Therapeutic Abortion Survey

Detailed information for 2006





Record number:


The purpose of the Therapeutic Abortion Survey is to provide some basic indicators (for example, counts and rates) on induced abortions. Information from this database is also used in the calculation of pregnancy statistics, especially for teen pregnancies.

Data release - August 24, 2009 (For additional information regarding subsequent therapeutic abortion data, please consult the current CIHI website.)


The annual Therapeutic Abortion Survey collects information on demographic and medical characteristics of women obtaining an induced abortion legally in Canada. Some data were also collected on Canadian residents who obtain an induced abortion legally in selected American states prior to 2004.

The purpose of the Therapeutic Abortion Survey is to provide some basic indicators (for example, counts and rates) on induced abortions. Information from this database is also used in the calculation of pregnancy statistics, especially for teen pregnancies.
The title of the survey retains the word 'therapeutic' for the sake of historical continuity. When the survey began in late 1969, a woman could only obtain an abortion for health reasons. In 1988, with the removal of abortion from the Criminal Code, a reason for obtaining an abortion was no longer required.

At Statistics Canada, the term Therapeutic Abortion Survey is the official title of the program for the collection, compilation and dissemination of induced abortion statistics. When referring to the actual data files, the term Therapeutic Abortion Survey database may be used. At CIHI, both the program and the actual database are referred to as the Therapeutic Abortions Database.

In 1995, the collection of the data for this survey was transferred from Statistics Canada to the Canadian Institute for Health Information (CIHI). Once the data were collected, they were transferred to Statistics Canada to be released in The Daily, Statistics Canada's main data release vehicle. This process has caused some confusion with the users of the data.

Statistics Canada has decided, in consultation with CIHI that the responsibility for disseminating these data should reside with CIHI. This move will bring the Therapeutic Abortion Survey in line with other programs at CIHI.

As a result, the 2006 data is the final year therapeutic abortion data will be released by Statistics Canada. For additional information regarding subsequent therapeutic abortion data, please consult the current CIHI website.

Reference period: Calendar year

Collection period: The annual file that the Canadian Institute for Health Information sends to Statistics Canada is ready for release approximately 25 months after the reference period.


  • Health
  • Health care services
  • Pregnancy and births

Data sources and methodology

Target population

The conceptual universe includes all induced abortions performed in Canada. The survey collects information on women obtaining an abortion in hospitals and clinics in Canada. Some information was also collected on Canadian residents who obtain an abortion in some American states, especially those along the border from 1998 to 2003.

Instrument design

Data in the Therapeutic Abortion Survey are collected partly from detailed administrative records and partly from a one-page direct survey. This section on instrument design pertains only to the one-page survey.

The Canadian Institute for Health Information sends a one-page survey to private clinics in Canada who supply aggregate counts only. The survey form requests the following information:

1) Clinic information such as name, address and telephone number. (This information is not included in the database.)
2) Total number of induced abortions performed
3) Total number of induced abortions to Canadian residents
4) Total number of induced abortions to Canadian residents by province of residence
5) Authorization to publish data

From 1998 to 2003, CIHI also sent a form to the departments of health of selected American states requesting the counts of induced abortions performed on Canadian residents in their jurisdictions. These counts are aggregated by age group but do not specify the province or territory of residence, or whether the abortion was done in a clinic or in a hospital.


This survey is a census with a cross-sectional design.

Data sources

Responding to this survey is voluntary.

Data are extracted from administrative files.

Statistics Canada was responsible for data collection from late 1969, when the survey began, until the end of the 1994 data year. As of the 1995 data year, the Canadian Institute for Health Information assumed this responsibility.

There are several data sources and formats used in the creation of the Therapeutic Abortion Survey database. CIHI obtains data from provincial/territorial departments of health, or directly from hospitals and clinics. Depending upon the source, the format can also vary from a single sheet of paper with aggregate counts to detailed electronic records submitted through CIHI's Discharge Abstract Database (DAD).

Listed below are the data elements that are collected or derived for inclusion in the Therapeutic Abortion Survey database maintained at Statistics Canada. The following information, however, is not available for all induced abortions reported to the survey. Some respondents supply only aggregate counts. Records are generated for these counts and included in the database, but most of the data element fields remain blank.

