Using Evidence To Improve Public Health Infrastructure:

Let the evidence guide our actions

January 7, 2004

Jonathan E. Fielding, M.D., M.P.H., M.B.ADirector of Public Health and Health Officer

L.A. County Department of Health ServicesChair, CDC Task Force of Community Preventive Services

Professor of Public Health and PediatricsUniversity of California, Los Angeles

If we did not respect the evidence, we would have very little leverage in our quest for truth

Carl Sagan

Quality of the Evidence

We hear about it oftenTV NetworksUsually related to high profile trials

Public Health evidence is different than legal evidenceIt is the available information on a

particular questionWe want the best available evidence

in making decisions

Decisions and Evidence

Evidence takes many formsOpinion of leadersOpinion of “experts” Studies without controlsStudies with controls Studies of variable quality in design

and execution

Decisions and Evidence

Our commitment: Improve public health

Health problems well defined Our job:

Make a difference through policies and programs

Inaction is not an option Hard to identify best evidence to

inform decision making

Decisions and Evidence

Decisions on policies and programs are often made based on:Personal experienceWhat we learned in formal trainingWhat we heard at a conferenceWhat a funding agency required/

suggested What others are doing

Evidence and Public Health Decision Making

Good news Strong evidence on the effect of

many policies/ programs aimed to improve public health

Major efforts underway to assess the body of evidence for wide range of public health interventions

What works to improve the public’s health?

Bad news Many public health professionals are unaware

of this evidence Some who are aware don’t use it Many existing disease control programs have

interventions with insufficient evidence –while others use interventions with strong evidence of effectiveness

Lack of use of effective interventions can adversely affect fulfilling mission and getting public support

How do we know what works in improving the health of populations? Background Many community health improvement efforts

have not achieved desired results Interventions often chosen based on opinions

and personal preferences Evidence based medicine---Clinical

Preventive Services Task Force –mid 80s Evidence based population health ---

Community Preventive Services Task Force– mid 90s

Systematic Search for the Best Evidence

U.S. Community Preventive Services Task Force Appointed by CDC Director in 1996

Non-Federal independent task force of experts in multiple relevant disciplines Epidemiology Public Health Practice Behavioral Sciences Evidence based medicine/ public health Other relevant areas of expertise

Goals

Conduct careful analytic reviews of acceptable evidence for population health interventions and make related recommendations

Use peer reviewed literature Standard rules of evidence Standard rules for translating

evidence into recommendations for interventions

Systematic Reviews of Public Health Interventions are Useful

Methods first developed by social scientists (e.g., Glass, ‘76)

Distill and summarize large and diverse bodies of evidence

Reduce errors and biases in interpretation Make assumptions explicit

Systematic Reviews Are Not:

Limited to randomized controlled trials Limited to healthcare interventions Restricted to a “biomedical model” of

health

- Petticrew, 2001

Task Force on Community Preventive Services

Members

Jonathan E. Fielding, Chair Patricia Dolan Mullen, Vice-chair Noreen M. Clark John M. Clymer Mindy T. Fullilove Alan Hinman

George J. Isham Robert L. Johnson Garland Land Patricia A. Nolan Dennis E. Richling Barbara K. Rimer Steven Teutsch

ConsultantsRobert S. LawrenceJ. Michael McGinnis

Lloyd F. Novick

Who Is the Audience?

People who plan, fund, or implement public health services and policies for communities and healthcare systems Public health departmentsHealthcare systems and providersPurchasers Government agenciesCommunity organizations

Community Guide Topics

Environmental Influences Sociocultural Environment

Physical Environment

Risk Behaviors Specific Conditions Tobacco Use Alcohol Abuse/Misuse Other Substance Abuse Poor Nutrition Inadequate Physical Activity Unhealthy Sexual Behaviors

Vaccine Preventable Disease Pregnancy Outcomes Violence Motor Vehicle Injuries Depression Cancer Diabetes Oral Health

Methods for Systematic Reviews of Effectiveness Evaluations

Develop conceptual framework Search for and retrieve evidence Rate quality of evidence Summarize evidence Translate strength of evidence into finding

Strongly recommended Recommended Insufficient evidence

Logic Framework: Vaccine Preventable Disease

Treatmentof Vaccine-Preventable

Diseases

Morbidity

andMortality

Population

Attendance in

Public, Private, or

Joint Healthcare

Systems

Exposure toVaccine-

Preventable Disease

Intervention Types

Determinants

Intermediate Outcomes

Public Health Outcomes

Reviewed

Not ReviewedReducingExposure

Environment

VaccinationCoverage

Vaccine-

PreventableDisease

Provider-Based

Interventions

IncreasingCommunityDemand for

Vaccinations

Enhancing Access to

Vaccinations

Reduced Disease Instance

Standardized Analysis Process

Systematic review of literature Abstracting of relevant studies Grading of evidence

Study design Execution

Translating from quality of evidence to recommendations

Economic analysis Other benefits and harms

How Does the Task Force Define Suitability of Study Design?

