Description  | Specific Indicators |  Corresponding Outcomes from the Ontario Public Health Standards |  Corresponding Indicators from Other Sources |  Data Sources  | Survey Questions | Alternative Data Source(s) | Method of Calculation |  Basic Categories |  Indicator Comments |  Cross-References to Other Indicators | Cited References |  Changes Made

Description

• The number of deaths in a population caused by cigarette smoking.^1,2

Specific Indicator

• Disease group-specific smoking attributable mortality
• Adult smoking attributable mortality
• All-cause smoking attributable mortality

• Passive-smoking attributable mortality
  

Corresponding Outcome(s) from the Ontario Public Health Standards (OPHS)

The Ontario Public Health Standards (OPHS) establish requirements for the fundamental public health programs and services carried out by boards of health, which include assessment and surveillance, health promotion and policy development, disease and injury prevention, and health protection. The OPHS consist of one Foundational Standard and 13 Program Standards that articulate broad societal goals that result from the activities undertaken by boards of health and many others, including community partners, non-governmental organizations, and governmental bodies. These results have been expressed in terms of two levels of outcomes: societal outcomes and board of health outcomes. Societal outcomes entail changes in health status, organizations, systems, norms, policies, environments, and practices and result from the work of many sectors of society, including boards of health, for the improvement of the overall health of the population. Board of health outcomes are the results of endeavours by boards of health and often focus on changes in awareness, knowledge, attitudes, skills, practices, environments, and policies. Boards of health are accountable for these outcomes. The standards also outline the requirements that boards of health must implement to achieve the stated results.

Link to OPHS:  http://www.ontario.ca/publichealthstandards

Outcomes Related to this Indicator

• Societal Outcome (Chronic Disease Prevention): There is increased adoption of behaviours and skills associated with reducing the risk of chronic diseases of public health importance.
• Societal Outcome (Chronic Disease Prevention): An increased proportion of the population lives, works, plays, and learns in healthy environments that contribute to chronic disease prevention
• Board of Health Outcome (Chronic Disease Prevention): Priority populations adopt tobacco-free living.

• The board of health shall conduct epidemiological analysis of surveillance data…in the area of comprehensive tobacco control (Chronic Disease Prevention).

• None

Data Sources

Numerator: Mortality Data
Original source: Ontario Office of Registrar General (ORG)
Distributed by: intelliHEALTH ONTARIO, Ontario MOHLTC
Suggested citation (see Data Citation Notes):
Ontario Mortality Data [years], intelliHEALTH ONTARIO, Ontario MOHLTC, Extracted: [date]

Numerator & Denominator: Canadian Community Health Survey (CCHS)
Original source: Statistics Canada
Distributed by: Ontario MOHLTC
Suggested citation (see Data Citation Notes):
Canadian Community Health Survey [Year], Statistics Canada, Share File, Ontario MOHLTC

Survey Questions 

The Canadian Community Health Survey contains a Smoking module (core content in 2000-01, 2003, 2005 and 2007-08).

The CCHS has a derived variable, type of smoker, which groups the type of smoker into six categories.
Please note that the derived variable definition for each category varies slightly from the National Advisory Group on Monitoring Evaluation’s (NAGME)¹¹ recommended indicator definitions in that the CCHS derived variable categories do not all include the 100 cigarette in lifetime criterion.

Below is a detailed description of how each of the CCHS type of smoker categories is derived:
Daily smoker = Smokes daily at present time (SMK_202=1)
Occasional smoker (former daily smoker) = Smokes occasionally at present time (SMK_202=2) AND has previously smoked cigarettes daily (SMK_05D=1)
Occasional smoker (never a daily smoker or has smoked less than 100 cigarettes) = Smokes occasionally at present time (SMK_202=2) AND has never smoked cigarettes daily (SMK_05D=2 or NA)
Former daily smoker (non-smoker now) = Does not smoke at present time (SMK-202=3) AND has previously smoked cigarettes daily (SMK_05D=1)
Former occasional smoker (at least one whole cigarette, non-smoker now) = Does not smoke at present time (SMK-202=3) AND [has never smoked cigarettes daily (SMK_05D=2) AND {has smoked 100 cigarettes in lifetime (SMK_01A=1) OR has smoked a whole cigarette (SMK_01B=1)}]
Never smoked (a whole cigarette) = Does not smoke at present time (SMK-202=3) AND has never smoked 100 cigarettes in lifetime (SMK_01A=2) AND has never smoked a whole cigarette (SMK_01B=2)

The Canadian Community Health Survey contains an Exposure to Second-Hand Smoke (ETS) module (core content in 2003, 2005 and 2007-08).

