Hospital Variation in Response to CDC Sepsis Program Core Elements Assessment Tool

INTRODUCTION

Sepsis, defined as life-threatening organ dysfunction caused by dysregulated host response to infection, is the leading cause of hospital mortality in the USA.^1,2 Prompt, accurate identification and early management of sepsis are key to improved outcomes.³ Hospital sepsis programs provide a structured organizational approach to coordinate and monitor multidisciplinary processes implemented by hospitals to achieve optimal sepsis care. Recently, the Centers for Disease Control and Prevention (CDC) released Hospital Sepsis Program Core Elements to assist hospitals with developing effective sepsis programs.⁴ The Sepsis Program Core Elements (SPCE) complement existing sepsis guidelines and facilitate implementation of best practices. The SPCE Assessment Tool is a companion survey to document program infrastructure. However, there is limited evidence describing how hospitals should use the SPCE assessment to monitor and improve their sepsis program. As part of a larger study to understand contextual factors that are causally linked to optimal sepsis diagnosis, we assessed survey-reported SPCE across diverse hospitals and examined response variation between and within hospitals. We compared three strategies to aggregate disparate survey responses within hospitals.

METHODS

We conducted a survey study at 20 hospitals across four states (AL, GA, MI, NC) between March and August 2024. We recruited health professionals (e.g., physicians, nurses, clinical leaders) directly involved in sepsis management at each hospital. Individuals who agreed to participate completed the CDC SPCE Assessment Tool⁴ electronically via REDCap. We analyzed the frequency and proportion of affirmative responses to 52 questions across seven domains (Hospital Leadership Commitment, Accountability, Multi-Professional Expertise, Action, Tracking, Reporting, Education). We fit analysis-of-variance models to assess between- and within-hospital response variation and used chi-square tests to examine response differences by role. We tested three strategies to aggregate multiple within-hospital responses into a summary classification of each SPCE present (Method 1: ‘Yes’ if all responses=Yes; Method 2: ‘Yes’ if ≥ 0.70 responses=Yes; Method 3: ‘Yes’ if any response=Yes). We used chi-square tests to compare differences between classification results obtained by each method. The study was approved by University of Michigan Institutional Review Board (IRB#00241324).

RESULTS

Eighty-five respondents from 19 hospitals completed the SPCE Assessment Tool (overall response rate= 85/206, 41%; median responses per hospital= 4, IQR= 3–5). Respondents included attending physicians (n= 37), bedside nurses (n= 14), hospital administrators (Chief Medical or Nursing Officers, n= 8), nurse managers (n= 6), pharmacists or antimicrobial stewardship representatives (n= 6), quality improvement specialists (n= 5), and sepsis program coordinators (n= 4). Affirmative responses indicating the presence of SPCE ranged from 38% (Q46 [Reporting Domain]) to 94% (Q38 [Tracking Domain]; Table 1). Individual question responses frequently varied among respondents from the same hospital, contributing to similar within- and between-hospital variation observed for 39 of 52 SPCE. Statistically significant between-hospital differences were observed for 13 SPCE, including elements from six of seven domains. The proportion of affirmative responses differed by participant role type (range: 48% [nurse managers] to 84% [bedside nurses]; p< 0.01). Three methods for aggregating multiple responses within hospitals yielded statistically different hospital-level classifications for the presence of 46 of 52 SPCE (Table 2).

Table 1.