Data Elements Collected:

- Province of report
- Month of report
- Survey year
- Province of residence of the patient
- Age or Age Group
- Marital status (see note at end of list)
- First day and month of last normal menses, or reported number of weeks of gestation
- Date fetus removed
- Days of inpatient care
- Number of previous deliveries
- Number of previous spontaneous abortions
- Number of previous induced abortions
- Initial surgical procedure
- Subsequent surgical procedure
- Type of sterilization
- Complication (up to 3 complications can be reported)

Note on Marital Status: As of data year 2001, information on marital status is significantly reduced. The Discharge Abstract Database, the major source for hospital abortion data, no longer collects marital status as of fiscal 2001/2002. As well, detailed records from Quebec no longer contain this information. In 2003, only 3% of total records (hospital and clinic abortions) contain a known marital status. In 2004, only Manitoba hospitals submitted marital status for January to March of 2004. As of April 2004, all Manitoba data were downloaded from the Discharge Abstract Database. As of 2005, no marital status data are reported anywhere.

Derived Data Elements:

- Type of report (that is, data source)
- Hospital / Clinic flag (indicates type of facility performing the abortion)
- Age Group
- Generated record flag (indicates if the record was generated)
- Gestation period expressed in days
- Gestation period expressed in weeks completed
- Gestation period expressed in weeks rounded up
- Gestation period expressed in ranges of weeks (for example, 9-12 weeks)
- Indication of residence (indicates if province of report = province of residence)

Error detection

The Canadian Institute for Health Information submits records to an edit process that checks for internal consistencies, compatibilities and completeness of each data item reported. There are 30 edits and 40 cross-edits. The edit system is reviewed and updated periodically. Invalid codes are reformatted to default values.


Generated records

Beginning in the data year 1995, records are generated in the database for abortion cases that were reported only as aggregate counts. For example, in data year 2006, records were generated for 49,360 abortions (or 54% of the database). Some data elements on these generated records are assigned values of 'unknown' or 'not available', while other data elements fields are left blank.

Area of residence

British Columbia clinics reported either Canadian or non-Canadian residency, or did not report a place of residence at all. All counts that indicated Canadian residency, or where no residency was reported, were imputed to residents of British Columbia. As of 2004, hospital data for British Columbia are downloaded through the Discharge Abstract Database which provides residency information.

Age group

In data year 2006, age group was the only other data element for which a value other than unknown or unspecified was imputed. Age (in single years) is usually reported for hospital abortions, but an age or age group is often not available for many abortions performed in clinics.

Some data suppliers who provide counts rather than detailed records supply these counts aggregated by age group. Using this information, age groups are imputed on the records generated for these counts.

Other data sources (that is, some private clinics) provide only counts. In such cases, the provincial age group distribution observed for abortions performed in hospitals in the province is used to impute age groups on records generated for clinic counts. In the case of Prince Edward Island residents who obtained an abortion in another province, the Canadian age distribution for hospital abortions is used to impute age groups.


Teen age subgroups (15 to 17 and 18 to 19)

Induced abortion data are required in the compilation of pregnancy statistics, especially for teenage pregnancies. Traditionally, teen pregnancy statistics provide information by the subgroups 15 to 17 and 18 to 19. These subgroups are not available or imputed on the Therapeutic Abortion Survey database. Therefore, these subgroups must be estimated when calculating pregnancy counts and rates. For abortion cases where age in single years is not provided, the estimation of the subgroups is based upon the provincial age distribution observed for those cases that did have a reported age in single years.

Quality evaluation

Upon receipt of an annual file and supporting documentation from the Canadian Institute for Health Information (CIHI), Statistics Canada carries out a series of quality checks that include: 1) carefully reviewing the record layout and data processing reports, 2) producing basic tabulations to ensure that STC and CIHI obtain the same results, 3) checking for internal consistencies, for example, running frequencies on certain data elements, and 4) comparing the most recent data year with past data years to detect any unusual or unexpected changes. After Statistics Canada approves the final file, a public release is announced in The Daily.

CIHI and STC communicate continually throughout the year on matters concerning the Therapeutic Abortion Survey.

The document 'Data quality documentation: Executive Summary -- Data Limitations Therapeutic Abortions Database 2003 is available on the CIHI website (

Disclosure control

Statistics Canada is prohibited by law from releasing any data which would divulge information obtained under the Statistics Act that relates to any identifiable person, business or organization without the prior knowledge or the consent in writing of that person, business or organization. Various confidentiality rules are applied to all data that are released or published to prevent the publication or disclosure of any information deemed confidential. If necessary, data are suppressed to prevent direct or residual disclosure of identifiable data.

