Estimating Distance to Health Services Using Probabilistic Sampling
Author Information
Author(s): Kirsten MM Beyer, Audrey F Saftlas, Anne B Wallis, Corinne Peek-Asa, Gerard Rushton
Primary Institution: Institute for Health and Society, Medical College of Wisconsin
Hypothesis
Can a probabilistic sampling method improve the estimation of geographic distance to health services when only the region of residence is known?
Conclusion
The probabilistic sampling method improves the precision of distance estimates to health services when complete residential address information is unavailable.
Supporting Evidence
- The PSM provides a distribution of possible distances rather than a single estimate.
- Using the median distance minimizes the impact of outliers in distance calculations.
- Confidence intervals for the PSM distances indicate the range of likely distances to health services.
Takeaway
This study created a new way to figure out how far people have to travel to get to health services, even when we only know their ZIP Code.
Methodology
The study used a probabilistic sampling method to estimate distances from census blocks within a ZIP Code to the nearest health facility, using Monte Carlo sampling techniques.
Potential Biases
Potential biases include the use of ZIP Code centroids that may not accurately represent population centers and the reliance on outdated Census data.
Limitations
The study's distance estimates may not reflect actual choices women make in seeking services, and demographic factors beyond age and gender were not considered.
Participant Demographics
Participants were women aged 18-44 years seeking abortion services.
Statistical Information
Confidence Interval
93%
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website