Preventing RSV bronchiolitis in infants

In young infants, bronchiolitis caused by respiratory syncytial virus (RSV) is a common cause of hospitalization, with an annual incidence of 17.9 hospitalizations per 1,000 infants two months or younger in the United States. Although 4 in 5 children hospitalized for RSV have no risk factors, until 2023, prevention strategies for healthy term newborns were limited. That year, the US Food and Drug Administration approved two new options: the monoclonal antibody nirsevimab and maternal vaccination with a bivalent RSV vaccine, both of which are discussed in a STEPS new drug review in American Family Physician. (In June 2025, a second monoclonal antibody, clesrovimab, became available in the United States.)

In a multicenter randomized trial involving more than 8,000 infants, RSV hospitalization occurred in 0.3% of the group that received nirsevimab during the first winter of life compared with 1.5% of the usual care group (NNT = 83). A study of pediatric emergency departments in Spain found that universal RSV prophylaxis with nirsevimab beginning in the 2023-2024 season was associated with decreases in acute bronchiolitis, related hospital admissions, and pediatric intensive care unit admissions of 59%, 63%, and 63%, respectively. Similarly, infants up to 6 months of age whose mothers received RSV vaccine had a lower likelihood of medically attended severe RSV (0.5% vs 1.8% in the control group; NNT = 81). A Cochrane review of six randomized, placebo-controlled trials confirmed the benefits of RSV vaccination during pregnancy, and a retrospective cohort study of more than 54,000 pregnancies found no increases in preterm birth, premature rupture of membranes, or hypertensive disorders of pregnancy.

The American Academy of Pediatrics, American Academy of Family Physicians, and American College of Obstetricians and Gynecologists recommend routine maternal vaccination or monoclonal antibody administration to prevent RSV infections in infants. Two studies at different institutions found that total uptake (vaccine or nirsevimab) was more than 80%, with acceptance of nirsevimab even in individuals who declined other maternal or infant vaccines. In one of the studies, 43% of parents who refused RSV, influenza, and Tdap vaccines during pregnancy consented to nirsevimab for their newborns. As a result, RSV hospitalization rates during the 2024-25 season were 41% to 51% lower than expected rates and rates in prior years, with the largest benefits observed in infants younger than 2 months.

Presented with two equally recommended prevention options, parents may ask whether one is more effective than the other. Although no head-to-head trials have occurred, a population-based cohort study in France found that compared with infants who were protected by maternal vaccination, those who received nirsevimab had lower risks of RSV hospitalization (adjusted hazard ratio = 0.74) and severe respiratory outcomes, including pediatric intensive care unit admission and the need for oxygen therapy and mechanical ventilation. It is unclear how to apply this finding to the United States, where neither prenatal nor well-child care are universally guaranteed. In a pregnant patient between 32 and 36 6/7 weeks gestation from September 1 through January 31 who is willing to receive RSV vaccine, it makes more sense to vaccinate than to wait to give a monoclonal antibody to the newborn.

**

This post first appeared on the AFP Community Blog.

Safely stopping screening for cervical cancer

All cancer screening tests have diminishing benefits and increasing harms as people age. An article in the December 2025 issue of American Family Physician reviewed considerations for discontinuing cancer screenings in older adults. For cervical cancer, the authors observed that 21% of new cases and 38% of deaths occur after 65 years, the usual age to stop screening. Notably, this stopping age applies only to patients without conditions that put them at high risk of cervical cancer (eg, HIV infection) who have had adequate prior screening, defined as “three consecutive negative cytology results or two consecutive negative cotesting results within 10 years.”

In a recently published guideline, the American Cancer Society (ACS) clarified that the last negative test before exiting cervical cancer screening should occur “at an age no younger than 65 years,” with primary human papillomavirus (HPV) testing at 60 and 65 years of age being preferred. The ACS reiterated its previous recommendation that clinicians continue screening beyond 65 years in those with a history of grade 2 or greater cervical intraepithelial neoplasia within the past 25 years.

An analysis of data from the 2013 and 2015 National Health Interview Surveys found that nearly 1 in 5 women 61 to 65 years of age had not been screened for cervical cancer with a Papanicolaou test in the preceding 5 years. Even in those with a more recent test, determining eligibility to stop screening after 65 years can be challenging. A retrospective study of electronic health records of 42,000 patients older than 65 years in two health systems in Dallas and Boston determined that 69% were not eligible to stop screening because “too few screening tests were documented”; nonetheless, 84% of these patients received no further screening. Of the 16% who continued screening, 2% had high-grade precancerous lesions detected, and 0.7% were diagnosed with cervical cancer.

Clinical studies of cervical cancer screening did not enroll women older than 65 years, so the evidence for the stopping age is derived from a modeling study commissioned by the US Preventive Services Task Force (USPSTF) to support its 2018 recommendations. (The USPSTF released a draft recommendation update in December 2024 that continues to support stopping screening at 65 years in women with adequate prior screening who are not otherwise at high risk of cervical cancer.) A large cohort study sponsored by the National Institutes of Health aims to add to this evidence base by estimating the effects of stopping vs continuing screening after 65 years on cancer outcomes and medical complications resulting from diagnostic procedures and surgical interventions.

**

This post originally appeared on the AFP Community Blog.

DIY medicine - the good, the bad, and the ugly

In a time when artificial intelligence chatbots can answer all of your medical questions, and it's possible to access laboratory testing and imaging scans without a health professional's order, who needs actual doctors?

