Provincial Health Indicators Work Group

April 11 and 12, 2002
Committee Room 4, North York Civic Centre, Toronto
 
MINUTES


Chair: Mary-Anne Pietrusiak
Present: Sherri Ennis, Jennifer Pennock, Sue Schultz, Jo-Anne Peterson, Nam Bains, Jessie Wong, Carol Paul, Brenda Coleman, Heather Schouten (recorder), Julie Stratton, Doug Manuel

1.0 Introductions

2.0 Approval of minutes of September 17-18, 2001
Approved

3.0 Business arising from the minutes
 
       3.1 Terms of Reference
Reviewed and approved

3.2 Web site student
Hired in cooperation with CEHIP. It is not determined where the funding will come from at this point as Elizabeth Rael has just returned to the MOHLTC. Sten Ardal has done the work in hiring the student and has reassured Mary-Anne that the funding will be available and to continue with the work for the website.

3.3 Help from public health librarians
MeSH terms should be used where possible under keywords. Keywords will be reviewed by PH librarians and MeSH terms will be provided as appropriate. The decision was made not to limit the keywords to MeSH terms since that may limit the capacity for those who are not familiar with MeSH terms to conduct fruitful searches. As well, some indicators (e.g., health-related quality of life) might not have up-to-date MeSH terms. The search capability should be robust.

3.4 Citations
Recommendations exist for both numerator and denominator sources in the indicator itself. Citations will also be repeated in the data sources section.

3.5 Chart Book Pages
Sherri Ennis is chairing (and Mary-Anne Pietrusiak is co-chairing) a committee of HIU, Ministry, APHEO and DHC people to produce chart book pages, some of which will be based on the Core Indicators. Other indicators will come from the DHC System Monitoring Project. It has not been determined who is doing what work, although there is involvement from all HIUs. Some HIUs have already set their priorities and find it difficult to take on this project. Chart book pages will be presented at the public health unit/DHC level.
This is a different product than what is currently available from Statistics Canada. More information is available from the chart book page, as in where the data has come from, data limitations, etc.
PHIWG needs to identify priority indicators that the Chart Book Page Committee should start with. Things to consider when deciding priority indicators: timeliness of data, demand for data/information (how often and by whom), indicators in F/P/T list (Carol Paul to provide list).
 
Round-Table Updates: 
ICES
  • The OHA report (sent to PHIWG in e-mail) highlights the need to consider how new indicators can be added (standards/performance measures for hospital sector).
  • There has been a request by the MOHLTC to develop performance measures for public health
  • Doug and others at ICES are working on bringing together researchers on population health indicators to develop national consensus.
  • Attributable risk indicators are in development stage and may be of interest to this project.
  • Canadian Population Health Initiative – group from Dalhousie is doing some interesting work that will also contribute to this project. Suggested that we need to do further collaboration with Federal/National initiatives (Jason Gilmore may be helpful).
  • Atlas reports – health service reports like Mental Health, colorectal screening
    •
Health Canada: Portal Project
Centre for Surveillance Coordination is currently developing a public health portal to provide access to public health information.

RRFSS
Currently developing modules for perinatal health (infusion of money for surveillance of perinatal health)
Student developed process for analyzing and presenting data for RRFSS. Currently seeking permission to post “core” data from all participating health units. There are approximately 22 health units participating.
Indicator Sections

Potential Years of Life Lost (Doug Manuel, Sherri Ennis)
Some issues were not resolved in the last meeting. PYLL appears throughout the document. As a result there was a suggestion to have PYLL only once in the mortality section. Doug feels it is good to have a health gap reference, but PYLL is his least favorite. Reasons include the cut off age, the meaningless of the numbers that are calculated. Most use is in comparing causes of death in a ranking.
Other gap measures include standard expected life lost (DALYS) – gap between age of death and when you would be expected to die. For example, if you are 40 your life expectancy would be 80, if age was 75 life expectancy would be 84 etc. Use a life table to make the calculation. WHO is currently using this approach. Can use a ranking to display the results.
Decision – use PYLL as one indicator in the mortality section, and delete from other sections. Clarify that PYLL should only be used for comparisons of different disease groups – the numbers are meaningless on their own. Compose a list of future indicators and add DALY.