Alternative Data Source(s)

Rapid Risk Factor Surveillance System (RRFSS) contains a module called Tobacco - Use by Respondent (core module since 2001-2008) and a Tobacco-Exposure to Environmental Tobacco Smoke (ETS) module (optional content since 2004). The Tobacco-Exposure to Environmental Tobacco Smoke (ETS) module is only asked of former (t1=1 and t2=5) and non-smokers (t1=5).

• It is recommended that public health units use the Share File provided by the Ministry of Health and Long-Term Care rather than public use file (PUMF) provided by Statistics Canada. The Share File has a slightly smaller sample size because respondents must agree to share their information with the province to be included; however, the share file has more variables and fewer grouped categories within variables. The Share File is a cleaner dataset for Ontario analysis because all variables that were not common content, theme content or optional content for Ontario have been removed.
• There may be slight differences between results from the share file and data published on the Statistics Canada website for the Health Indicators. Rates calculated for Health Indicators use the master CCHS data file and include not stated respondents (refusals, don't knows and/or not stated respondents) in the denominator. 
• Not applicable respondents should be excluded; however, it is important to understand who these respondents are from the questionnaire skip patterns to be able to describe the relevant population. 
• Users need to consider whether or not to exclude the ‘Refusal, 'Don't Know' and ‘Not Stated' response categories in the denominator. Rates published in most reports, including Statistics Canada's publication Health Reports generally exclude these response categories.  In removing not stated responses from the denominator, the assumption is that the missing values are random, and this is not always the case.  This is particularly important when the proportion in these response categories is high. 
• Estimates must be appropriately weighted (generally the share weight for the CCHS) and rounded.
• Users of the CCHS Ontario Share File must adhere to Statistics Canada's release guidelines for the CCHS data when publishing or releasing data derived from the file in any form. Refer to the appropriate user guide for guidelines for tabulation, analysis and release of data from the CCHS.  In general, when calculating the CV from the share file using the bootstrap weights, users should not use or release weighted estimates when the unweighted cell count is below 10.  For ratios or proportions, this rule should be applied to the numerator of the ratio.  Statistics Canada uses this approach for the tabular data on their website.  When using only the Approximate Sampling Variability (CV) lookup tables for the share file, data may not be released when the unweighted cell count is below 30.  This rule should be applied to the numerator for ratios or proportions.  This provides a margin of safety in terms of data quality, given the CV being utilized is only approximate. 
• Before releasing and/or publishing data, users should determine the CV of the rounded weighted estimate and follow the guidelines below:  
  • Acceptable (CV of 0.0 - 16.5) Estimates can be considered for general unrestricted release. Requires no special notation.
  • Marginal (CV of 16.6 - 33.3) Estimates can be considered for general unrestricted release but should be accompanied by a warning cautioning subsequent users of the high sampling variability associated with the estimates. Such estimates should be identified by the letter E (or in some other similar fashion).
  • Unacceptable (CV greater than 33.3) Statistics Canada recommends not to release estimates of unacceptable quality. However, if the user chooses to do so then estimates should be flagged with the letter F (or in some other fashion) and the following warning should accompany the estimates: "The user is advised that...(specify the data)...do not meet Statistics Canada's quality standards for this statistical program. Conclusions based on these data will be unreliable and most likely invalid". These data and any consequent findings should not be published. If the user chooses to publish these data or findings, then this disclaimer must be published with the data. 
• Caution should be taken when comparing the results from 2000/01 (Cycle 1.1) to subsequent years of the survey, due to a change in the mode of data collection.  The sample in Cycle 1.1 had a higher proportion of respondents interviewed in person, which affected the comparability of some key health indicators.  Please refer to http://www.statcan.gc.ca/imdb-bmdi/document/3226_D16_T9_V1-eng.pdf for a full text copy of the Statistics Canada article entitled "Mode effects in the Canadian Community Health Survey: a Comparison of CAPI and CATI". 

Mortality

• Use Deaths data source from the Vital Statistics folder in IntelliHEALTH, select # ON Deaths measure (number of deaths for Ontario residents who died in Ontario). Note: deaths for Ontario residents who died outside the province are not captured in Vital Statistics.
• A report has been created in the Shared FoldersPHU folder in IntelliHEALTH that provides the SAM cause of mortality data by prompted calendar year and public health unit, as well as for Ontario.
• Select appropriate geography from Deceased Information folder (public health unit or LHIN). Include other items, depending on your requirements (ICD10 Chapter, Lead Cause Group, age group, sex, etc.). 

