What are the recommended number of clinical pharmacist specialists to bed size of various clinical care areas (ED, Medical ICU, Trauma ICU, NICU, AMS, internal medicine, etc)? What are the recommended number of staff pharmacists per facility bed size?

Comment by InpharmD Researcher

While there is no universal pharmacist-to-patient ratio applicable across all practice settings, much of the literature investigating optimal pharmacist-to-patient ratios has focused on the intensive care unit (ICU) patient population. In the neonatal intensive care unit (NICU) setting, guidelines require that at least 1 pharmacist with experience in neonatal and/or pediatric pharmacology be readily available on-site or via telehealth 24/7. In the broader ICU setting (encompassing both adult and pediatric patients), a general ICU patient-to-pharmacist ratio of ≤20:1 has been recommended, although the optimal ratio will be highly dependent upon the specifics of the patient population in each ICU type. In the antimicrobial stewardship program (ASP) setting, increasing the full-time pharmacist involvement in ASP alongside physicians has been associated with a 1.48-fold increase in the odds of effective interventions, and a range of 1 to 3 pharmacists is recommended based on hospital bed size.

Background

The 2023 Standards for Levels of Neonatal Care II, II, and IV consensus document from the American Academy of Pediatrics outlines a comprehensive framework for stratified neonatal care and resources/staffing recommendations for Level II through IV neonatal intensive care units (NICUs). The document meticulously delineates the required components and standards for each care level, ranging from Special Care Nursery (SCN, Level II) to complex subspecialty services including surgery (Level IV). All three levels require at least 1 registered pharmacist with experience in neonatal and/or pediatric pharmacology who is available for consultation on-site or by telehealth/telephone 24 hours a day and 7 days per week, has completed continuing education specific to pediatric and neonatal pharmacology, and participates in multidisciplinary care. Level II SCN status permits participation "as needed," while Levels III and IV NICUs require active participation in patient care rounds. All levels must also have pharmacy policies and procedures to address drug shortages, verify appropriate allocation of medications to the corresponding unit (Level II SCN, Level III NICU, or Level IV NICU), and confirm neonatal competency for pharmacy staff who support or prepare medications for neonatal patients. Level III and IV NICUs have a further requirement that neonatal-appropriate total parenteral nutrition (TPN) consulting be available 24 hours a day and 7 days per week, supported by a written policy and procedure governing its proper preparation and delivery, a standard not specified for Level II SCN facilities. [1]

A 2025 interprofessional, multidisciplinary, multiorganizational consensus statement developed and validated standards and guidance on best practices for integrating critical care pharmacists into patient care teams using a modified Delphi approach involving a 21-member interprofessional expert panel of critical care pharmacists. A total of 10 consensus recommendation statements were provided, with corresponding guidance on implementation approaches, outcomes measurement, and future advocacy and research efforts. The final consensus statements acknowledged the importance of adequate staffing to ensure that each critical care patient has a dedicated pharmacist, with a recommended general ICU patient-to-pharmacist ratio ≤20:1, recognizing that a universal patient-to-pharmacist ratio applicable across all ICUs has not yet been identified. The panel recommended use of flexible pharmacist positions to meet this staffing need. At minimum, the panel indicated that ICU patients need critical care pharmacist services for one shift daily, but the goal would be continuous coverage 24 hours a day and 7 days a week, on the level of other critical care personnel who are considered essential to patient care (e.g., intensivists and critical care nurses). [2]

Literature Review

A search of the published medical literature revealed 6 studies investigating the researchable question:

Level of evidence

C - Multiple studies with limitations or conflicting results Read more→

Please see Tables 1-6 for your response.

Optimal pharmacist-to-patient ratio in intensive care unit (ICU) setting as outlined by various organizations
Source	Pharmacist to patient ratio
The Hospital of the University of Pennsylvania	1:8-1:14
United Kingdom Clinical Pharmacy Association	1:10-1:20
Joint Faculty of Intensive Care Medicine of Ireland	1:10-1:20
Allied Health Professionals (AHP) and Healthcare Scientists (HCS)	1:10-1:20
The Society of Hospital Pharmacists of Australia (SHPA) Practice Standards	1:12
Horn et al	1:20-1:30

