Anna Sarkadi1, Filipa Sampaio2, Michael P Kelly3, Inna Feldman2. 1. Department of Women's and Children's Health, Uppsala University, 751 85 Uppsala, Sweden. Electronic address: Anna.Sarkadi@kbh.uu.se. 2. Department of Women's and Children's Health, Uppsala University, 751 85 Uppsala, Sweden. 3. National Institute of Health and Care Excellence, Centre for Public Health, London, SW1A 2BU UK; Institute of Public Health, General Practice and Primary Care Research Unit, University of Cambridge, Cambridge, CB2 0SR UK.
Abstract
OBJECTIVES: To provide an analytical framework within which public health interventions can be evaluated, present its mathematical proof, and demonstrate its use using real trial data. STUDY DESIGN AND SETTING: This article describes a method to assess population-level effects by describing change using the distribution curve. The area between the two overlapping distribution curves at baseline and follow-up represents the impact of the intervention, that is, the proportion of the target population that benefited from the intervention. RESULTS: Using trial data from a parenting program, empirical proof of the idea is demonstrated on a measure of behavioral problems in 355 preschoolers using the Gaussian distribution curve. The intervention group had a 12% [9%-17%] health gain, whereas the control group had 3% [1%-7%]. In addition, for the subgroup of parents with lower education, the intervention produced a 15% [6%-25%] improvement, whereas for the group of parents with higher education the net health gain was 6% [4%-16%]. CONCLUSION: It is possible to calculate the impact of public health interventions by using the distribution curve of a variable, which requires knowing the distribution function. The method can be used to assess the differential impact of population interventions and their potential to improve health inequities.
OBJECTIVES: To provide an analytical framework within which public health interventions can be evaluated, present its mathematical proof, and demonstrate its use using real trial data. STUDY DESIGN AND SETTING: This article describes a method to assess population-level effects by describing change using the distribution curve. The area between the two overlapping distribution curves at baseline and follow-up represents the impact of the intervention, that is, the proportion of the target population that benefited from the intervention. RESULTS: Using trial data from a parenting program, empirical proof of the idea is demonstrated on a measure of behavioral problems in 355 preschoolers using the Gaussian distribution curve. The intervention group had a 12% [9%-17%] health gain, whereas the control group had 3% [1%-7%]. In addition, for the subgroup of parents with lower education, the intervention produced a 15% [6%-25%] improvement, whereas for the group of parents with higher education the net health gain was 6% [4%-16%]. CONCLUSION: It is possible to calculate the impact of public health interventions by using the distribution curve of a variable, which requires knowing the distribution function. The method can be used to assess the differential impact of population interventions and their potential to improve health inequities.