• Tax Free Savings Accounts at a Glance : Linkage of Tax Free Savings Accounts to Personal Income Tax Files
• Economic Impact of the Industrial Regional Benefits (IRB) Procurement Program (069-2012)
• Amendment to: Long-Term Income and Employment among Childhood, Adolescent and Young
  Adult Cancer Survivors: Linkage of the British Columbia Cancer Registry and the T1 Family File.
• Workers' Compensation Benefits in Ontario and British Columbia
• Understanding risk factors associated with hospital utilization and mortality – Canadian Community Health Survey to Hospitalization and Mortality Data Linkage
• Air Pollution Study: Linkage of 1991 Census of Population, Canadian Mortality Database and Canadian Cancer Database Follow-Up Study
• Expanding the Longitudinal Administrative Databank to Include Tax Free Savings Account Information
• Health outcomes of adolescents and young adults in relation to childhood exposures and experiences
• Industry Canada: Economic Impact of Venture Capital, 1999 to 2010
• Community Futures Program’s Regional Economic Contribution: Linkage of Client List to Business Tax and Employment Data, 1999 to 2011
• Atlantic Canada Opportunities Agency – Update of Business Performance Evaluation Report
• Linkable File Environment
• The Burden of Obesity, Osteoarthritis and Rheumatoid Arthritis in Ontario – Addition of Data Years and Extension of Retention Period
• Socioeconomic Influences on Use of Physicians in Ontario: Linkage of Canadian Community Health Survey and Ontario Medical Services Data – Addition of Data Years and Extension of Retention Period
• The Role of Technological Innovation in the Increased Concentration of Health Care Expenditure on the Elderly in Ontario – Addition of Data Years and Extension of Retention Period
• Study of Mortality among North American Women in the Synthetic Rubber Industry, 1950-2002 – Extension of Retention Period
• Sherritt International Mortality Study: 1954-2003 Update – Extension of Retention Period
• Perinatal Outcomes Study: Live Birth, Infant Death and Stillbirth Records Linked to 20% Sample Data from the 1996 and 2006 Censuses of Population
• Maternal Mortality and Infant Outcomes in Canada, 2001 to 2008
• Workers in the synthetic rubber industry: mortality follow-up 1950 to 2009
• Ongoing Longitudinal Linkage of Child and Spousal Support Cases in the Survey of Maintenance Enforcement Programs
• Linkage of the Survey of Maintenance Enforcement Programs to the T1 Family File
• Inter-provincial Workers: Linkage of Administrative Data to Examine the Number and Characteristics of Individuals Who Reside in One Province But Work in Another
• Evaluation of Blood Transfusion on Mortality Outcomes of Cardiovascular Patients Following Hip Fracture Repair: Linkage to Canadian Mortality Database, 2004 to 2009
• Linkage of the Census of Population 2006 to the Discharge Abstract Database and the Canadian Mortality Database for Purposes of the Longitudinal Health and Administrative Data Initiative
• Linkage of the Canadian Cancer Registry to the Discharge Abstract Database and the National Ambulatory Care Reporting System for Purposes of the Longitudinal Health and Administrative Data Initiative
• Canadian Perinatal Surveillance System: Linkage of Birth and Death Records for Infant Mortality Surveillance, Births 2006 to 2010
• Longitudinal Perspectives on Employment, Income and Health: Linkage of the Longitudinal Worker File, 1991 Census, Canadian Mortality Database and Canadian Cancer Database
• Estimating Distributions that Characterize the Natural History of Influenza Infections
• Study of the incomes of the affluent: The role of private corporations
• Re-contact with the Justice system
• Linkable File Environment
• Community Futures Program's Regional Economic Contribution: Linkage of Client List to Business Tax and Employment Data, 1999 to 2011
• Validation of frailty index
• Expansion of Household Survey Frame Service

---

Tax Free Savings Accounts at a Glance : Linkage of Tax Free Savings Accounts to Personal Income Tax Files

Purpose: With an aging population and recent developments in financial markets, the Canadian retirement income system continues to be under intense scrutiny. Pension coverage and contributions to RRSPs are indicators of Canadians preparedness for retirement. Only recently has the topic of Tax Free Savings Accounts (TFSA) entered the retirement picture. Are TFSAs being utilized and ultimately maximized by Canadians? The state of health of the Canadian retirement system has implications not only for the economic security of individuals and families as the population ages, but also for the overall stability of the Canadian financial system.

Statistics Canada has significant information on pensions and RRSP contributions, however, information on Tax Free Savings Accounts is more scarce. Income Statistics Division has recently received the TFSA files from CRA. This information is a key component of the current retirement income system and data users and public policy analysts are requesting this information. Primary external users of retirement statistics include the federal and provincial pension supervisory authorities, federal government departments such as Human Resources and Skills Development Canada (HRSDC) and Canada Revenue Agency (CRA), the Office of the Chief Actuary, the Office of the Superintendent of Financial Services (OSFI), as well as private consulting firms, insurance companies and the research community. Requests for statistics on this topic have increased across the board.

In order to fill the information gaps identified, Policy Committee is asked to approve a record linkage in which information received from the TFSA files can be linked to personal income tax dataT1 Family Files on an annual, on-going basis.

In addition to addressing this and other key research questions, the linkage will improve the quality of the Canadian retirement income system and enhance coherence between data sources.

Description: The TFSA account files and the T1 Family File (T1FF) will be linked on an on-going annual basis (provided funding is available for future years), starting with reference year 2009. The TFSA files contain information on the TFSA accounts, age of individual holder, contributions and withdrawals etc. while the T1FF contains characteristics (e.g. income, age, sex, etc) of individual tax-filers and their families. Both files contain Social Insurance Numbers (SIN), which are essential to carry out the linkage. The linkage will commence with data from the 2009 tax year and be repeated annually on an indefinite basis provided future funding is available to purchase the TFSA files from CRA.

Output: Only aggregate statistics and analysis conforming to the confidentiality provisions of the Statistics Act will be released outside of Statistics Canada. The linked TFSA analysis file and linking key file will be retained indefinitely by Statistics Canada, or until no longer required, at which time they will be destroyed. Access to these files will be restricted to Statistics Canada employees whose assigned work activities require access.

Economic Impact of the Industrial Regional Benefits (IRB) Procurement Program (069-2012)

Purpose: This project contributes to the evaluation of the IRB Procurement Program by producing statistics that objectively compare the economic performance of IRB funded firms with similar firms that have not received funding, and the economic performance of firms receiving IRB funding during the time of the funding to the periods immediately preceding and following the receipt of IRB funds. The performance of firms will be measured with many indicators of development and profitability such as: survival rate (various periods), revenue and profitability, R&D investment, growth in equity and employment, etc.

The information produced above will be used by Industry Canada to evaluate the effectiveness of the IRB Procurement Program, and in a broader context, help Industry Canada determine whether the IRB Procurement Program is consistent with its approach of developing and implementing policies, initiatives, and services to encourage innovation, international competitiveness, and economic growth. In summary, the project aims to inform the development of public policy.

Description: A list of firms that received IRB funding in the 2000 to 2010 period will be linked to data from the Business Register, T2 corporate tax, and Export Register. The records will be linked deterministically using statistical enterprise numbers. This is a one-time linkage.

Output: Only non-confidential aggregate statistical outputs and analyses that conform to the confidentiality provisions of the Statistics Act will be released outside of Statistics Canada. The information will be presented in the form of tables of regression results and summary statistics related to the projects's goal of ascertaining the impact of IRB funding.

The linked file will be retained until December 2017. All direct business identifiers will be removed from the analysis file once linkage is complete, and placed in a separate linkage key file. The linkage key file will be retained until December 2017, or sooner if no longer required, at which time it will be destroyed.

---

Amendment to: Long-Term Income and Employment among Childhood, Adolescent and Young
Adult Cancer Survivors: Linkage of the British Columbia Cancer Registry and the T1 Family File.

Purpose: To assess the long-term income and employment experience of survivors of cancer diagnosed during childhood, adolescence or young adulthood. This information will be linked to the existing approved analysis file.

Description: This is a request from the British Columbia Cancer Agency for an amendment to an existing linkage agreement of the T1 Family File (T1FF) with a subset of the BC Cancer Registry. This amendment seeks first to update the number of years of T1FF data originally requested (1982 to 2007), so as to include more recent years of data, specifically the years 2008 to 2010. Secondly, this amendment requests permission to obtain the industrial sector where an individual worked – BusinessRootNumber and NAICS/SIC – for the period 1982 to 1997, by linking Business Register and T4 information to the T1FF.

Output: All access to the linked microdata file will be restricted to Statistics Canada staff whose work activities require access. Only aggregate data conforming to the confidentiality provisions of the Statistics Act will be released outside of Statistics Canada. Statistics Canada will retain the linked file, stripped of identifiers, until it is determined that there is no further need for it.

---

Workers' Compensation Benefits in Ontario and British Columbia

Purpose: To compare the outcomes of two different designs of workers' compensation programs for work-related injuries, namely those of Ontario and British Columbia. Findings from the study will be used to evaluate the effectiveness of the programs in reducing the adverse financial consequences of wage-loss on Canadians who suffer work-related injuries, and their families. Because workers' compensation programs for injured workers share similarities across provinces, the results of the study will be relevant nation-wide.

The study has two components. Part 1 will determine whether there are significant differences in labour market re-entry and post-injury earnings for individuals sustaining permanent impairments and receiving benefits from the different programs. Part 2 will examine individuals experiencing a temporary disability arising from a work-related injury to determine if there are long-term labour-market earnings losses resulting from these temporary disabilities, and which factors impact on long-term earnings. In both parts of the study, the earnings dynamics at both the family and individual level will be investigated, and regional as well as gender differences will be explored.

Description: The project involves a one-time linkage of administrative data on short- and long-term disability beneficiaries from the Ontario Workplace Safety & Insurance Board and the British Columbia Workers' Compensation Board, for selected injury years between 1986 and 2002, linked to the 20% Longitudinal Administrative Databank (LAD) for selected years in the 1982 to 2003 period. The files will be linked deterministically using Social Insurance Number, but the number will not be stored on the linked file.

Output: Only aggregate data that conform to the confidentiality provisions of the Statistics Act will be released outside of Statistics Canada. The availability of the analysis files will be announced in The Daily. Findings will be disseminated in Statistics Canada publications, to the Ontario Workplace Safety and Insurance Board, the British Columbia Workers' Compensation Board, the Institute for Work and Health, the U.S. National Institutes of Safety and Health, and in peer-reviewed international scientific journals. The linked files, stripped of identifiers, will be retained until May 2011, after which they will be destroyed.

---

Understanding risk factors associated with hospital utilization and mortality – Canadian Community Health Survey to Hospitalization and Mortality Data Linkage

Purpose: The purpose of this linkage project is to better understand and quantify the association between risk factors (behavioural, socio-economic, and environmental), hospital utilization, and health outcomes at the individual and population level. Behavioural risk factors such as smoking, stress and obesity and socioeconomic factors such as low income and education and long-term exposure to air pollution have been shown to be associated with poorer health outcomes. However, there is currently less information regarding the direct impact of these risk factors on the use of hospital services and related outcomes as well as associated costs.

Specifically, the linked data will be used to address the following research objectives: to understand the association between behavioural risk factors and the use of hospital services and related outcomes; to understand the interaction between socio-economic and behavioural risk factors and their effect on the use of hospital services and related outcomes; to understand the extent to which differences in the prevalence of risk factors in Canada explains the variation in the use of hospital services and to examine the interaction between risk factors, ambient air pollution exposures, mortality, and the use of hospital services.

Description: Our ability to understand these relationships is currently limited due to data gaps. Administrative health data provide comprehensive information regarding the use of hospital services but provide limited information on the characteristics of individual patients. Conversely, health survey data provide comprehensive risk factor information but contain limited information regarding the use of healthcare services and mortality outcomes. This data linkage project will bring together hospital administrative data, vital statistics, and health survey data to fill this data gap.

This project will link the records of respondents to four cycles of the Canadian Community Health Survey (CCHS) (2000/2001 to 2011/2012) to the following databases: Discharge Abstract Database (DAD), 1996/1997 to 2015/2016; Canadian Mortality Database (CMDB), 2000 to 2015; and Historical Tax Summary File (HTSF), 1990 to 2015. Linkages would occur only for those CCHS respondents who have given consent to link information to their survey data. The CCHS provides comprehensive information regarding the behavioural risk (e.g. smoking, alcohol consumption, obesity, and diet) and socio-economic status (e.g. income, education) of respondents. The DAD provides comprehensive information regarding the use hospital services including diagnosis, treatment, and use of resources which can be used to derive costs. The CMBD will provide information regarding mortality outcomes, primary cause of death, and allow calculation of loss to follow-up. The HTSF will be used to assist in record linkage and no tax data will be retained on the linked file. This study will also make use of the air pollution exposure data assembled for the previously-approved Air Pollution, Mortality and Cancer Study. The final analysis file will not contain direct personal identifiers such as names, health information numbers or death registration numbers or tax information.

