P. La Rochelle · Journal of the Royal Society of Medicine 2013 · interrupted time-series analysis (SARIMA model) · n=?

How dramatic were the effects of handwashing on maternal mortality observed by Ignaz Semmelweis?

Cited 13 times in the scientific literature.

Level 4 - case-series / case-control

Historically controlled interrupted time-series analysis of historical observational data.

OpenAlex W2047101827 · doi:10.1177/0141076813507843 · record verified 2026-08-28

What was done

Authors conducted an interrupted time-series analysis of monthly maternal mortality rates from a Vienna maternity hospital between January 1841 and February 1849. They modeled the data using a Seasonal AutoRegressive Integrated Moving Average (SARIMA) model to evaluate the effect of introducing handwashing in June 1847 while accounting for seasonality, autocorrelation, and moving average errors.

What was found

The average monthly mortality rate was 10.65% before handwashing and 1.98% after its introduction, representing a risk ratio of 5.38. The SARIMA model estimated a statistically significant absolute reduction in maternal mortality of 8.87% (P = 0.0042), equivalent to one death prevented for every 11 women examined.

Why it matters

This analysis formally confirms with modern time-series modeling that the dramatic reduction in maternal mortality observed by Semmelweis was statistically robust and unlikely to be explained by seasonal variation or random chance.

Limits

The abstract provides aggregate monthly percentage rates rather than absolute participant counts (n is not reported). The study relies on historical controls from a single center, does not disaggregate data between the hospital's two clinics, and does not evaluate earlier historical periods prior to routine postmortem examinations.

Cited by