GreatestProspective with concurrent comparison

ModerateMultiple before-and-after measurements

but no concurrent comparison ORRetrospective

LeastSingle group before-and-afterCross-sectional

What Factors Determine Quality of Execution? Description of intervention and study

population Sampling procedures Exposure and outcome measurements Approach to data analysis Interpretation of results

Follow-upConfoundingOther bias

Other issues

How Does the Task Force Draw an Overall Conclusion About the Strength of a Body of Evidence?

Number of studies Design suitability Quality of execution Consistency Effect size

Evidence of Effectiveness

Quality of Execution

Design Suitability

Number of Studies

Consistent Effect Size

1. Strong

Good Greatest > 2 Yes Sufficient

Good Greatest or Moderate

> 5 Yes Sufficient

Good or Fair Greatest > 5 Yes Sufficient

Meet criteria for sufficient evidence Large

2. Sufficient

Good Greatest 1 -- Sufficient

Good or Fair Greatest or Moderate

> 3 Yes Sufficient

Good or Fair Greatest, Moderate or Least

> 5 Yes Sufficient

3. Insufficient Insufficient design or execution Too few No Small

Physical Activity:Review of One Intervention

Goal: increase the amount of time students spend doing moderate or vigorous activity in PE class through curricular change

Interventions reviewed included changing the activities taught (e.g., substituting soccer for softball) or modifying the rules of the game so that students are more active (e.g., in softball, have the entire team run the bases together when the batter makes a base hit). Many interventions also included health education.

14 acceptable studies; in all students’ physical fitness improved.

5 studies measured activity levels during PE class; all found increases in amount or percentage of time moderately/

vigorously active and/or intensity level of physical activity during class.

Median estimates--modifying school PE curricula as recommended will result in an 8% increase in aerobic fitness

School Curricular Interventions

to Improve Physical Fitness

School Curricula to Improve Physical Activity

Modifying school P.E. curricula was effective across diverse racial, ethnic, and socioeconomic groups, among boys and girls, elementary- and high-school students, and in urban and rural settings.

In a separate literature review, having students attend school PE classes was not found to harm academic performance.

Economic analysis pending.

Different Ways to Improve Public Health

Focus on diseases/ injuries e.g. sexually transmitted diseases, intentional injury, diabetes

Focus on risk factors e.g. tobacco use, nutrition, physical activity

Focus on underlying factors that impact multiple dimensions of health e.g. poverty, social isolation

SocialEnvironment

GeneticEnvironment

IndividualResponse

• Behavior• Biology

Healthand

Function

Diseaseand

Injury

Healthand

Medical Care

Well-Being Prosperity

PhysicalEnvironment

Underlying Health Determinants

Impact of Early Childhood Home Visitation Programs

Program can prevent child maltreatment in high-risk families. In studies reviewed, home visiting resulted in a

40% reduction in child maltreatment episodes. Longer duration programs produce larger

effects; programs of less than 2 years duration did not appear to be effective.

Professional home visitors may be more effective than trained paraprofessionals but longer-duration programs with trained

paraprofessionals can also be effective.

Impact of Early Childhood Home Visitation Programs

All programs reviewed were directed at families considered to be at high risk of child maltreatment, (e.g., single or young mothers, low-income households, families with low birth weight infants).

Other benefits Health benefits for premature, low birth weight

infants and for disabled and chronically ill children

Improved maternal educational attainment, reduced public support, improved child educational performance, reduction in drug use and contact with juvenile justice etc.

Tenant-based Rental Voucher Programs

Background Tenant-based vouchers allow very low income

families to rent safe, decent, and affordable privately owned housing in neighborhoods of their choice.

Rental voucher programs, known as “housing mobility programs,” work with landlords and tenants to find rental property outside of neighborhoods of concentrated poverty and relocate families to neighborhoods of greater prosperity.

Tenant-based Rental Voucher Programs

Findings from the Systematic Review 6 studies: rental voucher programs resulted in

decreases in victimization of tenants or their property Families enrolled in rental voucher programs who

moved to better areas were• 6% less likely to have a household member victimized • 15% less likely to experience neighborhood social

disorder. Changes in victimization in both urban and suburban

settings. Other benefits: substantially reduced symptoms of

maternal depression, boys’ behavioral problems in school, and childhood illnesses and accidents requiring medical attention.

