presented at Evaluation Considerations: Measures & Methods Shrikant I. Bangdiwala, PhD Professor of Research in Biostatistics Injury Prevention Research Center University of North Carolina at Chapel Hill, USA 1 WHO VIP Webinar 2011 Content Purpose of evaluation Cycle of program planning & evaluation Indicators Study designs Statistical modeling Challenges 2 WHO VIP Webinar 2011 What are we ‘evaluating’? Actions, programs, activities Conducted in a community setting, over a period of time Aimed at reducing deaths, injuries, and/or events and behaviors that cause injuries 3 WHO Safety VIP2010 Webinar London 2011 Example: Suwon, South Korea area of ‘safety promotion’ http://www.phs.ki.se/csp/safecom/suwon2.htm 4 WHO VIP Webinar 2011 Why do we ‘evaluate’? To know ourselves what works and if we are doing some good In performing some activity In the community In the country To convince funders and supporters that their investment is worthwhile To convince the community about the benefits of the multiple activities and actions carried out 5 WHO VIP Webinar 2011 Main purposes of evaluation Evaluation helps determine: How well a program/policy works relative to its goals & objectives Why a program/policy did or didn’t work, relative to planned process How to restructure a program/policy to make it work, or work better Whether to change funding for a program 6 WHOChicago VIP Webinar NSC 20102011 6 Methodological complications Multiplicities Multiple components of a program Multiple populations at risk Multiple study designs Multiple types of effects/impacts/outcomes & severities Multiple audiences/objectives of ‘evaluation’ Multiple methods for conducting evaluation 7 WHO VIP Webinar 2011 When should evaluation be considered? Evaluation needs to begin in, and be part of, the planning process… Otherwise, “if you do not know where you are going, it does not matter which way you go, and you will never know if you got there or not!” Lewis Carroll (1872) Alice in Wonderland 8 Adapted from M. Garrettson WHO VIP Webinar 2011 8 Types of evaluation depending on program phase Program Planning Phase Program Implementation Phase Post Program Phase 9 Formative Evaluation How can the program activities be improved before implementation? Process Evaluation How is/was the program (being) implemented? Impact / Outcome Did the program succeed in achieving the intended impact or outcome? WHOChicago VIP Webinar NSC 20102011 9 Cycle of program planning and evaluation 10 Adapted from C Runyan WHO VIP Webinar 2011 Identify population & problem Surveillance data Other needs assessment strategies key informant interviews focus groups surveys evaluations of past programs literature consultation with peers other info… 11 WHO VIP Webinar 2011 11 Disseminate Identify problem & population Evaluation: •Process •Impact •Outcome Identify resources Implement Test, Refine, Implement Test & refine implementation 12 Define target audience Evaluation: •Formative Choose strategies WHOChicago VIP Webinar NSC 20102011 Set goals/ objectives 12 Define target audience To whom is the program directed? Whose injuries need to be reduced? Who is the target of the program? • • • • • 13 at risk persons care givers (e.g. parents) general public media decision makers WHO VIP Webinar 2011 13 Understand target audience What are their characteristics? Special needs (e.g. literacy) Interests, concerns, priorities Attitudes & beliefs re: problem & solutions to problem Cultural issues 14 WHO VIP Webinar 2011 14 Disseminate Identify problem & population Evaluation: •Process •Impact •Outcome Identify resources Implement Test, Refine, Implement Test & refine implementation 15 Define target audience Evaluation: •Formative Choose strategies WHO VIP Webinar 2011 Set goals/ objectives 15 Identify resources Community partners interest in topic working on similar projects On-going activities Sources of financial support Interests in community 16 WHO VIP Webinar 2011 16 Disseminate Identify problem & population Evaluation: •Process •Impact •Outcome Identify resources Implement Test, Refine, Implement Test & refine implementation 17 Define target audience Evaluation: •Formative Choose strategies WHO VIP Webinar 2011 Set goals/ objectives 17 Set goals & objectives Goal broad statement of what program is trying to accomplish Objectives Specific Measurable Time-framed 18 WHO VIP Webinar 2011 18 Disseminate Identify problem & population Evaluation: •Process •Impact •Outcome Identify resources Implement Test, Refine, Implement Test & refine implementation 19 Define target audience Evaluation: •Formative Choose strategies WHO VIP Webinar 2011 Set goals/ objectives 19 Choose Strategies Identify existing strategies/programs Literature: evidence based? promising practice? WHO manuals Successes from other communities-regionscountries Develop new strategies: Logic model (how would it work) Haddon matrix 20 WHO VIP Webinar 2011 20 Haddon Matrix Person Vehicle/ vector Physical Environ. Social