This Week in Family Medicine — Aug 17, 2026

Welcome to This Week in Family Medicine. This week we're covering nine notable papers spanning diagnostic restraint and overtreatment in primary care, prevention gaps in the populations we serve least well, and new evidence on integrating and scaling care for chronic disease and mental health. Let's dive in.

We'll start with the theme of doing less, and doing it deliberately. JAMA Internal Medicine published a clinical review of shoulder pain in primary care by Haas and colleagues, and its central message is one of restraint [1]. Shoulder pain is the third most common musculoskeletal presentation we see, and the review argues for abandoning the tangle of overlapping labels — impingement syndrome, rotator cuff tendinopathy, subacromial bursitis — in favour of the single term subacromial pain, because those diagnostic labels are inconsistently defined and rarely change what we do. Once you have excluded the rare serious causes through history and examination — infection, malignancy, fracture, dislocation, referred pain, systemic inflammatory disease — first-line management is essentially the same regardless of the label: explain the favourable natural history, offer symptom relief and activity modification, and watch and wait. Early imaging is explicitly not indicated without significant trauma or red flags, because structural abnormalities on imaging often do not correlate with symptoms and set patients on a path to overdiagnosis and overtreatment. And the authors describe the evidence that subacromial pain does not benefit from surgery as high certainty. That is a clear licence to reassure rather than refer. A very different study makes a related point about testing. In the Scandinavian Journal of Primary Health Care, Lykkegaard and colleagues used Danish registers covering more than 860,000 out-of-hours consultations to ask whether being seen by a general practitioner who reaches often for C-reactive protein point-of-care testing changes a patient's later health-seeking behaviour [3]. Testing rose from twelve percent of consultations in 2014 to nineteen percent in 2017, and patients seen by a doctor with at least double the average testing propensity were slightly but significantly more likely to come back to out-of-hours care over the following years. The effect size is small — a hazard ratio just under one point zero four — and the authors are careful to say this could reflect the test itself, correlated doctor characteristics, or both. But it is a signal worth holding onto: a test ordered for reassurance may quietly teach patients that reassurance requires a test.

Our second theme is prevention gaps, and where they concentrate. Also in JAMA Internal Medicine, Roldan and colleagues analysed National Health and Nutrition Examination Survey data from 2017 to 2023 to estimate serologic immunity to hepatitis A and B across United States adults [2]. Just under forty percent of adults had evidence of hepatitis A immunity, and only about twenty-seven percent had hepatitis B immunity, with roughly a quarter showing vaccine-derived protection. The gaps were worst precisely where the stakes are highest: among people with metabolic dysfunction and alcohol-related liver disease, only about a quarter of patients were immune to hepatitis A, and among people with chronic kidney disease, only fourteen percent had vaccine-derived hepatitis B immunity. Older adults were far less likely to be protected than younger ones. The practical implication is to make hepatitis A and B serology and catch-up vaccination a routine part of chronic liver disease, chronic kidney disease, diabetes, and immunosuppression care rather than an afterthought. PLOS Medicine extends the equity theme with a population-based analysis by Yu and colleagues of mortality among adults with intellectual disability in England [9]. Among more than thirteen hundred adults with severe or profound intellectual disability, the median age at death was fifty-seven point nine years, compared with sixty-five for those with mild or moderate intellectual disability and eighty-two in the general population — an adjusted difference of about seven years even after accounting for other factors. Two fifths of those deaths, thirty-nine point five percent, met the definition of avoidable, meaning death before seventy-five from a preventable or treatable cause, amounting to more than fifteen thousand years of life lost. Dysphagia was present in sixty-three percent and epilepsy in thirty-six percent, which points to concrete targets: aspiration risk, seizure management, and vision. Alongside these, the New England Journal of Medicine published a review by Shelley and colleagues on tobacco cessation in low- and middle-income countries, where more than eighty percent of the world's 1.3 billion tobacco users live and where only thirty-one countries meet the World Health Organization best-practice standard of funded behavioural support plus pharmacotherapy [4]. Using case studies from India and Vietnam, the authors make the case for systematic screening paired with brief clinician advice, quitlines and digital supports, and the Ask-Advise-Connect model, combined with essential-medicines pharmacotherapy such as cytisine and nicotine replacement. The Ask-Advise-Connect framework is a reminder that our job is not to deliver the whole intervention but to reliably make the connection.

