You Cannot Function in “Overwhelm”: Helping Primary Care Navigate the Slow End of the Pandemic

SUMMARY

Goal:

This study was developed to explicate underlying organizational factors contributing to the deterioration of primary care clinicians’ mental health during the COVID-19 pandemic.

Methods:

Using data from the Larry A. Green Center for the Advancement of Primary Health Care for the Public Good’s national survey of primary care clinicians from March 2020 to March 2022, a multidisciplinary team analyzed more than 11,150 open-ended comments. Phase 1 of the analysis happened in real-time as surveys were returned, using deductive and inductive coding. Phase 2 used grounded theory to identify emergent themes. Qualitative findings were triangulated with the survey’s quantitative data.

Principal Findings:

The clinicians shifted from feelings of anxiety and uncertainty at the start of the pandemic to isolation, lack of fulfillment, moral injury, and plans to leave the profession. The frequency with which they spoke of depression, burnout, and moral injury was striking. The contributors to this distress included crushing workloads, worsening staff shortages, and insufficient reimbursement. Consequences, both felt and anticipated, included fatigue and demoralization from the inability to manage escalating workloads. Survey findings identified responses that could alleviate the mental health crisis, namely: (1) measuring and customizing workloads based on work capacity; (2) quantifying resources needed to return to sufficient staffing levels; (3) promoting state and federal support for sustainable practice infrastructures with less administrative burden; and (4) creating patient visits of different lengths to rebuild relationships and trust and facilitate more accurate diagnoses.

Practical Applications:

Attention to clinicians’ mental health should be rapidly directed to on-demand, confidential mental health support so they can receive the care they need and not worry about any stigma or loss of license for accepting that help. Interventions that address work–life balance, workload, and resources can improve care, support retention of the critically important primary care workforce, and attract more trainees to primary care careers.

INTRODUCTION

The United States is experiencing a mental health crisis and an overburdened mental health system (Kuehn, 2022). The COVID-19 pandemic harmed mental health by upending routines and disrupting connections (Office of the Surgeon General, 2023). Those on healthcare’s front lines faced uncertainty while caring for sick or dying patients, navigating evolving protocols, and dealing with fears of becoming ill or transmitting the virus to loved ones (Firew et al., 2020; Sasangohar et al., 2020). These stressors were unprecedented.

Before the pandemic, clinicians (i.e., physicians, physician assistants/associates, and nurse practitioners) dealt with job stressors that affected their mental health, including electronic health record systems (DeChant et al., 2019; Lou et al., 2022), documentation requirements (Khullar et al., 2021; Tseng et al., 2018), suboptimal teamwork (Edwards et al., 2018), and toxic cultures (Willard-Grace et al., 2014), with accompanying high rates of burnout, depression, and suicide (Peckham, 2018; Pospos et al., 2018). Studies have demonstrated that physicians’ suicide rates are higher than the general population’s (Center et al., 2003; Duarte et al., 2020), and U.S. physicians are 1.3 times more likely to die by suicide than their global counterparts (Dutheil et al., 2019). Concomitantly, healthcare workers (HCWs) such as physicians and advanced practice clinicians (nurse practitioners, in particular) experienced higher levels of burnout (Grace & VanHeuvelen, 2019).

These HCWs historically have reported high levels of psychological distress from caring for patients during pandemics (Kisley et al., 2020), with paradoxically low uptake of mental health services (Loiseau et al., 2022; Weiner, 2023). Physicians, in particular, are reluctant to seek help due to perceptions of stigma (Brower, 2021). Early data on the effects of the COVID-19 pandemic showed vast increases in anxiety, depression, and post-traumatic stress disorder (Firew et al., 2020; Schwartz et al., 2020; Shechter et al., 2020). Workplace factors associated with mental health symptoms during the health emergency included shortages of personal protective equipment, insufficient training for redeployment, minimal access to mental health support, and an unmet desire for greater recognition by leadership—along with the effects of caring for patients who died from COVID-19 (Khajuria et al., 2021; Schwartz et al., 2020).

