key: cord-0943664-fafqgzot authors: The COVID-19 Healthcare Personnel Study,; DiMaggio, C.; Abramson, D.; Susser, E.; Hoven, C.; Chen, Q.; Andrews, H.; Herman, D.; Kreniske, J.; Ryan, M.; Susser, I.; Thorpe, L.; Li, G. title: The COVID-19 Healthcare Personnel Study (CHPS): Overview, Methods and Preliminary Report date: 2020-10-30 journal: nan DOI: 10.1101/2020.10.29.20222372 sha: 65f2245e9c6c463a5171694df8df2037b5e97746 doc_id: 943664 cord_uid: fafqgzot Introduction. The COVID-19 Healthcare Personnel Study (CHPS) was designed to assess and mitigate adverse short and long-term physical and mental health impacts of the COVID-19 pandemic on New York's health care workforce. Here we report selected baseline results. Methods. Online survey of New York State physicians, nurse practitioners and physician assistants registered with the New York State Department of Health. Survey-weighted descriptive results were analyzed using frequencies, proportions, and means, with 95% confidence intervals. Odds ratios were calculated for association using survey-weighted logistic regression. Results. Approximately 51.5% (95% CI 49.1, 54.0) of the survey-weighted respondents reported having worked directly or in close physical contact with COVID-19 patients. Of those tested for COVID-19, 27.3% (95% CI 22.5, 32.2) were positive. Having symptoms consistent with COVID-19 was associated with reporting a subsequent positive COVID-19 test (OR=14.0, 95% CI 5.7, 34.7). Over half of the respondents, (57.6%) reported a negative impact of the COVID-19 efforts on their mental health. Respondents who indicated that they were redeployed or required to do different functions than usual in response to COVID-19 were more likely to report negative mental health impacts (OR=1.3, 95% CI 1.1, 1.6). Conclusions. At the height of the COVID-19 pandemic in New York State in Spring 2020, more than half of physicians, nurse practitioners and physician assistants included in this study responded to the crisis, often at a cost to their physical and mental health and disruption to their lives. The sudden onslaught of the COVID-19 in 2020 pandemic placed severe demands on US health systems and the health care workforce, especially in New York State (NYS) and New York City (NYC), the American epicenter. Early in the course of the pandemic, hospitals ran the risk of exhausting their supplies of ventilators, ICU beds, and personal protective equipment (PPE); the capacity of the health care workforce to meet the added demand was equally strained. In the late winter and early spring of 2020, NYS and New York City bore a disproportional share of the burden, with approximately half of all confirmed cases in America. NYS instituted a number of strategies to expand hospital capacity and the workforce: the governor mandated that all hospitals increase bed capacity by 50%; specialized hospital facilities were constructed in large convention spaces; efforts to purchase and obtain donated PPE were accelerated, and; volunteer, retired, and student health care professionals were enlisted to supplement the workforce. Hospitals and health systems explored ways of repurposing and expanding their stock of critical equipment. During Spring 2020, a period of extreme system stress, when the nature and context of health care rapidly adapted and changed, the COVID-19 Healthcare Personnel Study (CHPS) was launched to longitudinally assess and mitigate the adverse health impacts of the COVID-19 pandemic on the NYS health care workforce. Between 28 April and 30 June 2020, the height of the pandemic in NYS, CHPS collaborators recruited members of multiple health care professions across NYS to participate. Professions included physicians, nurses and nurse practitioners, medical residents, physician assistants, and ancillary health professionals such as respiratory therapists, paramedics, emergency medical technicians, visiting nurses, home health care workers, and others. Participants were drawn from community-based settings, as well as hospitals and other institutional settings. . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted October 30, 2020. ; https://doi.org/10.1101/2020. 10.29.20222372 doi: medRxiv preprint Our initial goals were to help define the challenges and stressors experienced by healthcare workers (HCW), to assess the impact of the COVID-19 pandemic on their mental, physical, and social health and that of their families, and to evaluate the role of interventions such as just-intime training, counseling and childcare services implemented to reduce adverse health outcomes of the health care workforce and to facilitate their professional response to the pandemic. We present here initial results of CHPS, based on the first wave of responses from physicians, nurse practitioners and physician assistants. We report key characteristics of respondents, their exposures and risk factors, including regional distribution and type of practice, exposure to and experience with treating COVID-19 patients, reported impact of the pandemic on health and practice patterns, and association of treating COVID-19 patients with testing positive for SARS-CoV2. In addition to establishing a baseline for longitudinal assessments of this important cohort this report documents its impact on healthcare professionals during an uncharted period of extreme stress. . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted October 30, 2020. 1 Data were downloaded and read into the R statistical computing system, 2 checked for outliers and cleaned. We used raking procedure to assign a survey weight for each respondent to make the sample representative of the target population of physicians, nurse practitioners and medical assistants in age, gender and geographic location across 10 regions of New York State. 