2017

Hospital Readmission is Highly Predictable from Deep Learning

We develop and test a method that can be applied generally by hospitals to improve readmission prediction. Relying on state-of-the art machine learning algorithms, we predict the probability of 30-day readmission at initial admission and at discharge using administrative data on 1,633,099 hospital stays in Quebec between 1995 and 2012. Deep Learning produces excellent predictions of readmission province-wide, and Random Forest reaches similar levels.

Premature Deaths, Accidental Bequests and Fairness

While little agreement exists regarding the taxation of bequests in general, there is a widely held view that accidental bequests should be subject to a confiscatory tax. We reexamine the optimal taxation of accidental bequests in an economy where individuals care about what they leave to their offspring in case of premature death. We show that, whereas the conventional 100 %-tax view holds under the standard “utilitarian social welfare” criterion, it does not hold under the “ex post egalitarian” criterion, which assigns a strong weight to the welfare of unlucky short-lived individuals.

Physical Disability and Labor Market Discrimination: Evidence from a Field Experiment

We investigate labor market discrimination toward wheelchair users using data from a large-scale field experiment conducted in the province of Quebec (Canada). Applications were randomly sent to 1,477 private firms operating in two urban areas and advertising open positions requiring various skill levels. The applications were randomly generated to cover a broad spectrum of potential determinants of discrimination (gender, skill level, work history, workplace adjustment costs, etc.). We find average callback rates of disabled and non-disabled applicants of 14.4% and 31%, respectively.

Whistle While You Work: Job Insecurity and Older Workers’ Mental Health in the United States

We estimate the effects of job insecurity on older workers’ health outcomes using downsizing episodes and state-industry level changes in employment. We provide evidence that job insecurity, as measured by the self-reported probability of job loss, increases stress at work and the risk of clinical depression and lowers self-reported health status. We interpret our findings as evidence that job insecurity which is outside the control of workers may have large effects on mental health.

Long-term Care Insurance: Knowledge Barriers, Risk Perception and Adverse Selection

We conduct a survey where we ask respondents to rate various insurance products aimed at protecting against long-term care financial risks. We use price variation built into the survey, as well as objective care and death risks for Canada and respondents’ purchasing probabilities, to predict market equilibrium for long-term care insurance. We then investigate the low take-up of long-term care insurance and find that 44% of respondents without insurance were never offered any, and 31% of the overall sample had no knowledge of the product.

Partners

Industrielle Alliance, Groupe financier

2019 © Chaire de recherche Industrielle Alliance sur les enjeux économiques des changements démographiques