Many researchers have studied the experiences of HCWs in hospitals (Sasangohar et al., 2020) and outpatient facilities (Shechter et al., 2020; Van Wert et al., 2022) in the wake of the pandemic, but few have examined the primary care setting (Firew et al., 2020). This study explored the mental health of those caregivers and their perceptions of factors contributing to their struggles. Data were collected from March 2020 to March 2022 through the Larry A. Green Center for the Advancement of Primary Health Care for the Public Good’s serial national survey of primary care clinicians.

METHODS

We used an exploratory design to conduct mixed-methods analyses of data from multiple waves of the survey. The methods have been detailed elsewhere (Etz et al., 2023; Sullivan et al., 2023). Briefly, in March 2020, the Green Center began surveying primary care clinicians practicing family medicine, internal medicine, pediatrics, and other specialties in primary care settings. The method encompassed a rapid cycle survey of multiple cohorts, typically every month, to measure caregiver capacity and response.

Survey invitations were distributed through more than 150 professional societies and organizations, listservs, and practice-based research networks known to Green Center leaders. Participating physicians could sign up for an automated mailing directly to their inboxes. Snowball sampling encouraged them to forward invitations to their colleagues, thereby expanding the survey pool (Kirchherr & Charles, 2018).

Each survey took approximately 3–5 min to complete and consisted of core topics, demographics, and questions. Repeated themes included care priorities, resources, finances, staffing, telemedicine, patient health, and vaccine issues. A team of scholars skilled in health policy, biostatistics, family medicine, and social science conducted analyses as each survey closed.

A comment box asking, “Is there anything else you would like us to know about your experience?” in each survey generated more than 11,150 comments. Comments were coded by a qualitative researcher (R.E.), a statistician, a nurse practitioner (S.R.), and a finance analyst. Analyses used an a priori-established codebook (Miles et al., 2019). The inclusion of emergent themes led to a deeper understanding of clinician experiences, which were quickly used to inform future survey questions.

After the survey closed in March 2022, a second analytic team (E.E.S. and J.D.) performed an independent grounded-theory analysis of comments, prioritizing emerging themes as opposed to applying previously determined constructs (Glaser & Strauss, 2010). The team generated an initial codebook based on analysis of a subset of comments, refined the emergent codes, and tested the codes with a second data subset (i.e., with deductive coding). The initial codes included health inequities, leader/administrator support, requests for advocacy, financial resources, positive changes, mental health (both patients and clinicians), pandemic challenges, and workload.

Once the codebook was created and tested, all deidentified comments were entered into NVivo12 software and thematically indexed and coded for formal content analysis (Miles et al., 2019). Twenty percent of the comments were coded by two analysts (E.E.S. and J.D.); intercoder reliability was calculated using Cohen’s kappa coefficient. The resulting kappa (0.92) indicated substantial agreement. The analysts met weekly to review and compare coding, as well as identify new codes. For this study, data from the mental health, challenges, and workload codes were used to identify subcodes related to clinician mental health.

Quantitative summaries of survey responses provided descriptive statistics (i.e., prevalence) to confirm or disconfirm themes from qualitative analyses. When possible, the percentages of respondents endorsing comments during pandemic Years 1 and 2 were noted to emphasize evolving themes over time. Quantitative findings were triangulated with qualitative results for concurrent validity.

The study was determined to be exempt from human subjects research by Virginia Commonwealth University’s Institutional Review Board.

RESULTS

The survey was fielded 36 times from March 2020 through March 2022. More than 32,000 individual responses were received from approximately 8,100 unique individuals (i.e., 92% physicians and 8% advanced practice clinicians); physicians were predominantly family physicians (averaging 70% per survey). Among these, approximately 35% were based in small practices, and 40% were from hospitals and health system-owned practices. The remainder practiced in direct primary care, retail settings, government-owned health systems, skilled nursing facilities, and urgent care settings. Primary care clinicians in the study represented all 50 states and Washington, DC, with regional representation as follows (percentages are rounded): Northeast (17%), Midwest (18%), Southeast (23%), Rocky Mountains (7%), Southwest (14%), and Pacific (20%). Twenty-two percent of respondents practiced in rural areas.

The survey links used to distribute the survey included embedded alphanumeric codes that allowed us to anonymously track respondents who took more than one survey. The data reported here are from all survey responses received. Each survey generated a national cohort through convenience and snowball sampling. Though there was no intentional framework to generate a representative sample, the demographics of each survey instance, as well as those of all respondents, mirrored the demographics of primary care clinicians in the United States (Willis et al., 2020).

