Does COVID-19 harm mental health through stress, enforced isolation, or the virus itself?

Eight months ago, during the first wave of the pandemic, my American Family Physician editor colleague Dr. Jennifer Middleton discussed World Health Organization and Centers for Disease Control and Prevention (CDC)-recommended strategies and resources for optimizing mental health in health care workers, patients, and children. A CDC representative national survey conducted in late June found a strikingly high prevalence of symptoms of anxiety or depressive disorder (30.9%), trauma- and stressor-related disorder (26.3%) and new or increased substance use (13.3%). By comparison, a 2019 survey found that only 8.1% and 6.5% of people had symptoms of anxiety or depression, respectively. 1 in 10 respondents to this year's survey also reported having seriously considered suicide in the preceding 30 days, with disproportionately higher suicidality in younger adults (age 18-24 years), racial and ethnic minorities, essential workers, and unpaid adult caregivers.

In a recent commentary, Dr. Christine Moutier from the American Foundation for Suicide Prevention recommended several COVID-19-specific suicide prevention strategies that fuse clinical, health system, and policy interventions: reduce risk for people with mental illness or addiction; increase social connectedness; address risk at the moment of crisis; reduce access to lethal means; address COVID-19 increases in alcohol consumption and drug overdoses; mitigate financial strain; address domestic violence and unsafe environments; and prevent unsafe media and entertainment messaging on suicide.

The rise in anxiety, depression, stress, and suicidality coincided with the widespread conversion of office-based visits for behavioral and psychiatric conditions to telehealth, which may have restricted access to mental health care for existing and new patients. As three psychiatrists observed in a JAMA Viewpoint:

Patients with psychiatric disorders are particularly vulnerable to COVID-19 due to high rates of overweight, tobacco smoking, medical comorbidities, and poor self-care. ... Daily news of large-scale COVID-19–related disease and death in the community over months or years is almost certain to elevate psychiatric burden in the population. As such, the pattern of stress resembles that experienced by refugees or others exposed to chronic violence. ... A sustained increase in demand for psychiatric services may well exceed the existing capacity of the system over time and may last for years, depending on the course the pandemic takes.

Persons with prior psychiatric diagnoses may be at higher risk of death from COVID-19 infection. A cohort study of 1685 patients hospitalized with COVID-19 from February through April found that after controlling for demographics, medical comorbidities, and hospital location, patients with a psychiatric disorder were 1.5 times as likely to die as those with no psychiatric diagnosis. Citing a kinship network study that suggested that each COVID-19 death in the U.S. leaves nine bereaved close family members, some have suggested that primary care physicians screen relatives of persons who die from COVID-19 for symptoms of depression, prolonged grief, or post-traumatic stress disorder and provide evidence-based interventions if needed.

On the other hand, a diagnosis of COVID-19 may increase the risk for developing a mental health disorder. A retrospective cohort study that utilized electronic health record data from more than 62,000 U.S. patients between January 20 and August 1 found that COVID-19 survivors were more likely to have a first psychiatric diagnosis, a new psychiatric diagnosis, or a relapse of a previously stable diagnosis within 14 to 90 days than six other unrelated health events. However, this study design could not determine if these additional diagnoses were preexisting and unrecognized prior to COVID-19 infection or a direct consequence of the infection or medical interventions (including isolation at home or in the hospital).

**

This post first appeared on the AFP Community Blog.

Incarceration, restorative justice and health

Of the barrage of discouraging pandemic statistics - the quarter of a million U.S. deaths to date, the 93 daily new cases per 100,000 residents and current 20+ percent test positivity rate in my adopted state of Utah - one of the most striking is that of the 14 persons who died from COVID-19 in Texas county jails from April to September, 11 were awaiting trial and had not been convicted of a crime. A recent report of COVID-19 deaths in Texas correctional institutions from the University of Texas at Austin found that jail deaths represented only 6 percent of the 231 overall deaths during incarceration (including prison staff). Those serving time for a criminal conviction were, clearly, not sentenced to death by suffocation from a deadly virus. Since Texas accounts for approximately 9 percent of the U.S. population, my conservative estimate is that at least 100 Americans, mostly persons of color, have already died from COVID-19 while detained in jails awaiting trials, unable to physically distance or otherwise protect themselves.

