The top ten AFP Community Blog posts of 2020

 - Kenny Lin, MD, MPH

Unsurprisingly, 7 of our 10 most-viewed blog posts from this pandemic year were related to COVID-19. Beginning in February, when I wrote with concern about initial reports of a novel coronavirus that was rapidly spreading from its original epicenter in Wuhan, China, the pandemic accelerated, transmitting exponentially across the United States by mid-March and causing state governments to close schools and non-essential businesses. In addition to the worldwide death toll (which, per the Johns Hopkins University Coronavirus Resource Center, stood at nearly 1.8 million as of today), we forecasted the staggering mental health impacts of the virus and prolonged economic disruption and social isolation.

Contributing editor Allen Shaughnessy, PharmD and his colleague Dr. Andrea Gordon provided ten points to consider when applying evidence-based medicine in a pandemic. Dr. Middleton described one of the more unusual manifestations of COVID-19 infection, "COVID toes," and our 2019-20 Resident Representative, Dr. Enkhee Tuvshintogs, discussed how medical students were adapting to disruptions in their educational programs and personal lives. Finally, we reviewed fraudulent COVID-19 research publications and the summer of #BlackLivesMatter protests against police violence and the persistence of structural racism that produced, among other things, vast disparities in coronavirus infections and mortality among African Americans.

Of course, 2020 included other important medical stories. Personally, we mourned the passing of AFP executive editor Joyce Merriman, who capably led our professional editor team for 15 years. Two past AFP Podcast co-hosts reviewed 9 other highly ranked podcasts for family physicians for medical education or moving health narratives. And Dr. Middleton argued that the results of a recent randomized controlled trial could mean that intermittent fasting's time in the popular diet limelight may be up.

All of us at American Family Physician hope that you and your loved ones are having an enjoyable holiday season, and we look forward to 2021 being a less unprecedented and more optimistic year.

Here come the COVID vaccines

 - Jennifer Middleton, MD, MPH

Last month, I wrote about the process for vaccine approval by the United States (US) Food and Drug Administration (FDA). Since then, the Pfizer/BioNTech SARS-CoV-2 vaccine has received Emergency Use Authorization (EUA), joined by the National Institutes of Health (NIH)/Moderna vaccine at the end of last week. These two vaccines use a novel mRNA process to induce immunity; a third vaccine by Oxford/AstraZeneca, which instead uses a chimpanzee adenovirus vector ("viral vector"), has just published its interim analysis data.

The Oxford/AstraZeneca ChAdOx1 vaccine was tested in nearly 35,000 individuals across Brazil, the United Kingdom (UK), and South Africa in phase 2/3 trials. Importantly, only adults aged 18-55 were enrolled. When researchers received the manufactured vaccine in the UK, though, their measurement of the viral particle concentration was higher than expected. To maximize safety while their teams resolved this discrepancy, they chose to inoculate the first group of participants with a lower dose for their first injection. Once the researchers settled the issue with their quantification methods, all participants then received the originally planned full dose of the vaccine for their second injection. Later UK enrollees and participants at the other trial sites received the full dose for both injections. 

Interestingly, the low dose cohort had better protection against COVID-19 than the participants who received two full dose injections:

In participants who received two standard doses, vaccine efficacy was 62·1% (95% CI 41·0–75·7; 27 [0·6%] of 4440 in the ChAdOx1 nCoV-19 group vs71 [1·6%] of 4455 in the control group) and in participants who received a low dose followed by a standard dose, efficacy was 90·0% (67·4–97·0; three [0·2%] of 1367 vs 30 [2·2%] of 1374; pinteraction=0·010).

The researchers did not have an explanation for this surprising finding, though they rightly point out that the confidence intervals are quite wide for both values.

This vaccine is unlikely to achieve EUA status in the US soon. Oxford/AstraZeneca failed to promptly notify the FDA of the dosing snafu as well as the decision to pause trials in September after a participant was diagnosed with transverse myelitis. The FDA reportedly learned about these issues via the lay press and not the study investigators, which reportedly did not sit well with FDA leadership. The FDA didn't clear Oxford/AstraZeneca to resume vaccine trials in US until October, which put them behind Pfizer/BioNTech and NIH/Moderna in the race to generate phase 3 trial data. And, they have yet to produce any data regarding their vaccine's efficacy in adults over the age of 55, the population arguably at greatest risk of COVID-19 mortality.

Assuming it eventually catches up, the Oxford/AstraZeneca vaccine has several appealing properties. Unlike the two mRNA vaccines by Pfizer/BioNTech and NIH/Moderna, the Oxford vaccine does not require ultra-cold storage and is much less expensive to produce. For parts of the world where storing and/or procuring an mRNA vaccine is not affordable or practical, the Oxford/AstraZeneca vaccine offers an alternative. 

