Benefits of COVID-19 vaccination in persons with past infection

Since the first COVID-19 vaccine received authorization for emergency use in December 2020, physicians and the public have vigorously debated whether infection-acquired ("natural") or vaccine-mediated immunity provides better protection against future infection and severe illness. The answer may never be known for certain, as it's hard to imagine a research ethics committee approving a trial that intentionally exposes immunologically naive people to a potentially lethal infection. A more important clinical question is: does getting vaccinated after recovery from COVID-19 provide additional benefits? Currently, the Centers for Disease Control and Prevention recommends routine vaccination in all persons aged 5 years or older, regardless of their history of past infection.

Two large cohort studies published earlier this month provided the strongest evidence to date that the answer is yes. The first study used electronic medical records from a health care organization covering more than half of the population of Israel to identify 149,000 patients age 16 years or older who had recovered from documented SARS-CoV-2 infection at least 100 days earlier and had not yet received COVID-19 vaccination as of March 1, 2021. 56% of these persons received at least one dose of BNT162b2 (Pfizer-BioNTech) vaccine by November 26, 2021. 2,168 of those who remained unvaccinated (3.3%) were reinfected during the study, compared to 354 of the vaccinated patients (0.4%). After adjustment for sociodemographic factors and coexisting illnesses, the estimated vaccine effectiveness was 82% for patients aged 64 years or younger and 60% for patients aged 65 years or older. A secondary analysis showed no difference in protection between one or two vaccine doses.

A second study in a highly vaccinated cohort of 35,768 health care workers in the United Kingdom tracked primary infections and reinfections between December 7, 2020 and September 21, 2021. Most participants received two doses of BNT162b2 (Pfizer-BioNTech) vaccine; 8% received the single-dose ChAdOx1 nCoV-19 vaccine (AstraZeneca). In previously uninfected participants who received the second dose of BNT162b2 six weeks or more after the first dose, adjusted vaccine effectiveness was 85% up to 73 days after the second dose but declined to 51% after 200 days. In comparison, adjusted effectiveness of the ChAdOx1 nCoV-19 vaccine was only 58% up to 73 days. In 6,169 participants who had COVID-19 prior to the study, long-term (>1 year) protection against re-infection was 69% in unvaccinated persons but remained high at 94% in persons who received one or two doses of BNT162b2.

Acknowledging some differences between the populations and the predominant variants circulating during the respective study periods, the results support the following conclusions. First, re-infection in unvaccinated persons is relatively uncommon during the first 9 months after a primary infection (1 in 30 in the Israeli study) but becomes more likely after 1 year (per the U.K. study). Similarly, the effectiveness of the initial two doses of BNT162b2 vaccine in preventing COVID-19 declines after 6-7 months, supporting booster doses. However, patients with past infections who subsequently receive one or two doses of BNT162b2 have sustained high levels of protection ("hybrid immunity") against re-infection for at least one year. In an editorial in The Guardian, Dr. Eric Topol recently argued that these and other data support re-defining "fully vaccinated" to include recovery from past infection plus a single dose of an mRNA vaccine.

**

This post first appeared on the AFP Community Blog.

Screening for atrial fibrillation, revisited

In 2018, the U.S. Preventive Services Task Force (USPSTF) concluded that the evidence was insufficient to assess the balance of benefits and harms of screening for atrial fibrillation (AF) with electrocardiography. However, the proliferation of wearable devices capable of detecting brief episodes of cardiac arrhythmias raised the question of whether screening high-risk patients outside of the office, analogous to home blood pressure monitoring, might prove beneficial. In a scientific statement, the American Heart Association discussed the knowledge gaps regarding the risk of stroke and benefits and harms of initiating long-term anticoagulation in persons with subclinical AF.

Two randomized screening trials published in 2021 aimed to fill these gaps, but instead arrived at mixed conclusions. In the LOOP Study, 6004 Danish adults aged 70 to 90 years with stroke risk factors were randomized in a 1:3 ratio to receive an implantable loop recorder (ILR) or routine medical care. ILR participants were contacted if they had atrial fibrillation lasting for at least 6 minutes and recommended to start anticoagulation. Control participants received electrocardiography as needed from their primary care physicians. After a median follow-up of 64.5 months, 32% of patients in the ILR group and 12% of patients in the control group had atrial fibrillation detected, with similar proportions initiating oral anticoagulation. However, there was no significant difference in the primary outcome of stroke or systemic arterial embolism (4.5% of patients in the ILR group vs. 5.6% in the control group). Rates of major bleeding were not statistically different between the groups.

In the STROKESTOP trial, 28,768 Swedish adults aged 75 or 76 years were randomized to receive an invitation to screening with a handheld single-lead electrocardiogram twice daily for 2 weeks or usual care. After nearly 7 years of follow-up, a composite outcome of stroke, systemic embolism, hospitalization for bleeding, or all-cause mortality was slightly less likely to occur in the intervention group (NNT=93), but differences in individual outcomes were not statistically significant.

