Recreational use of prescription medications among Canadian young people: identifying disparities.
Pulver, Ariel ; Davison, Colleen ; Pickett, William 等
The non-medical use of prescription medications is recognized as a
pressing public health issue in Canada. (1) While there have been calls
for research and intervention development to manage this issue, (1)
there are limited epidemiological studies on patterns of use in Canada,
particularly studies that identify populations at highest risk,
including adolescents.
To date, knowledge about the patterns of recreational use of
prescription medications in subpopulations of Canadian youth is very
limited. Results from a US survey indicate a 212% increase in
non-medical use of prescription drugs in adolescents aged 12-17 between
1992 and 2003. (2) This was 2.6-fold higher than the increase among
adults, suggesting that adolescents are particularly vulnerable with
regard to non-medical use of prescription drugs. (2) Reported rates of
use among youth from grades 7 through 12 have varied from 5.9%,3 to
15.5%. (4)
One established determinant of non-medical use of prescription
drugs among youth is older age. Between the ages of 12 and 15, the risk
of engaging in non-medical use of prescription drugs increases by
between 88% and 130%. (5,6) Female sex is another commonly identified
determinant, females having 17%-50% greater risk. (7,8) Low
socio-economic status (SES) has also been significantly associated with
non-medical use of prescription drugs, the odds ratios ranging from 1.2
to 1.5.5, (7-9) Moreover, Canada is a country where immigrant youth
account for 9.2% of the population under 24 years. (10) While studies of
alcohol and illicit drug use indicate disparities, (11) it is unknown
whether non-medical prescription drug use varies by whether a young
person is Canadian-born or not or by length of residence in Canada.
The National Advisory Committee on Prescription Drug Misuse
released a report in 2013 recommending a pan-Canadian strategy,
including an emphasis on addressing knowledge gaps surrounding
geographically remote and rural populations and the non-medical use of
prescription drugs. (1) The focus on rural groups follows from studies
from the US that identify rurality as an important risk factor for
opioid pain relievers in particular, with odds ratios ranging from 1.22
to 5.69. (5,8,12)
A standard method for defining urban or rural areas for studies of
geographic disparities in health does not exist. Population size and
density are most commonly used, although there have been discussion and
development of additional demarcation methods. (13) One study found that
selected types of substance use increased with remoteness, for example,
and not just with smaller or less dense rural populations. (14) To our
knowledge, only one Canadian study has examined the role of geographic
location in the non-medical use of prescription drugs among youth. (4)
In that study, female adolescents who reported having used opioids
non-medically had 1.95 times the odds of living in rural areas than in
urban/suburban areas, as defined by population size.
The aim of this study was to characterize recreational prescription
drug use in subgroups of Canadian youth by age, sex, SES, immigrant
status and geographic location to identify disparities that may
systematically place subgroups at further disadvantage with respect to
their health. The definition of disparities here follows from
Braveman's definition: a particular type of difference in or
influence on one's health potentially shaped by policies. (15) The
findings may help identify directions for improving prescription
practices and highlight circumstances in which secure storage of
medications is most warranted.
METHODS
This was a descriptive, epidemiological study employing
crosssectional analyses of data describing reported experiences of young
adolescents in Canada. The primary focus was on variations in
recreational use of prescription medications in a disaggregated analysis
by age, sex, SES, immigration status and geographic status. These
factors reflect important health determinants that underlie potential
disparities in the non-medical use of prescription medications.
Medications of interest included pain relievers, stimulants and
sedatives/tranquilizers.
Data source and sample
Health Behaviour in School-aged Children (HBSC) is a health survey
of young people primarily aged 11 through 15 conducted in 43 countries
or regions in collaboration with the World Health Organization. The
purpose is to understand health behaviours and determinants of health in
young people. (16) The data source for the current study was Cycle 6 of
the Canadian HBSC conducted during the 2009/2010 school year in all
Canadian provinces and territories except Prince Edward Island and New
Brunswick. The Canadian HBSC sample was obtained using a two-stage
cluster sample design, in which schools were selected randomly and the
class was the basic cluster. Response rates were 84.6% at the
provincial/territorial level and 57.0% at the school level; 77.0% of
eligible students who were approached participated in the study. Active
or passive consent was obtained depending on the schools' or school
boards' policies for conducting classroom-based research. The final
sample comprised 10,429 Canadian students, primarily in grades 9-10. The
study protocol received ethics approval from the Queen's University
Research Ethics Board.
Study variables
Students indicated their birth year and month, the date of survey
completion, and their sex. A geographic location for each student was
ascertained according to the school postal code. Their geographic status
was then determined using Statistics Canada definitions. (13) Students
were classified as living in "urban" areas if their school was
in a census metropolitan (>100,000 population) or census
agglomeration (>10,000 population) area. Students were identified as
living in "rural or small town" areas if their school was not
in an urban area. Rural and small towns were then further classified
into Metropolitan Influenced Zones (MIZ). These are founded upon
principles of distance, adjacency and accessibility between urban
centres and rural and small town areas. (13) They measure the degree to
which urban centres influence rural and small town municipalities, as
determined by commuting flows. "Strong" Metropolitan
Influenced Zones are census subdivisions in which 30%-50% of the
employed labour force commutes to work in an urban centre.