Respondents reported that they acutely felt the effects of the pandemic, with 48% stating that burnout was at an all-time high and 45% reporting detriments to their psychological well-being. Other concerns included physical (36%) and family well-being (33%). Interest in early retirement or leaving practice entirely was seen in 11% of participants throughout the two survey years (and climbed to 17% in an additional anniversary survey in March 2023). Qualitative analysis identified three emergent themes related to the deterioration of clinician mental health: stress and “overwhelm;” workload and capacity; and alienation from work.

Stress and Overwhelm

Early in the pandemic, respondents attributed feelings of stress and anxiety to high degrees of uncertainty as COVID-19 forced a shift to telemedicine and the closing (temporary or permanent) of practices. Primary care clinicians struggled with caring for patients in this new world. They associated early pandemic stress with the struggle to acquire and maintain adequate personal protective equipment and keep up with changing care protocols. One respondent noted, “We are now trying to convert in-person visits to virtual visits as much as possible. It has been stressful, as most of our clinicians have not done telehealth before.”

For many clinicians, there were constant worries related to the financial viability of their practices and how they and other team members would be paid. Those working in health systems shared a similar concern. One respondent said, “The incredible financial pressure from [the] health system [is] causing [pressure] to make up the lost revenue for the system.” Later in the pandemic, stress was associated with preparing for COVID-19 surges or dealing with staffing shortages. Quantitative results confirmed that these concerns were widespread (Table 1). Nearly half of the respondents feared they did not have sufficient cash on hand to stay open for 4 weeks, and 25% skipped or deferred their salaries.

TABLE 1.

Quantitative Survey Findings Support the Emergent Stress and “Overwhelm” Theme

Survey Series*	Survey Item	Percentage of Respondents**	Sample Size***
3–13, 15	Lack of personal protective equipment	54	1,265
20, 23, 24	Difficulty getting/lack of/inadequate amount of testing supplies	23	1,149
23	Video visits covered at parity with in-person visits	56	1,472
23	Phone visits covered at parity with in-person visits	28	1,472
5, 7, 9, 15, 17	Likelihood of having cash on hand to stay open for the next 4 weeks	56	1,953
23–25	Clinicians redeployed to other locations	19	1,341
30	Struggle to keep up with patient health needs—pent-up demand	55	1,263
15, 22, 23, 26	Negative health burden due to delayed/inaccessible care	57	925
9	Patients’ deaths from lack of access to care during pandemic	9	735
11–17	Clinician salaries skipped/deferred	26	584

Note. Data from the Larry A. Green Center for the Advancement of Primary Health Care for the Public Good’s serial national survey of primary care clinicians, March 2020–March 2022.

^*Surveys were numbered Series 1–36. Numbers here indicate the survey series from which the data were derived.

^**Where the percentage of respondents refers to multiple series, the number displayed is an average across those series.

^***Where data represent multiple surveys, the sample size displayed is the average number of respondents across the series.

The Shift From Stress to Overwhelm

Occasionally as early as 6 months into the pandemic but consistently at months 9–12, respondents’ feelings of stress shifted to overwhelm, and burnout was mentioned more frequently. One respondent summarized this shift, explaining, “The stress is still very high but it is different. My workload feels more overwhelming and the difficulty keeping up with chronic and preventive care is demoralizing.” The main sources of overwhelm were volume (“just too many patients to see in one day”) and acuity. As one clinician explained, “Patients delayed care, and now chronic conditions (e.g., kidney disease, diabetes, and hypertension) are poorly controlled, plus they are anxious/depressed and have lists of unrelated concerns. Trying to fit all their concerns into a visit is overwhelming.”

Piled on to the financial stress was an increased resignation to and acceptance of the reality that primary care was an especially difficult discipline and a growing sentiment that the challenges and lack of support were the results of a flawed health system. One respondent noted, “We do not get paid enough in primary care to provide all the care and coordination that our patients need.”

Workload and Capacity

While the workload, with its lack of capacity to care for patients, started as a stressful situation early in the pandemic, it was a constant source of overwhelm by Year 2—albeit not a new phenomenon in primary care, as suggested in this comment: “Once again, we are being asked to do too much with too little.” Exactly what was too much and what was too little varied, but in general, respondents reported insufficient staff support, too many administrative tasks, and too little reimbursement. These findings were supported by responses to structured survey questions (Table 2).

TABLE 2.

Quantitative Survey Findings Support the Emergent Workload and Capacity Theme

Series*	Item	Percentage of Respondents**	Sample Size***
24	March/April 2020: Practice paid for <50% of work done.	30	1,485
33, 35	Current billing is denied or well overdue from insurers or health plans.	28	795
34	Our clinicians are doing a larger amount of unreimbursed work to keep the office running.	53	630
18, 24, 30	The complexity of patient visits has increased.	58	1,096
18	Patients have a larger number of complaints per visit than they did prepandemic.	38	540
24	Patient questions have increased.	80	1,485
31	Staff turnover has increased.	74	983
35	Practice clinician/staffing ratio has changed (i.e., fewer staff per clinician).	52	847
30, 33	We have clinician positions we cannot fill.	31	1,002
30, 33	We have staff positions we cannot fill.	61	1,002
14	I have struggled to know when I can end my workday.	41	763

Note. Data from the Larry A. Green Center for the Advancement of Primary Health Care for the Public Good’s serial national survey of primary care clinicians, March 2020–March 2022.

^*Surveys were numbered Series 1–36. Numbers here indicate from which survey series the data were derived.

^**Where the percentage of respondents refers to multiple series, the percentage displayed is an average across those series.

^***Where data represent multiple surveys, the sample size displayed is the average number of respondents across the series.

Staffing

As the pandemic continued, respondents noted challenges in keeping practices fully staffed due to people calling in sick and a shrinking workforce. They also described the turnover of critical support staff (e.g., social workers, patient navigators, and medical assistants). Other respondents noted that budget cuts led to more work with less staff, thus increasing burnout. One summarized the situation by saying, “I work at work. I work at home. I work in the car. I work at the grocery store. Refills, messages, referrals … I’m tired.”

Administrative Burden

The increased administrative burden was felt to be the result of a confluence of factors, notably telemedicine, along with its increased communication from patients via a variety of channels, more forms for patients and insurers to complete and file, and a lack of infrastructure to manage the shift in care modalities that emerged in the pandemic. One respondent described the burden this way: “I continue to be pressured to do so much virtual work: patient e-mails, phone calls, and forms, in addition to nonstop telemedicine visits—with volume pressure. Telemedicine has aggravated how much admin stuff I do.” Many respondents mentioned that pent-up pandemic demand led to an increase in administrative tasks for worsening chronic conditions or diagnosing new conditions, which meant reviewing more labs and managing more medication refills.

Time

Finally, respondents noted that they did not have enough time, whether with patients or just time to think. They worried about how to manage their cognitive load. They resented the time they spent on nonclinical, nonreimbursable activities. Explained one respondent: “Patients need information, reassurance, and guidance—in person—via telemedicine, or on the telephone, but none of that time is reimbursed.” Others added that some days, they would finish the day by responding to more than 20 portal messages and completing their notes. So, while 32 hr of direct patient care translated to 50–60 hr per week of total work time, only 32 hr of direct care were compensated. Respondents also pointed out that because much of their nonbillable administrative work took up time after hours and on weekends, their feelings of exhaustion and burnout increased.

Alienation From Work

Clinicians became increasingly disconnected from their work. This feeling started early in the pandemic with the shift to virtual operations and fewer clinical practice staff. Respondents commented on the lack of in-person connection with coworkers, which was necessary to keep themselves, their patients, and their families safe. Some also noted that they missed their coworkers.

Year 2 showed a marked shift in the relationships between primary care clinicians and their patients. As one respondent said, “In the beginning, patients were very grateful. Now they are angry.” Another respondent shared, “We are seeing increasingly aggressive behaviors from our patients due to mask mandates, vaccination status, denied access to nonapproved medications, visitation restrictions, and distrust of the medical system.”

By the end of Year 2, comments were indicative of the damaged physician–patient relationships, with breaches of trust largely attributed to vaccination conversations. Clinicians felt disconnected from, and injured by, the public health and healthcare systems. While 72% felt valued by their patients, only 49% felt valued by hospitals and health systems, and only 34% felt valued by insurers.

Survey participants expressed frustration with being left out of the planning and execution of COVID-19 vaccination campaigns because vaccination is a core competency in primary care. They said they could have helped with vaccination prioritization and dissemination but were disregarded until the end, with only 1 in 5 noting they were eventually designated by state public health or local hospital systems as vaccination providers. Eventually, most were asked to vaccinate the hardest-to-reach populations and offered few to no resources for that work. At this point, respondents also reported compassion fatigue and were having a harder time feeling empathy for their patients. In a time of profound isolation for most people, primary care clinicians felt demonstrably alienated from their patients and their healthcare systems.

DISCUSSION

This study enhances knowledge of the effects of COVID-19 on the mental health of HCWs by focusing on primary care clinicians. The data illustrate the steady decline of mental health over 2 years, with contributors in three areas: stress and overwhelm, workload and capacity, and alienation from work. Clinicians’ comments can guide future improvements, specifically how organizations can restore and reaffirm their commitments to those on the front lines—the primary care clinicians who provide foundational healthcare support for all Americans. Table 3 summarizes key problems identified by respondents and solutions healthcare leaders can consider to remediate the decline in clinician mental health.

TABLE 3.

Problems and Solutions Related to Mental Health Issues, Stress, and “Overwhelm” in Primary Care

Problem	Proposed Solution
Feelings of overwhelm	Recognize when stress turns to overwhelm and then intervene. Create more flexible personal leave options or sabbaticals. Provide sufficient vacation time for all staff.
Unmanageable workload	Adjust workload to fit individual clinicians. Some can do more, while others must do less.
Doing more with less	Stop accepting this as an organizational option. Emergency funds for COVID-19 must be spent on primary care.
High volume of patients, less support staff	Retain/hire staff and start internal training programs for entry-level roles. Massively build up and invest in supportive work cultures.
Crushing administrative burden	Add administrative support roles to care teams. Advocate for state and federal policies to minimize the administrative burden created by insurers and electronic health record systems.
Time	Make appointments longer than 15–20 min, giving clinicians time to think, listen, and perform high-quality clinical work.
Alienation from patients	Take concrete steps to restore patient–clinician relationships. Follow evidence-based pathways for trust-building, including caring for clinicians in need and monitoring and providing resources to protect patient safety. Enact codes of conduct for clinicians to follow when they deal with disappointed patients.
Disconnection from coworkers	Create space for teams to hold team meetings; codify their team workflows and processes in new, hybrid work environments; and build relationships with each other.
Mental health support	Provide safe, confidential on-site, off-site, and on-demand mental health support that does not jeopardize anyone’s medical license. Healthcare organizations can work to reduce any stigma regarding mental illness among clinicians by rewording annual credentialing questions or through self-awareness campaigns (e.g., “If you feel something, say something.”).

Note. Data from the Larry A. Green Center for the Advancement of Primary Health Care for the Public Good’s serial national survey of primary care clinicians, March 2020–March 2022.

To help primary care clinicians, the first step is to distinguish stress from overwhelm. Stress is a common state in medicine, frequently accompanied by an inability to control the pace of work. Most clinicians are used to functioning under stress and have been able to modulate demands via controls of the work environment such as creating patient visits of different lengths (Linzer et al., 2014). But during COVID-19, primary care clinicians moved beyond stress into overwhelm, a state associated with slowed cognition and a reduced ability or motivation to act on one’s surroundings. This can have devastating consequences, as it paralyzes individuals and immobilizes care systems.

As Brené Brown explains:

When we say we’re overwhelmed, it’s really telling our body, “Things are happening too fast, we can’t handle them. Shut down.” When we’re stressed, we feel a lot of pressure, but we can handle it. We know what actions to take … Overwhelm means an extreme level of stress and emotional and/or cognitive intensity to the point of feeling unable to function … So, [while] we can function in stress, we really can’t function in overwhelm. (Brown, 2021)

These adverse outcomes affect cognition and the ability to make diagnoses (Linzer et al., 2022), and can interrupt care for prolonged periods until support systems are reestablished to modulate workflow. This situation became profoundly apparent in primary care during the pandemic.

