Pitt EPIDEM 2670 - Before and After Studies in Injury Research

Unformatted text preview:

Before and After Studies in Injury ResearchAimsSlide 3Before-and-After StudiesBefore-and-After Study DesignBackgroundSlide 7Before and After StudiesSlide 9Before-and-After Group Data ExampleHow do we know if the rate of bicycle head injuries would not have gone down anyway, even without the new law?Problems with Before and After DesignsQuestions to askTemporal ChangeRegression to the MeanAnnual Increase in offences with firearmsDid the Amnesty work?Evaluation of the California Child Passenger Safety InitiativeCPSI GoalsCPSI InterventionCPSI EvaluationObservationsObservational Study LocationsKnowledge of Booster Seat LawCare Seat Use: Self-reported vs. ObservedKey Lecture PointsBefore and After Studies in Injury ResearchThomas Songer, PhDUniversity of [email protected]•Recognize the role that before and after studies play in injury research•Recognize strengths and weaknesses of before and after studiesStudy DesignsCase reportCase seriesDescriptiveEpidemiologyDescriptiveRCTBefore-AfterstudyCross-sectionalstudyCase-CrossoverstudyCase-ControlstudyCohort studyAnalyticEcologic studyBefore-and-After Studies•A “quasi-experimental” design that surveys exposures and disease status before and after an interventiontimeIntervention starts hereTime periodbeforeTime periodafterBefore-and-After Study Design•A quasi-experimental design in which there is a pretest and posttest, but no comparison group.•Instead of comparing subjects who received treatment to those who did not, we compare subjects before and after they got the treatment (or compare time periods before and after the intervention).Background•Traditionally most interventions have been evaluated using a pre-test post-test or before and after design.•Participants are tested treated and then tested again any improvements are attributable to the intervention.•Currently this is probably the most POPULAR evaluative method in most fields.David TorgersonBefore-and-After Study Design•Advantage:–overcomes ethical concerns with randomized designs–Low cost, convenience, simplicity•Disadvantage:–weaker than RCT with respect to establishing a cause and effect relationship between the exposure and the diseaseBefore and After Studies•Individual level data–Surveys of individual research subjects•Group level data–A widely applied design for injury prevention policy evaluationTwo approaches are used in the literatureTrends in Alcohol Related and Non Alcohol Related Traffic Fatalities Persons Age 16-20 U.S. 1982-20025,2442,78356%42%Non Alcohol Related FatalitiesAlcohol Related FatalitiesSource: U.S. Fatality Analysis Reporting System3,6852,366Federal 21 Drinking Age Law21 the Legal Drinking Age in All 50 StatesHingsonBefore-and-After Group Data Example•Evaluating a bicycle helmet law–A new law goes into effect in 2000. How do we evaluate it’s effect?–Could look at number of bicycle head injuries before and after the lawWarrenHow do we know if the rate of bicycle head injuries would not have gone down anyway, even without the new law?WarrenProblems with Before and After Designs•Problems include:–Identifying individuals to test–Temporal changes;–Regression to the mean.David TorgersonQuestions to ask•Did the study identify a representative sample of individuals to test at points before and after the intervention? Did they test the same persons before and after the intervention?•Are there any identifiable time trends (apart from the intervention) which are likely to distort the results? Can these be measured and adjusted for?Temporal Change•Self-learning occurs.•Example: As children mature they will become better at learning.•Any intervention or treatment is mixed up with these temporal changes. It is difficult to disentangle the two effects.David TorgersonRegression to the Mean•As well as temporal changes before and after studies are confounded by a statistical phenomenon known as ‘Regression to or towards the mean’David TorgersonAnnual Increase in offences with firearmsAmnestyDavid TorgersonDid the Amnesty work?•Unclear, the year preceding the amnesty had a large, unexpected, increase in offences, we would expect through regression to the mean that in the following year the rate of increase would ‘regress’ back to towards the ‘average’ annual increase.David TorgersonEvaluation of the California Child Passenger Safety InitiativeJill Cooper, M.S.W.Kara E. MacLeod, M.A.David Ragland, Ph.D., M.P.H.UC Berkeley Traffic Safety CenterTransportation Research BoardJanuary 12, 2005CPSI Goals1. To increase CPS use among families using selected public health care sites.2. To decrease the rate of CPS misuse among these families.3. To increase awareness of the then-new CA “booster seat” law.CPSI Intervention•Focus on health care providers:–~10,000 restraints distributed to low-income families–>6,600 low-income parents shown how to fit children into restraints–11,000 parents/caregivers educated about CPS–>Almost 200 CPS check-ups held•Focus on parents/guardians:–>700 public health and children’s workers trained in CPSCPSI Evaluation•Quasi-Experimental Design with cross-sectional data collection before/after the interventions at sites to measure any changes in use, misuse and parental knowledge of law•Interview and observational componentObservations •Collected for parents/guardians who had traveled by private vehicle to the hospital or clinic on the day of the interview and who had children with them. • •Observers accompanied families to their vehicles and documented:–vehicle types –presence of air bags in the front passenger seats–children’s ages and weights–restraint types and location in vehicle and position of children in restraints and vehicle.Observational Study Locations•Should be consistent with study purpose•Data collectors must not influence subjects under observation•Factors that influence behavior–Physical conditions–Environment–Heavy traffic–Enforcement activityKnowledge of Booster Seat Law•79.4% of adults in “before” sample reported they knew about the new law.•In “after” interviews, 75% reported they new about the law. (p<0.05)Care Seat Use: Self-reported vs. Observed•Self-reported use did not change much between before and after (not statistically significant)–Car seat use stayed at 83%; booster seat use dipped from 57-56%. •However, observed use of car seats increased from 89-94%


View Full Document

Pitt EPIDEM 2670 - Before and After Studies in Injury Research

Download Before and After Studies in Injury Research
Our administrator received your request to download this document. We will send you the file to your email shortly.
Loading Unlocking...
Login

Join to view Before and After Studies in Injury Research and access 3M+ class-specific study document.

or
We will never post anything without your permission.
Don't have an account?
Sign Up

Join to view Before and After Studies in Injury Research 2 2 and access 3M+ class-specific study document.

or

By creating an account you agree to our Privacy Policy and Terms Of Use

Already a member?