That's the unwritten premise behind an unsettling article in STAT News that documents the growing phenomenon of empowered patients bypassing doctors and ordering direct-to-consumer tests that they think they need. The article begins with the story of a middle-aged man who shelled out $200 for dozens of lab tests that he didn't think his doctor would want to order, in the hope of obtaining more information to manage his metabolic conditions and lower his risk of developing dementia.

“This is the stuff that if I went to my doctor [with], my doctor would kind of look at me like I’m weird,” he said of the tests. ... Sidaway didn’t want to wait “20 years for the long-term studies and double-blind, placebo-controlled studies for them to say X, Y, and Z works,” he said. “Doctors are going to have to get used to this and navigate this.”

Do-It-Yourself, or DIY medicine, isn't a unique phenomenon among highly trained professions. I'm sure that there are building contractors who roll their eyes at homeowners who get in over their heads with DIY renovations, or trial lawyers who wince when they see people without training try to represent themselves competently in court. Some days, I am tempted to point to the diplomas hanging on my home office wall and proclaim that if a patient believes that a chatbot or Dr. Google can equal my decades of medical training and practice, they are welcome to put their lives in DIY medicine's hands.

On the other hand, I know that the traditional medical system does not serve many people well, and I support the democratization of health information and broadening of access to care - provided that it is effective care. One example that might fall into the DIY medicine bucket is being able to self-screen for cervical cancer from the privacy of one's home. The Department of Health and Human Services (HHS) announced this week that starting in 2027, it will mandate insurance coverage of FDA-approved self-sampling tests for patients who are not at high risk for cervical cancer. That's a win for women who dread going to the doctor for a speculum exam and Pap smear, and it's based on good evidence, synthesized more than a year ago by the U.S. Preventive Services Task Force, whose work has, ironically, been suspended indefinitely by RFK Jr. and the Trump administration.

One particular sentence in the HHS announcement got under my skin: "These additional steps demonstrate the Trump administration’s strong commitment to advancing women’s health and preventing chronic illnesses, including cancer." Really? They are so committed to preventing chronic illnesses and cancer that they have immobilized the panel of highly credentialed experts whose mission it is to review the evidence to identify effective screening tests? In addition, RFK Jr.'s DIY approach to childhood vaccinations (which HHS misleadingly and unethically calls "shared clinical decision making") is likely to increase the incidence of certain cancers. 

Parents who follow the new, expertise-free HHS guidance and disregard their pediatrician or family physician's advice to give their children hepatitis B vaccine at birth, or at all, will expose them to the risk of acquiring chronic viral hepatitis, which will put them on a path to developing liver cancer. (Hepatitis B is treatable with antivirals, but not curable.) And dropping the second dose of the human papillomavirus (HPV) vaccine based on promising but incomplete evidence from other countries who have done so mainly for cost reasons could put more people at risk for not only cervical, but oropharyngeal and anorectal cancers caused by HPV. (The U.S. Food and Drug Administration, which is part of HHS, has not approved a single dose HPV vaccine regimen for cancer prevention.) Finally, cutting $1.1 trillion from Medicaid in last year's budget reconciliation bill is projected to lead to "over 1 million missed [breast, lung, and colorectal] cancer screenings and hundreds of avoidable deaths within 2 years," according to a just-published analysis.

America's experiment with DIY medicine is underway, for better or for worse. Almost certainly, for worse.

De-implementation as quality improvement in primary care

One of my roles as a residency faculty member is to make sure that every resident completes a quality improvement (performance improvement) project by the time they graduate. Nearly all of these projects focus on getting clinicians to do more of something that improves health outcomes: lowering blood pressure, prescribing controller medications for asthma, controlling blood glucose and cholesterol levels in diabetes, providing guideline-consistent postpartum care. But quality improvement also includes doing less of things that don't improve (and may worsen) patients' health: vitamin D screening and supplementation, antireflux medications for healthy infants, cancer screening in older adults with limited life expectancies.

The American Board of Internal Medicine Foundation's Choosing Wisely campaign, which concluded in 2023 but remains active in many countries outside of the U.S., identified hundreds of "low value care" interventions, such as preventive care that is provided too soon or too often (e.g., screening colonoscopy). A 2020 paper reviewed more than 400 opportunities for stopping or scaling back unnecessary services in primary care and narrowed the field to 37 valid, high-priority recommendations. But identifying health care services to do less often or stop doing is the easy part; as "too much medicine" is often deeply embedded in systems, de-implementation is the real challenge.

A systematic review and meta-analysis of randomized trials of primary care de-implementation strategies identified 140 studies that aimed to reduce antibiotic use (54%), other drug treatments (30%), imaging (12%), and/or laboratory testing (11%). The authors found moderate certainty evidence that provider education combined with audit and feedback reduced targeted low value care, while provider education, audit and feedback, and/or patient education alone had limited or no benefits. Interventions that incorporated multiple strategies were the most effective, reducing relative risks of low value care provision by 30% to 35% over a median of 287 days of follow-up.

Compared to the much larger research base on intensifying care - I suspect there are hundreds, if not thousands, of studies on lowering blood pressure alone - the foundation for de-implementing low value care clearly needs to be expanded and strengthened. But there's more than enough here for a motivated learner - say, a family medicine resident - to build on.