Reviewing these trial results and additional data, the USPSTF recently updated its 2018 statement and concluded that the evidence remains insufficient to make a recommendation. An accompanying editorial in JAMA Internal Medicine by Drs. John Mandrola and Andrew Foy (who also authored a 2019 editorial on the downsides of detecting asymptomatic atrial fibrillation) noted that the potential benefits of widespread cardiac rhythm monitoring on cardiovascular and stroke risk could be offset by  "misdiagnosis and downstream cascades of care" and that the "work-up of [arrhythmias] can lead to anxiety, iatrogenic harm, and excess health care costs."

**

This post first appeared on the AFP Community Blog.

**

Update (3/2/22): VITAL-AF, a cluster randomized controlled trial of >30,000 patients aged 65 years or older in Massachusetts General Hospital-affiliated primary care clinics, tested screening for AF with a handheld single-lead ECG (KardiaMobile) during vital sign assessments. Compared to the usual care group, the screened group had a marginally higher incidence of newly diagnosed AF after 1 year, but no difference in the proportion of individuals treated with oral anticoagulants.

How much administration does health care really need?

During my residency in family medicine, one of the faculty was tasked with the challenge of teaching all of us the "practice management" curriculum, trying earnestly to make us understand how much of our time and energy in clinical practice would be consumed by administrative tasks. Even then, it struck me as borderline crazy how non-procedural physicians are paid. At the time I graduated from residency (2004) and for many years afterward, the financial value of an outpatient "evaluation and management" (E/M) visit was determined not only by the complexity, acuity, and number of medical issues addressed, but also by the number of body "systems" discussed and physically examined. Why the latter should have any relationship to how much money a patient and/or her health insurance company pays for an office visit was, and remains, beyond my comprehension.

My colleague and fellow family physician Dr. Ranit Mishori has written for Medscape about why she and many other clinicians have come to hate the review of systems (ROS). Like many others over many years, she pointed out that a comprehensive (rather than focused on the chief complaint or condition being treated) ROS was usually useless and often led to bloated documentation. Nonetheless, each health care organization I've worked for over the years has employed a small army of people whose job it is to make sure that physicians include the ROS and enough physical examination elements to justify coding a higher-level (i.e., higher charge) visit, and to let us know if we're not doing it right.

Last year, the Centers for Medicare & Medicaid Services (CMS) finally simplified their guidelines for E/M visit codes, essentially eliminating the requirement to document more information than clinically useful about the history of present illness, ROS, and physical examination. In FPM, the American Academy of Family Physicians' practice management journal (on whose editorial board I served from 2010-2014), this change was celebrated with articles with titles such as "Outpatient E/M Coding Simplified" and "A Step-by-Step Time-Saving Approach to Coding Office Visits." That a series of several articles was required to explain the simplified guidelines to physicians and common situations such as "Combining A Wellness Visit with a Problem-Oriented Visit" means that practice management curricula aren't going away any time soon, even though the physician who taught me has gone on to a well-deserved retirement. For while some administrative burdens have been lessened, others have been increased, as payment incentives for providing high quality of care require physicians to document other things, such as the reason my 70 year-old patient with end-stage renal disease on dialysis and chronic heart failure doesn't need to have any more screening mammograms. More primary care practices are using medical scribes, in-person or virtual, to document office visits. While that's good to prevent burnout, it remains faintly ridiculous that armies of such people are being employed for a task that adds little actual value to the heath care encounter.

Is there really too much administration in health care? From an objective viewpoint, the answer is absolutely yes, as a recent JAMA commentary observed:

A typical US services industry (for example, legal services, education, and securities and commodities) has approximately 0.85 administrative workers for each person in a specialized role (lawyers, teachers, and financial agents). In US health care, however, there are twice as many administrative staff as physicians and nurses [emphasis mine], with an estimated 5.4 million administrative employees in 2017, including more than 1 million who have been added since 2001.

The commentary went on to discuss the findings of a new report that concluded, amazingly, that "an estimated $265 billion, or approximately 28% of annual administrative spending, could be saved without compromising quality or access by implementing about 30 interventions that could be carried out in the next 3 years." This without any major structural changes in the U.S. health care system - no need to transition to "Medicare for All" or a single-payer model (though either could lead to even greater efficiencies and savings). Companion commentaries explained why U.S. health care administrative expenses are so high (efforts to control care utilization and prices through market-based forces give rise to activities that require the hiring of more administrators) and how administrative simplification can be economically incentivized (because, apparently, wasting $265 billion per year on unnecessary administration isn't incentive enough).