References:
[1] [1] Newsome AS, Smith SE, Jones TW, Taylor A, Van Berkel MA, Rabinovich M. A survey of critical care pharmacists to patient ratios and practice characteristics in intensive care units. J Am Coll Clin Pharm. 2020;3:68-74. https://doi.org/10.1002/jac5.1163

Interrelationships among Mortality Rates, Drug Costs, Total Cost of Care, and Length of Stay in United States Hospitals: Summary and Recommendations for Clinical Pharmacy Services and Staffing
Design	Severity of illness-adjusted multiple regression analysis N= 934-1029 hospitals
Objective	To evaluate interrelationships and associations among mortality rates, drug costs, total cost of care, and length of stay in United States hospitals, and to explore the relationships between these variables and the presence of clinical pharmacy services and pharmacy staffing
Study Groups	All hospitals (n= 934-1029)
Inclusion Criteria	Hospitals for which variables could be matched from the 1992 American Hospital Association's Abridged Guide to the Health Care Field, the 1992 National Clinical Pharmacy Services database, and 1992 Health Care Finance Administration mortality data
Exclusion Criteria	Not specified
Methods	A database was constructed using data from the 1992 American Hospital Association's Abridged Guide to the Health Care Field, the 1992 National Clinical Pharmacy Services database, and 1992 Health Care Finance Administration mortality data. Severity of illness-adjusted multiple regression analysis was employed to determine relationships and associations
Duration	Data from 1992 National Clinical Pharmacy Services database
Outcome Measures	Primary: Mortality rates, drug costs, total cost of care, length of stay Secondary: Association of clinical pharmacy services and staffing with these outcomes
Baseline Characteristics		All hospitals (n= 934-1029)
	Clinical pharmacists/occupied bed, 10th percentile	0.34/100
	Clinical pharmacists/occupied bed, 90th percentile	3.23/100
Results		Slope	p-value
	Mortality rate reduction	-43%	<0.001
	Drug protocol management	-1.30	0.008
	Pharmacist participation on medical rounds	-1.71	<0.001
	Clinical pharmacists/occupied bed	-26.59	<0.001
Adverse Events	Not applicable
Study Author Conclusions	The presence of clinical pharmacy services and increased clinical pharmacist staffing levels are associated with positive outcomes, including reduced mortality rates, drug costs, and length of stay. Enhancing clinical pharmacy services and staffing could significantly improve hospital outcomes.
Critique	The study effectively highlights the positive impact of clinical pharmacy services on hospital outcomes. However, it relies on data from 1992, which may not reflect current practices or advancements in healthcare. The study's retrospective nature and reliance on existing databases may limit the ability to control for all confounding variables.

References:
[1] [1] Bond CA, Raehl CL, Franke T. Interrelationships among mortality rates, drug costs, total cost of care, and length of stay in United States hospitals: summary and recommendations for clinical pharmacy services and staffing. Pharmacotherapy. 2001;21(2):129-141. doi:10.1592/phco.21.2.129.34105

ICU Patient-to-Pharmacist Ratios: A Prospective, Multicenter Time-Motion Study
Design	Prospective, multicenter, time-motion study across adult intensive care unit (ICU), pediatric ICU (PICU), and neonatal ICU (NICU) settings N=128 ICU pharmacists
Objective	To evaluate the relationship among ICU patient-to-pharmacist ratio, perceived quality of patient care, and pharmacist burnout
Study Groups	ICU pharmacists (N= 128)
Inclusion Criteria	ICU clinical pharmacists
Exclusion Criteria	None specified
Methods	Pharmacists recorded time associated with direct/indirect patient care activities, Maslach Burnout Inventory survey scores, and perceived quality of patient care on a 5-point Likert scale. Scores of 4 (moderately high) and 5 (high) were categorized as “higher perceived quality of care,” while scores of 1 (low), 2 (moderately low), and 3 (moderate) were categorized as “lower perceived quality of care.” Total pharmacist time on direct and indirect patient care activities were measured per shift
Duration	May 1, 2022, to February 28, 2023
Outcome Measures	Primary: Relationship between ICU patient-to-pharmacist ratio and perceived quality of patient care Secondary: Pharmacist burnout frequency, time spent on direct/indirect patient care activities
Baseline Characteristics		ICU Pharmacists (N= 128)
	Direct patient care time, hours per shift	5.9 ± 1.9
	Indirect patient care time, hours per shift	3.3 ± 1.7
	Assigned ICU patients per shift	19.3 ± 12.9
	Assigned total patients per shift	23.4 ± 17.9
Results		Result	p-value
	Burnout frequency	38.1%	--
	Burnout associated with total patients assigned	OR 1.03 (95% CI 1.01–1.05)	--
	Burnout associated with overtime worked	OR 1.18 (95% CI 1.03–1.35)	--
	Higher vs lower quality of care with smaller ICU patient assignments	17.0 ± 7.4 vs 30.6 ± 23.8	<0.001
	Additional direct patient care time per patient predictive of high quality care	OR 1.53 (95% CI 1.03–2.05)	--
	Burnout inversely associated with high quality care	OR 0.59 (95% CI 0.36–0.96)	--
Adverse Events	Not applicable
Study Author Conclusions	The ICU patient-to-pharmacist ratio range between 16:1 and 19:1 may optimize quality of care and burnout risk at large academic institutions.
Critique	The study provides valuable insights into the optimal ICU patient-to-pharmacist ratios for balancing quality of care and pharmacist burnout. However, the study's observational nature and reliance on self-reported data may introduce reporting bias. Additionally, the study does not account for variability in ICU settings and patient acuity, which may affect generalizability and pharmacist staffing needs.