Smoking Attributable Fraction (SAF):

SAFs for each disease (listed in the above table) and sex are derived from the following formula:

SAF = [(p0 + p1(RR1) + p2(RR2)) - 1] / [p0 + p1(RR1) + p2(RR2)]

The product of the smoking population attributable fraction and the number of deaths in the population yields an attributable mortality count (smoking-attributable mortality or SAM).²

SAM = Number of deaths X SAF 

Calculate SAM for each disease group (listed in Table), sex and age groups (35-64, 65+ for adult SAM).
The number of all-cause SAM is the sum of SAM from all disease categories.³
To assist with calculations, use the Adult SAM Calculation Examples spreadsheet.

Passive smoking attributable mortality (PSAM):  calculated for deaths associated with ischemic heart disease and lung cancer. ^4,5

Population Attributable Risk (PAR) = p(IDR - 1)/p(IDR - 1) + 1 ⁶

Relative Risks

  

Passive Smoking Incidence Density Ratios (IDR)/Relative Risk
accessed:  June 9, 2009
Disease Category	Both Sexes
Lung Cancer	1.21
Ischemic Heart Disease	1.24
Reference:  Baliunas et. al. 20075, De Groh et. al. 20026

Basic Categories

• Adult SAM age groups: 5 year age groups, 35 to 85+ for mortality data;  35-64 and 65+ for smoking prevalence data
• Sex: male, female and total.

• ICD-10 codes in table

• Geographic areas: public health unit, LHIN, province. 

Indicator Comments

Smoking-Attributable Mortality, Morbidity and Economic Costs (SAMMEC) software

• The SAM method from the Centers for Disease Control and Prevention’s (CDC) SAMMEC software was used to calculate Adult SAM. 

• The SAMMEC application is designed to estimate the overall disease impact of smoking in a population. It is not to be used as a surveillance tool.2

• The SAFs used in SAMMEC are derived by using relative risks (RRs) for each cause of death from the American Cancer Society's Cancer Prevention Study-II (CPS-II;1982--1988) and current and former cigarette smoking prevalence for two age cohorts: persons aged 35-64 years and persons aged >65 years.  For ischemic heart disease and cerebrovascular disease mortality, RR estimates were stratified by age (35–64 years and >65 years). ²

• The SAMMEC software was revised on the basis of findings from the 2004 Surgeon General's report on diseases caused by smoking. The list of smoking-attributable diseases now includes stomach cancer and acute myeloid leukemia and excludes hypertension.⁷

• CDC does not recommend the use of SAMMEC for analyses of populations of fewer than a few hundred thousand. The statistical validity of the results will be in question. ²

• SAMMEC users can generate their own Adult estimates for their health jurisdiction by visiting the SAMMEC website and changing or editing each data element directly from the menu.  Register for SAMMEC by visiting the CDC’s website at http://apps.nccd.cdc.gov/sammec/index.asp.

• The CPS-II has been criticized for not being generalizable to the entire US population.  Participants in CPS-II tend to overrepresent the middle class, have more education, and a disproportionate number of them are white. ^4,5,8,9

• The CPS II relative risks used a definition of at least one cigarette per day for at least 1 year for former smokers.  Thus, the CCHS former smoker definition probably overestimates the population attributable fraction in the Canadian population since CPS II required greater exposure, thereby leading to higher relative risk estimates.¹
   