"Moderate" MIZ (5.0% to 30% commuting flow) and
"Weak" MIZ (0.1% to <5.0%) were also identified. In a
"No Metropolitan Influenced Zone", none of the employed labour
force commuted to work. For the current study, Weak and No Metropolitan
Influence Zones were combined into one group.
SES was determined at the individual student level using a 5-point
student self-report Likert-like scale pertaining to how well off
students perceived their family to be. Responses were then categorized
into three groups: 1) Low (not at all well off and not very well off),
2) Average and 3) High (quite well off and very well off). This item has
been employed and investigated in previous Canadian and international
studies, has demonstrated reliability and validity based on other
measures of SES, (14) and has been shown to be a stronger predictor of
adolescent health outcomes than area level measures of SES. (17)
Immigrant status was determined by asking a student the country in
which they were born and how long they had lived in Canada. Data
corresponding to these items were categorized into Born in Canada;
immigrant > 5 years; or immigrant < 5 years.
Recreational use of prescription medications. Using a categorical
item with close-ended response categories, students were asked to
indicate how frequently they had used pain relievers, stimulants and
sedatives/tranquilizers "to get high" in the previous year.
Specific examples of drugs within each classification were provided.
Response categories ranged from never to 40 times or more. Responses
were subsequently grouped into "no use" and "ever
use". Those who reported past-year recreational use of one or more
medications > 3 times were further categorized as frequent users, and
those who reported using 1-2 times were categorized as infrequent users.
This categorization has been previously used to identify problematic
substance use in the Ontario Student Drug Use and Health survey. (18)
Survey weights and statistical analysis
Data were weighted by grade and province/territory to ensure that
the survey was nationally representative. If a specific grade group in a
specific province or territory was over-represented, those student
responses were given a weight of < 1, and under-representation was
corrected by weights of > 1 (weights ranged from 0.017 to 3.655).
Cross-tabulations were conducted to estimate the proportion of youth
within predefined subgroups who reported recreational use of medications
and to identify proportions of youth using more than one type of drug.
Proportions of infrequent and frequent users by subgroups were also
estimated. Multi-level and multivariable Poisson regression was used to
estimate the strengths of associations between the exposure variables of
interest and reported prescription medication outcomes in a fully
adjusted model. Adjusted relative risks (RRs) as well as corresponding
95% confidence intervals (CIs) were estimated. The model specified the
hierarchical sampling design, accounting for the nested and clustered
nature of the study sample, with students nested within schools. Random
intercepts were assumed for schools and fixed effects for the
determinants of interest.
RESULTS
Recreational use of prescription medication
A description of the study sample can be found in Table 1. Table 2
displays the proportion of youth who reported recreational use of any
prescription medication in the previous year, adjusted RR estimates and
a p test for linear trend in variables with more than two categories.
Table 3 contains a breakdown of this information by specific medication
types.
Older age was associated with increased risk of recreational use of
prescription drugs. Proportions of past-year use of any drug were 5.5%,
6.7% and 7.6% for students <14 years old, 15 years old and [greater
than or equal to] 16 years old, respectively ([p.sub.trend] < 0.01).
This difference was particularly notable for stimulant and sedative
medications, for which risk of use among students 16 years and older was
1.7 and 2.0 times greater than for those in the youngest age group.
Girls reported greater use of pain relievers than boys (5.5% vs.
4.6%; RR 1.25, 95% CI: 1.04-1.51), whereas boys reported slightly higher
use of stimulant medications than girls (2.6% vs. 2.2%).
Sedative/tranquilizer use did not differ between the sexes.
Lower SES students reported higher overall use as compared with
high SES students (13.0% versus 5.5%, RR 2.41, 95% CI: 1.94-2.99;
[p.sub.trend] < 0.01). For pain relievers, 10.0% of low SES students
reported past-year use, as compared with only 4.2% of high SES students
(RR 2.32, 95% CI: 1.81-2.98; [p.sub.trend] < 0.01). Of low SES
students 5.4% reported recreational use of stimulants, compared with
2.0% of their high SES counterparts (RR 2.70, 95% CI: 1.91-3.81;
[p.sub.trend] < 0.01). Use of sedative/tranquilizer medications was
least common; however, low SES students were 3.05 times more likely to
report using them recreationally than high SES students (95% CI:
1.92-4.88; [p.sub.trend] < 0.01). Proportions of use were similar
among those born in Canada, new immigrants and those living in Canada
for more than 5 years.
Compared with youth living in urban areas, those living in Strong
Metropolitan Influenced Zones (with 30%-50% commuting) were 2.39 times
more likely to use any prescription drug (95% CI: 1.03-5.55) and 3.13
times more likely to use pain relievers (95% CI: 1.23-8.01). Reports of
recreational use of prescription drugs did not differ between the more
remote geographic categories and urban areas.