The AAFP held a virtual town hall last week about the COVID-19 vaccines, and the Centers for Disease Control and Prevention (CDC) has also outlined its allocation recommendations for the initially limited supply of vaccine. We'll also continue to update the AFP By Topic on Coronavirus Disease 2019 (COVID-19). 

New guideline for managing acute pain from musculoskeletal injuries

 - Kenny Lin, MD, MPH

In a 2017 practice guideline based on a systematic review of noninvasive treatments, the American College of Physicians (ACP) recommended superficial heat, massage, acupuncture, and spinal manipulation as initial treatment options for patients with acute low back pain, in addition to a nonsteroidal anti-inflammatory drug (NSAID) or skeletal muscle relaxant if desired. But is a similar approach effective for treating pain resulting from acute musculoskeletal injuries not involving the lower back? To answer this question, the American Academy of Family Physicians (AAFP) joined the ACP in developing another practice guideline on management of acute pain from non-low back, musculoskeletal injuries in adults, a synopsis of which appeared in Practice Guidelines in the December 1 issue of American Family Physician. These are some key practice points from the guideline:

• Topical NSAIDs are the most effective intervention for acute musculoskeletal pain other than low back pain.

• Although oral NSAIDs and acetaminophen are effective for acute pain relief, combining them does not improve effectiveness.

• Although moderately effective for pain relief, opioids increase gastrointestinal and neurologic adverse effects and lead to long-term use in 6% of people treated.

• Acupressure and transcutaneous electrical nerve stimulation techniques are effective nonpharmacologic options for acute pain.

In an accompanying editorial, Dr. David O'Gurek and I, who represented the AAFP on the guideline committee, and Dr. Melanie Bird, AAFP Clinical and Health Policies Manager, discussed some of the guideline's highlights and limitations. A systematic review and network meta-analysis of randomized, controlled trials provided direct and indirect comparisons of various treatment options on outcomes that included pain relief and physical functioning, symptom relief, treatment satisfaction, and adverse events.

Topical NSAIDs improved all efficacy outcomes with minimal adverse effects, while oral NSAIDs and acetaminophen improved fewer outcomes and were more likely to cause adverse events. We suggested against using opioids, including tramadol, for acute musculoskeletal injury pain due to their poor adverse effect profile and the risk of prolonged use, ranging from 6% in low-risk to 27% in high-risk populations. We also noted that "equitable coverage and affordability of first-line treatments" are essential to reduce well-known disparities in pain management; for example, though a topical NSAID is now available over-the-counter, it costs significantly more than oral NSAIDs and acetaminophen and may not be covered by health insurance plans.

The polypill: promise & perils

 - Jennifer Middleton, MD

The promise of a simple "polypill" to reduce cardiovascular disease (CVD) morbidity and mortality has been tantalizing researchers for decades. A 2012 analysis suggested that "[p]olypill use by US adults age ≥ 55 years is projected to potentially prevent 3.2 million CHD events and 1.7 million strokes over 10 years." The third iteration of an international research effort examining a combination of 40 mg of simvastatin, 100 mg of atenolol, 25 mg of hydrochlorothiazide, and 10 mg of ramipril known as TIPS-3 (The International Polycap Study 3) demonstrates promise, but a significant minority of enrollees struggled to tolerate it.

TIPS-3 enrolled over 7500 adults identified at intermediate to high risk of CVD using a 2x2x2 factorial design. Most of the participants were from the Philippines or India. Enrolled men were at least 50 years old, and enrolled women were at least 55 years old. Participants were predicted to be at intermediate or high risk of CVD events by their INTERHEART risk score. They completed a 3 to 4 week run-in period, during which time researchers assessed their adherence to lower doses of the full regimens. Notably, 9.5% of initially enrolled participants did not complete the run-in due to complaints of side effects, and another 7.4% could not achieve the minimum 80% adherence required to be randomized into the trial. 

The primary outcome differed by group: for the polypill and polypill + ASA groups, it was a composite of CVD death, myocardial infarction (MI), CVA, cardiac arrest, heart failure, and/or cardiac revascularization; for the ASA group, the primary outcome was only a composite of CVD death, MI, and/or CVA. Participants in the polypill vs placebo group had a hazard ratio (HR) of its primary outcome of 0.79 (95% confidence interval [CI] 0.63, 1.00). The ASA vs placebo group had an HR of its primary outcome of 0.86 (95% CI 0.67, 1.10), and the polypill + ASA group had a HR for its primary outcome of 0.69 (95% CI 0.50, 0.97). The researchers had intended to continue their enrollment and data analysis through 2020, but the COVID-19 pandemic forced a halt to their plans.