Environ. Preevent Event Postevent 21 Haddon 1970 Am J Public Health WHO VIP Webinar 2011 21 3-dimensional Haddon Matrix Other?? Feasibility Preferences Stigmatization Equity Freedom Cost Event Post-event Effectiveness 22 Runyan 1998 Injury Prevention WHO VIP Webinar 2011 22 Disseminate Identify problem & population Evaluation: •Process •Impact •Outcome Identify resources Implement Test, Refine, Implement Test & refine implementation 23 Define target audience Evaluation: •Formative Choose strategies WHOChicago VIP Webinar NSC 20102011 Set goals/ objectives 23 Formative Evaluation Why it’s useful Questions it answers What is the best way to influence the target population? Will the activities reach the people intended, be understood and accepted by target population? How can activities be improved? 24 Improves (pilot-tests) program activities before full-scale implementation May increase likelihood program or policy will succeed May help stretch resources WHO VIP Webinar 2011 24 * Modified from Thompson & McClintock, 2000 Disseminate Identify problem & population Evaluation: •Process •Impact •Outcome Identify resources Implement Test, Refine, Implement Test & refine implementation 25 Define target audience Evaluation: •Formative Choose strategies WHOChicago VIP Webinar NSC 20102011 Set goals/ objectives 25 Implementation As planned, with attention to detail Documented clearly so others can replicate if appropriate 26 WHO VIP Webinar 2011 26 Disseminate Identify problem & population Evaluation: •Process •Impact •Outcome Identify resources Implement Test, Refine, Implement Test & refine implementation 27 Define target audience Evaluation: •Formative Choose strategies WHOChicago VIP Webinar NSC 20102011 Set goals/ objectives 27 Process evaluation Purpose is to address: What was done? How was it implemented? How well was it implemented? Was it implemented as planned? 28 WHO VIP Webinar 2011 28 Process evaluation – examples of questions • • • • Who carried out intervention? Was this the appropriate person/group? Who supported and opposed intervention? What methods/activities were used? 29 WHO VIP Webinar 2011 29 Process evaluation - why is it useful? • • 30 Allows replication of programs that work. Helps understand why programs fail. WHO VIP Webinar 2011 30 * Modified from Thompson & McClintock, 2000 The intervention cannot be a black box… It must be clearly understood Idea 31 ? WHO VIP Webinar 2011 Outcome 31 Disseminate Identify problem & population Evaluation: •Process •Impact •Outcome Identify resources Implement Test, Refine, Implement Test & refine implementation 32 Define target audience Evaluation: •Formative Choose strategies WHOChicago VIP Webinar NSC 20102011 Set goals/ objectives 32 Impact evaluation Purpose is to address changes in: 33 knowledge attitudes beliefs/ values skills behaviors / practices WHO VIP Webinar 2011 33 Using impact measures for Establishing effectiveness Suppose we have a public safety campaign as our strategy Need to show Campaign Outcome If we already have demonstrated that Behavior Behavior Outcome We simply need to show Campaign 34 Behavior WHO VIP Webinar 2011 34 Outcome evaluation Purpose is to address changes in: injury events (e.g. frequency, type, pattern) morbidity (e.g. frequency, severity, type) mortality (e.g. frequency, time to death) cost (e.g. direct and indirect) 35 WHO VIP Webinar 2011 35 Example: Bike helmets Intervention Physician counseling parents Enforcement of helmet law Impacts Outcomes Parental attitudes toward child helmet use Head injury in bike crashes Purchase of helmets Deaths from head injury in crashes Media campaign Use of helmets by children 36 WHOChicago VIP Webinar NSC 20102011 36 Evaluation examples of questions for local policy of smoke alarms Did the local policy of smoke alarms in apartments… Get passed Where people aware of it? Did people have access to smoke alarms? Did people get them installed properly? Do people keep them maintained? Lead to a reduction in the number or rates of: events (e.g. apartment fires) injuries deaths 37 costs (e.g. burn centerWHOcosts, family burden, property loss) VIP Webinar 2011 37 Evaluation – selection of measures ‘Quantitative Indicators’ Process Impact Outcome Health related Financial 38 WHO VIP Webinar 2011 38 Choice of measure or indicator We need to choose appropriate impact and outcome measures ‘Soft’ (more difficult to measure) outcomes – Perceptions constructs: fear, insecurity, wellbeing, quality of life Knowledge, attitudes and behaviors constructs Hard outcomes – Deaths, hospitalizations, disabilities due to injuries and violence Societal impacts – local development indicators Economics outcomes – Direct $/€/£/¥, indirect DALYs, QALYs, opportunities lost, burdens 39 WHO VIP Webinar 2011 Evidence of effectiveness Obtain qualitative ‘evidence’ to complement the quantitative ‘evidence’ Ex. Are “multisectorial collaborations and partnerships” friendly and functioning well? Ex. Is “community participation” optimal? Incorporate