The Therapeutic Abortion Survey database maintained at Statistics Canada does not contain personal identifiable information such as patient name, date of birth, health number, or facility names or numbers. Geographic information on the location of residence of the patient and the location where the abortion was performed is available at the provincial and national levels only.

Before releasing annual data to the public, Statistics Canada requests authorization from those provinces and territories that supplied data for three or fewer hospitals or three or fewer clinics. Private clinics sign an 'Authorization to Publish Data' at the same time as they submit their counts directly to the Canadian Institute for Health Information.

In regard to client requests for cross-classified data, Statistics Canada does not supply any tables that could identify a person or facility. As a guideline, a cell with a count of under 5 is suppressed, and in order to avoid residual disclosure, other cells are also chosen for suppression.

At CIHI, data disclosure is determined by their Privacy Principles and Policies in conjunction with the provincial/territorial ministries of health. A Privacy Impact Assessment of the Therapeutic Abortions Database is available on CIHI's website (

Revisions and seasonal adjustment

Induced Abortions - time series footnotes

Caution should be taken when comparing data for 1999 and later with data from earlier years. As of 1999, the Ontario Ministry of Health and Long-term Care no longer maintain a system for the collection of detailed information on abortions performed in clinics in Ontario. Ontario now uses the billing system of the Ontario Health Insurance Plan (OHIP) to provide counts of clinic abortions to the Therapeutic Abortion Survey. As a result, information is no longer available for clinic abortions performed on non residents of Ontario or on Ontario residents who do not submit a claim to OHIP. A comparison of the data collected using both sources for the years 1995 to 1998 shows that the new data source underestimated clinic abortions performed on Ontario residents by an average of 5.4% (950 cases) per year, or approximately 1% of all abortions performed in Canada. The new source does not include abortions performed on non-resident that averaged 70 cases per year or .4% of total clinic abortions performed in Ontario.

Similarly, data submitted by the Quebec Ministry of Health includes data on Quebec residents only. As such, it is possible that the number of abortions reported is lower than the actual number of abortions performed.

Data accuracy


For data years 1970 to 1987, the coverage of the survey was considered to be 100% of all induced abortions performed in Canada.

In January 1988, the Supreme Court of Canada struck down the 1969 abortion law which had contained a clause requiring the collection of data on all induced abortions performed in Canada. With the absence of any legal requirement for reporting, some respondents began to submit only counts of abortions rather than detailed records on each abortion. CIHI has estimated that as of the 2000 data year the Therapeutic Abortion Survey database represents approximately 90% of all abortions performed in Canada on Canadian residents (Canadian Institute for Health Information. Privacy impact assessment of the Therapeutic Abortion Database. June 2003. Page 4.)

At the national level, the percentage of abortions performed on non-residents of Canada that are included in the database is not known. The percentage of abortions performed in the United States on Canadian residents that are represented in the database is also not known.

There is a possibility of some over-coverage in the Therapeutic Abortions Database. It is not possible to track a patient once they have been discharged. As such, if a patient is released but later requires another procedure as a result of complications, for example an incomplete abortion, it is possible that she would be counted twice in the database. This issue is particularly relevant with the growing popularity of medical abortions in Canada, which have a higher rate of incomplete abortions than do surgical abortions. If a woman seeks treatment in a clinic and then requires further intervention elsewhere, that case will be counted twice in the database.

Response rates

There are few instances of complete non-response. The response rate for hospitals is 100%. However, recently, there have been occasions when clinics did not report. The Manitoba clinic has not responded after 2003; one clinic in British Columbia did not provide information in 2003, while another did not respond in 2005; and finally in 2006, in addition to Manitoba, one clinic in New Brunswick and two in British Columbia provided no data. The missing information was imputed for the British Columbia clinics in 2003 and 2005, but no imputation was ever done for missing Manitoba data, nor for any of the other clinics that did not respond in 2006 -- see Imputation.

For the most part, the survey obtains at least some response from identified respondents. The type of response, however, ranges from the provision of aggregate counts of abortions performed to detailed records for each abortion case. This range in type of response is the major limitation of the Therapeutic Abortion Survey. In data year 2006, complete detailed records were available for only 46% of all reported abortions.

For more details, see the document 'Data Quality in the Therapeutic Abortion Survey' in the Documentation section.


Date modified: