Evelyn L Forget1. 1. Community Health Sciences, Faculty of Medicine, University of Manitoba, 750 Bannatyne Ave., Winnipeg, MB R3E 0W3, Canada. Electronic address: evelyn.forget@med.umanitoba.ca.
Abstract
OBJECTIVES: This study investigates whether administration data from universal health insurance can yield new insight from an old intervention. Specifically, did a guaranteed annual income experiment from the 1970s, designed to investigate labor market outcomes, reduce hospitalization rates? METHOD: The study re-examined the saturation site of a guaranteed annual income experiment in Dauphin, Manitoba (CANADA) conducted between 1974 and 1979 (MINCOME). We used health administration data generated by the universal government health insurance plan to identify subjects (approximately 12,500 residents of Dauphin and its rural municipality). We used propensity-score matching to select 3 controls for each subject from this database, matched on geography of residence, age, sex, family size and type. Outcome measures were hospital separations and physician claims. RESULTS: Hospital separations declined 8.5% among subjects relative to controls during the experimental period. Accident and injury codes and mental health codes were most responsible for the decline. CONCLUSIONS: Even though MINCOME was designed to measure the impact of a GAI on the number of hours worked, one can re-visit old experiments with new data to determine the health impact of population interventions designed for other purposes. We determined that hospitalization rates declined significantly after the introduction of a guaranteed income.
OBJECTIVES: This study investigates whether administration data from universal health insurance can yield new insight from an old intervention. Specifically, did a guaranteed annual income experiment from the 1970s, designed to investigate labor market outcomes, reduce hospitalization rates? METHOD: The study re-examined the saturation site of a guaranteed annual income experiment in Dauphin, Manitoba (CANADA) conducted between 1974 and 1979 (MINCOME). We used health administration data generated by the universal government health insurance plan to identify subjects (approximately 12,500 residents of Dauphin and its rural municipality). We used propensity-score matching to select 3 controls for each subject from this database, matched on geography of residence, age, sex, family size and type. Outcome measures were hospital separations and physician claims. RESULTS: Hospital separations declined 8.5% among subjects relative to controls during the experimental period. Accident and injury codes and mental health codes were most responsible for the decline. CONCLUSIONS: Even though MINCOME was designed to measure the impact of a GAI on the number of hours worked, one can re-visit old experiments with new data to determine the health impact of population interventions designed for other purposes. We determined that hospitalization rates declined significantly after the introduction of a guaranteed income.
Authors: Erin Hobin; Anders Swanson; Gillian Booth; Kelly Russell; Laura C Rosella; Brendan T Smith; Ed Manley; Wanrudee Isaranuwatchai; Stephanie Whitehouse; Nicole Brunton; Jonathan McGavock Journal: BMJ Open Date: 2020-02-18 Impact factor: 2.692
Authors: Nav Persaud; Hannah Woods; Aine Workentin; Itunu Adekoya; James R Dunn; Stephen W Hwang; Jonathon Maguire; Andrew D Pinto; Patricia O'Campo; Sean B Rourke; Daniel Werb Journal: CMAJ Date: 2022-01-17 Impact factor: 8.262
Authors: Katharine B Burmaster; John C Landefeld; David H Rehkopf; Maureen Lahiff; Karen Sokal-Gutierrez; Sarah Adler-Milstein; Lia C H Fernald Journal: BMJ Open Date: 2015-08-03 Impact factor: 2.692