3 4 5 Raking used the population margins of age, race and geographic location to create weights such that the weighted sample distributions of these variables conform to their population distributions. Estimated extrapolations to population-level frequencies and proportions of physicians, nurse practitioners and medical assistants by age, gender and geographic locations were based on documentation from the Center for Health Workforce Studies. 6, 7, 8, and the New York State Department of Education 9 . In the case of physician assistants, population statistics were supplemented with national-level data. 10 . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted October 30, 2020. ; https://doi.org/10.1101/2020. 10.29.20222372 doi: medRxiv preprint Statistical analyses consisted of survey-weighted counts, proportions, means, and 95% confidence intervals using the R "survey" package. 11 Odds ratios for association were calculated using survey-weighted logistic regression models. A one-way ANOVA for the association of an 8-category clinical specialty variables (Primary Care, Pediatrics, Emergency Medicine, Critical Care, Non-Surgical Specialties, Surgery, Behavioral Health, and Other) was conducted on nonweighted data. The protocol was approved by the Institutional Review Boards (IRB) of Columbia University Medical Center, the New York State Psychiatric Institute, the City University of New York, and NYU School of Medicine. . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted October 30, 2020. ; https://doi.org/10.1101/2020.10.29.20222372 doi: medRxiv preprint Of the 139,109 emails sent, 38,090 (27.4%) reached an intended recipient and were opened. Of these, 2,076 physicians, nurse practitioners and physician assistants completed the survey, for a response rate of 5.5%. More female physicians responded (49% female physicians in the study sample vs 35% in the target population). Physicians responding to the survey were also skewed older (32% sample older than 60 vs 23% population). Physicians from upstate regions were represented slightly more than downstate regions. After the raking procedure, the weighted sample more closely approximated the physicians target population in age, gender and geographic distribution. (Table 1 ) Similar adjustments were seen with the Nurse Practitioner and Physician Assistant groups. After raking, the survey-weighted sample equaled a target population of 137,710 (95% CI 134422, 140999) physicians, nurse practitioners and physician assistants licensed in New York State. Our weighted sample had a mean age of 50.4 (95% CI 49.9, 51.0), of whom 46.1% (95% CI 43.7, 48.5) were female. (Table 2) The largest survey-weighted proportion of respondents (41.6%) practiced in New York City, followed by Long Island (19.3%), and the Hudson Valley (11.7%). (Table 3) A large majority of respondents had more than 5 years of practice experience, with nearly 27% (95% CI 29.08, 29.12) in practice for 25 years or more. The largest proportion of respondents (46.9%, 95% CI 46.88, 46.92) reported working in private practice or for a non-academic not for profit institution; 46.5% (95% CI 44.1, 49.0) of respondents practiced primarily in a hospital . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted October 30, 2020. ; https://doi.org/10.1101/2020.10.29.20222372 doi: medRxiv preprint setting. (Table 4 ) The largest proportion of respondents (30% 95% CI 30.18, 30.22) practiced in primary care. (Table 5 ) A survey-weighted 51.5% (95% CI 49.1, 54.0) or an estimated survey-weighted frequency of 69,586 (95% CI 65,572, 73,599) of NYS physicians, nurse practitioners and physician assistants reported having worked directly or in close physical contact with COVID-19 patients. Nearly a quarter of all respondents (24.1%; 95% CI 21.6, 26.5) reported changing their living arrangements because of concern about exposure to COVID-19. An estimated survey-weighted frequency of 1,159 (95% CI 586, 1,733) NYS MD/NP/PA's came out of retirement to work on the COVID-19 response. A survey-weighted one third (32.8%, 95% CI 30.5, 35.1) of respondents indicated that they were required to perform functions different than their usual practice in response to COVID-19. Of these, more than half (51.8%, 95% CI 45.9, 57.8) felt their new work setting placed them at increased risk for contracting COVID-19. A large survey-weighted proportion of respondents (43.3%; 95% CI 40.9, 45.8) reported reluctance to work directly with COVID-19 patients. The primary reasons for reluctance were fear of infecting oneself and fear of infecting others. ( . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted October 30, 2020. There was no statistically significant association between reporting working directly with COVID patients and reporting a positive test (OR=1.1, 95% CI 0.6, 1.9). Nor was there any statistically significant difference in a one-way ANOVA between clinical specialties among those reporting a positive COVID-19 test (p-value = 0.135). . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted October 30, 2020. ; https://doi.org/10.1101/2020.10.29.20222372 doi: medRxiv preprint Well over half of the respondents, (57.6%) reported a negative impact on their mental health as a result of COVID-19 efforts, and 43% felt it had negatively impacted personal relationships. A third of those reporting a negative impact felt it had affected their physical health and/or ability to work. (Table 7) There was no statistically significant association between working directly with COVID patients and negative mental health impacts (OR=0.7, 95% CI 0.4, 1.2). Respondents who indicated that they were redeployed or required to perform different functions in response to COVID-19 were more likely to report negative mental health impacts (OR=1.3, 95% CI 1.1, 1.6) There was a very similar association between reporting a hospital-based practice and reporting mental health symptoms (OR=1.3, 95% CI 1.1, 1.6). Having symptoms consistent with COVID-19 was associated with an adverse effect on mental health (OR=1.7, 95% CI 1.3, 2.1). . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted October 30, 2020. ; https://doi.org/10.1101/2020.10.29.20222372 doi: medRxiv preprint If the experience of New York State HCWs serves as a bellwether for the United States in the months and perhaps years to come, the COVID-19 pandemic will continue to be an unprecedented healthcare and public health challenge. We find that the COVID-19 pandemic increased the risk of infection, adversely impacted mental health and disrupted lives of HCWs. Despite this, over half of all physicians, nurse practitioners and physician assistants in this cohort provided direct patient care, indicating that they rose to the challenge despite such risks. A third changed from their usual practice or specialty, and a considerable number came out of retirement to meet the needs of their fellow New Yorkers. The literature on HCWs responses to previous epidemics is characterized by conflicting emotions of duty and fear. These findings are consistent with results of a study of 83 French HCWs regarding their state of preparedness to treat Ebola Hemorrhagic Fever in 2015, the majority of whom (73%) wanted to be "personally involved" in patient care. 12 Although the large majority (91%) of a group of Saudi Arabian nurses reported an "ethical duty" to respond to an outbreak of MERS-CoV outbreak that occurred in Jeddah, 96% "felt nervous and scared", and 92% had thought of quitting. 13 Interpersonal and mental health issues may prove to be among the most important impacts of COVID-19 on HCWs. The well-publicized suicide of an emergency physician in New York City during the fielding of this survey brought this issue to the fore. 14 Our study supports early reports from China that "a considerable proportion of health care workers (treating COVID-19 patients) reported experiencing symptoms of depression, anxiety, insomnia, and distress", 15 as well as a survey of Saudi HCWs who responded to the 2014 MERS outbreak in which "almost two thirds reported having psychological problems" 16 . Although we did not find an association of direct COVID-19 patient care with symptoms of mental health distress, we did find that redeployment . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted October 30, 2020. ; https://doi.org/10.1101/2020.10.29.20222372 doi: medRxiv preprint to non-familiar duties was associated with adverse mental health outcomes. Training and clinical knowledge may be key factors in addressing HCW concerns about responding to a pandemic. 17 Concern for children and childcare is a frequent issue. One metanalysis concluded that "Respondents living with children or having childcare obligations were one-third less likely to be willing to work compared with those without these obligations." 17 In a series of studies of the 2003 SARS epidemic in Canada, Canadian HCW's who developed SARS chiefly expressed concern about "the effects of quarantine and contagion on family members and friends." 22 It is vitally important to "alleviate the concerns and fears of HCWs and remove potential barriers to working" during a pandemic. 23 A 2009 study addressing this issue concluded that among the most important factors were "being ill, transportation, childcare (and) concern for family." 24 Our study results are consistent with these findings from studies of these earlier outbreaks. Fully two thirds of our respondents who expressed reluctance in treating COVID-19 patients cited fear of infecting themselves or others as the primary reason. There is evidence supporting specific measures to address distress among HCWs responding to pandemics. Among helpful factors reported by HCWs in coping with a MERS-CoV outbreak . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted October 30, 2020. ; https://doi.org/10.1101/2020.10.29.20222372 doi: medRxiv preprint that occurred in Jeddah, Saudi Arabia were a "Positive attitude from colleagues in your department" and financial compensation. 13 The authors of a series of studies of HCW attitudes toward responding to SARS concluded that "Reducing pandemic-related stress may best be accomplished through interventions designed to enhance resilience in psychologically healthy people," by providing psychological first aid, institutional training, support and leadership. 25 The same authors also pointed to the importance of practical and tangible support, 21 as well as the potential utility of computer-based tools to improve confidence and self-efficacy. 26 Our data support a high risk of contracting COVID-19 among HCWs carrying out their professional responsibilities, and describe important and meaningful challenges like a lack of personal protective equipment that contributed to that risk. 27 Our study's s results indicate that accessing a test for SARS-Cov2 at the height of the pandemic in NYS was difficult for clinicians. A large proportion of HCWs who did access a test, tested positive, with an alarming 27% reported test positivity rate. A recent study reports a nearly 12-fold increased risk of infection for HCW compared to the general population, but comparisons and estimates of relative risk based on our data are difficult. During a similar time period the overall NYS positivity rate was approximately 40%, 28 although tests for SARS CoV-2 in NYS were in short supply 29 and often reserved for the sickest patients, likely increasing the general positivity rate. Our study is subject to a number of important limitations. These results apply narrowly to physicians and advanced practice clinicians. The demographics of this group differ from other groups demonstrated to be at increased risk of COVID-19, including Blacks/African Americans who make up a large proportion of persons providing non-medical direct services to COVID-19 patients in NYC, but may not be well represented among physicians and advanced practice clinicians. 