Overall Survey Response and Source of Hospital-Level Variation

Sepsis Program Core Element Assessment Tool question	Response n			Source of variation1
	Yes	No	.	Mean (SD)	Between-hospital	Within-hospital	F	p
Hospital Leadership Commitment
Q1: Sepsis program leaders given sufficient time to manage hospital sepsis program	58	25	2	0.70 (0.46)	0.40	0.16	2.51	< 0.01
Q2: Sepsis program given sufficient resources like analytics and IT support to operate effectively	49	34	2	0.59 (0.49)	0.50	0.17	2.91	< 0.01
Q3: Staff from key clinical groups/support depts have sufficient time to contribute to sepsis activities	51	30	4	0.63 (0.49)	0.25	0.20	1.21	0.28
Q4: Our hospital has senior leader (e.g., CMO, CNO) who serves as Executive Sponsor for sepsis program	58	24	3	0.71 (0.46)	0.29	0.19	1.58	0.09
Q5: Sepsis identified as hospital priority by hospital leadership and priority communicated to staff	70	15	0	0.82 (0.38)	0.21	0.13	1.61	0.08
Q6: Hospital leadership communicates to staff and patients how hospital is addressing sepsis	50	34	1	0.60 (0.49)	0.41	0.20	2.06	0.02
Q7: Hospital leaders have regular meetings with sepsis program leaders to assess resources needed to accomplish goals for sepsis activities and outcomes	58	25	2	0.70 (0.46)	0.25	0.20	1.21	0.28
Q8: Hospital sepsis program activities are integrated into other QI and patient safety efforts	65	18	2	0.78 (0.42)	0.21	0.16	1.28	0.23
Q9: Hospital staff performance incentives are tied to achievement of targets for sepsis care or outcomes	34	47	4	0.42 (0.50)	0.31	0.23	1.33	0.20
Q10: Sepsis program-related duties are included in job description or performance reviews for hospital sepsis program leaders and key support staff	37	43	5	0.46 (0.50)	0.36	0.22	1.67	0.07
Q11: Hospital leadership supports external training/education for sepsis program leaders, key support staff	65	15	5	0.81 (0.39)	0.17	0.15	1.11	0.36
Q12: Hospital leadership supports internal training and education on sepsis for hospital staff and trainees	69	13	3	0.84 (0.37)	0.11	0.14	0.79	0.70
Q13: Hospital leadership supports participation in regional, national, and international sepsis QI collaboratives and initiatives	64	19	2	0.77 (0.42)	0.15	0.19	0.83	0.66
Accountability
Q14: Hospital has program or committee charged with monitoring and improving sepsis care and outcomes	70	14	1	0.83 (0.37)	0.21	0.12	1.76	0.05
Q15: Hospital has 1–2 leaders (physician, nurse) responsible for sepsis program or committee	62	22	1	0.74 (0.44)	0.30	0.17	1.84	0.04
Q16: Hospital sets ambitious-but achievable-goals for improving sepsis care and patient outcomes that are informed by review of hospital practices, sepsis outcomes, and clinical guidelines	63	18	4	0.78 (0.42)	0.19	0.17	1.11	0.37
Q17: Hospital assesses progress towards sepsis goals at regular intervals and updates goals periodically (e.g., annually) to promote continual improvement	64	18	3	0.78 (0.42)	0.26	0.15	1.75	0.05
Q18: Hospital has 1 physician and 1 nurse lead or champion to ensure physician and nursing engagement	55	27	3	0.67 (0.47)	0.28	0.21	1.35	0.19
Q19: Sepsis program activities and outcomes included in annual performance reviews for program leaders	42	37	6	0.53 (0.50)	0.26	0.25	1.02	0.45
Q20: Hospital has unit-level physician and nurse champions for sepsis activities	36	44	5	0.45 (0.50)	0.15	0.28	0.52	0.94
Q21: Sepsis program activities and outcomes reported to senior hospital leadership or board of directors	67	13	5	0.84 (0.37)	0.18	0.13	1.45	0.14
Multi-Professional Expertise
Q22: Hospital has sepsis coordinator, who oversees day-to-day implementation of sepsis program activities	46	36	3	0.56 (0.50)	0.38	0.21	1.79	0.05
Q23: Clinicians and leaders from ED, IP wards, and ICUs are fully engaged in sepsis program activities	56	26	3	0.68 (0.47)	0.23	0.22	1.05	0.42
Q24: Hospital sepsis program includes diverse multi-disciplinary representation	53	29	3	0.65 (0.48)	0.28	0.22	1.27	0.24
Q25: Hospital sepsis program has ongoing support from individuals with expertise and formal training in data management and analytics, IT, and QI and patient safety	60	20	5	0.75 (0.44)	0.19	0.19	1.01	0.46