A few years ago, as a member of the American Academy of Family Physicians' Commission on the Health of the Public and Science, I co-authored a position paper that articulated a family medicine perspective on the negative effects of mass incarceration on the health of justice-involved persons, their families, and their communities. We found that in 2016, the U.S. corrections system supervised 6.6 million people (1 in every 50 residents) in jails, prisons, or on probation or parole - the highest incarceration rate in the world and a nearly fivefold increase since 1978. Given these figures, a basic understanding of the justice system has become essential not only for family physicians and internists, but also pediatricians, who are increasingly likely to encounter justice-involved youth. The "Patients, Populations, and Policy" course that I co-direct at Georgetown includes a mandatory screening of the documentary 13th, which argues that a loophole in the 13th Amendment to the U.S. Constitution, which abolished slavery and involuntary servitude for African Americans "except as a punishment for crime," enabled institutionalized racism in policing and criminal sentencing that persists to this day.

This background explains why many progressive Americans are not celebrating the recent election of Senator Kamala Harris, a former prosecutor, to the office of Vice President of the United States. As the current Vice President pointed out during their October 7 debate, during Harris's tenure as California Attorney General, Black persons were much more likely to be prosecuted for minor drug offenses and were disproportionately incarcerated compared to their share of the general population. In a New York Times Magazine article, former felon Reginald Betts reflected on his mixed feelings about prosecutors and mass incarceration. Betts, who was convicted of carjacking and armed robbery and imprisoned from age 16 to 24, was shocked to learn after his release that his mother had been raped at gunpoint just weeks after his arrest. Naturally, even though some of the men with whom he served time were guilty of similar offenses, Betts "thought he [the rapist] should spend the rest of his years staring at the pockmarked walls of prison cells that I knew so well."

Betts observed that most Americans who oppose mass incarceration today imagine that most of the prison population is serving time for nonviolent drug-related crimes. Not so: "You could release everyone from prison who currently has a drug offense and the United States would still outpace nearly every other country when it comes to incarceration." What, then, is the responsibility of progressive prosecutors who, like Vice President-elect Harris, desire to address inequities in the justice system that result not only from unjust policing, but penalties for the crimes themselves? Betts responded:

The prosecutor’s job, unlike the defense attorney’s or judge’s, is to do justice. What does that mean when you are asked by some to dole out retribution measured in years served, but blamed by others for the damage incarceration can do? The outrage at this country’s criminal-justice system is loud today, but it hasn’t led us to develop better ways of confronting my mother’s world from nearly a quarter-century ago: weekends visiting her son in a prison in Virginia; weekdays attending the trial of the man who sexually assaulted her.

Ideally, our criminal justice system should serve two purposes: punishment and rehabilitation. That three-quarters of persons released from state prisons in 2005 were arrested again within 5 years suggests that the system fails miserably at the latter, and if spending time behind bars (punishment) is supposed to deter criminals from committing crimes again, failing at the former as well. As Betts wrote:

It always returns to this for me — who should be in prison, and for how long? I know that American prisons do little to address violence. If anything, they exacerbate it. If my friends walk out of prison changed from the boys who walked in, it will be because they’ve fought with the system — with themselves and sometimes with the men around them — to be different. 

Through the mystery novels of the late Tony Hillerman, I am superficially familiar with the Navajo Nation's concept of "restorative justice", described in a 1994 New Mexico Law Review article by former Navajo Nation Chief Justice Robert Yazzie. Yazzie characterized traditional American justice as an "adversarial" process administered by strangers:

Law, in Anglo definitions and practice, is written rules which are enforced by authority figures. It is man-made. Its essence is power and force. The legislatures, courts, or administrative agencies who make the rules are made up of strangers to the actual problems or conflicts which prompted their development. When the rules are applied to people in conflict, other strangers stand in judgment and police and prisons serve to enforce those judgments.

