What do you think?
Patient Care under Uncertainty
How cutting-edge economics can improve decision-making methods for doctors
Although uncertainty is a common element of patient care, it has largely been overlooked in research on evidence-based medicine. Patient Care under Uncertainty strives to correct this glaring omission. Applying the tools of economics to medical decision making, Charles Manski shows how uncertainty influences every stage, from risk analysis to treatment, and how this can be reasonably confronted.
In the language of econometrics, uncertainty refers to the inadequacy of available evidence and knowledge to yield accurate information on outcomes. In the context of health care, a common example is a choice between periodic surveillance or aggressive treatment of patients at risk for a potential disease, such as women prone to breast cancer. While these choices make use of data analysis, Manski demonstrates how statistical imprecision and identification problems often undermine clinical research and practice. Reviewing prevailing practices in contemporary medicine, he discusses the controversy regarding whether clinicians should adhere to evidence-based guidelines or exercise their own judgment. He also critiques the wishful extrapolation of research findings from randomized trials to clinical practice. Exploring ways to make more sensible judgments with available data, to credibly use evidence, and to better train clinicians, Manski helps practitioners and patients face uncertainties honestly. He concludes by examining patient care from a public health perspective and the management of uncertainty in drug approvals.
Rigorously interrogating current practices in medicine, Patient Care under Uncertainty explains why predictability in the field has been limited and furnishes criteria for more cogent steps forward.
Although uncertainty is a common element of patient care, it has largely been overlooked in research on evidence-based medicine. Patient Care under Uncertainty strives to correct this glaring omission. Applying the tools of economics to medical decision making, Charles Manski shows how uncertainty influences every stage, from risk analysis to treatment, and how this can be reasonably confronted.
In the language of econometrics, uncertainty refers to the inadequacy of available evidence and knowledge to yield accurate information on outcomes. In the context of health care, a common example is a choice between periodic surveillance or aggressive treatment of patients at risk for a potential disease, such as women prone to breast cancer. While these choices make use of data analysis, Manski demonstrates how statistical imprecision and identification problems often undermine clinical research and practice. Reviewing prevailing practices in contemporary medicine, he discusses the controversy regarding whether clinicians should adhere to evidence-based guidelines or exercise their own judgment. He also critiques the wishful extrapolation of research findings from randomized trials to clinical practice. Exploring ways to make more sensible judgments with available data, to credibly use evidence, and to better train clinicians, Manski helps practitioners and patients face uncertainties honestly. He concludes by examining patient care from a public health perspective and the management of uncertainty in drug approvals.
Rigorously interrogating current practices in medicine, Patient Care under Uncertainty explains why predictability in the field has been limited and furnishes criteria for more cogent steps forward.
184 pages, Hardcover
Published September 10, 2019
Ratings & Reviews
Friends & Following
Create a free account to discover what your friends think of this book!
Community Reviews
Displaying 1 - 1 of 1 review
November 18, 2020
A brief introduction to, and plea for, the use of formal decision theory in clinical decision-making, with simple toy examples, applications drawn from treatment, testing, and public health, mostly based on Manski's own work and that of his collaborators.
Manski's approach, "partial identification", based on the principles of using the fewest and most believable assumptions possible to derive bounds on unknown quantities, like the effectiveness of a drug when the trial data suffers from noncompliance or missing data, is reviewed here and its implications in medical settings drawn out. For long-time aficionados and readers of his previous books, this material is largely a rehash, a repackaging for a new audience relative to his previous books targeted more at public policy applications. The somewhat newer material is on decision theory.