 Health-Related Quality of Life (Doug Manuel)
Doug received some comments about being more specific about methods. A link to the CEHIP website will be added to the indicator.
Issue: The questions for some indicators (e.g., long term disability, activity limitation) have been revised in the new CCHS but it is unclear what the derived variables will be until the CCHS files are available.
Decision – Do not change the indicators until the CCHS is released and it is clear what the derived variables and questions are for these indicators. There should still be time to do the edits for this summer. Re-assess on the importance of longitudinal comparison. CCHS changes may limit the ability to do comparisons over time.

 Introduction (Mary-Anne Pietrusiak)
Still needs to be revised. Topics to include: history, description of project, frequently asked questions, list of previous and past members etc. This will be more in tune with a web site format.

The group suggested some FAQ’s:
  • How were the indicators developed?
  • Why another indicator project?
  • Who developed the indicators?
  • Where’s the data?
  • Why do you need to standardize rates?
  • Why did you choose the 1991 Canadian population as a standard?
  • How did it all start?
 Population (Sherri Ennis)

Issue: Aboriginal population: if we calculate, there is an issue with the denominator.
Decision: Use Statistics Canada indicator, which uses adjusted population counts.

Issue: The National Indicators are using population estimates from the MOHLTC, rather than Statistics Canada estimates. For the most part, there is very little difference between them (see Sherri’s table at end of minutes).
Decision: We will continue to use the Statistics Canada population estimates that we get from the MOHLTC for consistency. The MOHLTC estimates are also limited to 1995-1998 and we do analysis of data prior to these years.
Action: Carol Paul will talk to Jason Gilmore about trying to make population estimates consistent – persuade them to use the Statistics Canada estimates.

Other edits:
  •  Change census term to state “data is based on a 20% sample” to replace “data represents a 20% sample”.

 Social Environment and Health (Sherri Ennis)

Issue: Dropping average personal income household (#10 page 24 in draft). May not be useful as we already capture low income and income inequality.
Decision: Drop this indicator

Issue: Indicator #11 (Income inequality) Definition from national indicators is difficult to understand.
Decision: To clarify: If 50% of households have 50% income, there will be no inequality. Wording should change on the definition.
Action: Sherri to follow-up with Jason Gilmore on the Stats Can technical notes

Issue: Distribution source is not written consistently
Decision: Consistent format will be developed and sent out

Other edits:
  • More references are required for this section. Request possible reference documents from reviewers.

 Physical Environment and Health (Mary-Anne Pietrusiak)
  • Air Quality: change from “Air Quality Index” to simply “Air Quality” as more indicators have been added. The index is dependent on ozone and is not sufficient on its own. Mary-Anne needs to clarify with reviewers on how the indicators will be operationalized as there are two different types of indicators. May need to update this indicator as changes in monitoring systems develops
  • Water: Municipal water quality replaces drinking water quality as indicator is limited to municipal water supplies only.
  • Water advisories changed from boil water advisories since there can be boil water advisories or drinking water advisories.
  • Beach/bathing water: information added to provide more explanation to the definitions. No indicators at this point on pools – need is questionable.
 Mortality (Sherri Ennis)

Avoidable mortality – suggest removing as there are not a lot of deaths and it is more of a system indicator. Include on the emerging/future indicator list.

Issue: Name of first indicator, suggestion to change name to “cause of death”
Decision: Keep name as “all cause mortality”

Issue: SMR calculation – multiply by 100?
Decision: Get rid of the 100 in the indicator descriptions. Have a template for calculations in appendix.



