This Week in Family Medicine — Sep 6, 2026

Welcome to This Week in Family Medicine. This week we're covering 10 notable papers spanning respiratory and herpesvirus vaccination, cancer care in general practice, and a set of papers about the workforce, the evidence we read, and the medicines we prescribe. Let's dive in.

We start with the annual vaccine evidence updates, both published in JAMA, which together give you the talking points for autumn clinics. Lipson and colleagues screened nearly twelve thousand references and included 155 publications on COVID-19 vaccines licensed in the United States [1]. Updated vaccines were associated with roughly a halving of hospitalisation risk in adults aged sixty-five and over, and a similar effect in adults aged eighteen to sixty-four, though that younger estimate was far less precise. Two findings deserve particular emphasis in family medicine. Maternal vaccination cut emergency department and urgent care encounters in pregnant patients by more than half, and infants born to vaccinated mothers had about half the risk of COVID-related hospital contact in the first two months of life, regardless of which trimester the vaccine was given. In young children aged nine months to four years, vaccination was associated with roughly a three-quarters reduction in COVID-related emergency visits. Across adverse events of special interest, including myocarditis, stroke, and thrombosis, no new safety signals emerged. The companion influenza review from Senerth and colleagues drew on 69 studies [2]. Effectiveness against hospitalisation in older adults ranged from modest to moderate across seasons, with the low end of that range essentially indistinguishable from no benefit, so be honest with patients that flu vaccine performance varies year to year. But the paediatric signal is striking: across eight seasons of national surveillance, vaccine effectiveness against laboratory-confirmed influenza-associated death in children and adolescents was about eighty percent. High-dose vaccine outperformed standard dose against hospitalisation in older adults by a relative forty-four percent. And a self-controlled case series in nearly ten million people found no increased risk of serious adverse events, including Guillain-Barré syndrome, whether influenza vaccine was given alone or co-administered with respiratory syncytial virus or COVID-19 vaccines. That last point is practical: co-administration is defensible, and it saves a visit.

Staying with vaccine-preventable illness, Nature Reviews Disease Primers published a comprehensive review of varicella zoster virus by Bubak and colleagues [7]. More than ninety percent of the global population is infected, and after chickenpox the virus stays latent in ganglionic neurons for life. The clinically useful reminder here is that reactivation is not always a dermatomal rash. Vasculopathy, cranial neuropathies, myelopathy, and cardiovascular or gastrointestinal complications can occur without rash, or separated in time from it — which means zoster belongs on your differential for unexplained stroke in an older patient or an isolated cranial nerve palsy. Valacyclovir is the oral drug of choice, with intravenous acyclovir reserved for severe or disseminated disease, and prevention rests on the live attenuated varicella vaccine and the recombinant zoster vaccine.

The second theme is cancer, and the recurring message is that general practice is where the diagnostic and continuity work happens but is not where the system currently invests. BJGP Open published the protocol for the phase three ThinkCancer! trial from Walker and colleagues [5]. Nearly sixty percent of United Kingdom cancer diagnoses came through primary care referral in 2020, and late-stage diagnosis is the main driver of comparatively poor survival. This is a practice-level randomised trial of a team-based educational and behavioural intervention, with the primary care interval as the primary outcome and results expected at the end of 2027 — so nothing to change today, but worth knowing the trial exists. Complementing that from the other end of the pathway, BMC Primary Care carried a scoping review by Werner and colleagues on general practitioners in German cancer care [6]. Across 24 included studies, general practitioners were involved at every stage — diagnosis, active treatment, follow-up, and palliative care — acting as the patient's person of trust and the main source of counselling and emotional support, yet they were rarely included in specific treatment decisions, and communication and role clarification were repeatedly deficient. Evidence on survivorship care and comorbidity management in cancer patients was notably thin. The authors argue future guidelines should explicitly name the general practitioner's role in oncology pathways rather than leaving it to informal goodwill.