Output: The linked files will at all times remain on Statistics Canada premises. Linked data files will be made available through the Research Data Centres (RDCs) as per the Subsequent Use of Linked Data provision in the Directive on Record Linkage. Requests for access to the linked data will be conducted in accordance with established RDC application processes and guidelines. Only aggregate data that conform to the confidentiality provisions of the Statistics Act will be released outside of Statistics Canada. Existence of the linked file will be announced in the Statistics Canada Daily. Major findings will be used to prepare research papers for publication in peer-reviewed journals (including Statistics Canada's Health Reports) and presentation at workshops and conferences.

The linked analysis files, stripped of direct personal identifiers, will be retained until December 31, 2020, or, at which time their continued retention will be reviewed. The corresponding linkage key files housed in the Statistics Canada Privacy Office will also be retained until December 31 2020, at which time their continued retention will be reviewed.

---

Air Pollution Study: Linkage of 1991 Census of Population, Canadian Mortality Database and Canadian Cancer Database Follow-Up Study

Purpose: To assess the impact of long-term exposure to air pollution on human health, with the objective to inform the development of Canada-wide standards for key criteria pollutants. Linkage of separate sources of information is an important way in which Statistics Canada can meet identified data gaps on environmental data related to human exposure to air pollution. For example, Canada-wide standards for annual averages of either fine particulate matter or ozone have not been developed, largely due to lack of evidence from the Canadian population and uncertainties about the applicability of risk estimates generated in other countries to Canada.

The specific objectives of this study are: to determine whether deaths from all causes, from ischaemic heart disease, from cardiopulmonary disease, from respiratory cancer, and from all cancers combined are associated with long-term exposure to ambient air pollutants; to determine the air pollution risks for cancer incidence and the risks for specific cancer types; and, to examine the relationship of cancer incidence and causes of death to socio-demographic and neighbourhood characteristics over a 22-year period.

Description: For a previous approved record linkage (reference number 012-2001), a sample of 2.7 million Canadians was selected from respondents to the 1991 Census of Population long-form questionnaires and their Census information was linked to the 1991 Health and Activity Limitations Survey, the 1990 and 1991 Tax Summary Files and the 1991 to 2001 Canadian Mortality Database, for the development of indicators on health.

The current project will extend and expand the linked information on this 1991 Census sample, as follows:

• linkage to an additional 27 years of the Tax Summary Files, that is, from 1984 to 2012;
• linkage to an additional 10 years of the Canadian Mortality Database, up to 2011; and
• a new linkage to the 1969 to 2011 Canadian Cancer Database, for the period 1969 to 2011.

The linked files will contain only those data items required to conduct the study. Personal identifiers, such as name and social insurance number, will be used only for linkage purposes, then removed from the linked microdata file. Only a sample of individuals who completed the 1991 Census of Population long-form questionnaires are included on the file.

Output: All access to the linked microdata file will be restricted to Statistics Canada staff whose work activities require access. Only aggregate data that conform to the confidentiality provisions of the Statistics Act will be released outside of Statistics Canada. Availability of the linked file will be announced in The Daily. Major findings will be used to create research papers for publication in peer-reviewed journals (including Statistics Canada's Health Reports) and presentation at workshops and conferences.

The linked file, stripped of personal identifiers, will be retained until no longer required, at which time the file will be destroyed.

---

Expanding the Longitudinal Administrative Databank to Include Tax Free Savings Account Information

Purpose: The debut of the Tax Free Savings Account (TFSA) in 2009, introduced a new program that is separate from and is not captured by the T1 tax file, upon which the Longitudinal Administrative Databank (LAD) is based. The TFSA program may have an effect on the measurement of income, particularly when focusing on the retired population, as well as more generally on the tax and transfer system. The proposed addition of TFSA information to the LAD will permit Statistics Canada to improve its data reporting on income, and thereby assist our clients with their research in this area.

Description: The project involves linking on a continuing basis Statistics Canada's LAD file to the TFSA file from Canada Revenue Agency (CRA). The LAD and TFSA will be linked deterministically using the Social Insurance Number (SIN) of individuals. Following completion of the linkage, the SIN and other identifiers will then be removed and kept securely separate from the final linked file. No names or addresses will be used at any point in the linkage process or be on any of the files used. The linkage will be performed by Statistics Canada personnel.

Output: The output will be a linked database file combining LAD and TFSA information. Only aggregate data and analysis conforming to the confidentiality provisions of the Statistics Act will be released outside of Statistics Canada. The linked file will be retained at Statistics Canada.

---

Health outcomes of adolescents and young adults in relation to childhood exposures and experiences

Purpose: The purpose of this linkage project is to examine whether common adversities in childhood are associated with adverse health outcomes later in life. The specific outcomes targeted by this project are among the leading causes of disease burden in Canada: depression, substance misuse and mental distress (including anxiety). These issues have a substantial impact on quality of life and functioning for a large proportion of the general public.

Description: There was an integrated sample of children who were part of Cycle 1 of the both the National Population Health Survey and the National Longitudinal Survey of Children and Youth. However, the information collected by both surveys on the overlapping sample is stored in separate data files. Linking the two surveys (Cycle 1 of the NLSCY and cycles one to nine of the NPHS) will allow for content collected as part of the NLSCY to be used as predictive variables in models of health outcomes.

Output: The linked data file will be made available to the researchers at the Prairie Regional Data Centre at the University of Calgary and the COOL Research Data Centre at the University of Ottawa/Carleton. Only aggregate statistical outputs conforming to the confidentiality provisions of the Statistics Act will be released. Research findings will be disseminated through peer reviewed academic publications. In addition, the findings will be disseminated to decision makers within government and to health professionals.

---

Industry Canada: Economic Impact of Venture Capital, 1999 to 2010

Purpose: To support the evaluation of the Venture Capital (VC) by producing objective measures of its economic impact on the performance of small- and medium-sized enterprises. Key performance indicators, and value-added measures such as sales, profits, firm survival rate, and employment, will be calculated for VC-supported enterprises and for comparable enterprises which did not receive VC support, in order to measure the effectiveness of VC.

Description: A list of firms that were supported by Venture Capital in the period 1999 to 2010 will be linked to the Business Register to facilitate linkage to exporter, employment and tax data.

Records of VC-supported and non-supported enterprises will be linked to the Exporter Register, the Survey of Employment, Payroll and Hours, Longitudinal Employment and Analysis Program and the T2/General Index of Financial Information (T2/GIFI) for the period 1999 to 2010. The records will be linked deterministically using the Business Number and Statistical Enterprise Number and phone number. The resulting linked analysis file will enable longitudinal analysis of each cohort. This is a one-time linkage.

Output: Only non-confidential aggregate statistical outputs and analyses that conform to the confidentiality provisions of the Statistics Act will be released outside of Statistics Canada. The information will be presented in the form of separate summary tables of regression analysis results relating to the study hypotheses of the economic impact of Venture Capital, in addition to profiling tables. A methodology report will be prepared, explaining the file matching processes and constraints and key issues related to the quality of the data. An analytic report will be produced by Statistics Canada.

The linked analysis file will be retained until at least March 2017, at which time a decision will be made about its continued retention. All direct business identifiers will be removed from the analysis file once linkage is complete, and placed in a separate linkage key file. The linkage key file will be retained until at least March 2017, or until no longer required, at which time it will be destroyed.

---

Community Futures Program's Regional Economic Contribution: Linkage of Client List to Business Tax and Employment Data, 1999 to 2011

Purpose: To provide statistical information to support the assessment of the effectiveness of the Community Futures (CF) Program in assisting small- and medium-sized enterprises, by comparing the performance of enterprises that received financial assistance under the program to the performance of other unassisted enterprises in the same region. This information will be used by the regional development agencies (RDAs) which manage the CF programs to determine more effective means of providing assistance to their clients. Employment dynamics, enterprises entering and exiting, selected financial statistics, as well as measures of employment will be analyzed. Findings from this evaluation may be used by the regional agencies to improve assistance to enterprises.

Description: A list and updated lists of enterprises assisted by the Community Futures Program will be linked to the following files: 2004 to 2013 Business Register; 2009, 2010 and 2011 Longitudinal Employment Analysis Program database; and reference years 2004 to 2011 of the General Index of Financial Information. Unassisted enterprises will also be linked to these files to provide comparative data.

Output: The outputs released outside of Statistics Canada will be aggregate statistics and analyses that conform to the confidentiality provisions of the Statistics Act. The information will be presented in the form of statistical tables, broken down by RDA region, industry sector and enterprise size. The linked analysis file, containing the linkage keys and identifiers, will be retained until March 14, 2015, or until no longer required, at which time it will be destroyed.

---

Atlantic Canada Opportunities Agency – Update of Business Performance Evaluation Report

Purpose: To assess the effectiveness of the programs and activities of the Atlantic Canada Opportunities Agency (ACOA) and the usefulness of the Agency's efforts to assist small businesses, and to determine more effective means of providing assistance to the small business community. ACOA assists businesses by providing loans, as well as a broad range of programs and services, for purposes of establishing, expanding, or modernizing businesses, and for the development of human resources. Information resulting from the linkage will be used by ACOA to measure the performance of businesses which received financial assistance under the Agency's programs, and compare it to the performance of other firms in the Atlantic region. Employment dynamics, businesses entering and exiting, selected financial statistics, as well as measures of labour productivity will be analyzed. Findings from this assessment may be used by ACOA to improve assistance to businesses.

Description: A list of ACOA-assisted businesses will be linked to the following files: 2004 to 2011 Business Register, 2009 Longitudinal Employment Analysis Program file, 2004 to 2009 Corporate Tax-General Index of Financial Information, 2004 to 2009 Exporter Registry and 2004 to 2009 Research and Development in Canadian Industry database. The files will be linked using the Business Number, Statistical Enterprise Number and the legal/operating name.

Output: Only aggregate statistical outputs and analyses that conform to the confidentiality provisions of the Statistics Act will be released outside of Statistics Canada. These will be in the form of statistical tables at the business sector and business size level for Atlantic Canada; as well, research and development estimates will be produced at the Canada level. ACOA will publish these results in their annual performance report to Parliament, which will be available on the ACOA website, and in research studies on topics such as entrepreneurial start-ups, employment patterns and growth in Atlantic Canada. The linked analysis file will be retained for one year (until March 31, 2013) or until no longer required, at which time it will be destroyed.

---

Linkable File Environment

Purpose: Policy makers continue to have a keen interest in the role of small- and medium- sized enterprises in the economy. A Linkable File Environment (LFE) will be created that will serve the needs of policy researchers. Research requests will be accommodated according to demand, pending approval of the project proposals. The LFE will be used to construct and extract annually an SME database which will allow Statistics Canada to meet obligations to Industry Canada and the overall research community.

Description: The LFE will include information from the following databases and surveys: the Business Register, the Longitudinal Employment Analysis Program, the General Index of Financial Information, Research and Development in Canadian Industry, the Exporter Register, the Importer Register, the Value of Foreign Direct Investment, the Canadian Direct Investment Aboard, Trade in Services, tax data (PD7, T1 and T4, Goods and Services Tax), the Survey of Innovation and Business Strategy, the Survey of Electronic Commerce Technology, the Survey of Advanced Technology, the Survey of Commercialization of Innovation, the Survey on Financing Small and Medium Enterprises and the Survey of Employment, Payrolls and Hours. It will also include information from databases external to Statistics Canada: patents (Canadian Intellectual Property Office) and venture capital (Thompson Financial Venture Capital Database).

Output: The outputs released outside of Statistics Canada will be aggregate or modelled statistics and analyses that confirm to the confidentiality provisions of the Statistics Act.

---

The Burden of Obesity, Osteoarthritis and Rheumatoid Arthritis in Ontario – Addition of Data Years and Extension of Retention Period

Purpose: To improve treatment strategies for people suffering from obesity, Osteoarthritis (OA) and Rheumatoid Arthritis (RA), and to reduce the costs of treating them. The study will remedy the fact that little is known about the socio-demographic characteristics of the Ontario population with such health condition/illnesses, their quality of life, satisfaction with healthcare, healthcare resource utilization, and ensuing medical costs. The study will establish a cost per patient; identify the determinants of healthcare utilization and medical costs and, for obesity, provide a breakdown of health care utilization/medical costs per body mass index (BMI) levels. In determining utilization and costs, the study will take into account differences in individuals' characteristics and lifestyle.

In many cases of illnesses, physicians and health care policy makers are unaware of the true burden on the health care system and on sufferers. Burden of illness studies, such as this one, can provide a method for demonstrating the importance of a specific disease to society; provide a baseline against which treatment interventions can be assessed; help determine priorities for future medical research, and; help identify the cost drivers of the condition/disease. Additionally, burden of illness studies can further our understanding of the impact the condition/disease has on patients' quality of life and productivity. Findings may assist in the development of obesity, OA and RA policy models to evaluate different treatment and management strategies in Ontario. As well, the evidence generated in the area of obesity may be used by the Canadian Obesity Network which brings together researchers, health professionals, industry, policy makers and others across Canada.