Among youth who reported using prescription drugs recreationally,
15.9% reported using two types, and 11.4% reported using all three types
of medication. Youth who used sedatives showed the highest proportion of
co-use: 70.0% had also used pain relievers and 63.1% had also used
stimulants. Of stimulant users, 53.4% had also used pain relievers and
36.8% had also used sedatives; 25% of youth who used pain relievers had
also used stimulants, and 18.8% had used sedatives as well.
Frequent recreational use of prescription medications
Approximately half of students who reported using prescription
medications recreationally had done so at least three times in the
previous year, operationally defined as "frequent use" (see
Tables 4 and 5). Of the medications used frequently, stimulants were the
most common (57.6%) followed by sedatives (53.4%) and pain relievers
(43.5%). Age was not associated with frequent use of prescription
medications. Boys were more likely to report frequent recreational use
of prescription medications than girls (43% girls vs. 56% boys, adjusted
RR 0.77, 95% CI: 0.61-0.97). This gender-based pattern was most
pronounced for stimulant medications (47% girls vs 68% boys; adjusted
RR: 0.71, 95% CI: 0.49-0.99). Because of the complete or
quasi-separation of geographic and immigrant status variables for both
frequent stimulant and frequent sedative use, they were excluded from
these models.
DISCUSSION
Our study provides foundational information about the recreational
use of prescription medications by Canadians in their early adolescent
years. Older age, female sex, lower SES and living in rural areas with
more metropolitan influence were independently associated with increased
risk of reported recreational use of prescription drugs. Recreational
use of pain relievers was almost twice that of stimulants and three
times that of sedatives/tranquilizers.
Increasing reports of recreational prescription drug use by age
confirm findings from earlier studies. (3-6,8,19) However, the
prevalence of such use here was slightly lower than levels reported by
previous studies of Canadian youth. (4,6) This may because of our
younger adolescent sample. There is a need to identify patterns specific
to this young age group, as US evidence suggests that the mean age for
initial non-medical prescription drug use may be as early as 13 years
old. (9) Our examination of frequent use of these drugs was unique,
however, and we did not identify strong age-related patterns. While age
is an important predictor of drug experimentation, (20) substance abuse
disorders may not emerge until early adulthood (19-21 years), (21)
perhaps explaining why more problematic use was not apparent in our
relatively youthful sample.
Reported patterns of recreational drug use by males and females
differed by type of medication. Use of pain reliever medications was
higher among females, whereas males reported slightly more use of
stimulant medications. Females are more likely to be prescribed opioid
medications than males (22) and therefore may use their own
prescriptions recreationally more often. (23) Our gender-based finding
for increased non-medical stimulant use is consistent with that of a
study conducted in the Atlantic provinces, where males reported more
non-medical use. (24)
Students of low SES reported the highest rates for using all three
types of medication recreationally, supporting findings from a recent
Canadian study of recreational opioid use among youth. (4) Youth living
in poorer socio-economic conditions may confront greater barriers when
faced with decisions about engaging in drug use, as they may have fewer
opportunities for structured recreation, more deviant peers, less
parental supervision and more stressful life events. (25-28)
The geographic patterns highlighted in this study point to
substantial intra-rural variability with respect to non-medical use of
opioid medications. Students living in rural areas subject to strong
urban influence reported use of these drugs most commonly. While these
results build upon earlier findings that emphasize rural drug use
patterns, (4,5,8) the patterns highlighted here emphasize an urban-rural
connection that possibly relates to access to prescription medications.
Contrary to a past study that suggested an urban-rural gradient, (29) we
found that the highest levels occurred in rural areas that were more
proximal and accessible to urban settings.
There is some evidence that health service access and utilization
by people in rural areas proximal to urban centres is different from
those in other rural areas. (30,31) Others have reported that rural
residents living adjacent to urban centres are more likely to have a
regular medical doctor than those living more remotely (OR: 0.62, 95%
CI: 0.53-0.74). (30) Rates of specialist physician consultations are
also higher in these rural areas than in those further away. (30) Access
to prescriptions for controlled medications may follow this pattern, so
that such medications may be more readily available for rural residents
living adjacent to urban cores. This may also relate to the number and
influence of illicit suppliers in urban areas. (32)
There are other explanations for the excess recreational use of
prescription medications in certain types of geographic communities.
First, because use of some medical services is greater in these proximal
rural areas there may be a greater volume of unused medications in home
medicine cabinets. (30) Second, older adults are the most likely group
to receive controlled medications for chronic conditions and pain, (33)
and therefore a greater volume of controlled medications may be present
in some rural areas with their relatively older population structures.
Third, when rural residents do obtain prescriptions, they may be more
likely to save or stockpile excess amounts for future use because of
higher dispensing fees in rural areas. (34) Fourth, rural youth may
spend more of their time in unsupervised activities and thus may be at
greater risk of drug use. (35) All of these ideas are speculative, and
further investigation into the root causes of this geographic pattern is
clearly warranted.