30 This is an area in which additional results from the CHPS addressing nursing and non-medical providers can provide insights. . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted October 30, 2020. ; https://doi.org/10.1101/2020.10.29.20222372 doi: medRxiv preprint While response rate was low, it was based on a single-request and is in line with recent studies of physicians. 31 Low response rates are increasingly a feature of modern surveys, and less than optimal response rates are not necessarily indicative of poor survey quality or bias; in some instances lower response rates have been associated with less bias. 32 We utilized statistical procedure to align our survey-adjusted sample with important demographics of the target population, 6, 7, 8, 9, 10 but our sample may be biased. One or more additional factors associated with the domains we measured may have affected the likelihood of responding to the survey, and could have biased the results. The direction of that bias is difficult to discern. Respondents may have been more severely impacted by the pandemic and thus more motivated to respond than a representative population, in which case our results may be overestimated. Alternatively, during a time when much activity in the state, including routine patient care, had been curtailed there may have been more responses from clinicians less impacted than those healthcare workers who were busier providing direct patient care to COVID-19 patients. Our study adds to increasing evidence on the effect of the COVID-19 pandemic on the U. S. healthcare workforce. Our study finds that at the height of the COVID-19 pandemic in New York State in early 2020, the majority of physicians, nurse practitioners and physician assistants responded, often having to change their usual practice, and at a cost of physical and mental health and disruption to their lives. Future research should address the evolution of these impacts to facilitate the ability and willingness of HCWs to respond to the pandemic, expanding the target population to include other workers in the healthcare workforce, understanding the ways that the COVID-19 pandemic is affecting clinical decision-making and practice behavior, and documenting the impact of altered standards of care and triage decisions on vulnerable and socially marginalized groups. . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. We thank Barbara Lang for administrative support, Dr. Jennifer Norton for data cleaning and preparation, and Dr. Howard Zucker and the office of the New York State Commissioner of Health for inviting the NYS Healthcare workforce to participate in this survey. . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted October 30, 2020. ; https://doi.org/10.1101/2020.10.29.20222372 doi: medRxiv preprint . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted October 30, 2020. . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted October 30, 2020. . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted October 30, 2020. ; https://doi.org/10.1101/2020.10.29.20222372 doi: medRxiv preprint . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted October 30, 2020. ; https://doi.org/10.1101/2020.10.29.20222372 doi: medRxiv preprint . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted October 30, 2020. ; https://doi.org/10.1101/2020.10.29.20222372 doi: medRxiv preprint . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted October 30, 2020. ; https://doi.org/10.1101/2020. 10.29.20222372 REDCap R: A language and environment for statistical computing. R Foundation for Statistical Computing survey: analysis of complex survey samples Analysis of complex survey samples Metrics to Guide Reopening The New York Health Workforce Data System A Profile of New York State Nurse Practitioners A Comparative Analysis of New York's Active Nurse Practitioners and Physician Assistants American Academy of Physcian Assistants Analysis of complex survey samples Preparedness of healthcare workers at French Ebola referral centres International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity Healthcare Workers Emotions, Perceived Stressors and Coping Strategies During a MERS-CoV Outbreak Doctor Who Treated Virus Patients Dies by Suicide New York Times Factors Associated With Mental Health Outcomes Among Health Care Workers Exposed to Coronavirus Disease Attitudes and Behaviours of Healthcare Workers in the Kingdom of Saudi Arabia to MERS Coronavirus and Other Emerging Infectious Diseases Healthcare workers willingness to work during an influenza pandemic: a systematic review and meta-analysis. Influenza and Other Respiratory Viruses International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity The copyright holder for this preprint this version posted October 30, 2020. ; https://doi.org/10.1101/2020. 10 . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted October 30, 2020. . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted October 30, 2020. . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted October 30, 2020. ; https://doi.org/10.1101/2020.10.29.20222372 doi: medRxiv preprint