Description: The 2000/2001, 2003 and 2005 Canadian Community Health Survey will be linked to the Medical Services files and the Discharge Abstract Database In-Patient and Day Procedures for the years 1999/2000, 2000/2001, 2001/2002, 2002/2003, 2003/2004, 2004/2005, 2005/2006, 2006/2007 and to the National Ambulatory Care Reporting System Day Procedures files for the years 2003/2004, 2004/2005, 2005/2006, 2006/2007, using a deterministic match on an encrypted health number. A validation procedure, carried out by MOHLTC, will ensure that only valid health card numbers are found on the cohort file, and are encrypted. A probabilistic matching based on birth date, sex, and postal code will be used to resolve incomplete linkage results. Only cases where informed consent was received from the survey respondents will be linked.

Output: Only aggregate data and analyses conforming to the confidentiality provisions of the Statistics Act will be released outside of Statistics Canada, in the form of a release in The Daily, articles in peer-reviewed journals, and through presentations at local, national and international conferences. The linked file will be retained until December 31, 2016, at which time it will be destroyed.

---

Socioeconomic Influences on Use of Physicians in Ontario: Linkage of Canadian Community Health Survey and Ontario Medical Services Data – Addition of Data Years and Extension of Retention Period

Purpose: To examine whether the socioeconomic status of patients influences their use of general practitioner (GP) services, their referral patterns to specialists, and their joint use of different physicians' services. Actual, rather than self-reported, measures of physician utilization will be employed. More specifically, the research will:

(i) assess whether use of GP services, measured by the number of visits, the type of services and the related expenditures, varies with a patient's socioeconomic status (measured by income and education), after taking health care needs (measured by self-reported health status) into account;

(ii) model the pathway between a patient's use of GP services and their use of specialist services, taking into account socioeconomic status differences and health care needs; and,

(iii) determine whether certain categories of survey respondents systematically over- or under-report the number of physician visits they made in the year prior to the survey.

These findings will provide valuable information to the medical community and to health policy-makers. The results may indicate that certain population groups are disadvantaged by the current delivery of health care, in which case the study will indicate where changes should be made.

Description: Data from respondents to the Canadian Community Health Survey (CCHS), cycle 1.1 (2000/2001), cycle 2.1 (2003) and cycle 3.1 (2005) will be linked to administrative information on visits to physicians and health services received in Ontario. Only those cases where informed consent was received from survey respondents will be linked.

The administrative databases used in this research project are the Medical Services files – based on Ontario Health Insurance Program (OHIP) claims for 8 fiscal years: 1999/2000, 2000/2001, 2001/2002, 2002/2003, 2003/2004, 2004/2005, 2005/2006, and 2006/2007.

The data will be linked using deterministic matching on an encrypted health number. A validation procedure, carried out by the Ontario Ministry of Health and Long-Term Care (MOHLTC), has made sure only valid health card numbers for CCHS records are found on the cohort file, and are encrypted. Health card numbers have been similarly encrypted on the health administrative databases. Personal identifiers will be removed from the files and a number assigned by Statistics Canada will be appended to records in the administrative databases that link to CCHS records.

The study is part of a pilot project between Statistics Canada, the Ontario MOHLTC and McMaster University, aimed at enhancing access to Ontario health information by the research community. The creation of an analytical file, as well as aggregation and analysis of the data, will be carried out in the Statistics Canada Research Data Centre (RDC) at McMaster University. The researcher accessing the data in the RDC will do so as a deemed employee of Statistics Canada.

Output: Only aggregate statistical outputs conforming to the confidentiality provisions of the Statistics Act will be released outside of Statistics Canada. Research findings will be disseminated through a release in The Daily, research papers, peer-reviewed journal articles or through presentations at national or international conferences.

Statistics Canada will retain the linked files for a period of five years, that is, until December 31, 2016, at which time they will be destroyed.

---

The Role of Technological Innovation in the Increased Concentration of Health Care Expenditure on the Elderly in Ontario – Addition of Data Years and Extension of Retention Period

Purpose: By examining the relationship between the patient's age and the medical treatment received for a given health condition, this study will provide a basis for more accurate predictions of future health care expenditures in Ontario, taking into account the rate at which new technologies are adopted. The effect of numerous patient characteristics on health care expenditures, such as health conditions, region of residence, income, and education, will be examined by using health survey data. Findings may help decision makers to assess the benefits of increased health care expenditures. More specifically, the study will answer the following questions: 1) How much service do individuals receive at each age, and of what quality? 2) Do older people receive lower intensity care and/or care that is less state-of-the-art than their younger counterparts? 3) If yes, How large is the age-gap in treatment for a given condition, and how has it changed over time? Are there gains (or potential gains) in quality of life when older patients obtain costly procedures? 4) What have been the drivers of the differences (if any): co-morbidities, general health status, etc.? 5) How has medical innovation affected total health care expenditure and its increased concentration on older patients?

Description: Data for respondents to the National Population Health Survey (1996/1997 cycle) and to the Canadian Community Health Survey, cycle 1.1 (2000/01), cycle 2.1 (2003) and cycle 3.1 (2005) will be linked to administrative information on diagnoses, expenditures, and procedures received by those respondents who visited physicians or stayed in hospitals. Only those cases where informed consent was received from survey respondents will be linked.

Three administrative databases will be used in this research project: 1) Medical Services Files – based on Ontario Health Insurance Program (OHIP) claims for 1994/1995, 1995/1996, 1996/1997, 1997/1998, 1998/1999, 1999/2000, 2000/2001, 2001/2002, 2002/2003, 2003/2004, 2004/2005, 2005/2006, 2006/2007, 2007/2008, 2008/2009, 2009/2010; 2) Discharge Abstract Database (DAD) In-Patients for 1994/1995, 1995/1996, 1996/1997, 1997/1998, 1998/1999,1999/2000, 2000/2001, 2001/2002, 2002/2003, 2003/2004, 2004/2005, 2005/2006, 2006/2007, 2007/2008, 2008/2009, 2009/2010; and Day Procedures for 1994/1995, 1995/1996, 1996/1997, 1997/1998, 1998/1999, 1999/2000, 2000/2001, 2001/2002 and 2002/2003; and 3) National Ambulatory Care Reporting System (NACRS) Day Procedures for 2003/2004, 2004/2005, 2005/2006, 2006/2007, 2007/2008, 2008/2009, 2009/2010. The databases were made available by the Ontario Ministry of Health and Long-Term Care (MOHLTC) to Statistics Canada, who will do the linkage.

The data will be linked using a deterministic matching on an encrypted health number. A validation procedure, carried out by MOHLTC, has made sure only valid health card numbers for NPHS and CCHS records are found on the cohort file, and are encrypted. Health card numbers have been similarly encrypted on the health administrative data files. Personal identifiers will be removed from the files and a number assigned by Statistics Canada will be appended to records in the administrative databases, that link to CCHS and NPHS records.

The study is part of a pilot project between Statistics Canada, the Ontario MOHLTC and McMaster University, aimed at enhancing access to Ontario health information by the research community. The creation of an analytical file, as well as the aggregation and analysis of the data, will be carried out in the Statistics Canada Research Data Centre at McMaster University.

Output: Only aggregate data and analysis conforming to the confidentiality provisions of the Statistics Act will be released. Results of the research will be presented at conferences, appear as research papers, and be submitted to journals of health economics for possible publication.

Statistics Canada will retain the linked files for a period of five years, that is, until December 31, 2016, after which they will be destroyed.

---

Study of Mortality among North American Women in the Synthetic Rubber Industry, 1950-2002 – Extension of Retention Period

Purpose: This study, the Canadian portion of a North American investigation, seeks to link the records of approximately 2,400 female synthetic rubber workers to the 1950 to 2002 Canadian Mortality Data Base (CMDB). It will, for the first time, evaluate the overall cause-specific mortality experience of female workers in the synthetic rubber industry, relative to that of general population groups within Canada and the United States. Previous studies of male synthetic rubber workers have shown that relatively high exposure to the chemical 1,3-butadiene (BD) was positively associated with leukemia mortality. Results from these previous studies have been used by regulatory agencies in the United States and Canada to reduce the permissible exposure level (PEL) of BD in the workplace. Results have also been used to implement manufacturing controls in the areas of polymer manufacturing and petroleum refining, benefiting not only workers, but the public in general.

Description: The Polysar cohort file containing name, date of birth, sex, last known residence, dates of employment hire and termination will be linked to the 1984-2003 summary tax file, using Social Insurance Number, in order to determine the vital status of the workers. No income data will be used from this file. A random Statistics Canada number will be assigned to each individual record. The final stage of the linkage process will add the mortality data up to 2002 from the Canadian Mortality Database. The University of Alabama will also provide to Statistics Canada the Polysar work history file. The original study numbers will be removed from this file, a Statistics Canada random number will be assigned. Both files will be returned to researchers at the University of Alabama, who will then append the work history file to the Polysar cohort file using the Statistics Canada random number.

Output: An analysis file without names or identifiers will be released to the University of Alabama with the written consent of the provincial and territorial Vital Statistics Registrars. Results from this study will be published in peer reviewed journals, and a final written report will be presented to the International Institute of Synthetic Rubber Producers, Inc.

The linked file, stripped of all identifiers, will be retained until January 2021. The extension of the retention period of the linked file for this study was approved given the researcher's plan to update the study and integrate it with the male counterpart study with the same retention period.

---

Sherritt International Mortality Study: 1954-2003 Update – Extension of Retention Period

Purpose: This study has the potential of directing the industry to the safest process for refining nickel, thereby protecting the health of current and future nickel workers. It will evaluate the risks of developing lung and nasal sinus cancer, among a group of workers exposed to nickel concentrate dust and metallic nickel powder at Sherritt nickel refining facilities in Fort Saskatchewan, Alberta. Previous studies of the group of workers hired between 1954 and 1978, found no association between exposure to metallic nickel and the development of lung or nasal sinus cancers. Updating the analysis using more current mortality data and including a second cohort of workers hired between 1978 and 1994, will enhance and refine the measures used in the study, as well as the statistical certainty with regards to its conclusions.

Description: The Sherritt International Corporation files for this group of workers, containing the name, sex, data of birth, last known residence, dates of employment hire and termination, will be linked to the 1984-2003 summary tax files. No income data will be used from this tax file. This is only done to assist in the evaluation of the death search by determining the status of the individuals (dead, alive or emigrated). This portion of the linkage will be done deterministically using Social Insurance Number and validated with names and date of birth. The final stage of the linkage process will add the mortality data up to 2003 from the Canadian Mortality Database through probabilistic methods using the Generalized Record Linkage System (GRLS).

Output: Aggregate tables, conforming to the confidentiality provisions of the Statistics Act, will be released to Sherritt International Corporation who will carry out the analysis. The study's findings will be published in peer-reviewed journals. A report and communication will be presented to Sherritt Gordon management and union representatives, to Sherritt Gordon employees, to the Strathcona County Medical Officer of Health, Alberta Occupational Health Department and to the local media. Results of the study will also be presented to the Nickel Institute for distribution to the industry and regulatory agencies worldwide.

The linked file, stripped of all identifiers, will be retained until December 2016. The extension of the retention period of the linked file was approved given Sherritt International Corporation's intention to request an update to this study.

---

Perinatal Outcomes Study: Live Birth, Infant Death and Stillbirth Records Linked to 20% Sample Data from the 1996 and 2006 Censuses of Population

Purpose: To assess perinatal outcomes in Canada according to risk factors related to socioeconomic position, ethnocultural background and environmental exposures. Perinatal outcomes to be examined include preterm and small-and large-for-gestational-age birth, plus birth weight-specific and gestational age-specific fetal and infant mortality.

More specifically, the objectives of this study are: to estimate differences in major perinatal outcomes in Canada by multiple measures of socioeconomic position and ethnocultural background, including maternal and paternal education and occupation, income and housing characteristics of the family, and the ethnicity and nativity of each parent; to understand the role of contextual (neighbourhood) effects on perinatal outcomes after controlling for individual characteristics; to examine the effects of mothers' birth histories on perinatal outcomes; to estimate the effects of exposure to ambient air pollution on perinatal outcomes; and to estimate the extent to which all of the above have changed over the time periods covered by the study.

Description: The previously-linked Canadian Live Birth, Infant Death and Stillbirth Database combines information from the Canadian Live Birth Database, the Canadian Mortality Database, and the Canadian Stillbirth Database, beginning in 1985. However, it fails to account for the interrelationship of successive births to the same mother, and it contains minimal if any information on parental socioeconomic, ethnocultural and environmental health risks and protective factors which are known to affect perinatal health outcomes. Creation of an internally-linked, longitudinally-oriented file linking successive births to the same mother will identify sets of multiple births, permit the systematic examination of the extent to which a woman's successive pregnancy outcomes tend to repeat, and permit the assessment of the risk of adverse birth outcomes conditioned on her previous birth outcomes. Linkage of a sample of births to census data will provide much of the missing information, for a reasonably large sample of births in two recent periods, while avoiding the need for additional data collection. This study will also make use of the air pollution exposure data assembled for the previously-approved Air Pollution, Mortality and Cancer Study.