Achieving this logical next step in administrative simplification will, of course, be an uphill battle. To paraphrase Dr. Don Berwick, what some call health care "waste," others call "income." The same certainly goes for those armies of people looking over my E/M coding, the armies of scribes, and the armies of quality of care managers and utilization and cost controllers. The armies of people working in health care today who facilitate transactions and documentation instead of improving the patient's care experience and health outcomes.

Newer glucose-lowering drugs also treat obesity and heart failure

Last June, the U.S. Food and Drug Administration (FDA) approved a weekly semaglutide (Wegovy) subcutaneous injection for chronic weight management in adults with obesity or overweight with at least one weight-related condition, based on randomized controlled trial (RCT) evidence that it produces substantial weight loss in persons with a body mass index of 27 or greater without diabetes. A lower dose of semaglutide (Ozempic), a glucagon-like peptide-1 (GLP-1) receptor agonist, had previously been approved by the FDA as a second-line therapy for patients with type 2 diabetes that reduced risk of major adverse cardiac events (MACE) in patients with established cardiovascular disease (CVD).

A 2021 BMJ clinical practice guideline examined the benefits and harms of GLP-1 receptor agonists and sodium-glucose cotransporter 2 (SGLT-2) inhibitors and made recommendations for use of these two drug classes in persons with type 2 diabetes and different levels of CVD risk with and without chronic kidney disease, summarized in Patient-Oriented Evidence That Matters in the January issue of American Family Physician. A related editorial by Dr. Sandy Robertson discussed how evidence can inform when to recommend starting a diabetes drug from one of these two CVD risk-lowering classes:

Current data strongly support a reduction in MACE and all-cause mortality with SGLT-2 inhibitors and GLP-1 agonists in patients who have diabetes with established CVD or kidney disease. These patients should be offered one of these medications in the absence of contraindications, regardless of glucose control. Because there is no significant reduction in cardiovascular outcomes in patients who have diabetes without established CVD, patient-centered shared decision-making about adding an SGLT-2 inhibitor or a GLP-1 agonist for cardiovascular benefit is important and ... based on other established benefits such as weight control (moderate net weight loss for GLP-1 agonists and small weight loss for SGLT-2 inhibitors) against risks of genital infections (SGLT-2 inhibitors) or gastrointestinal disturbances (GLP-1 agonists).

Although GLP-1 agonists are associated with greater weight loss than SGLT-2 inhibitors, a population-based cohort study using Medicare and two U.S. commercial claims data sets found that starting a SGLT-2 inhibitor, compared to starting a GLP-1 agonist, reduced the relative risk of hospitalization for heart failure by about 30 percent in patients with and without CVD. A recent international RCT (622 centers in 23 countries) found that adding the SGLT-2 inhibitor empagliflozin (Jardiance) to usual therapy for patients with heart failure with preserved ejection fraction improved a composite outcome of CVD mortality and hospitalization, regardless of the presence of diabetes (RRR=19%, NNT=31). Similarly, a systematic review and meta-analysis of 8 earlier RCTs found that in heart failure patients without diabetes, SGLT-2 inhibitor treatment reduced the risk of this composite outcome by 20 percent.

Based on this new data, should family physicians consider adding a SGLT-2 inhibitor to the standard combination of drug therapies for their patients with heart failure? I think the evidence supports doing so, but it may take some time before generalists become comfortable with this practice change, as I explained in a Medscape news story. A qualitative study of Australian general practitioners suggested that knowledge gaps, drug adverse effects, and a preference for subspecialists to initiate SGLT-2 inhibitor therapy may be obstacles to increased prescribing of these drugs, which remain very expensive in the U.S.

**

A slightly different version of this post first appeared on the AFP Community Blog.

Preventing mumps and COVID-19: don't let perfect be the enemy of good

The lower efficacy of COVID-19 vaccines against the SARS-CoV-2 Omicron variant due to viral mutations and waning immunity of the initial series has unfortunately fueled an anti-vaccination narrative that a less-than-perfectly protective vaccine has little clinical value. A comparison to a well-established childhood vaccine exposes the flaw in this argument. According to a 2020 Cochrane review, the effectiveness of measles, mumps, and rubella (MMR) vaccine in preventing mumps is 72% after one dose and 86% after two doses. After the two-dose MMR vaccine was added to the U.S. childhood immunization schedule, though, the number of reported cases of mumps fell from >150,000 in 1968 to 231 in 2003. In the past two decades, outbreaks have occasionally increased the incidence to several thousand reported cases per year.

A recent study in Pediatrics examined the epidemiology of mumps in U.S. children and adolescents from 2007 to 2019. It found that 87% of children diagnosed with mumps during this period had received at least one dose of MMR vaccine, including most of the 2% of children who required hospitalization. Also, only 2% of cases were associated with international travel. The authors concluded that "clinicians should suspect mumps in patients with parotitis or mumps complications, regardless of age, travel history, and vaccination status."

