Using the numerical method in 1836, James Jackson bridged French therapeutic epistemology and American medical pragmatism.

OBJECTIVES

To review James Jackson's analysis of bloodletting among pneumonitis patients at the newly founded Massachusetts General Hospital, in which he implemented the numerical method advocated by Pierre-Charles-Alexandre Louis.

STUDY DESIGN AND

SETTING

The study sample included 34 cases of clinically diagnosed pneumonitis admitted to Massachusetts General Hospital between April 19, 1825, and May 10, 1835, and discharged alive. Patient data were extracted from meticulously kept case books. Jackson calculated mean number of venesections, ounces of blood taken, and days of convalescence within groups stratified by day of the disease when first bloodletting occurred. He also calculated average convalescence within groups stratified by age, sex, prior health, vesication, and day of the disease when the patients were admitted to the hospital.

RESULTS

To Jackson's surprise, it "seemed to be of less importance, whether our patients were bled or not, than whether they entered the hospital early or late" after the onset of the pneumonitis. Bloodletting was ineffective. Our multivariate reanalysis of his data confirms his conclusion. Outstandingly for his time, Jackson ruled out unwarranted effects of covariates by tabulating their numerical relations to the duration of pneumonia.

CONCLUSION

Using novel gathering of patient clinical data from hospital records and quantitative analytical methods, Jackson contributed results that challenged conventional wisdom and bridged French therapeutic epistemology and American medical pragmatism.