Our third theme is the day-to-day machinery of primary care: quality metrics, common infections, evidence summaries, and the workforce. The British Journal of General Practice published a retrospective cohort from Gao and colleagues covering more than 468,000 patients with multimorbidity in England [3]. The headline is counterintuitive. Higher attainment on quality indicators was generally associated with slightly more secondary care use — more outpatient visits, more emergency admissions, more emergency department attendances — which likely reflects case-finding and appropriate onward referral rather than failure. However, higher attainment specifically on multimorbidity-tailored quality indicators was associated with lower odds of elective admission over the following twelve months, and in patients with three or more conditions it was associated with fewer elective admissions and fewer outpatient visits. The signal, then, is that generic disease-by-disease metrics and multimorbidity-specific metrics behave differently, and it is the latter that appears to reduce downstream hospital use.

On recurrent urinary tract infection, BJGP Open published a systematic review from Miller and colleagues covering ten studies and just under three thousand participants [4]. Confidence in the evidence was low to very low. The only behaviour with reasonable support was increased fluid intake, backed by a single randomised trial of 140 women who had low baseline intake. Everything else — wiping technique, post-coital voiding, cranberry-adjacent advice — sat inside multi-component interventions and was never evaluated on its own. So the honest conversation is: push fluids if intake is low, offer the rest of the advice as reasonable and low-harm rather than proven, and acknowledge that practical and social barriers often matter more than knowledge. Then, from the Journal of the American Board of Family Medicine, Guthmann and colleagues compared large language model summaries against 24 matched Patient-Oriented Evidence that Matters summaries from American Family Physician [8]. Using a thirteen-item scoring tool and two blinded raters, the model summaries scored about one and a half points higher, at a similar length, and contained errors in two of twenty-four summaries versus nine of twenty-four for the expert-written ones. Reviewers generally preferred the machine output. The expert summaries most often fell short on contextual background and limitations; both approaches often missed clinical applicability. This is 24 summaries from one enterprise model with a carefully crafted prompt, so it is not a licence to outsource your reading — but it does suggest prompted, supervised machine summarisation is now a credible part of keeping up.

Finally, two papers about the people and the pressures. The Journal of the American Board of Family Medicine published a qualitative study from Byrd and colleagues interviewing 25 racially and geographically diverse early-career women family physicians drawn from the American Board of Family Medicine National Graduate Survey [9]. Three themes emerged: biased communication patterns within clinical teams, patients invalidating physicians' authority based on appearance, and the emotional labour of navigating both. Gender-based discrimination is reported by around three quarters of early-career women physicians, falling with seniority but never disappearing. The authors frame this as a workforce sustainability issue, not just an equity one, and put the responsibility on leadership to recognise and interrupt microaggressions in real time. And in Science, Hagenaars and Schmidt offer a commentary on the World Health Organization's first global guidelines recommending GLP-1 receptor agonists for obesity [10]. Their concern is the commercial environment into which those recommendations land — pharmaceutical competition, food companies chasing the GLP-1 market, and wellness firms selling supplements and telemedicine as a "GLP-1 lifestyle" — all of which may undermine the stated goal of fair access for the one billion people affected worldwide. The guidelines also call for population-level policy on food environments, a point easily lost when the conversation is only about prescribing.

If you only have time for one paper this week, make it the COVID-19 vaccine effectiveness and safety review in JAMA [1]. The maternal and infant data give you a concrete, quantified answer to the question you will be asked most often in the coming months, and it comes with a clean safety record.

Here are the key takeaways from this week in Family Medicine. First, updated COVID-19 vaccination roughly halves hospitalisation risk in adults and protects infants through maternal vaccination in any trimester, with no new safety signals. Second, influenza vaccine effectiveness against hospitalisation in older adults is variable across seasons, but paediatric protection against influenza death is substantial, high-dose vaccine beats standard dose in older adults, and co-administration with COVID-19 or respiratory syncytial virus vaccines appears safe. Third, keep zoster on the differential for stroke, cranial neuropathy, and myelopathy even when there is no rash. Fourth, for recurrent urinary tract infection, fluid intake is the only behaviour with trial support — offer other advice with appropriate humility. Fifth, multimorbidity-specific quality indicators, not generic ones, are the ones associated with fewer elective admissions in complex patients. And finally, well-prompted large language model summaries of research matched or beat expert-written ones in a small head-to-head comparison, with fewer errors.

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 if you have a paper you have been meaning to read, upload the PDF, or paste any link, at audioscholar dot C C. We will turn it into audio like this one, in any of thirty-one languages.

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