The antidote for healthcare organizations is to foster system-level support while facilitating recovery as a dynamic combined system- and individual-level process that allows resources to be restored (Fleuren et al., 2021). Creating workplaces where HCWs can detach from work, where they can mentally disengage or “switch off” as needed, promotes recovery (Bennett et al., 2018; Steed et al., 2019). In addition to efforts to restore meaning, a sense of autonomy, and manageable administrative burden (Edwards et al., 2018, 2021), developing workflows that support the ability to switch off coincides with team members’ requests for worry-free time away (Sullivan et al., 2022) and may prevent a shutdown as the innate response.

To facilitate recovery and retention, organizations need to intervene when clinicians are operating in overwhelm. Interventions can include flexible vacation policies, short but meaningful break periods ranging from minutes to days, convenient spaces at work for quiet reflection, and easy-to-arrange personal leaves with benefits or sabbatical options. Mental health support can range from critical incident support teams to wellness consults, on-site and on-demand counseling, “warmlines” staffed 24/7, and buddy programs (Albott et al., 2020).

While health systems necessarily focused attention on high-volume inpatient care during the pandemic, primary care—the typical setting for most acute care—was under-resourced, which then increased stress on emergency and hospital sectors (Stange, 2021). While hospitals received financial assistance to handle the high volume of inpatients, less than 5% of the support for the healthcare sector went to primary care, thus failing to acknowledge the financial stress on smaller practices in particular (Abbasi, 2022). Historically under-resourced, primary care clinicians found their work disrupted by the pandemic and were asked to do things for which they were not paid or trained. After the health emergency passed, they had to make up the millions of hours lost on chronic illness care and delayed or missed acute care without additional support.

The data from this research identify things that can be done with a sense of urgency to make up for time lost. Primary care cannot continue to take on more work with fewer resources without severe consequences for clinicians’ mental health. The necessary human and technological infrastructure to address this crisis includes resources to hire sufficient staff and shift electronic health record system workflows to nonclinical staff. Concomitant changes in state and federal policies can reduce the administrative burden associated with excessive documentation and measurement. While these changes are being enacted, mental health support for those within primary care is urgently needed.

In addition, healthcare leaders can encourage the restoration of relationships within primary care (National Academies of Sciences, Engineering, and Medicine, 2021). Clinicians who choose to practice primary care are guided by professional expectations to provide longitudinal, continuous care, and that expectation was shattered during the pandemic. Healing the broken trust in patient–doctor relationships can be addressed by enabling clinicians to spend more time with their patients. Indeed, the ability to care and the time to do so are important burnout mitigators (Burstein et al., 2022).

Study Strengths and Limitations

The evolving survey method used here, with questions developed based on real-time data analyses, allowed responsiveness to the rapidly developing pandemic. The triangulation of qualitative and quantitative findings enhanced concurrent validity, as quantitative analyses would miss embedded feelings. As a limitation, a snowball sample, while useful for an emerging pandemic, may diminish generalizability because it prohibits the calculation of a formal response rate.

The diversity of our sample suggests that no single group might be able to dominate a question response and create a particular bias. While there may be a response bias based on those who felt they had time to respond, the high amount of stress and burnout in our sample resonated with findings in other concurrent clinician surveys.

CONCLUSION

Healthcare leaders must provide primary care clinicians with a sense of hope—hope for manageable work conditions, hope for staff who are engaged and supported for long-lasting collegiality, and hope that society will value their work. Hope must be addressed in tangible and meaningful ways to allow strong practices to emerge that are sufficiently resourced for the work of primary care. In brief, healthcare organizations will be more likely to succeed if they can address the current sources of work distress, anxiety, depression, and overwhelm.

The thousands of clinicians who shared their feelings and comments in this study provide a prescription: Match workload to capacity for work; stop asking people to do more with less; advocate for support staff as integral team members; reduce administrative burden; allow sufficient time for visits; create personal leave options for all staff; rebuild clinician–patient relationships and collegiality among HCWs; and provide safe, confidential, and on-demand mental health support. If healthcare organizations pay attention to the responses presented in this study, the future of primary care can be made considerably brighter.