process indicators Incorporate narratives & testimonials 40 WHO VIP Webinar 2011 Disseminate Identify problem & population Evaluation: •Process •Impact •Outcome Identify resources Implement Test, Refine, Implement Test & refine implementation 41 Define target audience Evaluation: •Formative Choose strategies WHOChicago VIP Webinar NSC 20102011 Set goals/ objectives 41 Dissemination Dissemination not done well Not attempted Not based on research about how to disseminate information to intended audience Dissemination done well Defining audience How to access audience How best to communicate change message to them Presentation of clear, straightforward messages 42 WHO VIP Webinar 2011 42 Evaluation measures Lead to evidence of effectiveness But only if the research and study methodologies, and the statistical analyses methodologies, are appropriate to convince the funders and supporters, the skeptics, the stakeholders, the community 43 and understandable WHO VIP Webinar 2011 43 Research methodology approach: Evidence of effectiveness Obtain quantitative ‘evidence’ that favors the hypothesis that the intervention is effective as opposed to the (null) hypothesis that the intervention is not effective. How? Experimental study designs - randomized clinical trials, grouped randomized experiments, community-randomized studies Quasi-experimental study designs - non-randomized comparative studies, before-after studies Observational studies - cohort studies, case-control studies and comparative cross-sectional studies 44 WHO VIP Webinar 2011 Randomized controlled trial (RCT) / Experiment ‘strongest’ evidence Intervention Group O X O O X’ O Randomize Control Group 45 WHO VIP Webinar 2011 45 Quasi-experimental designs ‘qualified’ evidence Intervention Group Comparison Group 46 O O X O O WHO VIP Webinar 2011 46 One group pre/post ‘weak’ evidence Intervention Group 47 O X O WHO VIP Webinar 2011 47 One group – multiple pre / multiple post better ‘weak’ evidence Intervention Group 48 O O O X O O O O O WHO VIP Webinar 2011 48 One group, post only ‘basically ignorable’ evidence Intervention Group 49 X O WHO VIP Webinar 2011 49 Observational designs - cohort study evidence? Self-chosen Intervention Group Self-chosen Non-intervention Group 50 X X X O O O WHO Safety VIP2010 Webinar London 2011 50 Observational designs - case-control study evidence? X Cases O X Controls O 51 WHO Safety VIP2010 Webinar London 2011 51 Observational designs - cross-sectional study evidence? X X O Injured X O O X Non-injured O O O 52 WHO Safety VIP2010 Webinar London 2011 52 Statistical analysis methodologies Choice - often guided by what has been done previously, or what is feasible to do, or easy to explain Choice should be tailored to the audience & their ability to understand results; but also on the ability of the presenter to explain the methodologies 53 WHO VIP Webinar 2011 Statistical analysis Determined by research question(s) Guided by study design – experimental or observational Guided by whether outcome is studied at a single time point or multiple time points Group randomized controlled experiment Non-randomized comparison study Single site pre/post; surveillance study Retrospective or cross-sectional Time series analyses Guided by audience VIP Webinar 2011 Visual and descriptiveWHO appreciation 54 Visual and descriptive analysis – longitudinal time series Example: Espitia et al (2008) Salud Pública Mexico 55 WHO VIP Webinar 2011 Visual and descriptive analysis – comparisons over time Example: www.gapminder.org 56 WHO Safety VIP2010 Webinar London 2011 Statistical analysis - challenge But what we as a field have not done as well as other fields, is to draw strength from numbers develop collective evidence Combine results from multiple studies Systematic reviews (of observational studies) Meta analysis (of experimental & observational studies) Meta regression (of heterogeneous studies) Mixed treatment meta regression (for indirect comparisons) 57 WHO VIP Webinar 2011 Systematic reviews A protocol driven comprehensive review and synthesis of data focusing on a topic or on related key questions 58 formulate specific key questions developing a protocol refining the questions of interest conducting a literature search for evidence selecting studies that meet the inclusion criteria appraising the studies critically synthesizing and interpreting the results WHO VIP Webinar 2011 Example – Systematic review Shults et al (2001) Amer J Prev Med 59 WHO VIP Webinar 2011 Systematic reviews Of particular value in bringing together a number of separately conducted studies, sometimes with conflicting findings, and synthesizing their results. Zaza et al (2001) Amer. J Preventive Medicine – motor vehicle To this end, systematic reviews may or may not include a statistical synthesis called meta-analysis, depending on whether the studies are similar enough so that combining their results is Green (2005) Singapore Medical Journal meaningful 60 WHO