For physicians who have never seen a patient with this infection, a 2014 American Family Physician article on salivary gland disorders noted that mumps typically causes bilateral pain and edema of the parotid glands, otalgia, and trismus. Rarely, it can cause meningitis and encephalitis, as illustrated in a recent BMJ case report. Mumps spreads through airborne droplets (salivary, nasal, and urinary secretions) and is highly contagious. The diagnosis should be confirmed with reverse transcriptase-polymerase chain reaction (RT-PCR) or viral culture of a sample obtained with a buccal swab.

In fully vaccinated individuals, giving a third or "booster" dose of MMR vaccine was shown to reduce the risk of mumps during a 2015-16 U.S. college outbreak and 3 separate outbreaks in Queensland, Australia in 2017-18. Based on these studies, in 2018 the Centers for Disease Control and Prevention's Advisory Committee on Immunization Practices recommended that MMR-vaccinated individuals who are "part of a group or population at increased risk for acquiring mumps because of an outbreak" should receive a third dose of MMR vaccine "to improve protection against mumps disease and related complications." Sound familiar? Clinicians should continue to strongly recommend that patients receive authorized COVID-19 vaccines and booster doses and not let the perfect be the enemy of the good.

**

This post first appeared on the AFP Community Blog.

Clinical prevention shorthand: Do, Don't Do, or Don't Know

Dr. Ned Calonge, the chairman of the U.S. Preventive Services Task Force (USPSTF) during my time as a medical officer at the Agency for Healthcare Research and Quality (AHRQ), liked to say that USPSTF recommendation letter grades boiled down to one of three things: Do, Don't Do, or Don't Know. Doctors like actionable guidelines. Do, or Don't Do. We don't like seeing Don't Know. When I left AHRQ at the end of 2010, about one-third of recommendation statements were rated "I," or insufficient evidence to determine the balance of benefits and harms. In other words, a whole lot of clinical prevention Don't Know.

A provision of the Affordable Care Act stipulated that the USPSTF submit an annual report to Congress identifying evidence gaps and recommending priority areas for prevention research. Its first report, which came out the year after I left, identified eleven clinical topics that had critical evidence gaps (resulting in "I" statements) "that if filled are likely to result in new recommendations." These topics were

1. Screening for coronary heart disease

2. Screening for colorectal cancer with fecal DNA testing and CT colonography

3. Screening for hepatitis C

4. Screening for hip dysplasia

5. Obesity: moderate- to low-intensity counseling

6. Interventions to prevent child abuse and neglect

7. Screening for illicit drug use

8. Screening for osteoporosis in men

9. Screening for depression in children

10. Screening and counseling for alcohol misuse in adolescents

11. Aspirin use in adults ages 80 years and older

I've bolded the four topics that, just over a decade later, are no longer "I" statements. #2, 3, and 7 are now Do's ("A" and "B" grades), and #11 is a Don't Do ("D" grade). That's 36 percent, which is something of a disappointment since the USPSTF called these "critical" evidence gaps, not evidence gaps that it felt researchers could fill at some indistinct future date. Moreover, two of these Do's, in my opinion, were highly questionable calls; no trials have shown that fecal DNA testing, CT colonography, or illicit drug use screening improve patient-oriented outcomes. Arguably, these changes were driven by changes in the composition of the Task Force, not new evidence.

A recent study examined characteristics of evidence and funding support for 11 USPSTF "I" statements that changed to a letter grade (most commonly a "B") between 2010 and 2019. The National Institutes of Health (NIH), the largest federal funder of biomedical research, supported a sizable percentage of the critical studies: 28.8%, to be exact. Were the researchers reading the USPSTF's annual reports to Congress when they wrote their grant requests? Perhaps a few did, but it seems unlikely to me. And for every completed study that filled a prevention evidence gap, many more studies were not done because NIH program officers didn't know there were other gaps that needed to be filled.

A new National Academy of Medicine report, "Closing Evidence Gaps in Clinical Prevention," aims to make it easier for the USPSTF and AHRQ to communicate research questions that need to be answered to the agencies that fund prevention research. An ad hoc committee developed a clinical prevention taxonomy and workflow designed to classify different types of evidence gaps (Foundational Issues, Analytic Framework, Dissemination and Implementation) and prioritize them into a research agenda that can be more easily shared. This is such a good idea that it's a wonder that no one thought of it before or, if they did, got decision-makers to pay attention. When 2032 rolls around, hopefully a much higher percentage of evidence gaps identified by the Task Force in its Eleventh Report to Congress (which, as a reflection of our times, involve health equity and disparities in cardiovascular disease and cancer topics) will be Do, or Don't Do, rather than Don't Know.