SAM Calculation Limitations and Considerations

• Different definitions of current and former smoker can yield different SAM estimates.¹
• The attributable-fraction methodology calculates smoking-attributable deaths using smoking prevalence and number of deaths for the current year. However, most smoking-attributable deaths are the result of smoking in previous decades, during which smoking rates were higher. During periods where smoking prevalence is declining, the attributable-fraction (AF) methodology will tend to underestimate the number of deaths caused by smoking. Conversely, when smoking prevalence is increasing, the AF formula may overestimate the number of deaths caused by smoking.²
• Issues associated with exposure measurement (smoking definitions) and categorization error (exposure, risk and outcome) may affect some local public health units more than others. According to the Tanuseputro et al. study¹, some health units had higher than expected SAM estimates which likely arose because lung cancer rates were high, despite current low smoking prevalence. Others had unusually high prevalence of smoking in the elderly for one survey time period, but not others. It is advised that, when analyzing local SAM estimates, people investigate their estimates more closely using the methods outlined in the study, if there are general concerns that the SAM estimates appear different from what was expected.
• Use closely matching definitions of exposure (current/former smoker) in the study (source of relative risk estimates) and target populations (source of risk exposure estimate, i.e., CCHS)^1,4  
• Given the availability of different definitions of exposures - smoking and environmental tobacco exposure, it is recommended that sensitivity analyses be undertaken to understand the impacts of these differences.¹
• Use age and sex-specific prevalence, mortality, and relative risk estimates when available.  For Canada, the prevalence of smoking and the smoking attributable fraction decrease with increasing age.  The use of age-pooled prevalence estimates then underestimates smoking attributable fraction for younger age groups and overestimates it for older age groups.  Since the number of deaths is consistently greater for older age groups, the degree of overestimation in SAM for older persons will be greater in absolute terms than the degree of underestimation for younger persons.¹
• SAM calculations do not account for interrelationship between risk factors i.e., alcohol use (confounding)²
• Fire deaths caused by smoking are not reflected in the smoking-attributable mortality estimates.
• SAM estimates understate deaths attributable to all tobacco use because estimates of deaths attributable to cigar smoking, pipe smoking, and smokeless tobacco use are not included.²
• A series of years of mortality data (5 years) should be used when calculating SAM to avoid, if possible, concerns about small numbers and year-to-year variation.
• It is recommended to calculate age-adjusted adult SAM rates. Since risk varies by demographic characteristics, such as age, populations with older age distributions will tend to have higher crude death rates. As a result, when one compares crude rates across groups (e.g., health unit) one cannot be sure whether apparent differences are due to differences in the age distribution of the groups or to actual differences in mortality risk. Age-adjustment is a statistical technique designed to reduce differences in crude rates that result from differences in populations' age distribution.²
• Smoking attributable mortality, maternal smoking can also be calculated.  However, it has not been included in this indicator.  Numbers may be too small at the local level to produce valid maternal smoking estimates.  Also, further investigation is required to determine the most appropriate data source (survey questions from CCHS vs. other data sources) to use for the calculation of maternal smoking estimates.

Smoking Status Estimates and Health Effects

• Surveys are generally thought to underestimate smoking rates because smokers may be reluctant to admit they smoke or they may be unable to accurately report the regularity of their smoking habit.
• Smoker is defined differently in the CCHS and RRFSS. In order to be considered a smoker in RRFSS, the respondent must have smoked at least 100 cigarettes in their lifetime. This is not a requirement in the CCHS.
• There is evidence that smoking tobacco is related to more than two dozen diseases and conditions. It is the leading cause of preventable death and has negative health impacts on people of all ages: unborn babies, infants, children, adolescents, adults, and seniors.¹⁰
• Smoking tobacco is the most important preventable cause of lung cancer, accounting for 85% of all new lung cancer cases in Canada. Smoking tobacco can lead to respiratory and upper digestive tract cancers and has been shown to be a contributing cause of leukemia and cancers of the bladder, stomach, kidney and pancreas. Female smokers have an increased risk for developing cervical cancer. Smoking tobacco is associated with Chronic Obstructive Pulmonary Disease (COPD) and cardiovascular diseases caused by smoking include coronary heart diseases, peripheral vascular disease and cerebrovascular diseases.¹⁰ There is evidence that people repeatedly exposed to environmental tobacco smoke are more likely to develop and die from heart problems, lung cancer, and breathing problems. Environmental tobacco smoke can also cause chest infections, ear infections, excessive coughing and throat irritation.¹⁰
  

Definitions

• Smoking attributable fraction (SAF) – the fraction of the disease in the population that would not have occurred if the effects associated with smoking were absent.^4,5
• Passive smoking attributable mortality (PSAM) – applying age- and sex-specific relative risk and rates of mortality from lung cancer and ischemic heart disease (IHD) to the  population who have never smoked but who are exposed to environmental tobacco smoke (ETS) from spouses and other sources.⁵
• Current smoker – daily smoker + occasional smoker
• Former smoker – smoked daily or occasionally before but currently does not smoke
• Daily smoker – smoking at least one cigarette per day
• Occasional smoker – does not have at least one cigarette per day
  

Cross-References to Other Indicators

• Smoking Status (Section 5A: Behaviour and Health)
• Non-Smoker Second-Hand Smoke Exposure (Section 5A: Behaviour and Health)
• Chronic Disease Mortality (Section 4A: Chronic Diseases)