Q26: Hospital sepsis program has at least ad hoc involvement of case management, microbiology, lab medicine, phlebotomy, outpatient clinicians, hospital epidemiology, infection prevention, patients, families	53	29	3	0.65 (0.48)	0.22	0.23	0.93	0.54
Action
Q27: Hospital implemented standard process to screen for sepsis on presentation, through hospitalization	62	20	3	0.75 (0.43)	0.26	0.17	1.58	0.09
Q28: Hospital has hospital guideline or standardized care pathway for management of sepsis	67	18	0	0.79 (0.41)	0.28	0.14	2.00	0.02
Q29: Hospital has order sets for management of sepsis tailored to patient populations served	73	11	1	0.87 (0.34)	0.11	0.12	0.90	0.58
Q30: Hospital has structures and processes in place to facilitate prompt delivery of antimicrobials	74	10	1	0.88 (0.33)	0.11	0.10	1.10	0.38
Q31: Hospital has structures and processes in place to support effective hand-offs of patients with sepsis	53	28	4	0.65 (0.48)	0.17	0.25	0.71	0.79
Q32: Hospital rapid response team is trained in sepsis recognition and care	63	18	4	0.78 (0.42)	0.24	0.16	1.50	0.12
Q33: Hospital has ‘code sepsis’ protocol for facilitating prompt recognition and team-based care of sepsis	55	28	2	0.66 (0.48)	0.37	0.19	1.99	0.02
Q34: Hospital completes peri-discharge evaluations of patients after sepsis to screen for new or worsening limitations, cognitive impairment, posttraumatic stress disorder/anxiety symptoms, and chronic conditions	35	41	9	0.46 (0.50)	0.37	0.21	1.73	0.06
Q35: Hospital has post-discharge care coordination and anticipatory guidance to optimize sepsis recovery	43	31	11	0.58 (0.50)	0.28	0.24	1.19	0.30
Q36: Hospital has prevention of HAI, HO sepsis that follow facility infection prevention recommendations	76	7	2	0.92 (0.28)	0.11	0.07	1.54	0.11
Tracking
Q37: Hospital monitors hospital sepsis epidemiology, such as number of hospitalizations	63	18	4	0.78 (0.42)	0.15	0.18	0.84	0.65
Q38: Hospital monitors hospital sepsis management, such as time to antibiotic delivery	77	5	3	0.94 (0.24)	0.04	0.06	0.63	0.86
Q39: Hospital monitors sepsis outcomes, such as in-hospital mortality and length of hospitalization	74	9	2	0.89 (0.31)	0.08	0.10	0.76	0.74
Q40: Hospital assesses use, usability, and impact of sepsis tools to inform their ongoing improvement	60	20	5	0.75 (0.44)	0.30	0.16	1.89	0.03
Q41: Hospital monitors progress towards achieving hospital goals for sepsis management and/or outcomes	73	8	4	0.90 (0.30)	0.11	0.08	1.29	0.23
Q42: Hospital completes near real-time chart reviews for the purpose of clinician feedback and education	53	28	4	0.65 (0.48)	0.40	0.18	2.24	0.01
Q43: Hospital completes chart reviews regularly for root cause analysis and process improvement	69	13	3	0.84 (0.37)	0.15	0.13	1.19	0.29
Reporting
Q44: Hospital reports sepsis treatment/outcome data to nursing, physician, unit-based, hospital leadership	65	18	2	0.78 (0.41)	0.16	0.18	0.90	0.58
Q45: Hospital provides feedback to individual clinicians regarding the care of recent patients with sepsis	58	24	3	0.71 (0.46)	0.31	0.18	1.74	0.06
Q46: Hospital has and maintains a live dashboard to report sepsis treatment and outcomes in real-time	31	51	3	0.38 (0.49)	0.33	0.21	1.53	0.11
Education
Q47: Hospital provides sepsis specific training and education in the hiring or on-boarding process	61	22	2	0.73 (0.44)	0.21	0.19	1.05	0.42
Q48: Hospital provides annual sepsis education to clinical staff	50	30	5	0.63 (0.49)	0.41	0.19	2.18	0.01
Q49: Hospital provides written and verbal sepsis education to patients, families/caregivers before discharge	50	26	9	0.66 (0.48)	0.48	0.18	2.66	< 0.01
Q50: Hospital posts information on recognition of sepsis in prominent areas for patient-facing staff	46	34	5	0.58 (0.50)	0.29	0.21	1.41	0.16
Q51: Hospital holds lectures (e.g., grand rounds) or an annual meeting focused on sepsis	34	44	7	0.44 (0.50)	0.30	0.23	1.29	0.23
Q52: Hospital includes sepsis recognition and treatment in annual nursing competencies	48	25	12	0.66 (0.48)	0.24	0.23	1.05	0.42

