This Week in Family Medicine — Jun 30, 2026

Welcome to This Week in Family Medicine. This week we are covering nine notable papers spanning maternal-fetal and neonatal health, geriatric cardiovascular prevention, and emerging diagnostic and therapeutic frontiers in oncology. Let us dive in.

We begin with reassuring news for prenatal care and maternal counseling, addressing a common clinical dilemma in family medicine. For years, observational studies have suggested an association between prenatal acetaminophen exposure and an increased risk of autism spectrum disorder and attention-deficit/hyperactivity disorder in children. This has created significant anxiety for pregnant patients and clinicians when discussing and prescribing analgesia in pregnancy, a topic explored in a ten-minute consultation guide in the BMJ [3]. To address this concern, a population-based cohort study published in JAMA Internal Medicine utilized a sibling-matched design to control for unmeasured familial factors [2]. Analyzing over one hundred and twenty-four thousand children for autism spectrum disorder and over ninety-seven thousand for attention-deficit/hyperactivity disorder in Hong Kong, the researchers compared siblings who had discordant prenatal exposure to acetaminophen. While conventional cohort analyses showed a positive association, the sibling-matched analyses revealed no association between prenatal acetaminophen exposure and the risk of autism spectrum disorder, with an adjusted hazard ratio of exactly one point zero zero, or attention-deficit/hyperactivity disorder, with an adjusted hazard ratio of one point zero one. Furthermore, negative control analyses of prepregnancy exposure also showed positive associations, confirming that the risks observed in older, conventional studies were likely due to residual familial confounding rather than the drug itself. This provides robust reassurance that indicated acetaminophen use during pregnancy does not increase these neurodevelopmental risks, offering invaluable clarity for clinicians during prenatal visits.

Moving from prenatal exposure to the management of neonatal health, another trial in JAMA examined the optimal treatment strategy for infants experiencing neonatal opioid withdrawal syndrome [9]. In a cluster, crossover randomized clinical trial across twenty-three United States hospitals, researchers compared a symptom-based dosing approach to a traditional scheduled opioid taper. Among infants managed with the Eat, Sleep, Console approach, symptom-based dosing significantly shortened the mean time from birth to medical readiness for discharge to nine point eighteen days, compared to eleven point sixty-one days in the scheduled taper group. This approach did not increase the risk of initiating pharmacologic treatment or the overall length of stay, and safety outcomes were comparable. However, thirty-five percent of infants in the symptom-based group eventually required scheduled opioid dosing due to withdrawal severity that was not controlled with intermittent dosing. Interestingly, in a secondary cohort managed with the Finnegan-based care scoring system, there were no significant differences in time to medical readiness for discharge or length of stay. This suggests that combining symptom-based dosing with the Eat, Sleep, Console framework can safely and effectively expedite discharge readiness for these vulnerable newborns, allowing family physicians to optimize neonatal care pathways.

Next, we turn to the management of older adults, starting with cardiovascular prevention in patients with type two diabetes. While statins are a cornerstone of primary prevention, clinical trials have historically underrepresented adults aged seventy-five and older. To fill this gap, a target trial emulation study published in PLoS Medicine evaluated the effectiveness and safety of statin therapy for primary prevention in older adults with type two diabetes in Hong Kong [6]. Among over thirty thousand matched person-trials aged seventy-five to eighty-four, statin initiation was associated with a thirty-one percent reduction in the incidence of cardiovascular disease and a thirty-five percent reduction in all-cause mortality. For nearly thirty-eight hundred matched person-trials aged eighty-five and older, the benefits were consistently observed, with a thirty-five percent reduction in cardiovascular events and a thirty-nine percent reduction in all-cause mortality. Crucially, the study found no substantially increased risk of muscle-related adverse events or liver dysfunction in either age group. These findings strongly support the initiation of statins for primary prevention in older diabetic patients, even those over eighty-five, provided the clinician engages in shared decision-making regarding individual goals of care.

In the realm of chronic infectious disease, a potential paradigm shift is on the horizon for the treatment of chronic hepatitis B virus infection. Currently, long-term nucleoside or nucleotide analogue therapy suppresses the virus but rarely cures it. Two replicate phase three randomized controlled trials published in The New England Journal of Medicine, B-Well One and B-Well Two, investigated the efficacy of bepirovirsen, an antisense oligonucleotide targeting viral transcripts [7]. Noncirrhotic adults receiving stable analogue therapy were randomized to receive weekly subcutaneous injections of three hundred milligrams of bepirovirsen or a placebo for twenty-four weeks, with analogue therapy discontinued at forty-eight weeks. At seventy-two weeks, a functional cure—defined as sustained hepatitis B surface antigen loss and viral DNA below the limit of quantification for at least twenty-four weeks—was achieved in twenty percent of patients in B-Well One and nineteen percent in B-Well Two, compared to zero percent in the placebo groups. While adverse events, particularly transient elevations in alanine aminotransferase, were more common in the active treatment group, this therapy represents a major step toward a finite, curative treatment for chronic hepatitis B.

