269090 Proposed Alternatives to the Data Analysis Techniques of a National Road Safety Program in Brazil: Application of the Population Attributable Fraction Methodology

Tuesday, October 30, 2012

Birchie Whitman, BA, MSPH , Department of International Health, Health Systems, Johns Hopkins University - International Injury Research Unit, Baltimore, MD
Background/Purpose: In 2011 JHU's International Injury Research Unit (IIRU) was contracted by the Global Road Safety Partnership (GRSP) to observe and propose an evaluation plan for their implementation program for road safety interventions in Brazil, the Proactive Partnership Strategy (PPS). As part of this proposal, suggestions were made on improvements to PPS's data management techniques, specifically in reference to their methods for attributing a road traffic crash to a risk factor.

Methods: The Levin formula for Population Attributable Fraction (PAF) was applied in varying manners to 85 injuries reported from Belo Horizonte during the first two months of 2011 to determine the potential involvement of important risk factors for traffic crashes.

Results/Outcomes: Re-attribution of crash risk factors using a PAF approach points to possible methodological issues in PPS' current system of crash data analysis. Application of the PAF methodology to the crashes in question resulted in significant changes in the percentage of accidents that are attributed to the main risk factors of speed and alcohol. Most notably, the proportion of crashes attributed to speed dropped from 67.9% of all crashes to 54.7% or 51.4% depending on the method used.

Conclusions: Alternative, systematic methodologies for attribution of a crash to a risk factor should be considered in programs similar to PPS in order to ensure the most accurate understanding of risk factor involvement that is possible given limited information of individual crashes. The results of such analysis have serious ramifications for the allocation of program resources.

Learning Areas:
Assessment of individual and community needs for health education
Conduct evaluation related to programs, research, and other areas of practice

Learning Objectives:
To compare the proportions of road safety health outcomes that are attributed to a range of risk factors for crash when using various, contrasting methods of attribution by redistributing the crashes recorded by a road safety program in Brazil using statistical and population attributable fraction (PAF) approaches.

Keywords: Injuries, Risk Factors

Presenting author's disclosure statement:

Qualified on the content I am responsible for because: I am a current student of international public health at the Johns Hopkins Bloomberg School of Public Health and received funding from the Center for Global Health to complete an international fieldwork experience as part of my degree. This funded experience led to the data that I present with this abstract. In past positions, I have assisted in clinical research trials relating to the development of a microbicide gel prophylaxis against HIV/AIDS.
Any relevant financial relationships? No

I agree to comply with the American Public Health Association Conflict of Interest and Commercial Support Guidelines, and to disclose to the participants any off-label or experimental uses of a commercial product or service discussed in my presentation.

Back to: 4171.0: Data and Methods posters