The census-linked birth records for this study will consist of a 20% sample of births in the two years previous to the 1996 census and in the two years previous to the 2006 census, plus records for subsequent sibling births up to two years after each census, and records for previous sibling births to the same mothers (a total of approximately 430,000 births). The birth records have previously been linked to corresponding infant deaths up to 12 months following each birth. The birth outcome information to be included in the linked files originate from data collected by Canadian provincial and territorial registries of vital statistics, which have been extensively processed, validated, and linked by Statistics Canada. The census data to be included in the linked files were collected, processed, and validated by Statistics Canada.

The linked files will contain only those data items required to conduct the study. Names will be used only for linkage purposes, then removed from the linked microdata analysis files. Only a sample of births–about 4.5% of the longitudinally-oriented birth file–will be linked to census data.

Output: The linked files will at all times remain on Statistics Canada premises. Access to the linked microdata will be restricted to Statistics Canada staff and deemed employees whose work activities require access. Only aggregate data that conform to the confidentiality provisions of the Statistics Act will be released outside of Statistics Canada. Existence of the linked file will be announced in The Daily. Major findings will be used to prepare research papers for publication in peer-reviewed journals (including Statistics Canada's Health Reports) and presentation at workshops and conferences.

The linked analysis files, stripped of direct personal identifiers, will be retained until 31 December 2024, at which time their continued retention will be reviewed. The corresponding linkage key files housed in the Statistics Canada Privacy Office will also be retained until 31 December 2024, at which time their continued retention will be reviewed.

---

Maternal Mortality and Infant Outcomes in Canada, 2001 to 2008

Purpose: To ascertain all women in Canada who died while pregnant or within a year following a live birth or fetal death in order to assess the magnitude of and to better understand the epidemiology, causes, and trends of pregnancy-associated deaths and pregnancy-related deaths in Canada; and to study the outcomes of infants whose mother has died.

Description: Records from the Canadian Mortality Database (CMDB) of women, aged 10 to 55, who died between 2001 and 2008 were linked to records of births (Canadian Birth Database) and stillbirths (Canadian Stillbirth Database) that occurred between 2000 and 2008 and to infant deaths (CMDB) that occurred between 2000 and 2009.

The final Maternal Mortality – Birth/Stillbirth/Infant Death Analysis File contains linked women mortality records as well as birth/infant mortality records and stillbirth records. All direct personal identifiers and addresses were removed from the analysis file and stored in a separate linkage key file following completion of the linkage. Coded Census Dissemination Area/Enumeration Area is the lowest level of geography on the linked analysis file. Each record includes a random Statistics Canada identification number.

Output: Only aggregate statistics and analytical output conforming to the confidentiality provisions of the Statistics Act will be released outside of Statistics Canada. Results of this study will be communicated through publication in peer-reviewed journals. The Canadian Perinatal Surveillance System plans to publish a report in 2013 on Maternal Mortality and Severe Morbidity in Canada. Also, the study findings will be disseminated through publications on the Internet and through presentations to associations.

The linked analysis file and linkage key file will be retained for at least 10 years, that is, until December 2023, or until they are no longer required, at which time they will be destroyed.

---

Workers in the synthetic rubber industry: mortality follow-up 1950 to 2009

Purpose: To determine the health risks associated with exposure to chemicals and processes used in the production of synthetic rubber at the Lanxess (formerly Polysar) plant in Sarnia, Ontario. This study is the Canadian component of a North American investigation. Previous studies of workers in the synthetic rubber industry have shown that work in jobs entailing relatively high exposure to the chemical 1, 3-butadiene (BD) was positively associated with leukemia mortality. This current study seeks to link the records of approximately 9,000 male and female synthetic rubber workers to the 1950 to 2009 Canadian Mortality Data Base (CMDB). It will evaluate the overall cause-specific mortality experience of workers in the synthetic rubber industry, relative to that of general population groups within Canada and the United States. Similar studies have been conducted in the past and the new results will enhance the findings of the previous investigations, thus increasing the statistical certainty with regards to the conclusions of the study. Results may be useful for risk assessment and regulatory actions by federal, provincial/territorial and state agencies in Canada, the United States and other countries, which would benefit current and future workers in the synthetic rubber industry through process changes and the reduction of hazardous exposures.

Description: The Lanxess cohort file, consisting of approximately 9,000 employees who worked at the Sarnia plant (men who worked for one year or more between January 1, 1943 and December 31, 1991, and women employed one day or more between August 20, 1940 and December 31, 2004) will be linked to the 1984 to 2008 historic tax summary file which does not contain income data. This linkage is carried out to assist in the evaluation of the death search by determining the status of the individuals (dead, alive or emigrated) at the end of the study period. The cohort file will then be linked to the 1950 to 2009 Canadian Mortality Database using probabilistic record linkage methods.

Output: A mortality analysis file linkable to the work history file of the study cohort, without names and personal identifiers, will be disclosed to the principal investigator at the University of Alabama at Birmingham, with the written consent of the provincial and territorial Vital Statistics Registrars, and at the discretion of the Chief Statistician. Results from this study will be published in peer reviewed journals, and a final written report will be submitted to the International Institute of Synthetic Rubber Producers, Inc. and the American Chemistry Council. All published information will be in the form of aggregate data conforming to the confidentiality provisions of the Statistics Act. The analysis file and linkage key files will be retained by Statistics Canada until December 31, 2021 or until they are no longer required, at which point they will be destroyed.

---

Ongoing Longitudinal Linkage of Child and Spousal Support Cases in the Survey of Maintenance Enforcement Programs

Purpose: To enable federal government policy analysts, provincial and territorial maintenance enforcement programs and the public in general to better understand child and spousal support cases through supplying longitudinal data from the Survey of Maintenance Enforcement Programs (SMEP), which allows for analysis on the evolution of case characteristics and outcomes over time. Previously, analysis on SMEP data was limited to one-year snapshots.

Description: Monthly records from the Survey of Maintenance Enforcement Programs (SMEP) pertaining to the same maintenance enforcement case will be linked across fiscal years using the case identification number. Random study identification number will be attached to each record in the linked file, to replace the case identification number.

Output: Only aggregate tabular statistics that conform to the confidentiality provisions of the Statistics Act will be released outside of Statistics Canada. The linked longitudinal files developed for products using longitudinal analysis of SMEP data will be retained for five years. Access to the linked analysis file will be restricted to Statistics Canada employees whose work activities require such access.

---

Linkage of the Survey of Maintenance Enforcement Programs to the T1 Family File

Purpose: Divorce and separation are a common occurrence in modern Canadian society. When family breakdown occurs, the continued financial support of the children, and in some cases an ex-spouse or partner, is a significant issue for the separating couple. In most situations where children are involved, a support arrangement is set up between the ex-partners that stipulates the amount of support to be paid and frequency of payment. If the support arrangement is court-ordered, or the agreement is registered with court, the ex-partners can enrol in a maintenance enforcement program (MEP). These programs provide administrative support to payors and recipients of child and spousal support and improve compliance with support obligations. It has been estimated that 60% of agreements are registered with a maintenance enforcement program.

This linkage will provide information on income and employment characteristics as well as family composition of persons enrolled in MEPs, information that is currently not available. With these data, governments and the public at large will have a clearer understanding of the socio-economic situations of child and spousal support payors, recipients and beneficiaries.

Description: Records from 2006 to 2009 from the Survey of Maintenance Enforcement Programs for recipients and for payors of child and spousal support will be linked to records in the T1 Family File. The names, addresses and date of birth of recipients and payors will be the primary variables used in the matching process.

No personal information will be included on the linked recipient analysis file or the linked payor analysis file. Random study identification numbers will replace the case identification number used by the MEP.

Output: Only aggregate tabular statistics that conform to the confidentiality provisions of the Statistics Act will be released outside of Statistics Canada. The linked analysis files will be retained until March 31, 2014, to support on-going analysis; once the file is no longer required, it will be destroyed. Access to the linked analysis files will be restricted to Statistics Canada employees whose work activities require such access.

---

Inter-provincial Workers: Linkage of Administrative Data to Examine the Number and Characteristics of Individuals Who Reside in One Province But Work in Another

Purpose: Through the 2000s, high world oil prices and the expansion of Canada's resource sector have made certain regions—such as northern Alberta—an attractive destination for job seekers. For some, employment in Alberta is accompanied by a permanent move to the province (i.e., inter-provincial migration). For others, employment in Alberta is undertaken on an ongoing basis, while a permanent residence is maintained elsewhere in Canada. For the latter, their province of employment and province of residence are not one and the same.

In spite of much anecdotal evidence regarding "inter-provincial workforces", comprehensive and reliable information remains scarce. In part, this is because of the many challenges faced when trying to enumerate a mobile population that often resides in remote areas and in temporary accommodations.

The size and characteristics of the inter-provincial workforce have many implications for public programs and policies. For example, the in-flow of workers into a province such as Alberta has implications for demands placed on local infrastructure (e.g., roads, recreation, police and emergency services), for the demand and price of goods and services (e.g., housing), for the inter-provincial allocation of income taxes, and for federal equalization payments.

This linkage project combines information on individuals' province of employment, obtained from T4 Statements of Remuneration with information on individuals' province of residence obtained from T1 personal tax returns to identify "inter-provincial workers". Among the key policy questions that can be addressed with the linked database are: How large is the inter-provincial workforce in specific provinces? Are inter-provincial workers primarily young, non-married men, or do many have families in their province of residence? To what extent is inter-provincial employment ongoing over many consecutive years? 

Description: The database for this analysis contains information from: the T4 Statement of Remuneration file, which contains information on the province of employment, on earnings received, and premiums/contributions paid; the Longitudinal Employment Analysis Program (LEAP), which contains firm-level information on industry and firm size of employment; the T1 Family File, which contain information on individual-level demographic and income characteristics (e.g., sex, age, total earnings, province of residence on December 31^st) as well as family-level demographic and income characteristics (e.g., family size, number of co-resident children, total family earnings); and information from the T1 Historical file, which contains information from T1 personal tax returns, including individuals who filed their T1 return late. This allows estimates of inter-provincial employment to take into account the possibility that inter-provincial workers, given the mobile nature of their employment, may be more likely than the general population to file their tax returns late. The linkage will contain data for the 2003 to 2010 tax years.

Random study identification numbers will be attached to each record in the pension coverage analysis file to replace Social Insurance Numbers and Business Numbers. The algorithm for generating these identification numbers will be kept separately from the analysis file and kept confidential.

Output: The data file will be used to estimate the number of inter-provincial workers in Canada and to examine their socio-economic characteristics. Only aggregate statistics and analysis conforming to the confidentiality provisions of the Statistics Act will be released outside of Statistics Canada. The linked file and linking key file will be retained by Statistics Canada until March 31, 2015, at which time they will be destroyed. Access to these files will be restricted to Statistics Canada employees whose assigned work activities require access.

---

Evaluation of Blood Transfusion on Mortality Outcomes of Cardiovascular Patients Following Hip Fracture Repair: Linkage to Canadian Mortality Database, 2004 to 2009

Purpose: The purpose of this study is to evaluate the impact of different red blood cell transfusion strategies on long-term mortality. It will show whether a liberal transfusion strategy is associated with a decreased rate of long-term mortality compared to a restrictive transfusion strategy, and will provide much needed clinical trial evidence to help guide transfusion practice.

Description: The study is a multi-site randomized clinical trial which enrolled anaemic patients following surgical repair of a fractured hip from 2004 through 2009. Patients with cardiovascular disease, or risk factors for cardiovascular disease, were randomized to receive either a liberal (100 g/L) or a restrictive (symptoms of anaemia or at physician discretion if the haemoglobin level was less than 80 g/L) transfusion strategy. The complete cohort consists of 2,016 patients who underwent surgery in either the United States or Canada. The Canadian component of the cohort consists of 794 patients who underwent surgery in one of twelve participating hospitals in Canada.

A file consisting of records of those patients who underwent surgery in Canada will be linked by Statistics Canada to the 2004 to 2009 Canadian Mortality Database. The principal investigator received consent for linkage of patient records to mortality information for those enrolled in the study.

Output: No analysis or publication of the results of this linkage will be conducted by Statistics Canada. A mortality output file will be produced, containing the clinical trial study number, complete date of death (month, day and year), province/territory of death, and cause(s) of death.

The mortality output file will be split by province or territory of death, and the records will be sent to the appropriate vital statistics registrars who, at their discretion, will release the information to the principal investigator at the Robert Wood Johnson Medical School in New Jersey, United States.

The principal investigator has undertaken to publish the study findings in the form of aggregate statistical outputs that will not result in the identification of individual patients. Results of the analysis of the data will be presented at medical meetings and papers will be submitted to peer-reviewed medical journals for publication.