The strengths of this study include its use of a nationally
representative sample that was of adequate size to detect most subgroup
differences, the uniqueness of our data and our emphasis on
disaggregation of the analysis by important subgroups of youth. The
limitations also warrant comment. No differences in use were detected
with respect to immigration status. This may be due to limitations with
our available immigrant measure, resulting in cultural and religious
heterogeneity within groups. Given the relative infrequency of use among
immigrant youth, it would be challenging to detect subgroup differences
by ethnicity or cultural heritage within the immigrant groups
themselves. We relied on self-reported drug use, which may be subject to
social desirability bias and result in some misclassification. We
believe, however, that these possible biases would be non-differential
among subgroups, thereby potentially underestimating, and not
overestimating, effect estimates. Another limitation is that young
people living on First Nations reserves, incarcerated youth,
home-schooled students, students who did not have consent, those who
were absent on the day of the survey and those attending private schools
were excluded. This limits our comparisons of groups that may be
particularly vulnerable to recreational use of these drugs. We also do
not have any information about the method or dose of drug
administration, information that would be helpful in signalling more
problematic use. All of these limitations point to the need for more
refined study of this important and emerging public health issue for
adolescent Canadians.
CONCLUSIONS
The non-medical use of prescription medications is an important
public health issue in Canada. Nearly 7% of Canadian youth reported
recreational use of prescription medications in the previous year, and
approximately half reported use of them three or more times.
Recreational use of pain relievers was most common and was highest among
youth living in rural areas proximal to urban centres. Findings from
this study could help inform preventive interventions, such as efforts
to promote parental vigilance and other strategies to restrict access to
leftover medications, particularly in rural settings. Future research
should consider the diversity of rural communities and particularly the
risk factors that may be in place in vulnerable rural locations.
Received: August 11, 2013
Accepted: January 26, 2014
Correspondence: Colleen Davison, Department of Public Health
Sciences, Carruthers Hall, Queen's University, Kingston, ON K7L
3N6, E-mail:
[email protected]
Acknowledgements: This study was supported by research grants from
the Canadian Institutes of Health Research (CIHR) (operating grants MOP
9762 and PCR 101415). Ariel Pulver was supported by the Queen's
Graduate Award and a CIHR Strategic Training Fellowship Program in
Public Health and the Agricultural Rural Ecosystem. The Health Behaviour
in School-aged Children (HBSC) survey, a World Health Organization
European Region collaborative study, was funded in Canada by the Public
Health Agency of Canada and Health Canada. International Coordinator of
the HBSC is Candice Currie (University of Edinburgh). The principal
investigators of the 2009/2010 Canadian HBSC survey were William Pickett
and John Freeman, and its national coordinator was Matthew King. We
thank Mr. Andrei Rosu for his assistance with the geographic aspect of
this study.
Conflict of Interest: None to declare.
REFERENCES
(1.) National Advisory Committee on Prescription Drug Misuse. First
Do No Harm: Responding to Canada's Prescription Drug Crisis.
Ottawa, ON: Canadian Centre on Substance Abuse, 2013.
(2.) CASA National Advisory Commission on Diversion and Abuse of
Controlled Prescription Drugs. Under the Counter: The Diversion and
Abuse of Controlled Prescription Drugs in the U.S. New York, NY:
National Center on Addiction and Substance Abuse at Columbia University,
2005.
(3.) Currie CL, Wild TC. Adolescent use of prescription drugs to
get high in Canada. Can J Psychiatry 2012; 57(12):745-51.
(4.) Fischer B, Ialomiteanu A, Boak A, Adlaf E, Rehm J, Mann RE.
Prevalence and key covariates of non medical prescription opioid use
among the general secondary student and adult populations in Ontario,
Canada. Drug Alcohol Rev 2013; 32(3):276-87.
(5.) Havens JR, Young AM, Havens CE. Nonmedical prescription drug
use in a nationally representative sample of adolescents: Evidence of
greater use among rural adolescents. Arch Pediatr Adolesc Med 2011;
165(3):250-55.
(6.) Brands B, Paglia-Boak A, Sproule BA, Leslie K, Adlaf EM.
Nonmedical use of opioid analgesics among Ontario students. Can Fam
Physician 2010; 56(3):256-62.
(7.) Nakawaki B, Crano WD. Predicting adolescents'
persistence, non-persistence, and recent onset of nonmedical use of
opioids and stimulants. Addict Behav 2012; 37(6):716-21.
(8.) Ford JA. Nonmedical prescription drug use among adolescents:
The influence of bonds to family and school. Youth Soc 2009;
40(3):336-52.
(9.) Wu L, Ringwalt CL, Mannelli P, Patkar AA. Prescription pain
reliever abuse and dependence among adolescents: A nationally
representative study. J Am Acad Child Adolesc Psychiatry 2008;
47(9):1020-29.