References:
[1] [1] Buckley MS, Smith SE, Birriel B, et al. ICU Patient-to-Pharmacist Ratios: A Prospective, Multicenter Time-Motion Study. Crit Care Med. 2025;53(4):e863-e873. doi:10.1097/CCM.0000000000006605

ASHP National Survey of Pharmacy Practice in Hospital Settings: Clinical Services and Workforce—2024
Design	Prospective, survey-based study N=250 hospitals (16.7% survey response rate)
Objective	To present the results of the 2024 ASHP National Survey of Pharmacy Practice in Hospital Settings
Study Groups	All respondents (pharmacy directors of N=250 hospitals of various staffed bed sizes represented)
Inclusion Criteria	A sampling frame of 4,926 general and children's medical-surgical hospitals in the United States was created from the IQVIA database. Respondents from the 6 prior issuings of this survey were also included. The final sampling frame had 1,497 hospitals, including 353 hospitals with <50 beds, 138 hospitals with ≥600 staffed beds, and ~200 hospitals in each of the other hospital size categories.
Exclusion Criteria	Specialty, federal, and Veterans Health Administration hospitals
Methods	The questionnaire was developed by evaluating questions from previous surveys that pertained to topics of interest in this issue of the survey for clarity and response. A mixed-mode survey method of contact by email and mail was used. Survey completion was done online using Qualtrics. Pharmacy directors in the sample were contacted up to 7 times during the survey period. During data analysis, weights were assigned to respondents to adjust their contribution to the population estimate. The weights were 31.59 for hospitals with fewer than 50 staffed beds, 25.11 for hospitals with 50 to 99 beds, 28.42 for hospitals with 100 to 199 beds, 18.09 for hospitals with 200 to 299 beds, 11.97 for hospitals with 300 to 399 beds, 10.03 for hospitals with 400 to 599 beds, and 4.18 for hospitals with 600 or more staffed beds.
Duration	September 2024 through December 20, 2024
Outcome Measures	Hospital characteristics, extent of pharmacist coverage across clinical service areas (defined as dedicated staffing for at least 8 hours per day for at least 5 days per week), extent of tracking pharmacy-specific and hospital-specific outcome metrics
Baseline Characteristics		Respondents (N=250)
	Staffed beds, n (%) <50 50-99 100-199 200-299 300-399 400-599 ≥600	56 (15.9) 27 (13.4) 36 (17.5) 34 (17.0) 31 (15.7) 33 (16.4) 33 (23.9)
	Region West Midwest South Northeast	40 (16.9) 93 (18.9) 73 (14.0) 44 (17.7)
	Ownership For-profit Nonprofit	9 (4.4) 241 (18.7)