Access at Statistics Canada to the identifiers, linking keys and mortality output files will be restricted to employees whose assigned work requires such access. The linkage key file and mortality output file will be retained until December 31, 2018, or until no longer required, at which time these files will be destroyed.

---

Linkage of the Census of Population 2006 to the Discharge Abstract Database and the Canadian Mortality Database for Purposes of the Longitudinal Health and Administrative Data Initiative

Purpose: To meet the requirements of the Longitudinal Health and Administrative Data Initiative Research Agenda, the sample portion (2B) of the Census of Population 2006 will be linked to the Discharge Abstract Database and to the Canadian Mortality Database to investigate the hospitalization patterns among: 1) Aboriginal groups; 2) immigrant groups; and 3) older adults. As well, there will be an assessment of the validity of the linked file for use in health services research.

As hospitals comprise the single largest share of all health care expenditures and costs continue to rise, understanding their patterns of use is critical. In particular, better understanding of the patterns of use among key sub-groups such as Aboriginal peoples and immigrants, who otherwise cannot be identified in administrative data, could assist policy makers in identifying groups at high risk for hospitalizations including those risks that are potentially modifiable via adaptive health services, public promotion, and prevention strategies.

Furthermore, given the richness of the Census data, the data will provide a first ever look at the potential differences in use among Aboriginal groups living on and not on reserve. This is critical information for health care planners, including those at the federal level, responsible for the delivery of services to these communities.

Similarly, the linked data will provide a unique opportunity to investigate the critical differences of health services use patterns by country and region of birth, time since immigration as well as generational status. Policy makers require information on the utilization patterns of immigrants by all dimensions to inform their decision making.

Finally, better understanding the socio-demographic characteristics of older adults being hospitalized in acute care facilities, over and above the clinical factors related to their hospitalization, could inform policy makers on the need for more adaptive services outside of hospitals and inform healthcare planners on the potential resource load based on the demographic characteristics of the population they serve.

Description: The sample portion (2B) of the Census of Population 2006 will be linked to the Discharge Abstract Database (DAD), 2004-2005 to 2009-2010 and to the Canadian Mortality Database (CMDB), 2006 to 2010.

The linked Census/DAD/CMDB file will contain only those data items required to conduct the studies. All direct personal identifiers and addresses are removed from the analysis file. Personal identifiers used for linkage purposes, such as name, death registration number and health insurance number, are stored in separate files.

Output: The linked Census/DAD/CMDB file will remain within Statistics Canada. All access to the linked microdata file will be restricted to Statistics Canada staff whose work activities require access. Only aggregate data that conform to the confidentiality provisions of the Statistics Act will be released outside of Statistics Canada. Research papers based on analyses of the linked data will be submitted for publication in the Statistics Canada peer-reviewed quarterly, Health Reports, as well as in medical or epidemiological journals or released as a working paper in the Health Research Working Paper Series.

The linked analysis file will be retained until December 31, 2018, or until no longer required by Statistics Canada, at which point the continued retention of the file will be reviewed.

---

Linkage of the Canadian Cancer Registry to the Discharge Abstract Database and the National Ambulatory Care Reporting System for Purposes of the Longitudinal Health and Administrative Data Initiative

Purpose: The Longitudinal Health and Administrative Data Initiative (LHAD) Steering Committee, comprised of representatives of provincial/territorial ministries of health, Statistics Canada, the Canadian Institute for Health Information, the Canadian Council of Cancer Registries and the Vital Statistics Council for Canada, is interested in looking at improving the capacity to analyse surgical practice patterns for cancer patients. Such analyses would add to the evidence required to develop ever more effective strategies for the management and treatment of cancer.

Description: The Canadian Cancer Registry (CCR) will be linked to the Discharge Abstract Database (DAD) and the National Ambulatory Care Reporting System (NACRS) within the Longitudinal Health and Administrative Data environment at Statistics Canada. The CCR contains cancer diagnosis, mortality and cancer staging information but does not contain information on treatment. The DAD contains information on procedures performed in hospitals but does not contain tumour information. NACRS contains information on procedures performed in the Emergency Department and outpatient procedures. The linked file would allow analysis of cancer treatment procedures in hospitals while controlling for certain characteristics of the cancer such as multiple primary tumours, tumour size, metastases or nodal involvement.

Output: The linked CCR/DAD/NACRS file will remain within Statistics Canada. All access to the linked microdata file will be restricted to Statistics Canada staff whose work activities require access. Only aggregate statistics and analytical output conforming to the confidentiality provisions of the Statistics Act will be released outside of Statistics Canada. Research papers based on analyses of the linked data will be submitted for publication in the Statistics Canada peer-reviewed quarterly, Health Reports, as well as in medical or epidemiological journals or released as a working paper in the Health Research Working Paper Series.

The linked analysis file will be retained until December 2018, or until it is no longer required, at which point the continued retention of the file will be reviewed.

---

Canadian Perinatal Surveillance System: Linkage of Birth and Death Records for Infant Mortality Surveillance, Births 2006 to 2010

Purpose: The Canadian Perinatal Surveillance System (CPSS) is an ongoing surveillance program. Its mandate is to contribute to improved health for pregnant women, mothers and infants in Canada through ongoing monitoring and reporting on perinatal health determinants and outcomes. The birth-death linked file is a core file for the CPSS. It is used for ongoing reporting on some key perinatal health indicators such as gestational age, birth weight and maternal age specific mortality rates.

The purpose of the linkage is to associate mortality information for infants less than one year of age, such as age at death and cause, with single and multiple birth information such as birth weight and gestational age. This linkage is an update to earlier linkages. It will enable epidemiologists, public health specialists, scientists and researchers to identify trends or changes over time.

Description: Five annual linkages are included in this application, commencing with linkage of the 2006 Canadian Birth Database (CBDB) with the 2006-2007 Canadian Mortality Database (CMDB), and continuing for each consecutive year up to linkage of the 2010 CBDB with the 2010-2011 CMDB. The linked birth-death file will contain composite birth/death records, stillbirth records, and birth records which did not link to a death. Each record will be assigned a random Statistics Canada number.

Output: Birth-death linked analysis files without names or personal identifiers will be disclosed to the CPSS at the Public Health Agency of Canada with the consent of the provincial and territorial Vital Statistics Registrars and at the discretion of the Chief Statistician.

Results of the surveillance analysis will be communicated for action through the Canadian Perinatal Health Reports, fact sheets, web site, public health practice guidelines and published in peer-reviewed journals. Results may also be disseminated in Statistics Canada Health Reports.

Statistics Canada will retain the data produced during these linkages until at least 31 December 2020, prior to which a decision will be made about their continued retention, based on the program requirements of Statistics Canada and the CPSS at that time.

Longitudinal Perspectives on Employment, Income and Health: Linkage of the Longitudinal Worker File, 1991 Census, Canadian Mortality Database and Canadian Cancer Database

Purpose: The objective of this project is to create a new database that will support longitudinal analysis and outcome measures pertaining to employment, income and health. The database will be used to examine various issues pertaining to returns to education and training, the labour market outcomes of immigrants, retirement transitions, and changes in individual- and family-level earnings in the wake of layoffs or a cancer diagnosis. In addition, the file will be used to strengthen inputs into Statistics Canada's Population Health Model cancer modules and the Lifepaths micro-simulation model.

Description: This project builds on previous initiatives undertaken by Statistics Canada. First, in 2003, Statistics Canada's Policy Committee approved an initiative that drew a 15% sample of Canadians aged 25 or older from the 1991 Census 2B and 2D long forms and linked them to their 1991 and/or 1992 T1 tax returns, and subsequently to the Canadian Mortality Database. In 2009, this database was extended to cover a longer reference period and expanded to include information from the Canadian Cancer Database as well as postal code information on an annual basis.

Second, in 1999, Statistics Canada's Policy Committee approved the creation and annual update of the Longitudinal Worker File for 1983 onwards. An amendment was approved in 2007 to add additional variables from the T1 personal tax file. The Longitudinal Worker File (LWF) is comprised of a 10% sample of employed Canadians, and contains information drawn from the T1 Personal Tax file, T4 Statement of Remuneration, the Record of Employment, and the Longitudinal Employment Analysis Program. The LWF provide longitudinal information on employment and earning outcomes from 1983 to 2010.

These two initiatives have yielded large and complementary data bases – the first containing rich socio-demographic information (but little information on economic outcomes) and the second containing rich information on economic outcomes (but little socio-demographic information). Because of the large size of both data bases, the overlap between them yields a subsample comprised of 1.5% of Canadians who were aged 25 or older in 1991.

Four separate analytical files will be created and linkable with a randomly-generated Statistics Canada respondent number.

Longitudinal Worker File output file: This file contains the demographic and economic variables from the 1983-2010 LWF, individual-and family-level variables appended from the T1 Family File, and a randomly-generated Statistics Canada respondent number.

Census of Population output file: This file contains the socio-demographic variables from the 1991 Census of Population 2B and 2D (long forms) available in the original 1991 Census mortality cohort, as well as a randomly-generated Statistics Canada respondent number.

Mortality Output file: This file will contain the randomly-generated Statistics Canada number for each individual in the cohort, and the following mortality information: age, province/country of birth, underlying cause of death, nature of injury, province/ country of death, sex, postal code and standard geographic codes of residence (e.g., census subdivision), year, month, and day of death, derived person-years at risk, and mortality linkage weight.

Cancer Output file: This file will contain the randomly-assigned Statistics Canada number for each individual in the cohort, and the following information from the cancer database: sex, province and year, month and day of diagnosis, year of birth, age, province or country of birth, diagnostic information (diagnostic codes, morphology and topography, morphology code indicator, source of registration, method of diagnosis, laterality, primary site number), patient vital status, province of residence, postal code of residence at diagnosis, year and province of death (if applicable), postal code of death (if applicable), cause of death (if applicable), and the cancer incidence linkage weight.

Output: Only aggregate statistics and analysis conforming to the confidentiality provisions of the Statistics Act will be released outside of Statistics Canada. Methodological and analytical findings resulting from these linked data will be used to prepare research papers for publication in analytical reports, peer-reviewed scientific journals (including Statistics Canada's Health Reports), CANSIM, for presentation at conferences, workshops and meetings.

The output files will be retained by Statistics Canada until December 31, 2022, at which time the continued retention of the file will be reviewed. All linkage keys and identifiers will be removed from the output files are retained separately, with access limited to Statistics Canada employees whose assigned work requires access to the file.

---

Estimating Distributions that Characterize the Natural History of Influenza Infections

Purpose: This project will estimate statistical distributions that are critical for development of infectious disease models, and which have never been estimated reliably. More specifically, the research will:

• Use the hospital inpatient records in the DAD database to estimate the distribution of time intervals from hospital admission to influenza-related death in hospital.
• Use the OHIP fee-for-service database linked to the DAD database to estimate the distribution of time intervals from diagnosis to hospital admission, and to hospital discharge or death.

Mathematical and statistical models are used to project the future time course of infectious disease epidemics and the expected future burden on the health care system and the economy. More accurate parameterization of these models will lead to better forecasting, which will be of direct benefit to health and economic planners and indirect benefit to the public since resources will be better managed. For example, we will be in a much better position to predict hospital bed demands in the face of an influenza pandemic.

Description: Several administrative databases will be used in this study:

1. Medical Service Files based on Ontario Health Insurance Program (OHIP) claims; data years: 1994/1995, 1995/1996, 1996/1997, 1997/1998, 1998/1999, 1999/2000, 2000/2001, 2001/2002, 2002/2003, 2003/2004, 2004/2005, 2005/2006, 2006/2007, 2007/2008, 2008/2009, 2009/2010
2. Discharge Abstract Database (DAD) In-patient Files; data years: 1994/1995, 1995/1996, 1996/1997, 1997/1998, 1998/1999,1999/2000, 2000/2001, 2001/2002, 2002/2003, 2003/2004, 2004/2005, 2005/2006, 2006/2007, 2007/2008, 2008/2009, 2009/2010
3. Discharge Abstract Database (DAD) Day Procedures files; data years: 1994/1995,1995/1996, 1996/1997, 1997/1998, 1998/1999, 1999/2000, 2000/2001, 2001/2002 and 2002/2003
4. National Ambulatory Care Reporting System (NACRS) Day Procedures files; data years: 2003/2004, 2004/2005, 2005/2006, 2006/2007, 2007/2008, 2008/2009, 2009/2010.

The data will be linked using deterministic matching on an encrypted health number that has been generated by the Ontario Ministry of Health and Long-Term Care (MOHLTC) on all the files.

The study is part of a pilot project between Statistics Canada, the MOHLTC and McMaster University, aimed at enhancing access to Ontario health information by the research community. The creation of analytical files, as well as aggregation and analysis of the data, will be carried out in the Statistics Canada Research Data Centre (RDC) at McMaster University. The researchers accessing the data in the RDC will do so as deemed employees of Statistics Canada.