General Comments on Evidence Based Reviews

More evidence than sometimes expected, however

Insufficient evidence common outcome Very resource intensive process Quality of studies vary widely Economic data still uncommon Important to consider harms even though

uncommon

TASK FORCE REVIEWS AND RECOMMENDATIONS

www.thecommunityguide.orgResults of all reviews to dateFrequent updatingDownloadable Slide Sets

How Evidence Can Improve Public Health Infrastructure

Explore evidence underlying options to reach each public health goal

e.g. smoking control, reducing disparities in infant mortality, increasing physical activity, increasing immunization rates

Comprehensively review the best sources of evidence reviews

Community Guide (best source when topic of interest has been covered)

Recent review articles in peer reviewed journals Other meta-analyses funded by responsible federal

agencies Compare results and recommendations of different

sources

Use evidence to decide among possible interventionsHow does each possible intervention

suit the problem and the population? • E.g. was it tried on particular racial/

ethnic/ age/ gender groups?• Is there reason to belief it would not be as

effective for some of these on whom it was not tried?

• Is the problem now similar to what it was when the major studies took place?

How Evidence Can Improve Public Health Infrastructure

Deciding on Interventions

Single versus multiple component interventionsSingle component interventions easier

to develop, implement, control and assess, but

Multi-component interventions usually more effective

• E.g. Tobacco control in California

Deciding on Interventions

Consider both policies and programs Programs

Greater control over all aspects Organizational unit has primary

responsibility for design, implementation and outcomes

Policies Control varies: broad policies often made by

elected officials Policies have potential for greater public

health impact Credit needs to be shared e.g. LAUSD

Nutrition Policies, increase in tobacco tax

Deciding on Interventions

Consider effect sizeMedianConsistency

Consider breadth of target population Together effect size and target

population define the overall population effect

How Evidence Can Improve Public Health Infrastructure Use evidence to determine realistic goals by estimate effect

size (i.e.. how much you move the needle!)

Relative percentage changes in exposure to environmental tobacco smoke attributable to workplace smoking bans and restrictions from studies that qualified for inclusion in this review (“a” and “b” in Study names refer to first or second study by the same author in that year, included in this review)

Deciding on Interventions

What is the slope of the effect curve? • Larger initial effects with significant recidivism• Smaller initial effects with Increasing impact over

time

What is the time frame for observed health benefits?

• How long were the follow-up periods for the best studies?

• For equal benefit, shorter is better, but• Long term benefit is primary interest

Deciding on Interventions

What is the cost of the intervention? Personnel Dollars i.e. contracts Time to implement Likelihood of funding for sufficient period to

get effect Potential for dedicated or incremental

funding Opportunity cost (i.e. cost of not doing other

things)

Deciding on Interventions

Relative cost-effectiveness Cost effectiveness is dollar cost per health outcome

(including clear intermediate outcomes) e.g. smoker prevented

lead poisoning preventedSTD cureddrug treatment completed

Note: some interventions have multiple health benefits e.g. smoking affects CVD, some cancers, respiratory disease etc.

Deciding on Interventions

Who else needs to be involved to be successful? Within public health Within personal health services Voluntary agencies Health care organizations Health plans Employers

How difficult is it to get agreement on: Roles and responsibilities Interventions?

Time cost versus partnership benefit

How Evidence Can Improve Public Health Infrastructure

Use evidence to help decide on construction of intervention

Interventions with same name can be very different

Follow the design used in most successful interventions

Talking to those who did the studies is very helpful in refining intervention

Using the Evidence

Use evidence to frame objectives Use evidence to develop evaluation plan

and related evaluation Approach Measures Data collection plan

Develop internal evidence through performance measurement system

Frequent monitoring essential

What more is needed?

More research on public health practice; for many interventions---insufficient evidence

Increased funding for evidence based reviews using consistent methodologies

More training on appropriate sources and uses of evidence

in schools of public health and others training public health professionals

in public health practice settings e.g. state and local health departments

Political leaders and others who influence the decision making process to improve health

DHS Public Health Opportunities

Become sophisticated user of evidence based information and recommendations

Make use of best evidence key aspect of performance of program directors and key managers

Contribute to the literature on what works in public health practice

Presentation available on DHS Intranet

Also see Evidence-Based Public Health, Ed. Ross C. Brownson, Elizabeth Baker, Terry L. Leet etc.

Oxford University Press, 2003