Making Recommendations for Special Populations: Older Adults, Children

Slide Presentation from the AHRQ 2011 Annual Conference

Making Recommendations for Special Populations: Older Adults, Children, and Racial/Ethnic Minorities

Slide Presentation from the AHRQ 2011 Annual Conference

On September 19, 2011, Kirstin Bibbens-Domingo, Al Siu, Michael LeFevre and Evelyn Whitlock made this presentation at the 2011 Annual Conference. Select to access the PowerPoint® presentation (2 MB). Plugin Software Help.

Slide 1

Making Recommendations for Special Populations: Older Adults, Children, and Racial/Ethnic Minorities

September 19, 2011

Evelyn P. Whitlock, MD, MPH Associate Director Oregon Evidence-based Practice Center
Kaiser Permanente, Center for Health Research

Michael L. LeFevre, MD, MSPH Department of Family & Community Medicine
University of Missouri School of Medicine

Al L. Siu, MD, MSPH
The Brookdale Department of Geriatrics and Palliative Care
The Mount Sinai Medical Center

Kirsten Bibbins-Domingo, PhD, MD
Division of General Internal Medicine
University of California, San Francisco

Slide 2

Welcome

Slide 3

History of the U.S. Preventive Services Task Force (USPSTF): Primary Care & Prevention

• First convened by the U.S. Public Health Service (USPHS) in 1984.
• Sponsored by AHRQ since 1998.
• Commissions Evidence-based Practice Centers (EPCs) to perform systematic reviews of clinical preventive services.
• Bases clinical recommendations for primary care providers on evidence reviews.
• Guide to Clinical Preventive Services (online):
  • Over 100 screening, behavioral counseling, and chemoprevention recommendations.

Slide 4

Selection of Recent and Upcoming Topics

• Recent Screening Recommendations:
  • Screening for Depression in Adults (2009).
  • Screening for Obesity in Adults (in press).
  • Screening for Cervical Cancer (in press).
• Special Population Recommendations:
  • Screening for Child and Adolescent Depression (2009).
  • Screening for Obesity in Children and Adolescents (2010).
  • Interventions to Prevent Falls (upcoming).
  • Screening for Dementia (upcoming).
  • Screening for Hypertension in Children and Adolescents (upcoming).

Slide 5

Incidence of Cervical Cancer by Age and Race/Ethnicity

Image: A line graph shows incidence of cervical cancer by age and race/ethnicity.

Source: U.S. .Crude Cervical Cancer Incidence Rates by Age and Race/Ethnicity (SEER 2000-2007).
Rates are expressed as cases per 100,000 women. ICC only unless otherwise specified.
American Indian/Alaska Native statistics only include cases from the Contract Health Service Delivery Area (CHSDA) counties.
Hispanic and Non-Hispanic are not mutually exclusive from White, Black, American Indian/Alaska Native, and Asian or Pacific Islander.

Slide 6

USPSTF's Steps to an Evidence-based Recommendation

• Select highest priority topics.
• Set scope of topic/work plan: what questions are important to answer; what evidence to search for.
• Consult during protocol-based systematic review.
• Consider evidence synthesis results: overall benefits & harms, subgroups/effect modifiers, key contextual findings.
• Apply USPSTF rules of evidence and make related judgments to make a recommendation.

Slide 7

Overview of USPSTF Process for an Evidence-based Recommendation—Rules of Evidence

• What evidence does the USPSTF consider?
  • Key questions on both benefits and harms of all aspects of a preventive service are answered through a systematic review and sometimes a companion modeling exercise.
  • Rules of evidence provide framework for determining net benefit-way beyond "hierarchy of evidence" approach.

Reference: Sawaya G et al. Ann Intern Med 2007;147:871-875.

Slide 8

USPSTF Evaluation of Evidence for Each Key Question — Convincing, Adequate, or Inadequate?

• Appropriate study design(s)?
• High quality studies?
• Applicable studies to U.S.?
• Precision of results (number and size of studies).
• Consistency of results.
• Additional factors (biological plausibility, dose/response...).

Reference: Sawaya G et al. Ann Intern Med 2007;147:871-875.

Slide 9

USPSTF Evaluation of Net Benefit

• How does the USPSTF estimate certainty & magnitude of net benefit for the whole preventive service?
• Certainty and magnitude of net benefit together determine the recommendation letter grade (A, B, C, D, I).

Slide 10

USPSTF Evaluation of Body of Evidence: High, Moderate, or Low Certainty in Findings Overall

• Appropriate study design(s) for all key questions (KQ).
• High quality studies for all KQ.
• Applicable body of studies to U.S.
• Precision of results (number and size of studies).
• Consistency of results across key questions.
• Additional factors.

Reference: Sawaya G et al. Ann Intern Med 2007;147:871-875.

Slide 11

Magnitude of net benefit—Substantial, Moderate, Small, Zero-Negative

• "Quantitative" subtraction of average harms (immediate and downstream) from average benefits.
• Substantial: benefits substantially outweigh harms (large benefit in rare group or smaller benefit across prevalent condition).
• Zero-negative: harms equal or out-weigh benefits.
• Small: benefits slightly outweigh harms and may vary individually (preference, other factors).
• Judgment necessarily involved.