Output: Only aggregate statistical outputs conforming to the confidentiality provisions of the Statistics Act will be released outside of Statistics Canada. Research findings will be disseminated through research papers, peer-reviewed journal articles or through presentations at national or international conferences.

Statistics Canada will retain the linked files for a period of five years, that is, until December 31, 2017, at which time they will be destroyed.

---

Study of the incomes of the affluent: The role of private corporations

Purpose: The proposed research will provide significantly improved estimates of the trends and patterns of income inequality in Canada, most notably at the top of the income distribution. The key improvement is taking explicit account of private corporation income. In Canada, tax planning incentives are such that owners of small businesses and other private corporations are more likely to retain earnings within their corporations. As a result, levels of income inequality as measured using personal income tax returns alone could well be understated, though it is not clear in which direction trends in measured income inequality will be affected.

Description: The project involves matching selected corporate income tax return (T2) data (only for Canadian Controlled Private Corporations or CCPCs), T5 tax forms (for Investment Income) and T4 (employment income) for the years 2000 to 2011 to Statistics Canada Longitudinal Administrative Databank (LAD). The LAD, T2, T5 and T4 datasets will be linked deterministically using the Social Insurance Number (SIN) and corporation business number of individuals who have submitted a tax return. Following completion of the linkage, the SIN and business number identifiers will then be removed from the linked file. No names or addresses will be on the final linked analysis data file. The linkage will be performed by Statistics Canada personnel.

Output: Only aggregate data and analysis conforming to the confidentiality provisions of the Statistics Act will be released outside of Statistics Canada. Findings will be disseminated in academic research papers as well as presented to Canadian academic conferences, such as the Canadian Economics Association. Depending on requests, there may be wider dissemination of the research results. The linked file will be retained at Statistics Canada until 30 June, 2018 when the research project is scheduled to be completed and the information will no longer be needed, after which time it will be destroyed.

---

Re-contact with the Justice system

Purpose: To determine the types of unique information required to create and support high quality indicators of re-contact within three criminal justice sectors. Whereas contact is defined as a documented official intervention (e.g. charge) against a person by a criminal justice agency/organization, a re-contact is defined as a subsequent contact signifying a new, official intervention by the agency/organization during a specified follow-up period.

The project will attempt to establish baseline metrics on re-contact with the justice system which can serve as a comparison group for assessing the impact of policies and programs which may be implemented in a particular jurisdiction. It will also provide the potential to be able to track emerging patterns of re-contact which may appear to be unique within a jurisdiction at a local level yet are more systematic in nature when evaluated at a higher level (e.g. national) of analysis.

Description: The proposed project consists of three separate record linkage activities which will be used to support the development of re-contact indicators within the policing, courts and corrections sectors of justice.

The first linkage will use records collected under the Uniform Crime Reporting Survey (UCR2), in addition to supplemental personal identifiers provided by the Toronto Police Service for the years January 1, 2005 to December 31, 2010.

The second will link records collected from the Integrated Criminal Court Survey (ICCS) and additional personal identifiers provided by the Nova Scotia Department of Justice Policy, Planning and Research for the for the fiscal years 2006/2007 to 2010/2011.

The third will link records collected by the Integrated Correctional Services Survey (ICSS) in addition to other personal identifiers provided by the Saskatchewan Ministry of Corrections 1999/2000 to 2010/2011 and the Correctional Service of Canada for the years 2001/2002 to 2010/2011.

Output: Only aggregate statistics and analyses conforming to the confidentiality provisions of the Statistics Act will be released outside of Statistics Canada. Access to linking keys and linked analysis file will be restricted to Statistics Canada employees whose assigned work activities require such access.

High-level findings may be reported in the form of presentations to various National Justice Statistics Initiative partners.

Statistics Canada will retain the linked analysis files until March 31, 2017 or sooner if no longer required, at which time the linked analysis files will be destroyed.

---

Linkable File Environment

Purpose: Policy makers continue to have a keen interest in the role of small- and medium- sized enterprises in the economy. A Linkable File Environment (LFE) will be created that will serve the needs of policy researchers. Research requests will be accommodated according to demand, pending approval of the project proposals. The LFE will be used to construct and extract annually an SME database which will allow Statistics Canada to meet obligations to Industry Canada and the overall research community.

Description: The LFE will include information from the following databases and surveys: the Business Register, the Longitudinal Employment Analysis Program, the General Index of Financial Information, Research and Development in Canadian Industry, the Exporter Register, the Importer Register, the Value of Foreign Direct Investment, the Canadian Direct Investment Aboard, Trade in Services, tax data (PD7, T1 and T4, Goods and Services Tax), the Survey of Innovation and Business Strategy, the Survey of Electronic Commerce Technology, the Survey of Advanced Technology, the Survey of Commercialization of Innovation, the Survey on Financing Small and Medium Enterprises and the Survey of Employment, Payrolls and Hours. It will also include information from databases external to Statistics Canada: patents (Canadian Intellectual Property Office) and venture capital (Thompson Financial Venture Capital Database).

Output: The outputs released outside of Statistics Canada will be aggregate or modelled statistics and analyses that confirm to the confidentiality provisions of the Statistics Act.

---

Community Futures Program's Regional Economic Contribution: Linkage of Client List to Business Tax and Employment Data, 1999 to 2011

Purpose: To provide statistical information to support the assessment of the effectiveness of the Community Futures (CF) Program in assisting small- and medium-sized enterprises, by comparing the performance of enterprises that received financial assistance under the program to the performance of other unassisted enterprises in the same region. This information will be used by the regional development agencies (RDAs) which manage the CF programs to determine more effective means of providing assistance to their clients. Employment dynamics, enterprises entering and exiting, selected financial statistics, as well as measures of employment will be analyzed. Findings from this evaluation may be used by the regional agencies to improve assistance to enterprises.

Description: A list and updated lists of enterprises assisted by the Community Futures Program will be linked to the following files: 2004 to 2013 Business Register; 2009, 2010 and 2011 Longitudinal Employment Analysis Program database (LEAP); and reference years 2004 to 2011 of the General Index of Financial Information (GIFI). Unassisted enterprises will also be linked to these files to provide comparative data.

Output: The outputs released outside of Statistics Canada will be aggregate statistics and analyses that conform to the confidentiality provisions of the Statistics Act. The information will be presented in the form of statistical tables, broken down by RDA region, industry sector and enterprise size.

The linked analysis file, containing the linkage keys and identifiers, will be retained until March 14, 2015, or until no longer required, at which time it will be destroyed.

---

Validation of frailty index

Purpose: The specific objectives of the project are to:

1. To validate the application of a frailty index (FI) to the Canadian Community Health Survey;
2. To identify a valid cut-point at which seniors are at higher risk of adverse hospital events (all cause hospitalization, number of hospitalizations, emergency admissions, discharge to long-term care, in-hospital death)

Description: Frailty provides a means to understand health differences between similarly aged seniors. Independent of age, frailty is strongly predictive of death, hospitalization, institutionalization, falls, and worsening health status. Understanding the level of frailty among Canada's community-dwelling seniors is important information that can be used to inform aging-related policies and to estimate the need for health-care resources including homecare and residential care. Various measures of frailty have been developed that can be applied to large-scale population health surveys to assign a frailty score between 0 and 1 denoting lower versus higher levels of frailty. While FI scores provide information regarding the level of frailty at the individual level, cut-points are required to estimate the prevalence of frailty at the population level. Currently, there are no universally accepted frailty cut-points. To identify validated cut-points, data that link frailty scores to adverse events, such as hospitalizations, are required to identify ranges of frailty scores associated with different risks of experiencing these adverse events.

This study requires that records of respondents to the CCHS 65 years of age and older be linked to hospital records. The files from cycle 2.1 (2003) and cycle 3.1 (2005) have been linked to the Discharge Abstract Database (2002/03-2007/2008). Only those records of survey respondents who granted consent to link and share their data were used in the linkage.

The study sample will include CCHS respondents 65 years of age and older for whom an FI score could be derived. To assess the relationship between the FI score and adverse hospital events, individuals will be "followed" over time in the linked hospital data to identify those individuals who experienced an adverse hospital event (i.e. at least one hospital admission, emergency admission, discharge to long-term care, in-hospital death). Stratum-Specific Likelihood Ratios (SSLRs) will be calculated to compare the risk of hospital events across various levels or stratums of FI scores to identify significantly distinct bands or levels of frailty.

Output: The results of the validation study will be used to identify FI score cut-points to estimate the prevalence of frailty among community dwelling seniors in Canada. The findings of the research project will be submitted for publication to a peer reviewed journal.

---

Expansion of Household Survey Frame Service

Purpose: Statistics Canada has periodically expanded its Household Survey Frame Service to accommodate the needs of internal survey design and data collection since 1989. The Frame first developed as a list of likely valid dwellings in larger urban centers, but has been progressively expanded to a national register of dwellings, with contact information when possible. The expansion of Statistics Canada's Dwelling Frame Service will enable all of the Agency's household surveys and the Census to be supported.

Description: The expansion of the Household Survey Frame Service will further improve the cost-effectiveness of Statistics Canada's survey design and collection processes. Specifically:

1. Increased ability to do telephone contact for sampled dwellings will reduce collection costs;
2. The inclusion of more current indicators of socio-economic status at the dwelling level will be used to improve sample selection, weighting, and bias adjustment;
3. Consolidating the management of key administrative files used as input to the Service into one functional area will reduce costs, facilitate access and improve compliance to information management directives;
4. Finally, the expansion sets the stage for future improvement in the capacity to limit overlap between surveys and to build more comprehensive data-sets for analysis drawing on samples selected across several surveys and time periods.

Output: The Household Survey Frame Service is comprised of three components:

1. A national list of likely valid dwellings and their descriptors;
2. Contact information (currently telephone numbers) for dwellings where these can be found on publically accessible administrative files and internal Statistics Canada data;
3. A range of socio-economic indicators at the person, household, and small-area geography levels facilitating survey design, sampling, and bias control. The indicators are derived from internal Statistics Canada data and Canada Revenue Agency data obtained under standing Agreements.

Each component of the Frame is produced quarterly, but may contain up-dated information ranging from quarterly to annually depending on type.

The output components are only available for internal Statistics Canada use in survey design, sample selection, and data collection. The data are stored and maintained in compliance with all federal security and data management policies.

Statistics Canada neither encourages nor discourages the use of SEPH data for contract escalation purposes, but can offer advice of a purely statistical nature on the limitations associated with the use of its data.

The Survey of Employment, Payrolls and Hours (SEPH) earnings statistics are sometimes used in the public and private sectors to index various types of labour costs, usually through contractually set pricing formulae.

SEPH publishes a wide range of earnings estimates, many of which are not advisable for pricing purposes. Generally speaking, it is preferable to avoid using series for highly disaggregated industry groupings (4 digit of the NAICS), as well as those estimates specific to class of worker, that is hourly, salaried or other employees, as these series are based on a relatively small sample. Canada trend and level estimates for a given industry are usually more stable than their Provincial/Territorial counterparts. Users should always consider available quality indicators and the number of employees relevant to the earnings series they are interested in.

It is important to note that changes in average earnings reflect a number of factors, including wage growth, changes in composition of employment by industry, occupation and level of job experience, as well as average hours worked per week – not to mention sampling variability. As an attempt to observe earnings over time while controlling for changes in hours and employment, Statistics Canada produces Fixed Weighted Index (FWI) data. This approach is closer to the concept of a labour or employment cost index as used in other countries, but does not control for other factors that can impact on earnings. The FWI is only available at higher levels of aggregation, does not include overtime earnings and does not include earnings data from employees on commission or paid by a piece rate.

In addition, the use of month-over-month changes to escalate costs should be avoided in favour of year-over-year movements based on annual averages. If monthly calculations are required, moving averages of several months should be strongly considered.

Any indexing formula should also take into account the fact that the survey data undergoes periodic revision. Users should always use current and complete data series.

For more information, contact us (toll-free 1-800-263-1136; infostats@statcan.gc.ca).

The Monthly Survey of Manufacturing (MSM) publishes statistical series for manufacturers – sales of goods manufactured, inventories, unfilled orders and new orders. The values of these characteristics represent current monthly estimates of the more complete Annual Survey of Manufactures and Logging (ASML) data.

The MSM is a sample survey of approximately 10,500 Canadian manufacturing establishments, which are categorized into over 220 industries. Industries are classified according to the 2012 North American Industrial Classification System (NAICS). Seasonally adjusted series are available for the main aggregates.

An establishment comprises the smallest manufacturing unit capable of reporting the variables of interest. Data collected by the MSM provides a current ‘snapshot’ of sales of goods manufactured values by the Canadian manufacturing sector, enabling analysis of the state of the Canadian economy, as well as the health of specific industries in the short- to medium-term. The information is used by both private and public sectors including Statistics Canada, federal and provincial governments, business and trade entities, international and domestic non-governmental organizations, consultants, the business press and private citizens. The data are used for analyzing market share, trends, corporate benchmarking, policy analysis, program development, tax policy and trade policy.