Our final clinical theme explores how advanced therapeutics and artificial intelligence are reshaping oncology care. In the British Journal of General Practice, researchers demonstrated how large language models can improve the early detection of ovarian cancer by analyzing electronic health records [8]. Because early symptoms of ovarian cancer are often vague and documented only as free text in clinical notes rather than structured diagnostic codes, they are frequently missed in retrospective analyses and clinical decision alerts. Using a population-based case-control design in the United States, the study found that large language models successfully extracted seventeen prespecified symptoms from free-text notes far more frequently than structured codes alone. By combining coded and model-extracted data, fourteen clinical features were significantly associated with ovarian cancer, compared to only eight features when relying on codes alone. Thirteen features remained significantly associated when restricting the analysis to early-stage diagnoses. This demonstrates that natural language processing can capture an identifiable symptom signature for ovarian cancer—even in early-stage disease—and highlights the potential for artificial intelligence to support clinical decision tools in primary care.

Once cancer is diagnosed, novel therapies are showing remarkable long-term durability. The New England Journal of Medicine published ten-year follow-up data for patients with relapsed or refractory B-cell non-Hodgkin lymphomas treated with a single infusion of tisagenlecleucel, an anti-CD19 chimeric antigen receptor, or CAR, T-cell therapy [5]. At a median follow-up of over ten years, no relapses occurred beyond five point four years. The ten-year lymphoma-free survival was thirty-two percent for patients with large B-cell lymphoma and forty-seven percent for those with follicular lymphoma. While late adverse events included a cumulative second primary cancer incidence of twenty-one percent, these results confirm that CAR T-cell therapy offers long-term, potentially curative remissions for a substantial portion of patients with advanced hematologic malignancies.

Further refining our understanding of immunotherapy outcomes, a study in Nature Medicine used targeted metabolomics and metagenomics to profile over four thousand plasma samples from patients undergoing immune-checkpoint inhibitor therapy [4]. The researchers identified that plasma levels of the amino acid histidine were associated with favorable survival, whereas long-chain fatty acids and succinate were associated with poorer outcomes. In preclinical models and patient cohorts, histidine-rich diets improved progression-free survival, particularly in patients who lacked gut microbiota that catabolize histidine. This suggests that dietary modifications and microbiome profiling could eventually be used to personalize and enhance the efficacy of cancer immunotherapies.

As clinical evidence and therapeutic options expand, the guidelines we rely on must remain transparent and inclusive. To address this, the Annals of Internal Medicine published the RIGHT-MuSE checklist, a new international reporting standard designed to help guideline developers systematically document interest-holder engagement [1]. Consisting of eleven items, the checklist covers methods of engagement, representation of diverse groups, and the management of conflicts of interest. While not yet widely applied, this tool aims to improve the quality and trustworthiness of the clinical practice guidelines that shape our daily decisions.

If you only have time for one paper this week, make it the population-based sibling-matched cohort study on prenatal acetaminophen safety from JAMA Internal Medicine [2]. This study provides the high-quality, reassuring evidence we need to confidently counsel pregnant patients who require pain or fever management, effectively debunking the confounding-driven associations with autism and attention-deficit/hyperactivity disorder reported in previous observational literature.

Here are the key takeaways from this week in Family Medicine. First, prenatal acetaminophen use is not associated with an increased risk of autism spectrum disorder or attention-deficit/hyperactivity disorder in sibling-matched analyses, offering strong clinical reassurance for gestational prescribing. Second, in older adults with type two diabetes, starting a statin for primary prevention significantly reduces cardiovascular events and all-cause mortality, with clear benefits extending to those aged eighty-five and older. Third, for infants with neonatal opioid withdrawal syndrome managed with the Eat, Sleep, Console framework, symptom-based dosing safely reduces the time to discharge readiness by over two days compared to scheduled tapers. Finally, large language models can successfully extract undocumented, free-text clinical symptoms from electronic health records, paving the way for advanced clinical decision support tools to catch early signs of cancers like ovarian cancer.

That's your roundup for This Week in Family Medicine. The full transcript and references are available on the episode page in your AudioScholar library. This is an AI-curated summary — for clinical decisions, always consult primary sources and current guidelines. See you next week.

If this weekly briefing is useful, follow the show in your podcast app so new episodes arrive automatically. And for audio briefings on your own clinical questions and papers, visit audioscholar dot C C.

This is an automated summary generated by artificial intelligence, which can make mistakes. Always review the original source materials.