Results	Endpoint	Respondents (N=250)
	Areas to which pharmacists are assigned to provided drug therapy management services, n (%) Inpatient-medical-surgical Critical care Oncology Cardiology Infectious diseases/antimicrobial stewardship Emergency department Neonatal Solid organ transplant Pediatrics Psychiatry/behavioral health Obstetrics/gynecology Operating room/perioperative areas	183 (73.3) 171 (68.5) 142 (56.9) 121 (48.5) 120 (48.1) 116 (46.5) 87 (35.0) 81 (32.3) 80 (32.2) 78 (31.2) 74 (29.8) 65 (26.1)
	Pharmacy-specific outcome metrics by staffed beds per hospital, (%) All (n=242) <50 50-99 100-199 200-299 300-399 400-599 ≥600	23.5 16.4 37.5 20.6 21.2 29.0 39.4 31.3
	Pharmacy and hospital outcome metrics by staffed beds per hospital, (%) All (n=242) <50 50-99 100-199 200-299 300-399 400-599 ≥600	23.6 7.3 12.5 26.5 42.4 54.8 42.4 46.9
	No tracking of outcome metrics by staffed beds per hospital, (%) All (n=242) <50 50-99 100-199 200-299 300-399 400-599 ≥600	52.9 76.4 50.0 52.9 36.4 16.1 18.2 21.9

Adverse Events	Not applicable
Study Author Conclusions	Hospital pharmacy departments are facing many challenges, including worsening shortages of pharmacists and pharmacy technicians, drug shortages, reimbursement and formulary concerns, and regulatory compliance. Despite these challenges, inpatient and ambulatory care clinical pharmacy services continue to expand across the country, with most hospitals routinely providing care to a majority of patients in over three-quarters of hospitals. While progress toward other PAI 2030 focused initiatives has been mixed over the past 5 years, responsibilities of pharmacy technicians are trending toward advanced roles.
Critique	This survey-based study, in its 6th iteration in 2024, showed that pharmacists routinely provide clinical pharmacy services to a majority of patients in over 75% of hospitals. The results also showed the continuing expansion of ambulatory care pharmacy services. The major challenge in adequately staffing and/or expanding the scope of clinical practice areas provided in U.S. hospitals was identified as pharmacy staffing shortages.

References:
[1] [1] Pedersen CA, Naseman RW, Schneider PJ, Ganio MC, Scheckelhoff DJ. ASHP National Survey of Pharmacy Practice in Hospital Settings: Clinical Services and Workforce-2024. American Journal of Health-System Pharmacy. 2025;82(18):979-1005. doi:10.1093/ajhp/zxaf150