The TIPS-3 results are promising, but future research should focus on overcoming this study's shortcomings. Participants who did not experience disabling side effects and were able to adhere to their regimen lowered their risk of CVD events, but it's concerning that more than 1 in 6 enrolled participants could not get past the run-in period to enter the full trial. Physicians should exercise caution extrapolating results from this study to patients deemed at intermediate to high risk of CVD events using a different tool than INTERHEART (for example, the ASCVD calculator is used far more often in the United States). Lastly, a longer study period and larger number of participants may result in a better-powered study with more convincing confidence intervals; the HR of both polypill groups had wide confidence intervals perilously close to non-significance. With its lower HR, the polypill + ASA group appears to have been more effective than the polypill alone, but the confidence intervals of both groups overlap; this finding also conflicts with recent robust evidence showing that ASA is ineffective for primary CVD prevention among both average and higher risk persons. 

You can read more at the AFP By Topics on Coronary Artery Disease/Coronary Heart Disease and Stroke and TIA along with this review of an earlier polypill benefit analysis by Dr. Kenny Lin.

Newborn screening for metabolic conditions has long-term benefits

 - Kenny Lin, MD, MPH

State-mandated screening at birth for rare, serious medical conditions occurs in 4 to 5 million newborns and detects 5,000 to 6,000 affected infants each year. A 2017 American Family Physician article reviewed various conditions that are targeted by newborn screening: amino acid disorders, fatty acid oxidation disorders, organic acid disorders, hemoglobinopathies, endocrine disorders, and miscellaneous diseases (including congenital hearing loss and critical congenital heart defects). With a combined incidence of 1 out of every 1,500 births, inborn errors of metabolism are the most common conditions detected by newborn screening.

After tandem mass spectrometry made it possible to test for many conditions using a single blood sample, the federal Health Resources and Services Administration's Maternal and Child Health Bureau commissioned the American College of Medical Genetics (ACMG) to create a uniform list of conditions for newborn screening panels in 2005. However, the ACMG's recommended core panel of 29 conditions was criticized by the U.S. Preventive Services Task Force (USPSTF) for not taking an evidence-based approach. In a position paper, the USPSTF noted that the ability to detect a condition with high diagnostic accuracy was insufficient to include it in the panel:

A newborn screening program is not just a panel of screening tests. ... It is also parental education, follow-up, diagnosis, treatment and management, and program evaluation, and all of the various parts of the system must be in place and working well to realize the benefits of screening. ... Moreover, a newborn screening panel should be expanded only if the newborn screening program is fully prepared to make all the components of the complex system available for the new disorders. Expansion would be costly and might not be the best use of scarce health care resources, given the many other unmet child health needs.

Reinforcing the USPSTF's concerns, an analysis by the Centers for Disease Control and Prevention projected that if all 50 states expanded their newborn screening panels to align fully with the ACMG recommendations, "although such an expansion would have increased the number of children identified by 32% (from 4,370 to 6,439), these children would have had many rare disorders that require local or regional capacity to deliver expertise in screening, diagnosis, and management." A cross-sectional survey of Ontario primary care clinicians found that family physicians had limited knowledge of conditions identified by newborn screening tests, and many were not comfortable leading detailed discussions of abnormal results with parents or guardians.

The U.S. Secretary of Health and Human Services' Advisory Committee on Heritable Disorders in Newborns and Children (SACHDNC) subsequently developed a more rigorous framework to evaluate conditions nominated as additions to the uniform screening panel, requiring an independent systematic evidence review of key questions based on an analytic framework similar to those used for USPSTF reviews. In a separate document, the SACHDNC outlined questions for newborn screening long-term follow-up data systems to answer to make sure that programs achieve their goals of improved outcomes for children and families.

An observational study published last month in Pediatrics reported the clinical outcomes of 306 individuals with inherited metabolic diseases identified by a university hospital laboratory performing Germany's newborn screening panel from 1999 to 2016. The German national panel is less extensive than the ACMG's, consisting of 2 endocrine and 12 inherited metabolic diseases, and the nearly 2 million newborns screened during the study period represented 15 percent of Germany's live births. 28 individuals presented with metabolic symptoms prior to newborn screening results being available; the rest were successfully enrolled in specialized metabolic/nutritional therapy while still asymptomatic. Although nearly 1 in 4 individuals eventually developed irreversible disease-specific clinical signs, 88% had normal cognitive outcomes, and more than 95% showed normal development and attended regular kindergarten and primary schools.