Slide 12

USPSTF Recommendation Grid-Grading

Slide 13

General Versus Population-specific Recommendations

• General recommendations are not always appropriate for certain groups.
• Different natural history of disease and different health states can affect net benefits of preventive services.
• The average effect may be misleading:
  • One group may benefit, another may be harmed, and one may have no benefit or harms.
• Either approach (general or population-specific) has trade-offs.

Slide 14

Some Challenges in Making Recommendations in Special Populations

• Often under-represented in clinical trials.
• Important population-specific characteristics that differ from the general population affect prevention (e.g., limited life expectancy in older adults, cultural differences, proxy or caregiver involvement, tracking of risk factors into adulthood) are hard to factor in.
• Important outcomes also vary across population.
• Methodology:
  • No rules for extrapolation from other populations.
  • Incomplete approaches to deal with heterogeneity of treatment effects.
  • No clear guidance on when a population-specific approach is preferable.

Slide 15

Special Populations of Particular Interest to the USPSTF

• Racial and Ethnic Minorities (growing emphasis):
  • Blacks.
  • Asians / Pacific Islanders.
  • Hispanics.
  • Native Americans / Alaskan Natives.
• Children—Methods workgroup since 2002:
  • Children under 18 years of age.
• Older Adults—Methods workgroup since 2005 (recently published):
  • Adults over 65 years of age.

Reference: Leipzig R et al. Ann Intern Med 2010;153:809-814.

Slide 16

Purpose of Today's Session

• To discuss important barriers and potential solutions for using evidence and making recommendations for these special populations.
• To generate important areas for future work by USPSTF methods working groups focused on these special populations.
• To identify other important subpopulations for USPSTF consideration.

Slide 17

Racial/Ethnic Minorities—
Kirsten Bibbins-Domingo, PhD, MD, USPSTF Member

• Understudied:
  • Even if study participants are enrolled proportional to U.S. population, low power for group-specific results.
• Non-standard definitions for racial/ethnic groups and heterogeneity within defined groups (e.g., Asians).
• Health disparities may or may not represent the need for population-specific recommendations:
  • Possible differential factors: baseline risk, biology of disease, differential access to screening or treatment, differential response to treatment, cultural/acculturation issues.
• Different spectrum of disease (e.g., stroke vs MI in CVD) in R/E minorities may drive health outcome differences.

Slide 18

Starting Questions

• What is your impression of how well the USPSTF recommendations meet the needs of racial-ethnic minorities?
• What are the important methodological advances that need to occur in order to improve the development and implementation of recommendations for minorities?
• How can the USPSTF do a more apt job communicating how they review the evidence (or lack thereof) for racial-ethnic minorities?

Slide 19

Children—
Michael L. LeFevre, MD, MSPH, USPSTF Co-Chair

• Many insufficient evidence ("I" statements) for children's preventive services-some already well-accepted parts of pediatric practice:
  • Children are relatively understudied.
  • Service impact (e.g., anticipatory guidance) hard to measure.
• Relatively healthy populations:
  • Developmental issues more common than disease issues.
• Health outcomes may differ from adults:
  • Broader concerns (sensory function, school readiness, behavior).
  • Longer time frame to achieve health impact (adult disease).

Slide 20

Starting Questions

• What is your impression of how well the USPSTF recommendations meet the needs of children and adolescents?
• What are the important methodological advances that need to occur in order to improve the development and implementation of recommendations for kids?
• How can the USPSTF do a more apt job communicating how they review the evidence (or lack thereof) for children and adolescents?

Slide 21

Older Adults—
Al L. Siu, MD, MSPH, USPSTF Co-Chair

• Understudied, particularly older (75+) and oldest old (85+).
• Age captures only part of variation in health:
  • Unhealthy 65 year old may be more like a healthy 85 year old.
• Life expectancy is very important in determining net benefits, but hard to determine.
• Comorbidities are common, understudied, and modify prevention considerations:
  • Single focus prevention recommendations less relevant.
• Patient-important outcomes predominate (function, independence, quality of life more than mortality).

Slide 22

Starting Questions

• What is your impression of how well the USPSTF recommendations meet the needs of older adults?
• What are the important methodological advances that need to occur in order to improve the development and implementation of recommendations for older adults?
• How can the USPSTF do a more apt job communicating how they review the evidence (or lack thereof) for older adults?

Slide 23

Summary and next steps....

Image: A photograph of a mountain and lake is shown.

Slide 24

Contact Information

• Tracy Wolff, MD, MPH.
  Medical Officer, USPSTF Program.
  Tracy.wolff@ahrq.gov.hhs.
• http://www.preventiveservices.ahrq.gov.
• http://www.uspreventiveservicestaskforce.org.
• http://epss.ahrq.gov/PDA/index.jsp.

For more information stop by the Prevention and Chronic Care Program booth in the mAHRQet Place Café and join us tomorrow in Salon E at 330 for a Guided Tour of How a USPSTF Topic Becomes a Recommendation!

Current as of December 2011

Internet Citation:

Making Recommendations for Special Populations: Older Adults, Children, and Racial/Ethnic Minorities. Slide Presentation from the AHRQ 2011 Annual Conference (Text Version). December 2011. Agency for Healthcare Research and Quality, Rockville, MD. http://www.ahrq.gov/about/annualconf11/whitlock_lefevre_siu_bibbins/bibbens.htm