Abbreviations: IT, information technology; CMO, Chief Medical Officer; CNO, Chief Nursing Officer; QI, quality improvement; ED, emergency department; IP, inpatient; ICU, intensive care unit; HAI, healthcare-associated infection; HO, hospital-onset

¹Variation assessed using one way analysis of variance, with F-statistic used to compare mean sum of squares

Table 2.

Comparison of Hospital Classification by Three Aggregation Strategies

Sepsis Program Core Element Assessment Tool question	Method 1	Method 2	Method 3	p
	Yes, n (%)	Yes, n (%)	Yes, n (%)
Hospital Leadership Commitment
Q1: Sepsis program leaders given sufficient time to manage hospital sepsis program	10 (52.6%)	13 (68.4%)	19 (100%)	0.01
Q2: Sepsis program given sufficient resources like analytics and IT support to operate effectively	5 (26.3%)	10 (52.6%)	16 (84.2%)	< 0.01
Q3: Staff from key clinical groups/support depts have sufficient time to contribute to sepsis activities	7 (36.8%)	11 (57.9%)	19 (100%)	< 0.01
Q4: Our hospital has senior leader (e.g., CMO, CNO) who serves as Executive Sponsor for sepsis program	7 (36.8%)	12 (63.2%)	19 (100%)	< 0.01
Q5: Sepsis identified as hospital priority by hospital leadership and priority communicated to staff	11 (57.9%)	13 (68.4%)	18 (94.7%)	0.03
Q6: Hospital leadership communicates to staff and patients how hospital is addressing sepsis	6 (31.6%)	8 (42.1%)	18 (94.7%)	< 0.01
Q7: Hospital leaders have regular meetings with sepsis program leaders to assess resources needed to accomplish goals for sepsis activities and outcomes	7 (36.8%)	11 (57.9%)	19 (100%)	< 0.01
Q8: Hospital sepsis program activities are integrated into other QI and patient safety efforts	9 (47.4%)	15 (78.9%)	19 (100%)	< 0.01
Q9: Hospital staff performance incentives are tied to achievement of targets for sepsis care or outcomes	3 (15.8%)	4 (21.1%)	16 (84.2%)	< 0.01
Q10: Sepsis program-related duties are included in job description or performance reviews for hospital sepsis program leaders and key support staff	5 (26.3%)	6 (31.6%)	17 (89.5%)	< 0.01
Q11: Hospital leadership supports external training/education for sepsis program leaders, key support staff	10 (52.6%)	14 (73.7%)	19 (100%)	0.01
Q12: Hospital leadership supports internal training and education on sepsis for hospital staff and trainees	9 (47.4%)	14 (73.7%)	19 (100%)	0.01
Q13: Hospital leadership supports participation in regional, national, and international sepsis QI collaboratives and initiatives	7 (36.8%)	13 (68.4%)	19 (100%)	< 0.01
Accountability
Q14: Hospital has program or committee charged with monitoring and improving sepsis care and outcomes	12 (63.2%)	13 (68.4%)	19 (100%)	0.04
Q15: Hospital has 1–2 leaders (physician, nurse) responsible for sepsis program or committee	10 (52.6%)	13 (68.4%)	19 (100%)	0.01
Q16: Hospital sets ambitious-but achievable-goals for improving sepsis care and patient outcomes that are informed by review of hospital practices, sepsis outcomes, and clinical guidelines	11 (57.9%)	13 (68.4%)	19 (100%)	0.02
Q17: Hospital assesses progress towards sepsis goals at regular intervals and updates goals periodically (e.g., annually) to promote continual improvement	10 (52.6%)	15 (78.9%)	19 (100%)	0.01
Q18: Hospital has 1 physician and 1 nurse lead or champion to ensure physician and nursing engagement	8 (42.1%)	12 (63.2%)	19 (100%)	< 0.01
Q19: Sepsis program activities and outcomes included in annual performance reviews for program leaders	4 (21.1%)	6 (31.6%)	18 (94.7%)	< 0.01
Q20: Hospital has unit-level physician and nurse champions for sepsis activities	2 (15.8%)	3 (15.8%)	18 (94.7%)	< 0.01
Q21: Sepsis program activities and outcomes reported to senior hospital leadership or board of directors	12 (63.2%)	14 (73.7%)	19 (100%)	0.05
Multi-Professional Expertise
Q22: Hospital has sepsis coordinator, who oversees day-to-day implementation of sepsis program activities	7 (36.8%)	8 (42.1%)	19 (100%)	< 0.01
Q23: Clinicians and leaders from ED, IP wards, and ICUs are fully engaged in sepsis program activities	7 (36.8%)	11 (57.9%)	19 (100%)	< 0.01