1. Tanuseptutro, P. Manuel, DG. Schultz, SE. Johansen, H. Mustard, CA. Improving Population Atributable Fraction Methods :  Examining Smoking-attributable Mortality for 87 Geographic Regions in Canada. American Journal of Epidemiology Vol 161, No. 8. 2005.  pp. 787-798. Available online at: http://aje.oxfordjournals.org/cgi/content/full/161/8/787 (Accessed October 16, 2008).
2. Centers for Disease Control and Prevention. Smoking-Attributable Mortality, Morbidity, and Economic Costs (SAMMEC): Adult SAMMEC and Maternal and Child Health (MCH) SAMMEC software, 2007. Available at https://apps.nccd.cdc.gov/sammec/index.asp. 
3. Tanuseputro, P. Manuel, DG. Leung, M. Nguyen, K. Johansen, H. Risk factors for cardiovascular disease in Canada. Canadian Journal of Cardiology. Vol. 19, No. 11. October 2003. pp. 1249-1259 
4. U.S. Surgeon General. The Impact of Smoking on Disease and the Benefits of Smoking Reduction. The Health Consequences of Smoking. Chapter 7. 2004. pp. 853-893. Available online at: http://www.cdc.gov/tobacco/data_statistics/sgr/2004/pdfs/chapter7.pdf (Accessed October 14, 2008).
5. Baliunas, D. Patra, J. Rehm, J. Popova, S. Kaiserman, M. Taylor, B. Smoking-attributable mortality and expected years of life lost in Canada 2002 : Conclusions for prevention and policy. Chronic Diseases in Canada, Vol. 27, No. 2007 pp. 154-162. Available online at: http://www.phac-aspc.gc.ca/publicat/cdic-mcc/27-4/pdf/cdic274-3_e.pdf (Accessed October 14, 2008).
6. De Groh, M. Morrison, HI. Environmental tobacco smoke and deaths from coronary heart disease in Canada. Chronic Diseases in Canada. Vol. 23, No. 1. 2002. Available online at: http://www.phac-aspc.gc.ca/publicat/cdic-mcc/23-1/b_e.html (Accessed March 30, 2009
7. Centers for Disease Control and Prevention. Annual Smoking-Attributable Mortality, Years of Potential Life Lost, and Productivity Losses – United States, 1997-2001. MMWR, July 1, 2005 / 54(25);625-628. Available online at: http://www.cdc.gov/mmwr/preview/mmwrhtml/mm5425a1.htm (Accessed June 8, 2009).
8. Sterling, TD. Rosenbaum, Wl Weindam, JJ. Risk attribution and tobacco-related deaths.  American Journal of Epidemiology 1993; 138:128-139.
9. Malarcher, AM. Schulman, J. Epstein, LA. Thun, MJ. Mowery, P. Pierce, B et al. Methodological issues in estimating smoking-attributable mortality in the United States.  American Journal of Epidemiology 2000; 152:573-584.
10. Health Canada. Smoking and your body. Available online at: http://www.hc-sc.gc.ca/hc-ps/tobac-tabac/body-corps/index-eng.php (Accessed October 16, 2008).
11. Copley TT, Lovato C, O’Connor S. Canadian Tobacco Control Research Initiative. Indicators for Monitoring Tobacco Control: A Resource for Decision-Makers, Evaluators and Researchers. Toronto, ON: Canadian Tobacco Control Research Initiative on behalf of the National Advisory Group on Monitoring and Evaluation, 2006. Available online at: http://ctcri.ca/en/index.php?option=com_docman&task=doc_download&gid=8 (Accessed 2007).

• Flanagan, W. Boswell-Purdy, J. Le Petit, C. Berthelot, JM. Estimating summary measures of health: a structured workbook approach. Population Health Metrics 11 May 2005, 3:5 Available online at: http://www.pophealthmetrics.com/content/3/1/5 (Accessed March 30, 2009).
• He, J. Vupputuri, S. Allen, K. Prerost, M. Hughes, J. Whelton, P.K. Passive Smoking and the Risk of Coronary Heart Disease – A Meta-Analysis of Epidemiologic Studies. The New England Journal of Medicine 25 March 1999, 340:12 Available online at http://www.nejm.org/ (Accessed June 9, 2009).
• U.S. Department of Health and Human Services. The Health Consequences of Involuntary Exposure to Tobacco Smoke. A Report of the Surgeon General 2006. Available online at http://www.surgeongeneral.gov/library/secondhandsmoke/  (Accessed June 9, 2009).
• Single, E. Rehm, J. Robson, L. Truong, M.V. The Relative Risks and Etiologic Fractions of Different Causes of Death and Disease Attributable to Alcohol, Tobacco and Illicit Drug Use in Canada. Canadian Medical Association Journal 13 June 2000, 162:12. Available online at http://www.cmaj.ca/cgi/content/abstract/162/12/1669 (Accessed October 2008).
• Rehm, J. Baliunas, D. Brochu, S. Fischer, B. Gnam, W. Patra, J. et al. The Costs of Substance Abuse in Canada 2002. March 2006. Available online at http://www.risqtoxico.ca/documents/2006_Brochu_ReportCost.pdf (Accessed April 2011).

Changes Made