Essential Resources and Strategies for Antibiotic Stewardship Programs in the Acute Care Setting
Design	Cross-sectional survey N=244
Objective	To describe the resources and strategies necessary for effective antibiotic stewardship programs (ASPs) in acute care settings and to propose staffing structures based on survey results
Study Groups	All respondents (N=244)
Inclusion Criteria	Membership in the Infectious Diseases Society of America (IDSA), Society for Healthcare Epidemiology of America (SHEA), or Pediatric Infectious Diseases Society (PIDS) with patient care, epidemiology, or administration as a primary responsibility
Exclusion Criteria	Degrees other than MD, MBBS, DO, or PharmD; trainee status; members outside the US; employment in industry, public health, or other business; no affiliated facility listed
Methods	A 73-question electronic survey was distributed to members of IDSA, SHEA, and PIDS. Responses were analyzed using descriptive statistics and logistic regression models to evaluate the relationship between staffing levels and program effectiveness. Only 1 response per hospital was accepted. No incentives were provided for participation. Respondents working in stewardship at >1 hospital or as part of a health-system were instructed to answer the questions for the hospital where they spent the majority of their time. Definitions for cascade reporting of antibiotics, computerized decision support, antibiotic time-out, formulary restriction/priorauthorization, and prospective audit and feedback (PAF) are taken from the IDSA/SHEA guidelines for implementation of antimicrobial stewardship and the CDC Core Elements of Hospital Antibiotic Stewardship Programs.
Duration	June 29, 2016 through August 3, 2016
Outcome Measures	Primary: ASP effectiveness (cost savings, decreased antibiotic utilization, decreased multi-drug resistant organism, MDRO, rates) Secondary: Relationship between full-time equivalent (FTE) support and ASP effectiveness
Baseline Characteristics		Respondents (N=244)
	Location - Northeast	56 (23.0%)
	Location - Midwest	63 (25.8%)
	Location - South	62 (25.4%)
	Location - West	60 (24.6%)
	No. of beds at primary hospital - <100	15 (6.2%)
	No. of beds at primary hospital - 100–300	91 (37.3%)
	No. of beds at primary hospital - 301–500	82 (33.6%)
	No. of beds at primary hospital - 501–1000	45 (18.4%)
	No. of beds at primary hospital - >1000	11 (4.5%)
	Primary hospital’s teaching status - Major academic medical center	71 (29.1%)
	Primary hospital’s teaching status - Academic affiliation	114 (46.7%)
	Primary hospital’s teaching status - Nonteaching	54 (22.1%)
	Primary hospital’s teaching status - Other	5 (2.1%)
	Hospital-level stewardship	93 (48.7%)
	Emergency department antibiotic stewardship	135 (55.3%)
	ASP provides telestewardship	30 (12.3%)
Results	Endpoint	Univariate OR (95% CI)	Primary aOR (95% CI)	Sensitivity Analysis aOR (95% CI)
	Combined PharmD and MD FTE, 0.50 increase	1.60 (1.17–2.20)	1.36 (0.98–1.90)	1.30 (0.94–1.81)
	Antibiotic time-out	1.59 (0.71–3.56)	1.58 (0.68–3.67)	1.60 (0.68–3.77)
	Cascade reporting	0.82 (0.40–1.66)	0.68 (0.32–1.46)	0.70 (0.33–1.51)
	Restricted formulary with prior authorization	1.01 (0.41–2.48)	0.81 (0.31–2.14)	0.80 (0.29–2.16)
	Institutional guidelines	1.70 (0.80–3.59)	1.35 (0.59–3.07)	1.32 (0.58–3.02)
	Prospective audit and feedback	4.88 (2.21–10.79)	3.92 (1.66–9.30)	3.82 (1.60–9.13)
	ASP technology add-on	2.57 (1.25–5.28)	---	1.97 (0.91–4.22)
	Minimal FTE support, as PharmD and MD ratio, recommended by hospital bed size: 1.4 (100-300 beds, 1 PharmD and 0.4 MD) 1.6 (301-500 beds, 1.2 PharmD and 0.4 MD) 2.6 (501-1000 beds, 2.0 PharmD and 0.6 MD) 4.0 (>1000 beds, 3.0 PharmD and 1.0 MD)
Adverse Events	Not applicable
Study Author Conclusions	Prospective audit and feedback should be the cornerstone of ASPs, with both physician leadership and pharmacists with expertise in stewardship being crucial for success. An FTE-to-bed ratio is proposed to guide resource discussions with executive leadership.
InpharmD Researcher Critique	The study provides valuable insights into the staffing and resources needed for effective ASPs, but the low response rate and reliance on self-reported effectiveness may limit the generalizability of the findings. Additionally, the survey's focus on US-based programs may not fully capture global practices.

References:
[1] [1] Doernberg SB, Abbo LM, Burdette SD, et al. Essential Resources and Strategies for Antibiotic Stewardship Programs in the Acute Care Setting. Clin Infect Dis. 2018;67(8):1168-1174. doi:10.1093/cid/ciy255