(10.) Statistics Canada. 2006 Census of Population. Ottawa:
Statistics Canada, 2007. Catalogue no. 97-557-XCB2006013.
(11.) Gfroerer JC, Tan LL. Substance use among foreign-born youths
in the United States: Does the length of residence matter? Am J Public
Health 2003; 93(11):1892-95.
(12.) Young AM, Havens JR, Leukefeld CG. A comparison of rural and
urban nonmedical prescription opioid users' lifetime and recent
drug use. Am J Drug Alcohol Abuse 2012; 38(3):220-27.
(13.) McNiven C, Puderer H, Janes D. Census Metropolitan Area and
Census Agglomeration Influenced Zones (MIZ): A Description of the
Methodology. Ottawa: Geography Division, Statistics Canada, 2000. Report
No. 2000-2.
(14.) Simpson K, Janssen I, Craig WM, Pickett W. Multilevel
analysis of associations between socioeconomic status and injury among
Canadian adolescents. J Epidemiol Community Health 2005; 59(12):1072-77.
(15.) Braveman P. Health disparities and health equity: Concepts
and measurement. Annu Rev Public Health 2006; 27:167-94.
(16.) Currie C, Griebler R, Inchley J, Theunissen A, Molcho M,
Samdal O, Dur W. Health Behaviour in School-aged Children (HBSC) Study
Protocol: Background, Methodology and Background Items for the 2009/10
Survey. Edinburgh, Scotland: Child and Adolescent Health Research Unit,
and Vienna, Austria: Ludwig Boltzmann Institute for Health Promotion
Research, 2010. Available at: http://www.hbsc.org
(17.) Schnohr CW, Kreiner S, Due EP, Currie C, Boyce W, Diderichsen
F. Differential item functioning of a family affluence scale: Validation
study on data from HBSC 2001/02. SocIndicRes 2008; 89(1):79-95.
(18.) Paglia-Boak A, Adlaf EM, Mann RE. Drug Use Among Ontario
Students 19772011: Detailed OSDUHS Findings. Toronto, ON: Centre for
Addiction and Mental Health, 2011. Report No.32.
(19.) Sung H, Richter L, Vaughan R, Johnson P, Thom B. Nonmedical
use of prescription opioids among teenagers in the United States: Trends
and correlates. J Adolesc Health 2005; 37(1):44-51.
(20.) Patton G, McMorris B, Toumbourou J, Hemphill S, Donath S,
Catalano R. Puberty and the onset of substance use and abuse. Pediatrics
2004; 114(3):e300 e306.
(21.) Kessler R, Berglund P, Demler O, Jin R, Merikangas K, Walters
E. Lifetime prevalence and age-of-onset distributions of DSM-IV
disorders in the National Comorbidity Survey Replication. Arch Gen
Psychiatry 2005; 62:593-602.
(22.) Green TC, Grimes Serrano JM, Licari A, Budman SH, Butler SF.
Women who abuse prescription opioids: Findings from the Addiction
Severity IndexMultimedia Version Connect prescription opioid database.
Drug Alcohol Depend 2009; 103(1-2):65-73.
(23.) Back SE, Payne RL, Simpson AN, Brady KT. Gender and
prescription opioids: Findings from the National Survey on Drug Use and
Health. Addict Behav 2010; 35(11):1001-7.
(24.) Poulin C. From attention-deficit/hyperactivity disorder to
medical stimulant use to the diversion of prescribed stimulants to
non-medical stimulant use: Connecting the dots. Addiction 2007;
102(5):740-51.
(25.) Hanson MD, Chen E. Socioeconomic status and health behaviors
in adolescence: A review of the literature. Behav Med 2007;
30(3):263-85.
(26.) Pettit GS, Bates JE, Dodge KA, Meece DW. The impact of after
school peer contact on early adolescent externalizing problems is
moderated by parental monitoring, perceived neighborhood safety, and
prior adjustment. Child Dev 1999; 70:768-78.
(27.) Bartko WT, Eccles JS. Adolescent participation in structured
and unstructured activities: A person-oriented analysis. J Youth Adolesc
2003; 32:233-41.
(28.) Simantov E, Schoen C, Klein JD. Health-compromising
behaviors: Why do adolescents smoke or drink? Identifying underlying
risk and protective factors. Arch Pediatr Adolesc Med 2000; 154:1025-33.
(29.) Lambert D, Gale JA, Hartley D. Substance abuse by youth and
young adults in rural America. J Rural Health 2008; 24(3):221-28.
(30.) Sibley L, Weiner J. An evaluation of access to health care
services along the rural-urban continuum in Canada. BMC Health Serv Res
2011; 11(20).
(31.) Pong R, DesMeules M, Heng D, Lagace C, Guernsey J, Kazanjian
A, et al. Patterns of health services utilization in rural Canada.
Chronic Dis Can 2011; 31(supplement 1).
(32.) Firestone M, Fischer B. A qualitative exploration of
prescription opioid injection among street-based drug users in Toronto:
Behaviours, preferences and drug availability. Harm Reduct J2008; 5:30.
