The best doctors don’t simply offer more care: they deliver more effective care
Do some doctors provide better care than others? Despite the central role general practitioners play in healthcare systems around the world, we know surprisingly little about how much individual doctors differ in their ability to improve patient outcomes. Using rich administrative data from Norway and a natural experiment involving doctor reassignment, we estimate what’s known as “doctor value-added”—that is, the causal impact of general practitioner quality on patient survival. Our findings reveal significant variation in how effective doctors are, with major consequences for patient health. These differences are not explained by observable doctor characteristics or common practice styles, and patients themselves seem unable to tell which doctors are better at improving outcomes.
When a general practitioner retires or leaves their practice in Norway, patients are automatically reassigned to a new local doctor. This process is essentially random, creating a natural experiment that allows us to isolate the direct effect of individual general practitioner quality on health. We study tens of thousands of patients aged 55 and older who were reassigned in this way, and we link detailed patient and doctor data to measure general practitioner quality based on changes in patient mortality following the switch. This enables us to track the long-term impact of doctor quality on health outcomes.
The differences between doctors are not just statistically significant—they are substantial and meaningful. A one-standard-deviation increase in general practitioner quality reduces the risk of dying within two years by 15.7 percentage points. Patients randomly assigned to high-performing doctors live, on average, 3.5 months longer. These effects are striking, particularly in a universal healthcare system like Norway’s, which is designed to ensure equal access and reduce inequality. Even in such a setting, the effectiveness of individual doctors plays a powerful role in shaping patient outcomes.
So, what makes some doctors more effective than others? We examine a broad range of observable characteristics, including years of experience, specialization, and the number of patients seen, but none explain much of the variation in doctor effectiveness. Similarly, features of a doctor’s practice style, such as how long they spend with each patient or how often they refer patients to specialists, do not predict better outcomes. The most effective doctors aren’t simply providing more care: they are providing more effective care. Their patients are less likely to visit the emergency room and have lower mortality from diseases like cancer, where early detection is critical. This suggests that the most effective doctors make better clinical decisions early in the care process, even if those decisions aren’t immediately obvious to patients or policymakers.
If some doctors really are better, can patients identify them? Our study suggests not. Patients who voluntarily switch doctors don’t tend to move toward higher-quality general practitioners. Moreover, subjective online ratings—one of the few sources of publicly available information about doctor quality—are completely unrelated to actual general practitioner effectiveness. This points to a major information gap in the healthcare system. Patients can’t tell who the good doctors are, and the tools they have to guide their choices are misleading or irrelevant. The consequences of this gap extend beyond individual health. They touch on issues of efficiency, fairness, and the overall effectiveness of the healthcare system.
In the end, our findings raise difficult but important questions. Doctor quality varies more than we might think, it matters a great deal for patient health, and yet we don’t fully understand what makes a doctor effective. If neither patients nor health systems can reliably identify the best doctors, improving primary care through better doctor quality will require new strategies, whether through improved training, better performance metrics, or deeper research into the art and science of clinical decision-making.
© Rita Ginja, Julie Riise, Barton Willage, Alexander Willén
Rita Ginja is Professor at the University of Bergen, Norway, and IZA Research Fellow
Julie Riise, is Associate Professor at the University of Bergen, Norway, and IZA Research Fellow
Barton Willage is Associate Professor at University of Delaware
Alexander Willén is Professor at the Norwegian School of Economics, and IZA Research Affiliate
Please note:
We recognize that IZA World of Labor articles may prompt discussion and possibly controversy. Opinion pieces, such as the one above, capture ideas and debates concisely, and anchor them with real-world examples. Opinions stated here do not necessarily reflect those of the IZA.
Related IZA World of Labor content:
https://wol.iza.org/articles/economics-of-mental-health by Richard Layard
https://wol.iza.org/articles/relationship-between-recessions-and-health by Nick Drydakis
https://wol.iza.org/articles/disability-and-labor-market-outcomes by Melanie Jones
Foto by CDC on Unsplash