Characterizing Critical Care Pharmacy Services Across the United States
Design	Prospective, survey-based study across 1,220 U.S. hospitals; work product of the Society of Critical Care Medicine, Clinical Pharmacy and Pharmacology section's Practice Advancement Committee N=401 responses (35.4% response rate) representing 493 intensive care units (ICUs)
Objective	To survey assessed services and activities of ICU pharmacists
Study Groups	All respondents (N=401)
Inclusion Criteria	Pharmacy Director, Clinical Coordinator, or Residency Director/Coordinator at institutions identified as having acute care services in the 2018 American Hospital Association database, who then identified the corresponding ICU pharmacists for each ICU within their facility to complete the survey
Exclusion Criteria	Not specified
Methods	Questions were designed to address specifically the recommendations for pharmacy services and activities identified in prior opinion article. Questions categorized responses by hospital/ICU characteristics and resources (14 questions); pharmacist activities across practice, education, scholarship, and administration (10 questions); and allocation of effort, funding, and employment benefits (three questions). A draft survey was developed using Qualtrics and piloted by a panel of 8 ICU pharmacists, including external reviewers. The questions were modified and redistributed to the panel for further feedback until the 27-question survey was finalized. The survey instrument was distributed via email to the Pharmacy Director, Clinical Coordinator, or Residency Director/Coordinator at institutions identified as having acute care services in the 2018 American Hospital Association database. A link to the survey was embedded in the e-mail. The initial recipient was asked to complete questions pertaining to institutional and general ICU pharmacy services and provide emails for the pharmacists with the greatest involvement in each ICU of the in stitution, and Qualtrics then automatically generated another email to each new potential respondent. Only 1 survey was completed per ICU, but an individual pharmacist providing services to multiple ICUs may have completed surveys for each ICU if the initial survey recipient entered the pharmacist’s e-mail address for each ICU. Surveys for each section were excluded in the analyses if less than 25% of questions for the specific section were answered. Therefore, completed questionnaires may only represent institution-specific responses or ICU specific responses although it was expected that an ICU-specific survey would be completed for each insti tution-specific survey.
Duration	September 2018 through November 2018 (survey distributed three times approximately 4 weeks apart)
Outcome Measures	Description of hospital/ICU characteristics and resources; pharmacist activites across practice, education, scholarship, and administration; and allocation of effort, funding, and employment benefits
Baseline Characteristics		Respondents (N=401)
	Financial structure, % Not-for-profit nongovernment Nonfederal government For-profit nongovernment Federal government	75.3 10.7 9.5 4.7
	Types of institution, % Community-teaching hospital Community nonteaching hospital University/academic hospital	39.9 34.4 25.7
	Region, % East Midwest South Atlantic Mid-Atlantic West Midwest Mountain West New England South Central Pacific Southeast	19.2 15 15 12.7 10.5 7.5 7 6.5 6.5
	Median daily inpatient census (range)	295 (173 to 500)
	ICU demographics Median total critical care beds (range) Median number critical care units (range)	41 (21 to 80) 3 (1 to 5.8)
	Median inpatient pharmacist full-time equivalents (FTEs) (range) Overall Critical care	30 (18 to 53.8) 2.8 (1 to 6)
	ICU type, n (%) Medical Surgical Mixed Cardiovascular Neurosurgical Cardiothoracic Trauma Pediatric Burn Neonatal Other	126 (25.6) 103 (20.9) 84 (17) 48 (9.7) 34 (6.9) 30 (6.1) 24 (4.9) 18 (3.7) 11 (2.2) 6 (1.2) 9 (1.8)
	Median daily ICU census (range)	12 (6 to 20)
	Median pharmacy FTEs for each ICU (range)	1 (1 to 1.5)
	ICU physician staffing model, % Closed Open Mixed	54.6 26.2 19.3
Results	Direct clinical ICU pharmacy services were available in 70.8% of ICUs, with pharmacists attending rounds a median 5 days/week (range 4 to 5) and with a median patient to-pharmacist ratio of 17 (range 12 to 26). Services were less likely to occur overnight or on weekends, with rare use employment of telemedicine services. Dependent prescriptive authority (e.g., per protocol or via practice agreements) was available to 51.1% of pharmacists and independent prescriptive authority was provided by 13.4% of pharmacists. Median % time of a typical ICU pharmacist work week per domain: Direct ICU patient care: 50% (range 40% to 60%) Teaching: 10% (range 8% to 16%) Order processing: 8% (range 5% to 18%) Direct non-ICU patient care: 5% (range 0% to 20%) Administration: 5% (range 2% to 10%) Scholarship: 5% (range 0% to 10%) Drug distribution: 0% (range 0% to 5%)
Adverse Events	Not applicable
Study Author Conclusions	ICU pharmacists have diverse and versatile responsi bilities and provide several key clinical and nonclinical services. Initiatives to increase the availability of services are warranted.
Critique	The researchers described the validation and pilot testing process employed and also provided the full survey instrument, which enhances this study's external validity. Despite inclusion of various sizes of hospitals with differing ICU bed sizes, the results showed that generally facilities with larger ICU capacities responded to the survey, and so the results may not be fully generalizable to the average ICU practice in the U.S. Additionally, since only 1 pharmacist per ICU completed the survey, it was more likely that hospital employees responded vs affiliates or contract employees.

References:
[1] [1] MacLaren R, Roberts RJ, Dzierba AL, Buckley M, Lat I, Lam SW. Characterizing Critical Care Pharmacy Services Across the United States. Crit Care Explor. 2021;3(1):e0323. Published 2021 Jan 8. doi:10.1097/CCE.0000000000000323