1. Sales of goods manufactured

Sales of goods manufactured (formerly shipments of goods manufactured) are defined as the value of goods manufactured by establishments that have been shipped to a customer. Sales of goods manufactured exclude any wholesaling activity, and any revenues from the rental of equipment or the sale of electricity. Note that in practice, some respondents report financial transactions rather than payments for work done. Sales of goods manufactured are available by 3-digit NAICS, for Canada and broken down by province.

For the aerospace product and parts, and shipbuilding industries, the value of production is used instead of sales of goods manufactured. This value is calculated by adjusting monthly sales of goods manufactured by the monthly change in inventories of goods / work in process and finished goods manufactured. Inventories of raw materials and components are not included in the calculation since production tries to measure "work done" during the month. This is done in order to reduce distortions caused by the sales of goods manufactured of high value items as completed sales.

2. Inventories

Measurement of component values of inventory is important for economic studies as well as for derivation of production values. Respondents are asked to report their book values (at cost) of raw materials and components, any goods / work in process, and finished goods manufactured inventories separately. In some cases, respondents estimate a total inventory figure, which is allocated on the basis of proportions reported on the ASML. Inventory levels are calculated on a Canada‑wide basis, not by province.

3. Orders

a) Unfilled Orders

Unfilled orders represent a backlog or stock of orders that will generate future sales of goods manufactured assuming that they are not cancelled. As with inventories, unfilled orders and new orders levels are calculated on a Canada‑wide basis, not by province.

The MSM produces estimates for unfilled orders for all industries except for those industries where orders are customarily filled from stocks on hand and order books are not generally maintained. In the case of the aircraft companies, options to purchase are not treated as orders until they are entered into the accounting system.

b) New Orders

New orders represent current demand for manufactured products. Estimates of new orders are derived from sales of goods manufactured and unfilled orders data. All sales of goods manufactured within a month result from either an order received during the month or at some earlier time. New orders can be calculated as the sum of sales of goods manufactured adjusted for the monthly change in unfilled orders.

4. Non-Durable / Durable goods

a) Non-durable goods industries include:

Food (NAICS 311),
Beverage and Tobacco Products (312),
Textile Mills (313),
Textile Product Mills (314),
Clothing (315),
Leather and Allied Products (316),
Paper (322),
Printing and Related Support Activities (323),
Petroleum and Coal Products (324),
Chemicals (325) and
Plastic and Rubber Products (326).

b) Durable goods industries include:

Wood Products (NAICS 321),
Non-Metallic Mineral Products (327),
Primary Metals (331),
Fabricated Metal Products (332),
Machinery (333),
Computer and Electronic Products (334),
Electrical Equipment, Appliance and Components (335),
Transportation Equipment (336),
Furniture and Related Products (337) and
Miscellaneous Manufacturing (339).

Survey design and methodology

Concept Review

In 2007, the MSM terminology was updated to be Charter of Accounts (COA) compliant. With the August 2007 reference month release the MSM has harmonized its concepts to the ASML. The variable formerly called “Shipments” is now called “Sales of goods manufactured”. As well, minor modifications were made to the inventory component names. The definitions have not been modified nor has the information collected from the survey.

Methodology

The latest sample design incorporates the 2012 North American Industrial Classification Standard (NAICS). Stratification is done by province with equal quality requirements for each province. Large size units are selected with certainty and small units are selected with a probability based on the desired quality of the estimate within a cell.

The estimation system generates estimates using the NAICS. The estimates will also continue to be reconciled to the ASML. Provincial estimates for all variables will be produced. A measure of quality (CV) will also be produced.

Components of the Survey Design

Target Population and Sampling Frame

Statistics Canada’s business register provides the sampling frame for the MSM. The target population for the MSM consists of all statistical establishments on the business register that are classified to the manufacturing sector (by NAICS). The sampling frame for the MSM is determined from the target population after subtracting establishments that represent the bottom 5% of the total manufacturing sales of goods manufactured estimate for each province. These establishments were excluded from the frame so that the sample size could be reduced without significantly affecting quality.

The Sample

The MSM sample is a probability sample comprised of approximately 10,500 establishments. A new sample was chosen in the autumn of 2012, followed by a six-month parallel run (from reference month September 2012 to reference month February 2013). The refreshed sample officially became the new sample of the MSM effective in December 2012.

This marks the first process of refreshing the MSM sample since 2007. The objective of the process is to keep the sample frame as fresh and up-to date as possible. All establishments in the sample are refreshed to take into account changes in their value of sales of goods manufactured, the removal of dead units from the sample and some small units are rotated out of the GST-based portion of the sample, while others are rotated into the sample.

Prior to selection, the sampling frame is subdivided into industry-province cells. For the most part, NAICS codes were used. Depending upon the number of establishments within each cell, further subdivisions were made to group similar sized establishments’ together (called stratum). An establishment’s size was based on its most recently available annual sales of goods manufactured or sales value.

Each industry by province cell has a ‘take-all’ stratum composed of establishments sampled each month with certainty. This ‘take-all’ stratum is composed of establishments that are the largest statistical enterprises, and have the largest impact on estimates within a particular industry by province cell. These large statistical enterprises comprise 45% of the national manufacturing sales of goods manufactured estimates.

Each industry by province cell can have at most three ‘take-some’ strata. Not all establishments within these stratums need to be sampled with certainty. A random sample is drawn from the remaining strata. The responses from these sampled establishments are weighted according to the inverse of their probability of selection. In cells with take-some portion, a minimum sample of 10 was imposed to increase stability.

The take-none portion of the sample is now estimated from administrative data and as a result, 100% of the sample universe is covered. Estimation of the take-none portion also improved efficiency as a larger take-none portion was delineated and the sample could be used more efficiently on the smaller sampled portion of the frame.

Data Collection

Only a subset of the sample establishments is sent out for data collection. For the remaining units, information from administrative data files is used as a source for deriving sales of goods manufactured data. For those establishments that are surveyed, data collection, data capture, preliminary edit and follow-up of non-respondents are all performed in Statistics Canada regional offices. Sampled establishments are contacted by mail or telephone according to the preference of the respondent. Data capture and preliminary editing are performed simultaneously to ensure the validity of the data.

In some cases, combined reports are received from enterprises or companies with more than one establishment in the sample where respondents prefer not to provide individual establishment reports. Businesses, which do not report or whose reports contain errors, are followed up immediately.

Use of Administrative Data

Managing response burden is an ongoing challenge for Statistics Canada. In an attempt to alleviate response burden, especially for small businesses, Statistics Canada has been investigating various alternatives to survey taking. Administrative data files are a rich source of information for business data and Statistics Canada is working at mining this rich data source to its full potential. As such, effective the August 2004 reference month, the MSM reduced the number of simple establishments in the sample that are surveyed directly and instead, derives sales of goods manufactured data for these establishments from Goods and Services Tax (GST) files using a statistical model. The model accounts for the difference between sales of goods manufactured (reported to MSM) and sales (reported for GST purposes) as well as the time lag between the reference period of the survey and the reference period of the GST file.

Effective from the January 2013 reference month, the MSM derives sales of goods manufactured data for non-incorporated establishments (e.g. the self employed) from T1 files. A statistical model is used to transform T1 data into sales of goods manufactured data.

In conjunction with the most recent sample, effective December 2012, approximately 2,800 simple establishments were selected to represent the GST portion of the sample.

Inventories and unfilled orders estimates for establishments where sales of goods manufactured are GST-based are derived using the MSM’s imputation system. The imputation system applies to the previous month values, the month-to-month and year-to-year changes in similar firms which are surveyed. With the most recent sample, the eligibility rules for GST-based establishments were refined to have more GST-based establishments in industries that typically carry fewer inventories. This way the impact of the GST-based establishments which require the estimation of inventories, will be kept to a minimum.

Detailed information on the methodology used for modelling sales of goods manufactured from administrative data sources can be found in the ‘Monthly Survey of Manufacturing: Use of Administrative Data’ (Catalogue no. 31-533-XIE) document.

Data quality

Statistical Edit and Imputation

Data are analyzed within each industry-province cell. Extreme values are listed for inspection by the magnitude of the deviation from average behavior. Respondents are contacted to verify extreme values. Records that fail statistical edits are considered outliers and are not used for imputation.

Values are imputed for the non-responses, for establishments that do not report or only partially complete the survey form. A number of imputation methods are used depending on the variable requiring treatment. Methods include using industry-province cell trends, historical responses, or reference to the ASML. Following imputation, the MSM staff performs a final verification of the responses that have been imputed.

Revisions

In conjunction with preliminary estimates for the current month, estimates for the previous three months are revised to account for any late returns. Data are revised when late responses are received or if an incorrect response was recorded earlier.

Estimation

Estimates are produced based on returns from a sample of manufacturing establishments in combination with administrative data for a portion of the smallest establishments. The survey sample includes 100% coverage of the large manufacturing establishments in each industry by province, plus partial coverage of the medium and small-sized firms. Combined reports from multi-unit companies are pro-rated among their establishments and adjustments for progress billings reflect revenues received for work done on large item contracts. Approximately 2,800 of the sampled medium and small-sized establishments are not sent questionnaires, but instead their sales of goods manufactured are derived by using revenue from the GST files. The portion not represented through sampling – the take-none portion - consist of establishments below specified thresholds in each province and industry. Sub-totals for this portion are also derived based on their revenues.

Industry values of sales of goods manufactured, inventories and unfilled orders are estimated by first weighting the survey responses, the values derived from the GST files and the imputations by the number of establishments each represents. The weighted estimates are then summed with the take-none portion. While sales of goods manufactured estimates are produced by province, no geographical detail is compiled for inventories and orders since many firms cannot report book values of these items monthly.

Benchmarking

Up to and including 2003, the MSM was benchmarked to the Annual Survey of Manufactures and Logging (ASML). Benchmarking was the regular review of the MSM estimates in the context of the annual data provided by the ASML. Benchmarking re-aligned the annualized level of the MSM based on the latest verified annual data provided by the ASML.

Significant research by Statistics Canada in 2006-2007 was completed on whether the benchmark process should be maintained. The conclusion was that benchmarking of the MSM estimates to the ASML should be discontinued. With the refreshing of the MSM sample in 2007, it was determined that benchmarking would no longer be required (retroactive to 2004) because the MSM now accurately represented 100% of the sample universe. Data confrontation will continue between MSM and ASML to resolve potential discrepancies.

As of the December 2012 reference month, a new sample was introduced. It is standard practice that every few years the sample is refreshed to ensure that the survey frame is up to date with births, deaths and other changes in the population. The refreshed sample is linked at the detailed level to prevent data breaks and to ensure the continuity of time series. It is designed to be more representative of the manufacturing industry at both the national and provincial levels.

Data confrontation and reconciliation

Each year, during the period when the Annual Survey of Manufactures and Logging section set their annual estimates, the MSM section works with the ASML section to confront and reconcile significant differences in values between the fiscal ASML and the annual MSM at the strata and industry level.

The purpose of this exercise of data reconciliation is to highlight and resolve significant differences between the two surveys and to assist in minimizing the differences in the micro-data between the MSM and the ASML.

Sampling and Non-sampling Errors

The statistics in this publication are estimates derived from a sample survey and, as such, can be subject to errors. The following material is provided to assist the reader in the interpretation of the estimates published.

Estimates derived from a sample survey are subject to a number of different kinds of errors. These errors can be broken down into two major types: sampling and non-sampling.

1. Sampling Errors

Sampling errors are an inherent risk of sample surveys. They result from the difference between the value of a variable if it is randomly sampled and its value if a census is taken (or the average of all possible random values). These errors are present because observations are made only on a sample and not on the entire population.

The sampling error depends on factors such as the size of the sample, variability in the population, sampling design and method of estimation. For example, for a given sample size, the sampling error will depend on the stratification procedure employed, allocation of the sample, choice of the sampling units and method of selection. (Further, even for the same sampling design, we can make different calculations to arrive at the most efficient estimation procedure.) The most important feature of probability sampling is that the sampling error can be measured from the sample itself.

2. Non-sampling Errors

Non-sampling errors result from a systematic flaw in the structure of the data-collection procedure or design of any or all variables examined. They create a difference between the value of a variable obtained by sampling or census methods and the variable’s true value. These errors are present whether a sample or a complete census of the population is taken. Non-sampling errors can be attributed to one or more of the following sources:

a) Coverage error: This error can result from incomplete listing and inadequate coverage of the population of interest.

b) Data response error: This error may be due to questionnaire design, the characteristics of a question, inability or unwillingness of the respondent to provide correct information, misinterpretation of the questions or definitional problems.

c) Non-response error: Some respondents may refuse to answer questions, some may be unable to respond, and others may be too late in responding. Data for the non-responding units can be imputed using the data from responding units or some earlier data on the non-responding units if available.