In contrast, traditional Navajo peacemaking shuns a justice system based on "social control" in favor of pragmatic group problem-solving about "the means to live successfully." In a related article, Yazzie wrote:

Navajo peacemaking is about the effects of what happened. Who got hurt? What do they feel about it? What can be done to repair the harm? ... In Navajo peacemaking, offenders are brought in to a session involving the person accused of an offense and the person who suffered from it, along with the “tag-along” victims of the crime, namely the relatives of the accused and of the person hurt by the accused.  The sessions are moderated by a community leader called a “peacemaker.” The action is put on the table. People talk about what happened and how they feel about it. A harmful act is “something that gets in the way of living your life,” and Navajo peacemaking deals with such an act by identifying it, talking about it, and devising a plan to deal with it.

The recent execution of a Navajo man on federal death row for the carjacking-murder of two Navajos in 2001, despite the opposition of the Navajo Nation, highlighted the potential advantages of incorporating restorative justice into state and federal criminal justice systems. Although it's possible that the victims' loved ones gained some satisfaction from the execution (albeit 19 years after the murders), it's hard to argue that any harm was "repaired" or harmony restored by this second violent act. (Note: as a practicing Catholic, I believe that the death penalty is wrong regardless of the crime.) I don't believe that prisons should be abolished, any more than I believe that police departments should be defunded. But if the U.S. is going to continue to pour hundreds of billions of dollars into incarceration every year, a large chunk of those dollars ought to be devoted to peacemaking - making the offender whole and less likely to offend again - rather than punishment.

Mobile heart monitoring: advantages and limitations

During the first wave of the pandemic, when hydroxychloroquine was still thought to be an effective treatment for hospitalized patients with COVID-19 (subsequent studies have shown otherwise), some U.S. hospitals used personal electrocardiogram (ECG) devices to monitor these patients for drug-induced QT interval prolongation to conserve personal protective equipment and telemetry monitors. Such devices were already being used by patients with known cardiac conditions to monitor their heart rhythms in out-of-office settings.

As I discussed in a previous post, wearable devices such as the Apple Watch are also being studied to detect atrial fibrillation in asymptomatic primary care patients. In a 2019 study of more than 400,000 U.S. Apple Watch wearers with no self-reported history of atrial fibrillation, about 1 in 200 individuals received an irregular pulse notification and were scheduled for a telemedicine visit with a clinician to confirm study eligibility and triage those with urgent symptoms to the emergency department. Participants without urgent symptoms were mailed an ECG patch to wear for up to 7 days and then mail back to the study center.

Of the 450 participants who returned an ECG patch, 34% were confirmed to have atrial fibrillation, with a higher diagnostic yield in persons age 65 years or older. Of the 86 patients whose watches generated irregular pulse notifications when they were wearing the ECG patch, the positive predictive value for atrial fibrillation was 84%. In a survey completed at 90 days by patients who received an irregular pulse notification, 28% reported being prescribed a new medication, 33% were referred to a specialist, and 36% were recommended to have additional testing.

In the November 1 issue of American Family Physician, Dr. Madhavi Singh and colleagues reviewed the diagnostic test features of KardiaMobile, a $99 single-lead device that connects wirelessly to a smartphone app to generate an ECG tracing with automated interpretation. In studies of patients with known or suspected arrhythmias, KardiaMobile had greater than 90% sensitivity and specificity for atrial fibrillation or atrial flutter compared to a standard 12-lead ECG. However, no studies have compared its accuracy with a Holter or event monitor, and effects on patient outcomes are uncertain, particularly in populations at low risk for arrhythmias. (In 2018, the U.S. Preventive Services Task Force found insufficient evidence to assess the balance of benefits and harms of screening for atrial fibrillation, which has several potential downsides.) The authors concluded (and I agree) that similar to the Apple Watch, "further studies are needed before KardiaMobile can be recommended for use in seemingly healthy patients."

**

This post first appeared on the AFP Community Blog.