Beginning with a review of the Wald statistical decision theory framework based on decision rules evaluated by criteria over loss functions, there is a discussion of how the maximin and, especially, minimax regret approaches pair naturally with partial identification of clinically relevant parameters, resulting in defensible decision procedures even in settings where there does not exist a full probabilistic description of uncertainty. This is paired with commentary on existing approaches to developing clinical guidelines and FDA approval processes, which tend to rely on a mix of classical binary hypothesis testing and some degree of informal reasoning. The criticisms of hypothesis testing are well-worn, at least to anyone used to the decision theory approach. The criticisms of the more ad hoc parts like ratings scales are a little harder to assess, as while some components are hard to justify in a formal sense, they may in part be mechanisms for heuristic application of expert judgment, in which case, given the evidence he presents on the value provided by clinical guidelines, it may make sense to reserve judgment.
In terms of critiques, I did find odd the lack of reference to the now deep and extensive literature in computer science which relies on the regret-based approach he advocates, particularly in dynamic settings and with learning and experimentation. While perhaps only recently making a comeback in statistics and econometrics, in some parts of the academy these ideas are entirely mainstream. Stylistically, the book suffers from not quite targeting one audience. The body is fully informal, with few equations, but the discussion relies on mathematical concepts and examples, relegated to appendices, without which the text is unclear, and which, due to their separation from the text, are harder to follow on their own, and some proofs or explanations are deferred entirely with references to the original papers. The text alone then seems like it may not be clear enough to persuade the typical clinician at whom the book is ostensibly aimed, and a biostatistician or health economist with the requisite technical background would be able to follow more closely with a bit more in the way of math in the main text. However, the book is a quick read and the material does get across, if a bit drily.
Overall I would recommend this to economists looking into health applications or health researchers looking for a gateway to the powerful and principled ideas behind methods based on identification and decision rules. For economics and policy researchers new to the area, I would instead point them to Manski's earlier books, such as "Identification for Prediction and Decision."
Manski's approach, "partial identification", based on the principles of using the fewest and most believable assumptions possible to derive bounds on unknown quantities, like the effectiveness of a drug when the trial data suffers from noncompliance or missing data, is reviewed here and its implications in medical settings drawn out. For long-time aficionados and readers of his previous books, this material is largely a rehash, a repackaging for a new audience relative to his previous books targeted more at public policy applications. The somewhat newer material is on decision theory.
Beginning with a review of the Wald statistical decision theory framework based on decision rules evaluated by criteria over loss functions, there is a discussion of how the maximin and, especially, minimax regret approaches pair naturally with partial identification of clinically relevant parameters, resulting in defensible decision procedures even in settings where there does not exist a full probabilistic description of uncertainty. This is paired with commentary on existing approaches to developing clinical guidelines and FDA approval processes, which tend to rely on a mix of classical binary hypothesis testing and some degree of informal reasoning. The criticisms of hypothesis testing are well-worn, at least to anyone used to the decision theory approach. The criticisms of the more ad hoc parts like ratings scales are a little harder to assess, as while some components are hard to justify in a formal sense, they may in part be mechanisms for heuristic application of expert judgment, in which case, given the evidence he presents on the value provided by clinical guidelines, it may make sense to reserve judgment.
In terms of critiques, I did find odd the lack of reference to the now deep and extensive literature in computer science which relies on the regret-based approach he advocates, particularly in dynamic settings and with learning and experimentation. While perhaps only recently making a comeback in statistics and econometrics, in some parts of the academy these ideas are entirely mainstream. Stylistically, the book suffers from not quite targeting one audience. The body is fully informal, with few equations, but the discussion relies on mathematical concepts and examples, relegated to appendices, without which the text is unclear, and which, due to their separation from the text, are harder to follow on their own, and some proofs or explanations are deferred entirely with references to the original papers. The text alone then seems like it may not be clear enough to persuade the typical clinician at whom the book is ostensibly aimed, and a biostatistician or health economist with the requisite technical background would be able to follow more closely with a bit more in the way of math in the main text. However, the book is a quick read and the material does get across, if a bit drily.
Overall I would recommend this to economists looking into health applications or health researchers looking for a gateway to the powerful and principled ideas behind methods based on identification and decision rules. For economics and policy researchers new to the area, I would instead point them to Manski's earlier books, such as "Identification for Prediction and Decision."
Displaying 1 - 1 of 1 review