 Hospitalization (Sherri Ennis)

Issue: Less than one age-group it is mostly newborns. CIHI and ICES removes newborns from the less than one age-group calculations. Comment added under categories.
Decision: Sherri will add codes and information on analysis.
  • Calendar vs fiscal years: comment added
  • In-patient vs day-procedures: comment added.
  • Length of stay: different categories of stay and identified as separate attributes.
  • E codes: comment on how injury is treated differently in morbidity data vs mortality data added.
 Chronic Diseases (Jane Hohenadel, Jennifer Pennock)

Issue: Whether to include COLD coding (490-496) which includes asthma vs. COPD coding (490-492, 496) which excludes asthma. Asthma would then be separate category
Decision: Jennifer to review respiratory health report for rationale on definitions of COPD. “Respiratory Diseases in Canada”. Heather will follow-up at Health Canada for clarification from the Centre for Chronic Disease and the Centre for Surveillance Coordination. Julie looked up the differences between COPD and COLD that she came across in a text book – it indicated that the COLD coding is inappropriate.

Issue: Include prevalence of high blood pressure, arthritis and long-term disability?
Decision: Include in this section

 Cancer Incidence & Early Detection of Cancer (Heather Schouten, Jennifer Pennock)
  • Add note on the gaps in Ontario data for years 1994-1997
  • The reference to Health Indicators should be Volume 2001, No. 3, December 2001 for the most up-to-date information on cancer incidence.
  • Replace “k” with appropriate multiplier
 Injury Prevention & Substance Abuse Prevention (Jo-Anne Peterson, Julie Stratton)

Issue: Indicator #6 Use of medication.
Decision: Drop it
  • Poison Control Database: currently not in use by Poison Control Centres due to inability to pay for the implementation and maintenance of the surveillance that has been developed by the Centre for Surveillance Coordination, Health Canada
  • Indicator #9 MV Traffic Collisions: main indicator is all three categories combined and will be stipulated as such in specific indicator section.
  • Need to separate drivers into age-groups that are indicated by the MTO, but add category with standard age-groupings.
 Action: Julie Stratton to email Ed Adlaf to obtain comments on adolescent drug use indicator
Smoking

Issue: primary data source
Decision: Use CCHS as primary data source, and note where the questions have changed from the OHS/NPHS
  • Identify variable names in check-list (include questions in new category of Survey Questions)
  • CCHS – sharing files are supposed to be available in May
 Issue: OHS: check geography is consistent.
Decision: Recommend adding an appendix with geographic breakdown of local, regional areas. Action: Sherri and Nam will create this geography appendix.

Issue: should current and former smokers have their own indicator category?
Decision: Consider attributable risk – future indicator, but for now do not make separate categories for former and current
Categories: smoker (current daily and occasional, as stated in the stem question); daily (current daily).
  • Comment section will include notes on former smoker data.
    •
Alcohol

Issue: Should the age be changed to 19+ in the drinking and driving indicator? Age groups right now are 16-24 and 24+ for adults.
Decision: Data should be reviewed to determine if there is a break or defining line in ages. For now, calculate total as 16+, adults as 20+ to stay consistent with other OHS indicators.
May want to look at rationale for decisions for groupings in RRFSS. Consider grouping 18-24.

 Physical Activity
  • No issues reported
    •

Nutrition

Issue: BMI for children (persons below the age of 20)
Decision: Add to the list of future indicators as it is still in development. As a core indicator, use should be reserved for adults. When used for children or seniors, caution should be exercised (include in indicator comments). Also, to be consistent with National Indicators, people less than 3 feet, or greater than 6’11” should be excluded.

Issue: Fruit and Vegetable consumption. 
Decision: CCHS and serving derivation to be determined. There was conflicting information - CCHS nutritionist had initially indicated that number of servings could not be calculated. Will have to assess documentation.

