By Kenneth W. Lin, MD, MPH

One of the lessons I try to impart on my medical students and residents is the importance of humility regarding established medical knowledge. The old saying that up to half of what one learns in medical school will later turn out to be wrong isn’t far from the truth. Two axioms from my training in the late 1990s — that every postmenopausal woman should receive estrogen replacement therapy and most older adults should take a daily baby aspirin — were overturned by subsequent research.

The pace of scientific progress hasn’t slowed since I began writing a primary care column in 2015. In this, my 100th commentary, I would like to reflect on how three subjects I explored during that first year of my column have evolved over time.

Lung Cancer Screening

My very first column cautioned that low-dose CT screening for lung cancer in high-risk patients, which the US Preventive Services Task Force (USPSTF) first recommended in 2013 based on the positive results of the National Lung Screening Trial (NLST), could lead to more harm than good. I returned to this theme in a 2018 column that defended primary care clinicians who were not yet referring large numbers for screening. In it, I wrote, “The one-sided message that low-dose CT scans save lives and need to be done on more patients completely skips over the potential harms of lung cancer screening: false-positive results, incidental findings, radiation exposure, and overdiagnosis.” I highlighted two preliminary studies suggesting that false-positive rates and procedural complications were higher in community hospitals.

In 2020, I changed my mind on lung cancer screening, observing that “multiple studies have confirmed the NLST’s findings and the ability of community radiology centers to meet the same technical standards and low adverse event rates as academic institutions,” and that false-positive and overdiagnosis rates were lower than initial projections. I am not, however, ready to expand the pool of eligible patients to those who quit smoking more than 15 years ago, as the American Cancer Society now recommends. In a 2023 column on why I decided to stick with the USPSTF guidelines, I noted that “other than disease-modeling studies, no data have shown that conducting low-dose CT scans beyond the 15 years-since-quitting benchmark does more good than harm.”

Hypertension Management

Another 2015 column that I revisited over the years was my initial take on the SPRINT trial that has had an outsized influence on hypertension guidelines over the past decade. I pointed out that SPRINT — which found fewer cardiovascular events and lower mortality in persons treated to a blood pressure target of less than 120/80 mm Hg compared to 140/90 mm Hg — had restrictive eligibility criteria and measured blood pressure in ways that are not always practical in primary care.

When a 2017 guideline from the American College of Cardiology/American Heart Association endorsed lower blood pressure targets, I pushed back against what I viewed as a case of guideline creep, observing that the guideline’s “ major omissions included a discussion of the overall harms of more intensive treatment, and shared decision-making tools for physicians to weigh the harms and benefits in individual patients.” And in a 2018 discussion with Dr William Cushman, a member of the Joint National Committee 7 and 8 guideline panels, I addressed the challenges of measuring blood pressure accurately in the office and inconsistent patient access to ambulatory and home blood pressure monitoring. Of note, the American Academy of Family Physicians continues to endorse a blood pressure target of less than 140/90 mmHg for most patients, based on its 2022 guideline that I co-authored.

Primary Care Policies

Finally, I have written several columns on primary care training, payment, and outcomes. In a 2015 column, I discussed an analysis that showed that “Medicare [funding] mostly subsidizes graduate medical education in places that train mostly subspecialists,” and in 2016 I explored US medical students’ low interest in primary care careers in the annual residency Match compared to higher-income subspecialties. Unfortunately, this issue persists a decade later, as a record number of family medicine positions went unfilled.

In 2017, I reviewed evidence that access to primary care is the mechanism by which having health insurance saves lives. Then in 2019 and 2023, I examined the pros and cons of prospective primary care payment models designed as Medicare pilot programs; these models have had mixed success in reducing costs and improving outcomes, but unfortunately they have not been expanded to the entire Medicare population or widely duplicated in private insurance plans.

That’s too bad, because as I pointed out in a 2024 column, America’s primary care physicians have been doing an admirable job with their limited resources in making headway against some of the toughest issues in medicine. Fewer people are dying of drug overdoses. The prevalence of obesity has stabilized. And youth smoking rates — cigarettes and vaping — have been falling.