(33.) Simoni-Wastila L, Yang HK. Psychoactive drug abuse in older
adults. Am J Geriatr Pharmacother 2006; 4(4):380-94.
(34.) Ontario Ministry of Health and Long-Term Care. Ontario Drug
Benefit Program: Dispensing Fees. 2013. Available at:
http://www.health.gov.on.ca/
en/public/programs/drugs/programs/odb/opdp_dispensing_fees.aspx
(Accessed August 3, 2013).
(35.) Gage JC, Overpeck MD, Nansel TR, Kogan MD. Peer activity in
the evenings and participation in aggressive and problem behaviors. J
Adolesc Health 2005; 37(6):517.
Ariel Pulver, MSc, [1] Colleen Davison, PhD, [1-3] William Pickett,
PhD [1-3]
Author Affiliations
[1.] Department of Public Health Sciences, Queen's University,
Kingston, ON
[2.] Clinical Research Centre, Kingston General Hospital, Kingston,
ON
[3.] Department of Emergency Medicine, Queen's University,
Kingston General Hospital, Kingston, ON
Table 1. Characteristics of the 2009/2010 cycle of the
Canadian Health Behaviour in School-Aged
Children (HBSC) sample
Characteristic % (95% CI)
Age [less than or equal to] 14 35.0 (34.1-36.0)
15 45.9 (44.9-46.8)
[greater than or equal to] 16 19.1 (18.4-19.9)
Sex Boys 48.3 (47.3-49.2)
Girls 51.7 (50.8-52.7)
SES High 55.2 (54.2-56.1)
Average 35.7 (34.8-36.7)
Low 9.1 (8.6-9.7)
Immigrant Born in Canada 76.0 (75.2-76.9)
status Immigrant >5 yrs 19.7 (19.0-20.5)
Immigrant [less than or 4.3 (3.9-4.7)
equal to] 5 yrs
Geographic Urban 77.1 ( 76.2-77.9)
status Strong MIZ 0.7 (0.6-0.9)
Moderate MIZ 15.9 (5.2-16.6)
Weak or no MIZ 6.3 (5.8-6.8)
CI = confidence interval; SES = socio-economic status;
MIZ = Metropolitan Influenced Zone.
Table 2. Proportions and results from multiple Poisson
regression analysis* for recreational use of any type
of prescription medication, by demographic
characteristics, from the 2009/2010 Canadian
HBSC
Any medication RR (95% CI)
% (95% CI)
Overall 6.5 (6.0-7.0)
Age (years) [less than or 5.5 (4.6-6.4) 1.00
equal to] 14
15 6.7 (6.0-7.5) 1.23 (1.02-1.49)
[greater than or 7.6 (6.5-8.9) 1.41 (1.12-1.78)
equal to] 16
p trend <0.01
Sex Boys 6.1 (5.4-6.8) 1.00
Girls 6.8 (6.2-7.5) 1.13 (0.96-1.34)
SES High 5.5 (4.9-6.1) 1.00
Average 6.2 (5.5-7.1) 1.13 (0.95-1.32)
Low 13.0 (11.1-15.6) 2.41 (1.94-2.99)
p trend <0.01
Immigrant Born in Canada 6.6 (6.1-7.2) 1.00
status Immigrant >5 yrs 6.5 (4.0-8.8) 1.02 (0.83-1.24)
Immigrant 6.0 (5.4-7.6) 1.06 (0.69-1.61)
[less than or
equal to] 5 yrs
p trend 0.79
Geographic Urban 6.6 (6.1-7.2) 1.00
status Strong MIZ 14.2 (7.0- 24.0) 2.39 (1.03-5.55)
Moderate MIZ 5.4 (4.4- 6.7) 0.93 (0.66-1.29)
Weak or no MIZ 6.5 (4.7- 8.7) 1.01 (0.67-1.51)
p trend 0.79
* Model was adjusted for age, sex, SES, immigrant and geographic
status. CI = confidence interval; RR = relative risk; SES = socio-
economic status; MIZ = Metropolitan Influenced Zone.
