1995; what is shown instead are the changes from one body mass
index category to another. 
6. 
Percentages are based on the population with
a specific characteristic (one category) in a given cycle as the numerator
divided by the total population in 1994/1995 having all possible
characteristics (all categories) as the denominator, excluding missing values.

7. 
Bootstrapping techniques were used to produce
the coefficient of variation (CV) and 95% confidence intervals (CIs).

8. 
Data with a coefficient of variation (CV) from 16.6%
to 33.3% are identified by an (E) and should be interpreted with caution.

9. 
Data with a coefficient of variation (CV) greater
than 33.3% were suppressed (F) due to extreme sampling variability. 
10. 
Changes in body mass index of the 1994/1995 population
in 2004/2005 are based on the 2004/2005 longitudinal full
cycle 1 and cycle 6 subset of respondents (from 2004/2005 longitudinal
data file). This subset contains all panel members with a complete (full)
response in cycle 1 (1994/1995) and cycle 6 (2004/2005),
regardless of their response pattern in cycle 2 (1996/1997), cycle 3 (1998/1999),
cycle 4 (2000/2001) and cycle 5 (20022003). Number of
respondents for this table (excluding missing values): 6,382. 
Note(s): 
When comparing estimates, it is important to use confidence intervals
to determine if differences between values are statistically significant.
Confidence intervals describe sampling variability and give an indication
of the precision of a given estimate. Please note that confidence intervals
and coefficient of variations are provided in the CANSIM tables. 
Source(s): 
Statistics Canada, National Population Health Survey, 2002/2003 and 2004/2005 longitudinal
data file (CANSIM table 1047030).
