Robert Aronowitz on Risky Medicine

Episode Summary

Executive Summary: The conversation critiques routine medical screening and risk-based medicine, arguing that many tests are adopted before strong evidence exists and often create anxiety, false positives, and overtreatment. Aronowitz emphasizes that decisions should weigh benefits, harms, and patient peace of mind, while Roberts highlights the difficulty of interpreting imperfect evidence and the need for skepticism about both screening and the studies behind it.

Main Topics: Screening, uncertainty, and the ethics of disrupting peace of mind (Priority: 5/5): The discussion argues that screening healthy people should face a very high evidentiary and ethical bar because it can create anxiety and medicalization without clear benefit. Overdiagnosis and overtreatment in prostate cancer (Priority: 5/5): PSA screening is used as the main example of how a test can lead to biopsies, surveillance, and surgery driven partly by fear of uncertainty rather than improved survival. Personal refusal of routine prenatal ultrasound and medicalization (Priority: 4/5): Aronowitz describes how he and his wife opted out of routine ultrasounds in pregnancy to avoid incidental findings and false positives, illustrating resistance to unnecessary medical interventions. Framingham and the rise of risk-factor thinking (Priority: 4/5): The Framingham Heart Study is presented as historically influential but too focused on individual-level risk factors, helping create a broader medical culture centered on risk reduction. Mammography evidence, false positives, and public debate (Priority: 4/5): The postscript examines evidence on mammography, the limits of older randomized trials, and how difficult it is to interpret benefit versus harm estimates in screening policy. Behavioral economics, regret, and decision-making under uncertainty (Priority: 3/5): The speakers connect medical choices to anticipated regret, false positives, and the psychological preference to feel one has done everything possible. Medical education and statistical literacy (Priority: 3/5): The conversation closes with concern that physicians are not trained deeply enough in evidence interpretation, risk analysis, and the human consequences of testing.

Key Arguments: Routine screening of asymptomatic people should not be treated as self-evidently beneficial; it should require strong randomized evidence because it can produce harm through false positives, incidentalomas, and unnecessary follow-up. PSA screening often initiates a cascade: elevated test, anxiety, referral, biopsy, surveillance, and sometimes surgery, even when the underlying cancer may be slow-growing and unlikely to matter. Many patients choose surgery not primarily to prolong life but to escape the psychological burden of living with uncertainty and the 'sword of Damocles' effect of a diagnosis. The Framingham model advanced useful heart-disease prediction but also entrenched a narrow, individual-level framework that underplays social, political, and industrial causes of disease. Observed risk-factor associations in chronic disease epidemiology are often unstable and can be reversed by later studies, especially when effects are small. Shared decision-making is not a cure-all when evidence is weak; sometimes the more ethical choice is not to initiate testing at all. The evidence base behind common screening practices such as mammography is complicated, and listeners should not treat summaries as simple counts; sources, assumptions, and trial vintage matter. Medical and consumer culture often turns screening into a norm that reduces fear and creates social expectations, making it hard to reverse even when evidence is uncertain. Physicians and patients alike need better statistical and historical literacy to evaluate benefits, harms, and conflicts of interest in prevention and screening. There is a risk that future genetic and digital health data will intensify fear and create new markets for interventions before efficacy is established.

Data Points: Birth year of first child discussed: 1992 - Aronowitz and his wife describe declining routine prenatal screening when their son was born in 1992. PSA follow-up frequency: Every 6 months - He describes the surveillance routine that often follows an elevated PSA and biopsy. Prostate biopsy sample count: 20 to 25 needle specimens - Modern ultrasound-guided prostate biopsy can involve taking up to 20 or 25 samples. Mammography deaths averted: 10 deaths averted per 10,000 women - Russ Roberts cites a JAMA article discussing annual mammograms for women ages 50 to 59. Mammography false positives: 6,100 false positives per 10,000 screenings - In the postscript, Roberts summarizes a chart on the harms of mammography screening. Mammography biopsies with no cancer: Over 900 - Roberts says the cited evidence included more than 900 biopsies that found nothing. PSA randomized-trial evidence delay: First randomized screening results in 2009 - Aronowitz notes that widespread PSA testing long preceded randomized trial evidence. PSA treatment threshold: About 50 men treated for one life saved - He summarizes a rule-of-thumb estimate from prostate screening studies. Relative risk example from genetic testing: 3x diabetes risk, 2x heart disease risk, 50% lower testicular cancer risk - Aronowitz warns that future genetic reports may generate anxiety without actionable interventions. Framingham study start: 1949 - He discusses the historical origins of the Framingham Heart Study.

Pivotal Quotes: "I think we need the best possible evidence there is, which means investing in knowledge production, in the form of clinical trials" — Robert Aronowitz: Aronowitz argues for a high evidentiary bar before recommending screening and prevention. "I don't want to know that number because it's not a meaningful number." — Russ Roberts: Roberts explains why he wants to avoid routine PSA testing after discussing the harms of screening. "I'd rather go to my death knowing that all there was no stone unturned." — Robert Aronowitz: Aronowitz cites a historical breast cancer case to illustrate anticipated regret and the desire to do everything possible.

Implications: Listeners should treat routine screening and risk information with caution, checking evidence quality, trial design, and trade-offs. The medical system may need stronger thresholds before introducing new tests, especially genetic ones, to avoid fear-driven overtreatment.