VIP Webinar 2011 Meta analysis A method of combining the results of studies quantitatively to obtain a summary estimate of the effect of an intervention 61 Often restricted to randomized controlled trials Recently, the Cochrane Collaboration is ‘branching out’ to include both experimental and observational studies in meta analyses WHO VIP Webinar 2011 Meta analysis e.g. Liu et al (2008) Cochrane Collaboration 62 WHO VIP Webinar 2011 Meta analysis The combining of results should take into account: the ‘quality’ of the studies • Assessed by the reciprocal of the variance the ‘heterogeneity’ among the studies • Assessed by the variance between studies 63 WHO VIP Webinar 2011 Meta analysis – estimation of effect The estimate is a weighted average, where the weight of a study is the reciprocal of its variance In order to calculate the variance of a study, one can use either a ‘fixed’ effects model or a ‘mixed’/’random’ effects model Fixed effects model: utilizes no information from other studies var( Yi ) var( ei ) VYi Random effects model: considers variance among and within studies Yi i ei var( ) 2 var(Yi ) 2 VY*i WHO VIP Webinar 2011 64 Borenstein et al (2009) Introduction to Meta Analysis Meta analysis & meta regression Dealing with ‘heterogeneity’ among the studies - 2 Decompose the total variance into among and within components using mixed effects models for getting a more precise estimate of the intervention effect If there is still residual heterogeneity 65 Expand the mixed effects model to include study-level covariates that may explain some of the residual variability among studies meta regression WHO VIP Webinar 2011 Meta regression e.g. Yi 1 X 1i 2 X 2i i ei random error study random effect Overall mean X1 study variable – EU/USA X2 study variable – population type 66 WHO VIP Webinar 2011 Meta analysis Standard meta-analytical methods are typically restricted to comparisons of 2 interventions using direct, head-tohead evidence alone. So, for example, if we are interested in the Intervention A vs Intervention B comparison, we would include only studies that compare Intervention A versus Intervention B directly. Many times we have multiple types of interventions for the same type of problem, and we hardly have head-tohead comparisons We may also have multiple component interventions 67 WHO VIP Webinar 2011 Mixed treatment meta analysis Let the outcome variable be a binary response 1 = positive response 0 = negative response We can calculate the binomial counts r j:k out of a total number at risk n j:k on the kth intervention in the jth study rj:k We can then calculate p j:k n j:k the estimated probability of the outcome (risk of response) for the kth intervention in the jth study 68 Welton et al 2009 Amer J Epid WHO VIP Webinar 2011 Mixed treatment meta analysis Let each study have a reference ‘‘standard’’ intervention arm, sj, with study-specific ‘‘standard’’ log odds of outcome, j . The log odds ratio, j:k, of outcome for intervention k, relative to standard sj, is assumed to come from a random effects model with mean log odds ratio (d k d s j ) , and between-study standard deviation where dk is the mean log odds ratio of outcome for intervention k relative to control (so that d1 = 0). 69 Welton et al 2009 Amer J Epid WHO VIP Webinar 2011 Mixed treatment meta analysis This leads to the following logistic regression model: ln where 70 Welton et al 2009 Amer J Epid p j:k 1 p j:k j j j:k j:k ~ N[(dk d s ), ] j WHO VIP Webinar 2011 int s j int k Mixed treatment meta analysis - multiple-methods interventions If we have multiple methods in the ith intervention M1i , M 2i , M 3i ,... Plus we have multiple times when the outcome is assessed Yit i 1t 2 M1it 3 M 2it 4 X i eit Study effect Components 1 & 2 effects Time effect 71 WHO VIP Webinar 2011 Study covariable Error term Statistical analysis Methodology does exist for developing stronger collective evidence, evaluating the effectiveness of community based interventions, using different types of study designs and interventions Developing “practice-based evidence” 72 WHO VIP Webinar 2011 Dissemination We should not stop at developing the evidence We must work alongside economists in developing ways to effectively communicate ‘what works’ methodology and cost models do exist for estimating the “return on investment” Money talks !! 73 WHO Safety VIP2010 Webinar London 2011 Challenges – Evaluation requires Integration of evaluation from the beginning Appropriate measures, possible to be collected objectively, unbiasedly, easily and with completeness Appropriate qualitative and process information, to complement the quantitative information Concrete and convincing evidence of what aspects work in individual communities Formal methodological statistical evaluation of specific elements of programs Collective evidence of what common elements of programs work Effective dissemination strategies – “return on investment” 74 WHO VIP Webinar 2011 75 WHO VIP Webinar 2011 75