The extent of error due to imputation is usually unknown and is very much dependent on any characteristic differences between the respondent group and the non-respondent group in the survey. This error generally decreases with increases in the response rate and attempts are therefore made to obtain as high a response rate as possible.

d) Processing error: These errors may occur at various stages of processing such as coding, data entry, verification, editing, weighting, and tabulation, etc. Non-sampling errors are difficult to measure. More important, non-sampling errors require control at the level at which their presence does not impair the use and interpretation of the results.

Measures have been undertaken to minimize the non-sampling errors. For example, units have been defined in a most precise manner and the most up-to-date listings have been used. Questionnaires have been carefully designed to minimize different interpretations. As well, detailed acceptance testing has been carried out for the different stages of editing and processing and every possible effort has been made to reduce the non-response rate as well as the response burden.

Measures of Sampling and Non-sampling Errors

1. Sampling Error Measures

The sample used in this survey is one of a large number of all possible samples of the same size that could have been selected using the same sample design under the same general conditions. If it was possible that each one of these samples could be surveyed under essentially the same conditions, with an estimate calculated from each sample, it would be expected that the sample estimates would differ from each other.

The average estimate derived from all these possible sample estimates is termed the expected value. The expected value can also be expressed as the value that would be obtained if a census enumeration were taken under identical conditions of collection and processing. An estimate calculated from a sample survey is said to be precise if it is near the expected value.

Sample estimates may differ from this expected value of the estimates. However, since the estimate is based on a probability sample, the variability of the sample estimate with respect to its expected value can be measured. The variance of an estimate is a measure of the precision of the sample estimate and is defined as the average, over all possible samples, of the squared difference of the estimate from its expected value.

The standard error is a measure of precision in absolute terms. The coefficient of variation (CV), defined as the standard error divided by the sample estimate, is a measure of precision in relative terms. For comparison purposes, one may more readily compare the sampling error of one estimate to the sampling error of another estimate by using the coefficient of variation.

In this publication, the coefficient of variation is used to measure the sampling error of the estimates. However, since the coefficient of variation published for this survey is calculated from the responses of individual units, it also measures some non-sampling error.

The formula used to calculate the published coefficients of variation (CV) in Table 1 is:

CV(X) = S(X)/X

where X denotes the estimate and S(X) denotes the standard error of X.

In this publication, the coefficient of variation is expressed as a percentage.

Confidence intervals can be constructed around the estimate using the estimate and the coefficient of variation. Thus, for our sample, it is possible to state with a given level of confidence that the expected value will fall within the confidence interval constructed around the estimate. For example, if an estimate of $12,000,000 has a coefficient of variation of 10%, the standard error will be $1,200,000 or the estimate multiplied by the coefficient of variation. It can then be stated with 68% confidence that the expected value will fall within the interval whose length equals the standard deviation about the estimate, i.e., between $10,800,000 and $13,200,000. Alternatively, it can be stated with 95% confidence that the expected value will fall within the interval whose length equals two standard deviations about the estimate, i.e., between $9,600,000 and $14,400,000.

Text table 1 contains the national level CVs, expressed as a percentage, for all manufacturing for the MSM characteristics. For CVs at other aggregate levels, contact the Dissemination and Frame Services Section at (613) 951-9497, toll free: 1-866-873-8789 or by e-mail at manufact@statcan.gc.ca.

Text table 1: National Level CVs by Characteristic
Table summary
This table displays the results of text table 1: national level cvs by characteristic. The information is grouped by month (appearing as row headers), sales of goods manufactured, raw materials and components, goods / work in process inventories, finished goods manufactured inventories, unfilled orders and inventories, calculated using % units of measure (appearing as column headers).

Month	Sales of goods manufactured	Raw materials and components	Goods / work in process inventories	Finished goods manufactured inventories	Unfilled Orders
		inventories
	%	%	%	%	%
March 2012	0.89	1.37	1.80	1.38	2.97
April 2012	0.87	1.37	1.82	1.40	2.94
May 2012	0.89	1.29	1.76	1.46	2.89
June 2012	0.90	1.28	1.81	1.48	2.80
July 2012	0.86	1.32	1.91	1.51	2.22
August 2012	0.86	1.25	1.78	1.44	1.70
September 2012	0.82	1.27	1.78	1.43	1.42
October 2012	0.87	1.31	1.75	1.43	1.32
November 2012	0.87	1.31	1.80	1.40	1.27
December 2012	0.41	0.96	1.54	1.37	0.90
January 2013	0.43	0.96	1.39	0.91	0.86
February 2013	0.42	0.94	1.27	0.97	0.84
March 2013	0.43	1.04	1.20	1.16	0.89

 

2. Non-sampling Error Measures

The exact population value is aimed at or desired by both a sample survey as well as a census. We say the estimate is accurate if it is near this value. Although this value is desired, we cannot assume that the exact value of every unit in the population or sample can be obtained and processed without error. Any difference between the expected value and the exact population value is termed the bias. Systematic biases in the data cannot be measured by the probability measures of sampling error as previously described. The accuracy of a survey estimate is determined by the joint effect of sampling and non-sampling errors.

Sources of non-sampling error in the MSM include non-response error, imputation error and the error due to editing. To assist users in evaluating these errors, weighted rates are given in Text table 2. The following is an example of what is meant by a weighted rate. A cell with a sample of 20 units in which five respond for a particular month would have a response rate of 25%. If these five reporting units represented $8 million out of a total estimate of $10 million, the weighted response rate would be 80%.

The definitions for the weighted rates noted in Text table 2 follow. The weighted response and edited rate is the proportion of a characteristic’s total estimate that is based upon reported data and includes data that has been edited. The weighted imputation rate is the proportion of a characteristic’s total estimate that is based upon imputed data. The weighted GST data rate is the proportion of the characteristic’s total estimate that is derived from Goods and Services Tax files (GST files). The weighted take-none fraction rate is the proportion of the characteristic’s total estimate modeled from administrative data.

Text table 2 contains the weighted rates for each of the characteristics at the national level for all of manufacturing. In the table, the rates are expressed as percentages.

Text Table 2: National Weighted Rates by Source and Characteristic
Table summary
This table displays the results of text table 2: national weighted rates by source and characteristic. The information is grouped by characteristics (appearing as row headers), data source, response or edited, imputed, gst data and take-none fraction, calculated using % units of measure (appearing as column headers).

Characteristics	Data source
	Response or edited	Imputed	GST data	Take-none fraction
	%	%	%	%
Sales of goods manufactured	92.18	3.96	0.36	3.50
Raw materials and components	77.42	18.20	0.00	4.38
Goods / work in process	81.58	14.91	0.00	3.51
Finished goods manufactured	79.55	16.67	0.00	3.78
Unfilled Orders	89.88	6.78	0.00	3.34

Joint Interpretation of Measures of Error

The measure of non-response error as well as the coefficient of variation must be considered jointly to have an overview of the quality of the estimates. The lower the coefficient of variation and the higher the weighted response rate, the better will be the published estimate.

Seasonal Adjustment

Economic time series contain the elements essential to the description, explanation and forecasting of the behavior of an economic phenomenon. They are statistical records of the evolution of economic processes through time. In using time series to observe economic activity, economists and statisticians have identified four characteristic behavioral components: the long-term movement or trend, the cycle, the seasonal variations and the irregular fluctuations. These movements are caused by various economic, climatic or institutional factors. The seasonal variations occur periodically on a more or less regular basis over the course of a year. These variations occur as a result of seasonal changes in weather, statutory holidays and other events that occur at fairly regular intervals and thus have a significant impact on the rate of economic activity.

In the interest of accurately interpreting the fundamental evolution of an economic phenomenon and producing forecasts of superior quality, Statistics Canada uses the X12-ARIMA seasonal adjustment method to seasonally adjust its time series. This method minimizes the impact of seasonal variations on the series and essentially consists of adding one year of estimated raw data to the end of the original series before it is seasonally adjusted per se. The estimated data are derived from forecasts using ARIMA (Auto Regressive Integrated Moving Average) models of the Box-Jenkins type.

The X-12 program uses primarily a ratio-to-moving average method. It is used to smooth the modified series and obtain a preliminary estimate of the trend-cycle. It also calculates the ratios of the original series (fitted) to the estimates of the trend-cycle and estimates the seasonal factors from these ratios. The final seasonal factors are produced only after these operations have been repeated several times. The technique that is used essentially consists of first correcting the initial series for all sorts of undesirable effects, such as the trading-day and the Easter holiday effects, by a module called regARIMA. These effects are then estimated using regression models with ARIMA errors. The series can also be extrapolated for at least one year by using the model. Subsequently, the raw series, pre-adjusted and extrapolated if applicable, is seasonally adjusted by the X-12 method.

The procedures to determine the seasonal factors necessary to calculate the final seasonally adjusted data are executed every month. This approach ensures that the estimated seasonal factors are derived from an unadjusted series that includes all the available information about the series, i.e. the current month's unadjusted data as well as the previous month's revised unadjusted data.

While seasonal adjustment permits a better understanding of the underlying trend-cycle of a series, the seasonally adjusted series still contains an irregular component. Slight month-to-month variations in the seasonally adjusted series may be simple irregular movements. To get a better idea of the underlying trend, users should examine several months of the seasonally adjusted series.

The aggregated Canada level series are now seasonally adjusted directly, meaning that the seasonally adjusted totals are obtained via X12-ARIMA. Afterwards, these totals are used to reconcile the provincial total series which have been seasonally adjusted individually.

For other aggregated series, indirect seasonal adjustments are used. In other words, their seasonally adjusted totals are derived indirectly by the summation of the individually seasonally adjusted kinds of business.

Trend

A seasonally adjusted series may contain the effects of irregular influences and special circumstances and these can mask the trend. The short term trend shows the underlying direction in seasonally adjusted series by averaging across months, thus smoothing out the effects of irregular influences. The result is a more stable series. The trend for the last month may be subject to significant revision as values in future months are included in the averaging process.

Real manufacturing sales of goods manufactured, inventories, and orders

Changes in the values of the data reported by the Monthly Survey of Manufacturing (MSM) may be attributable to changes in their prices or to the quantities measured, or both. To study the activity of the manufacturing sector, it is often desirable to separate out the variations due to price changes from those of the quantities produced. This adjustment is known as deflation.

Deflation consists in dividing the values at current prices obtained from the survey by suitable price indexes in order to obtain estimates evaluated at the prices of a previous period, currently the year 2007. The resulting deflated values are said to be “at 2007 prices”. Note that the expression “at current prices” refer to the time the activity took place, not to the present time, nor to the time of compilation.

The deflated MSM estimates reflect the prices that prevailed in 2007. This is called the base year. The year 2007 was chosen as base year since it corresponds to that of the price indexes used in the deflation of the MSM estimates. Using the prices of a base year to measure current activity provides a representative measurement of the current volume of activity with respect to that base year. Current movements in the volume are appropriately reflected in the constant price measures only if the current relative importance of the industries is not very different from that in the base year.

The deflation of the MSM estimates is performed at a very fine industry detail, equivalent to the 6-digit industry classes of the North American Industry Classification System (NAICS). For each industry at this level of detail, the price indexes used are composite indexes which describe the price movements for the various groups of goods produced by that industry.

With very few exceptions the price indexes are weighted averages of the Industrial Product Price Indexes (IPPI). The weights are derived from the annual Canadian Input-Output tables and change from year to year. Since the Input-Output tables only become available with a delay of about two and a half years, the weights used for the most current years are based on the last available Input-Output tables.

The same price index is used to deflate sales of goods manufactured, new orders and unfilled orders of an industry. The weights used in the compilation of this price index are derived from the output tables, evaluated at producer’s prices. Producer prices reflect the prices of the goods at the gate of the manufacturing establishment and exclude such items as transportation charges, taxes on products, etc. The resulting price index for each industry thus reflects the output of the establishments in that industry.

The price indexes used for deflating the goods / work in process and the finished goods manufactured inventories of an industry are moving averages of the price index used for sales of goods manufactured. For goods / work in process inventories, the number of terms in the moving average corresponds to the duration of the production process. The duration is calculated as the average over the previous 48 months of the ratio of end of month goods / work in process inventories to the output of the industry, which is equal to sales of goods manufactured plus the changes in both goods / work in process and finished goods manufactured inventories.

For finished goods manufactured inventories, the number of terms in the moving average reflects the length of time a finished product remains in stock. This number, known as the inventory turnover period, is calculated as the average over the previous 48 months of the ratio of end-of-month finished goods manufactured inventory to sales of goods manufactured.

To deflate raw materials and components inventories, price indexes for raw materials consumption are obtained as weighted averages of the IPPIs. The weights used are derived from the input tables evaluated at purchaser’s prices, i.e. these prices include such elements as wholesaling margins, transportation charges, and taxes on products, etc. The resulting price index thus reflects the cost structure in raw materials and components for each industry.

The raw materials and components inventories are then deflated using a moving average of the price index for raw materials consumption. The number of terms in the moving average corresponds to the rate of consumption of raw materials. This rate is calculated as the average over the previous four years of the ratio of end-of-year raw materials and components inventories to the intermediate inputs of the industry.