Table 3. Proportions and results of Poisson regression analysis * for
any recreational use of pain relievers, stimulants and sedative
medications, by demographic characteristics, from the 2009/2010
Canadian HBSC
Pain relievers
% (95% CI) RR (95% CI)
Overall 5.1 (4.7- 5.5)
Age (years) [less than or 4.3 (3.7-5.1) 1.00
equal to] 14
15 5.5 (4.8-6.2) 1.37 (1.09-1.72)
[greater than or 5.5 (4.5-6.6) 1.49 (1.11-2.00)
equal to] 16
p trend <0.01
Sex Boys 4.6 (4.0-5.2) 1.00
Girls 5.5 (4.9-6.2) 1.25 (1.04-1.51)
SES High 4.2 (3.7-4.8)
Average 5.0 (4.3-5.8) 1.12 (0.91-1.36)
Low 10.0 (8.2-12.3) 2.32 (1.81-2.98)
p trend <0.01
Immigrant Born in Canada 5.1 (4.6-5.6) 1.00
status Immigrant >5 yrs 5.1 (3.5-8.2) 1.03 (0.82-1.30)
Immigrant 5.5 (4.2-6.1) 1.27 (0.81-1.98)
[less than or
equal to] 5 yrs
p trend 0.74
Geographic Urban 5.0 (4.6-5.6) 1.00
status Strong MIZ 14.2 (7.5- 24.7) 3.13 (1.23-8.01)
Moderate MIZ 4.5 (3.6-5.7) 1.03 (0.70-1.52)
Weak or no MIZ 5.7 (4.0-7.8) 1.16 (0.73-1.84)
p trend 0.61
Stimulants
% (95% CI) RR (95% CI)
Overall 2.4 (2.1-2.7)
Age (years) [less than or 2 (1.6-2.5) 1.00
equal to] 14
15 2.1 (1.7-2.5) 0.96 (0.69-1.34)
[greater than or 3.9 (3.1-4.9) 1.66 (1.17-2.36)
equal to] 16
p trend <0.01
Sex Boys 2.6 (2.1-3.1) 1.00
Girls 2.2 (1.9-2.7) 0.86 (0.66-1.12)
SES High 2.0 (1.6-2.5) 1.00
Average 2.2 (1.7-2.5) 1.19 (0.89-1.60)
Low 5.4 (3.1-4.9) 2.70 (1.91-3.81)
p trend <0.01
Immigrant Born in Canada 2.4 (2.1-2.8) 1.00
status Immigrant >5 yrs 2.4 (1.3-4.7) 0.90 (0.64-1.26)
Immigrant 2.4 (1.8-3.2) 1.14 (0.59-2.22)
[less than or
equal to] 5 yrs
p trend 0.74
Geographic Urban 2.5 (2.1-2.9) 1.00
status Strong MIZ 1.8 (0.2-8.4) 1.02 (0.14-7.39)
Moderate MIZ 2.1 (1.4-2.9) 1.07 (0.62-1.84)
Weak or no MIZ 2.2 (1.3-3.8) 1.00 (0.51-1.98)
p trend 0.99
Sedatives
% (95% CI) RR (95% CI)
Overall 1.4 (1.2-1.7)
Age (years) [less than or 1.3 (1.0-1.8) 1.00
equal to] 14
15 0.9 (0.6-1.2) 0.64 (0.41-0.99)
[greater than or 2.9 (2.1-3.7) 1.97 (1.28-3.04)
equal to] 16
p trend <0.01
Sex Boys 1.4 (1.1-1.8) 1.00
Girls 1.4 (1.1-1.8) 1.02 (0.72-1.46)
SES High 1.0 (0.8-1.3) 1.00
Average 1.5 (1.2-2.0) 1.52 (1.03-2.23)
Low 3.1 (2.1-4.5) 3.05 (1.92-4.88)
p trend <0.01
Immigrant Born in Canada 1.4 (1.1-1.7) 1.00
status Immigrant >5 yrs 1.4 (1.1-4.3) 1.09 (0.71-1.68)
Immigrant 2.2 (0.9-2.0) 1.71 (0.84-3.47)
[less than or
equal to] 5 yrs
p trend 0.39
Geographic Urban 1.5 (1.2-1.8) 1.00
status Strong MIZ 1.8 (0.1-8.4) 1.83 (0.24-14.19)
Moderate MIZ 1.1 (0.6-1.7) 0.73 (0.35-1.52)
Weak or no MIZ 1.5 (0.8-3.0) 1.12 (0.49-2.54)
p trend 0.68
* Model was adjusted for age, sex, SES, immigrant and geographic
status.
CI = confidence interval; RR = relative risk; SES = socio-economic
status; MIZ = Metropolitan Influenced Zone.
Table 4. Proportions and results of Poisson regression
analysis * for frequent recreational use of any
prescription drug, by demographic characteristics,
from the 2009/2010 Canadian HBSC
Frequent use of any
prescription drug
% (95% CI) RR (95% CI)
Overall 51.1 (44.9-52.8)
Age (years) [less than or 49.5 (42.3-56.7) 1.00
equal to] 14
15 44.3 (28.9-50.3) 0.79 (0.28-2.22)
[greater than 57.8 (61.2-81.3) 0.89 (0.68-1.16)
or equal to] 16
p trend 0.38
Sex Girls 43.4 (38.4-48.8) 0.77 (0.61-0.97)
Boys 55.6 (49.6-61.4) 1.00
SES High 43.8 (38.1-49.6) 1.00
Average 51.2 (44.6-58.1) 1.21 (0.94-1.57)
Low 56.1 (46.5-65.1) 1.34 (0.99-1.81)
p trend 0.04
Immigrant Born in Canada 50.6 (46.1-55.1) 1.00
status Immigrant >5yrs 40.2 (36.9-77.2) 0.77 (0.56-1.05)
Immigrant 56.7 (31.9-49.6) 1.01 (0.58-1.76)
[less than or
equal to] 5 yrs
p trend 0.13
Geographic Urban 48.7 (44.4-53.3) 1.00
status Strong MIZ 34.9 (13.7-72.6) 0.79 (0.28-2.23)
Moderate MIZ 54.6 (43.0-64.6) 1.06 (0.77-0.46)
Weak or no MIZ 42.1 (26.7-57.8) 0.88 (0.54-1.45)
p trend 0.83
* Model was adjusted for age, sex, SES, immigrant and geographic
status.
