This week in Family Medicine

This AudioScholar briefing is generated by artificial intelligence for healthcare professionals and trainees. It is not medical advice.

Welcome to This Week in Family Medicine. This week we're covering 10 notable papers spanning cardiovascular prevention, infections and the painful shoulder, cancer diagnosis and surveillance, and how primary care itself is organised. Let's dive in.

We start with cardiovascular prevention, where two large studies test how far better delivery and lower targets can move outcomes. In JAMA, Lopez-Lopez and colleagues ran a pragmatic cluster randomised trial across forty primary care clinics in Colombia and Jamaica, enrolling about seventeen hundred adults with uncontrolled hypertension, almost all of whom were already on treatment. Intervention clinics combined locally purchased single-pill combinations, provided to patients, with a team-based package of physician and nurse training, health coaching, home blood pressure monitoring, and audit and feedback. Over eighteen months, systolic pressure fell by about 19 millimetres of mercury with the intervention versus about 9 with standard care, a between-group difference of roughly ten points. The share of patients reaching a target below 130 over 80 was more than double, at roughly seventy percent against under thirty percent, and serious adverse events were similar between groups. The trial was unblinded and the pills were supplied free, so the result speaks most directly to resource-constrained public systems, but it is a large randomised test of implementation rather than efficacy, which is exactly the evidence gap the authors set out to fill [1]. In PLOS Medicine, Blais and colleagues followed more than 185,000 people starting a statin for primary prevention in Hong Kong, New Zealand and the United Kingdom. Compared with an LDL cholesterol at or above 2.6 millimoles per litre two years after starting, reaching 1.8 to 2.5 was associated with about a fifth fewer major cardiovascular events, with a similar estimate between 1.4 and 1.7. Below 1.4 the estimate was imprecise and not statistically significant, and fewer than one in twenty patients got there. Reassuringly, very low LDL or non-HDL cholesterol was not associated with higher all-cause mortality. This is observational work with acknowledged confounding by indication, but the authors argue it supports aiming for a non-HDL cholesterol below 3.4 after statin initiation in primary prevention [2].

Next, infections, where one paper prepares clinicians for a returning disease and another questions a new test. In the Annals of Internal Medicine, Lo offers a clinical review of measles, noting that cases and outbreaks are rising in the United States and other countries where it had been eliminated, that most physicians there have never treated it, and that adults as well as children are now presenting. The review emphasises prompt clinical suspicion, infection control and early notification of public health departments as the levers that support timely diagnosis and outbreak control [3]. In the British Journal of General Practice, Knüsli and colleagues took nasopharyngeal swabs from a Swiss primary care trial of lower respiratory tract infection and retrospectively ran molecular panels. Bacteria were detected in about two in five of the 345 patients analysed, yet median time to recovery was seven days whether or not bacteria were found, and this did not depend on whether antibiotics were prescribed. Cough lasted a median of about sixteen days in this population, a useful figure for counselling. As a secondary analysis it cannot settle the question, but the authors conclude it casts doubt on syndromic molecular testing for guiding these infections in general practice [4].

Two papers this week concern the shoulder, one about an injury we cause and one about a condition we treat. In JAMA Internal Medicine, Zheng and colleagues used natural language processing across nearly two million deltoid vaccinations at Kaiser Permanente Southern California to measure shoulder injury related to vaccine administration. The injury was rare after both vaccines, but more common after COVID-19 vaccination, at about 1.8 per ten thousand versus about 1.2 for influenza, roughly seven extra cases per hundred thousand vaccinations. In adults aged 65 or older, the risk did not differ significantly from high-dose influenza vaccine. The authors frame this small absolute risk as something to weigh alongside the established benefits of both vaccines [5]. In the BMJ, Meng and colleagues published a network meta-analysis of 189 randomised trials in adhesive capsulitis, covering almost thirteen thousand participants and 59 interventions. Most of the contributing trials were at high risk of bias. Moderate certainty evidence suggested extracorporeal shockwave therapy probably reduces short-term pain by about two points on a ten-point scale versus usual care, and that manipulation under anaesthesia, with or without intra-articular steroid, probably improves short-term flexion by around thirty degrees. Almost everything else rests on low or very low certainty evidence, mid-term effects are poorly established, and harms were reported too inconsistently to compare, so this review maps the uncertainty more than it resolves it [6].

Cancer diagnosis and follow-up form our next theme, with one paper questioning routine surveillance and another highlighting where diagnoses slip. In JAMA Internal Medicine, Richman and colleagues used SEER-Medicare data on almost ten thousand women diagnosed with breast cancer at 67 or older who were still being screened about five years later. Advanced second primary breast cancer was rare, at roughly two percent, and its incidence did not differ significantly between women who continued surveillance mammography and those who stopped, in both hormone receptor positive and negative groups. Retrospective data cannot exclude residual confounding, but the authors conclude the low absolute risk supports individualised rather than default surveillance in older survivors [7]. In the British Journal of General Practice, Price and colleagues analysed linked English primary care and registry data on more than 220,000 patients to find what they call interim diagnoses, non-cancer conditions recorded in the year before a cancer diagnosis. They identified 37 such diagnoses across seven cancer sites. Reflux, gastritis, oesophagitis and peptic ulcer, for example, were linked to worse outcomes in pancreatic, colorectal and ovarian cancer, though better outcomes in stomach cancer. These are observational associations, but the authors argue they mark points where deliberate safety-netting and follow-up may matter most [8].

Our final theme looks at the people and relationships behind primary care. In Family Practice, Young and colleagues describe an intensive primary care program in Tarrant County, Texas, where a family physician and a community health worker offered round-the-clock personal care to patients with frequent emergency or inpatient use. Among 128 patients with repeated contact, emergency visits fell about twenty percent over a year, while the small comparison group seen only once saw visits rise by a similar margin. Interviews pointed to addressing social needs, rebuilding trust and flexible medication access. The comparison is non-randomised and regression to the mean is a real concern, so this is promising rather than proven [9]. And in the Scandinavian Journal of Primary Health Care, Wendel and colleagues surveyed 191 primary care physicians about four eating disorders. Attitudes were similar across diagnoses, but only about one in six to one in five physicians reported enjoying this work, and familiarity with avoidant restrictive food intake disorder was low. Education was consistently associated with greater comfort and more positive attitudes, though a cross-sectional survey cannot show that training causes the difference [10].

If you only have time for one paper this week, make it the JAMA hypertension trial from Lopez-Lopez and colleagues [1]. It moves the single-pill combination question from whether these pills work to whether a whole primary care system can deliver them at scale, and it answers with a large randomised result.

That's your roundup for This Week in Family Medicine. The full transcript and references are available on the episode page. 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.