The third theme is integration — bolting care for one condition onto an existing service that patients already attend. The Lancet Global Health reports the InterCARE trial by Mosepele and colleagues, a pair-matched cluster-randomised trial across fourteen HIV clinics in Botswana enrolling more than 4,600 adults with both HIV and hypertension [8]. Clinics randomised to integrated care received provider training, coaching, community treatment partners, and use of the national electronic health record to support diagnosis and electronic prescribing during routine HIV visits. At twelve months, blood pressure control among patients on antihypertensives was sixty-seven percent in the intervention clinics versus fifty-two percent under standard care — about a thirty percent relative improvement — and electronic antihypertensive prescribing rose from eleven percent of encounters to forty percent. Critically, viral suppression stayed above ninety-eight percent in both arms and serious adverse events were rare and no different, so adding hypertension care did not dilute HIV outcomes. In the same spirit of scalability, PLOS Medicine published the RESPOND randomised trial by Roos and colleagues, testing a remotely delivered stepped-care programme for Polish migrant workers in the Netherlands with elevated psychological distress [5]. Two hundred and eighteen participants were randomised to care as usual or to guided online self-help followed, if distress persisted, by problem-solving therapy over videoconference — both delivered by non-professional helpers. At the two-month follow-up, combined anxiety and depression scores were about six and a half points lower in the intervention group, a moderate effect, with benefit already visible after the low-intensity first step. Notably, close to ninety percent of participants who completed the first step still met criteria to progress to the second, so the low-intensity step alone was rarely sufficient. Generalisability is limited: most participants were women who had lived in the Netherlands for years.

Finally, two infection papers worth knowing about. Science Translational Medicine reports a randomised placebo-controlled trial by Shankar-Hari and colleagues giving a single dose of thirteen-valent pneumococcal conjugate vaccine to 214 sepsis survivors at intensive care discharge, on the hypothesis that vaccination might accelerate immune recovery [7]. It did not. Over a year of follow-up there were more infection-related rehospitalisations or deaths in the vaccine arm than in the placebo arm, with a hazard ratio of one point two three and a confidence interval crossing one — so no benefit, and if anything a signal in the wrong direction, with higher reinfection rates and earlier antibiotic prescriptions in primary care. Antibody responses were highly variable between serotypes and individuals. There were no vaccine-related serious adverse events, but this is a negative trial: timing vaccination to ICU discharge in sepsis survivors is not supported. And Nature Reviews Disease Primers published a comprehensive primer on methicillin-resistant Staphylococcus aureus, a useful refresher on the spectrum from uncomplicated skin and soft tissue infection to bacteraemia and osteomyelitis, and on the fact that decolonisation and screening strategies vary in effectiveness depending on local prevalence [6].

If you only have time for one paper this week, make it the JAMA Internal Medicine review on shoulder pain [1]. It addresses one of the most common presentations in general practice and gives you explicit, high-certainty permission to skip the imaging and the surgical referral in the great majority of patients.

Here are the key takeaways from this week in Family Medicine. First, for non-traumatic shoulder pain without red flags, education, symptom relief, and watchful waiting are the treatment, and imaging is more likely to harm than help. Second, be aware that a habit of point-of-care testing may correlate with patients returning more often — a small effect, but a reason to test with a question in mind. Third, check and complete hepatitis A and B vaccination in patients with liver disease, kidney disease, diabetes, and immunosuppression, where susceptibility remains very high. Fourth, patients with severe or profound intellectual disability are dying more than twenty years early, with two fifths of deaths avoidable — screen actively for dysphagia, epilepsy, and vision problems. Fifth, integrating chronic disease care into an existing visit works, as blood pressure control improved substantially in Botswana's HIV clinics without harming HIV outcomes; and stepped mental health care delivered remotely by non-professionals can reach populations that avoid formal services. And finally, pneumococcal vaccination at intensive care discharge did not reduce reinfection or death in sepsis survivors.

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 a quick rating on Apple Podcasts or Spotify helps other physicians discover the show.

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