Sexual Health (Evelyn Crosse)

  • Non-response categories
  • Age-groupings should not include 12-14 as we do not have data on this age group
  • Need for more reviewers on this section – Jann Houston, Manager of Sexual Health, Planning and Policy may be an option.
  • End at age 59 as seniors typically do not respond.
Reproductive Health (Brenda Coleman)
  • Age-groupings are difficult to determine, and comments are added in the notes section to reflect considerations in age-groups.
  • Births from 10-14 are in numerator, but not included in the denominator.
 Issue: Indicator rate of multiple births – live births or all multiple births (live and still)?
Decision: Capture all multiple births (live and still) as the indicator, and then break it down for further analysis.
  • Analysis checklist notes: Use weight (500 grams).
  • Abortion geographical locations: recommended to not go lower than public health unit area (too much variability in census division from year to year). Questionable if the data can be pulled from OHIP (easy to get the data from CIHI, but it is not certain what data is provided from independent clinics).
  • Neural tube defects: total number of NTD required, although there may be double counting with multiple conditions diagnosed for one child. Elizabeth Rael was planning to review the data.
Action: Mary-Anne to follow up with Elizabeth.
  • Infant mortality rate: very small. Stillbirths come out of the early neonatal mortality.
  • Folic Acid Supplementation: Brenda to investigate timelines pre-post conception.
  • RRFSS and NLSCY are alternative data sources.
  • ISCIS: use of this data (pre-natal assessment, Parkin to identify high risk new moms) is limited at this point. It is has been identified that the data can be used for indicators – the ISCIS people at the Ministry are creating a program that will create a flat file with identified variables for each health unit (ISCIS doing to do the programming, each health unit can pull their information).
 Child and Adolescent Health (Ruth Sanderson)

Section not completed yet

 Mental Health (Jessie Wong)

Issue: Jessie resigning from working group.
Decision: Jessie will see if Jane Hohenadel can finish the remaining work.
  • Indicator for contacting mental health professionals – access to mental health services data may be available.
    •
Issue: Comment from reviewers was that prevalence of depression should not be used as an indicator in isolation as many people may have experienced depressive episodes, however not to the degree that it is an indicator of mental health.
Decision: Enhance indicator comments to reflect some of the issues around prevalence of depression and access/use of mental health services.

Issue: Low number of suicides reported. All potential suicides are reviewed by the Coroner’s Office and Ontario uses a more conservative decision tree than other provinces.
Decision: Carol Paul is reviewing how deaths are coded in the Coroners Office, and notes will be added in the indicator comments section.

Issue: Suicidal thoughts: data source is CCHS, however only 4-5 health units have chosen this question. If OHS90 is used, then everyone will have a rate, but the data is out of date.
Decision: Keep this section as it is in the revised Mandatory Guidelines, using CCHS.
  • Use rates per 100,000
  • Remove general references in this section and place in appendix (only indicator specific references should be placed in sections)
  • Thoughts and attempts should be separate indicators.
  • Work stress: age groups need to be consistent. Statscan work age starts at age 15, data available from the CCHS (include variable name in the analysis section).
  • Self-perceived Life Stress: Change indicator name to chronic life stress.
  • Jessie to make revisions as discussed and will submit to replacement to check for consistency.
 Infectious Diseases (Julie Stratton)
  • Include all polio (not just indigenous).
  • Keep specific indicator list as is, but develop a list of other indicators for other vaccine preventable diseases that are not part of a routine public health immunization program, e.g., chicken pox, meningitis.
  • Add RRFSS as alternative data source for vaccination coverage for certain vaccines (e.g., tetanus)
  • Data gaps: mom identified as heb B carrier and child is immunized.
  • Appendix: VAAE code list (VACART)
 Contact with Health Professionals (Nam Bains)

Issue: Should this section be kept given the small number of indicators.
Decision: Keep the section. Rename some of the indicators
  • Contact with a physician (GP/FP/Specialist)
  • Use of other health professionals (CCHS) as additional supplemental indicator (not core)
  • Analysis checklist: contact with physician in hospital is not included unless it is emergency care.
  • Other indicators (not core) that are barriers to access: insurance plans (do you have insurance coverage for medical services not covered by a public health insurance plan?)
 Appendices: 

Appendix 1: Members of PHIWG
Include everyone who has been a part of the process.