CI = confidence interval; RR = relative risk; SES = socio-economic
status;
MIZ = Metropolitan Influenced Zone
Table 5. Proportions and results of Poisson regression analysis* of
frequent recreational users of prescription drugs, by demographic
characteristics, from the 2009/2010 Canadian HBSC
Pain relievers
% (95% CI) RR (95% CI)
Overall 43.5 (39.1-47.8)
Age (years) [less than or 45.1 (37.0-53.3) 1.00
equal to] 14
15 41.5 (35.6-48.0) 0.92 (0.67-1.26)
[greater than or 45.9 (35.9-55.7) 1.00 (0.68-1.48)
equal to] 16
p trend 0.97
Sex Girls 39.9 (34.2-45.8) 0.80 (0.60-1.05)
Boys 48.3 (41.5-55.3) 1.00
SES High 39.5 (33.1-46.1) 1.00
Average 47 (39.7-54.8) 1.19 (0.88-1.61)
Low 46.9 (36.0-57.5) 1.23 (0.84-1.79)
p trend 0.20
Immigrant statBorn in Canada 46.3 (40.8-51.0) 1.00
Immigrant >5yrs 30.9 (28.8-71.2) 0.69 (0.47-1.03)
Immigrant 5 yrs 48.6 (22.6-41.5) 0.99 (0.53-1.86)
p trend 0.07
Geographic staUrban 43.1 (38.1-48.1) 1.00
Strong MIZ * 34.9 (13.7-72.6) 0.84 (0.29-2.39)
Moderate MIZ * 50.6 (38.1-61.9) 1.09 (0.74-1.59)
Weak or no MIZ * 36.6 (21.3-53.8) 0.89 (0.50-1.58)
p trend 0.94
Stimulants
% (95% CI) RR (95% CI)
Overall 57.6 (51.2-64.0)
Age (years) [less than or 60.1 (46.9-70.9) 1.00
equal to] 14
15 48.3 (37.7-58.3) 0.77 (0.49-1.19)
[greater than or 67.4 (55.6-77.7) 1.01 (0.66-1.55)
equal to] 16
p trend 0.51
Sex Girls 46.6 (37.8-56.4) 0.71 (0.49-0.99)
Boys 68.3 (59.4-76.6) 1.00
SES High 56.5 (47.1-66.2) 1.00
Average 55.6 (43.5-66.3) 0.98 (0.65-1.48)
Low 63.8 (47.3-75.7) 1.15 (0.74-1.80)
p trend 0.43
Immigrant statBorn in Canada 57.6 (50.2-64.9)
Immigrant >5yrs 53.5 (44.2-96.5)
Immigrant 5 yrs 77.6 (38.2-67.6)
p trend
Geographic staUrban 58.5 (51.1-65.4)
Strong MIZ * 100 (9.5-100.0)
Moderate MIZ * 57.8 (39.4-74.1)
Weak or no MIZ * 41.3 (18.8-70.4)
p trend
Sedatives
% (95% CI) RR (95% CI)
Overall 53.4 (45.0-62.0)
Age (years) [less than or 55.7 (40.2-69.5) 1.00
equal to] 14
15 52.4 (28.5-56.7) 0.95 (0.52-1.75)
[greater than or 52.0 (38.8-66.5) 0.72 (0.39-1.33)
equal to] 16
p trend 0.52
Sex Girls 49.2 (37.0-60.5) 0.83 (0.51-1.36)
Boys 58.0 (46.3-70.8) 1.00
SES High 52.6 (38.0-65.5) 1.00
Average 57.3 (42.4-69.9) 0.85 (0.47-1.53)
Low 45.7 (29.2-67.7) 0.85 (0.42-1.72)
p trend 0.73
Immigrant statBorn in Canada 51.5 (40.1-59.9)
Immigrant >5yrs 44.8 (62.9-1.00)
Immigrant 5 yrs 100 (26.0-64.4)
p trend
Geographic staUrban 53.3 (43.9-63.0)
Strong MIZ * 100 (9.5-100.0)
Moderate MIZ * 49.5 (23.9-71.5)
Weak or no MIZ * 55.2 (20.1-79.9)
p trend
* Model was adjusted for age, sex, SES, immigrant and geographic
status.
CI = confidence interval; RR = relative risk; SES = socio-economic
status; MIZ = Metropolitan Influenced Zone.
