Premium
Risk of death from human insulin
Author(s) -
Suissa S.,
Spitzer W. O.,
Abenhaim L.,
Downey W.,
Gardiner R. J.,
Fitzgerald D.
Publication year - 1992
Publication title -
pharmacoepidemiology and drug safety
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.023
H-Index - 96
eISSN - 1099-1557
pISSN - 1053-8569
DOI - 10.1002/pds.2630010403
Subject(s) - medicine , insulin , human insulin , cohort , population , cohort study , diabetes mellitus , cause of death , demography , pediatrics , environmental health , endocrinology , disease , sociology
Abstract In response to growing concerns surrounding the risks of human insulin, we conducted a 13‐year historical cohort study of 4925 insulin users less than 40 years of age. This cohort was formed from the population of Saskatchewan, Canada, from 1978–90, using the computerized databases arising from the universal health insurance plan of this Province of 1.1 million inhabitants. Of 101 deaths, 53 occurred within the defined one‐month time window of effect. The cohort resulted in 14,600 person‐years of exposure to insulin use. The overall rate of death was 3.6 per 1000 person‐years of insulin use (95% CI: 2.7–4.6). This death rate remained stable throughout the 13‐year study period and did not increase after 1984, the year of introduction of human insulin in Saskatchewan. We found 46 deaths for animal insulin (4.0/1000/year) and 7 deaths for human insulin (2.3/1000/year). Among the latter 7 deaths, 1 death occurred in the group treated solely with human insulin (1.2/1000/year), and 6 deaths in those who were switched from animal to human insulin (2.8/1000/year). These observed differences are not statistically significant. The data from this rapid response pharmacoepidemiological study, which took 8 months to complete, do not support allegations that treatment with human insulin increases the risk of death. They do not however exclude the possibility that human insulin is associated with an elevated risk of fatal or non‐fatal hypoglycaemia.