Appendix 2: Direct standardization (SRATES)
Appendix 3: Indirect standardization (SMRs, SIRs)
Nam will have a student take on this issue (background, methods, small numbers and how to deal with zero counts in some age groupings, age and sex standardization vs. age standardization, etc). Sue Bondy, Mary-Anne Pietrusiak, Brenda Coleman and Jennifer Pennock to review.

Appendix 4: Standard population: Canada 1991 
No change

Appendix 5: PYLL calculation
Sherri to review – change to PYLL 75 (for clarity)

Appendix 6: Public health units and health areas sampled in the 1990 OHS and 1996/97 OHS
No Change. Add another appendix for CCHS (Carol) 

Appendix 7:Methods for age standardizing the Ontario Health Survey
Sue Bondy to review

Appendix 8: Avoidable mortality from specific diseases
Remove

Appendix 9: How to calculate a moving average
Brenda Coleman to review

Appendix 10: Mandatory Health Programs and Services Guidelines: Objectives and Indicators
No change – indicate which version of Mandatory. Will need to be updated once new Mandatory is approved.

Appendix 11: Summary of Notes on Data Sources
General description of data sources. Mary-Anne and Carol to review. Julie to prepare notes on IRIS.

New Appendices:
References: General – Mary-Anne will pull together

Bootstrapping – mention briefly in analyst checklist as applicable as a method of calculation.

Methodology for life expectancy – link to ICES website.

Infectious diseases, VAAE – formatting is questionable. Julie will complete.

Fact sheet on RRFSS – Brenda Guarda, or a link to website.

Geography (public health unit, regions, etc.) – Sherri and Nam

 Next Steps and Wrap-Up

Priority indicators for the chart book:
Population – general on population by age and sex
Social and Physical Environment – pass
Cause of death; mortality from selected diseases – chronic diseases
Life expectancy – Sherri to follow-up with Doug
All cause hospitalization separations
Cancer – incidence and mortality (all-cancers and top cancers – e.g., breast, lung)
Prevalence of selected chronic diseases
Injury and substance – death from selected causes of injury
Behaviour and health – daily smoking, BMI
Teen pregnancy; low birth weight; infant mortality

Future indicators to consider
  • DALY instead of PYLL, or some other mortality gap measures
  • Attributable mortality to specific behaviours (eg smoking attributable mortality, obesity)
  • Avoidable mortality from specific diseases
  • Colorectal screening
  • BMI for children
  • Seeking help from mental health professional
  • Assistance with activities of daily living: basic and instrumental. Useful as measures of disability in the community and need for services. Long term disability questions are being used by the CCHS, but are in the process of being modified. Sue Schultz to investigate.
 Reviewers who have not sent in comments:
Follow-up by Mary-Anne

Other Important Points:
  • Indicate in all sections of a variable from CCHS and RRFSS is a core or optional question, and consistency with the way they are described.
  • Indicator comments – caveats of applying the indicator should be included in this section. For instance, cautionary notes of data sources (i.e., HELPS) as applicable.
  • The questions from CCHS, OHS, RRFSS, etc should go in a new section for each indicator “Survey Questions”.
  • Non response categories: Brenda Coleman to check the literature for best practice
  • Citations: Mary-Anne to review
 Flow of Sections:
Indicator Name
Description
Key words
Specific indicator(s)
Mandatory Objectives
Corresponding National Indicator
Corresponding Indicators from other sources
Data sources
Survey Questions (if relevant)
ICD codes (if relevant)
Alternative Data Sources
Analysis Check-List
Method of Calculation
Basic Categories
Indicator Comments
Definitions
Cross-References to Other Sections
References
Date of Last Revision

Next Step:
Members will complete edits to their sections and forward to Mary-Anne. Mary-Anne will be reviewing all sections for consistency.