Q24: Hospital sepsis program includes diverse multi-disciplinary representation	5 (26.3%)	8 (42.1%)	18 (94.7%)	< 0.01
Q25: Hospital sepsis program has ongoing support from individuals with expertise and formal training in data management and analytics, IT, and QI and patient safety	7 (36.8%)	12 (63.2%)	19 (100%)	< 0.01
Q26: Hospital sepsis program has at least ad hoc involvement of case management, microbiology, lab medicine, phlebotomy, outpatient clinicians, hospital epidemiology, infection prevention, patients, families	4 (21.1%)	8 (42.1%)	18 (94.7%)	< 0.01
Action
Q27: Hospital implemented standard process to screen for sepsis on presentation, through hospitalization	8 (42.1%)	13 (68.4%)	19 (100%)	< 0.01
Q28: Hospital has hospital guideline or standardized care pathway for management of sepsis	10 (52.6%)	14 (73.7%)	19 (100%)	< 0.01
Q29: Hospital has order sets for management of sepsis tailored to patient populations served	11 (57.9%)	16 (84.2%)	19 (100%)	0.01
Q30: Hospital has structures and processes in place to facilitate prompt delivery of antimicrobials	14 (73.7%)	17 (89.5%)	19 (100%)	0.14
Q31: Hospital has structures and processes in place to support effective hand-offs of patients with sepsis	4 (21.1%)	9 (47.4%)	19 (100%)	< 0.01
Q32: Hospital rapid response team is trained in sepsis recognition and care	9 (47.4%)	12 (63.2%)	19 (100%)	< 0.01
Q33: Hospital has ‘code sepsis’ protocol for facilitating prompt recognition and team-based care of sepsis	8 (42.1%)	13 (68.4%)	18 (94.7%)	< 0.01
Q34: Hospital completes peri-discharge evaluations of patients after sepsis to screen for new or worsening limitations, cognitive impairment, posttraumatic stress disorder/anxiety symptoms, and chronic conditions	5 (26.3%)	7 (36.8%)	16 (84.2%)	< 0.01
Q35: Hospital has post-discharge care coordination and anticipatory guidance to optimize sepsis recovery	5 (26.3%)	9 (47.4%)	18 (94.7%)	< 0.01
Q36: Hospital has prevention of HAI, HO sepsis that follow facility infection prevention recommendations	15 (79.0%)	18 (94.7%)	19 (100%)	0.17
Tracking
Q37: Hospital monitors hospital sepsis epidemiology, such as number of hospitalizations	8 (42.1%)	13 (68.4%)	19 (100%)	< 0.01
Q38: Hospital monitors hospital sepsis management, such as time to antibiotic delivery	16 (84.2%)	19 (100%)	19 (100%)	0.38
Q39: Hospital monitors sepsis outcomes, such as in-hospital mortality and length of hospitalization	13 (68.4%)	17 (89.5%)	19 (100%)	0.06
Q40: Hospital assesses use, usability, and impact of sepsis tools to inform their ongoing improvement	10 (52.6%)	14 (73.7%)	19 (100%)	< 0.01
Q41: Hospital monitors progress towards achieving hospital goals for sepsis management and/or outcomes	15 (79.0%)	17 (89.5%)	19 (100%)	0.30
Q42: Hospital completes near real-time chart reviews for the purpose of clinician feedback and education	8 (42.1%)	10 (52.6%)	19 (100%)	< 0.01
Q43: Hospital completes chart reviews regularly for root cause analysis and process improvement	13 (68.4%)	15 (78.9%)	19 (100%)	0.10
Reporting
Q44: Hospital reports sepsis treatment/outcome data to nursing, physician, unit-based, hospital leadership	8 (42.1%)	13 (68.4%)	19 (100%)	< 0.01
Q45: Hospital provides feedback to individual clinicians regarding the care of recent patients with sepsis	7 (36.8%)	10 (52.6%)	19 (100%)	< 0.01
Q46: Hospital has and maintains a live dashboard to report sepsis treatment and outcomes in real-time	1 (5.3%)	1 (5.3%)	14 (73.7%)	< 0.01
Education
Q47: Hospital provides sepsis specific training and education in the hiring or on-boarding process	7 (36.8%)	12 (63.2%)	19 (100%)	< 0.01
Q48: Hospital provides annual sepsis education to clinical staff	6 (31.6%)	10 (52.6%)	19 (100%)	< 0.01
Q49: Hospital provides written and verbal sepsis education to patients, families/caregivers before discharge	6 (31.6%)	11 (57.9%)	19 (100%)	< 0.01
Q50: Hospital posts information on recognition of sepsis in prominent areas for patient-facing staff	7 (36.8%)	8 (42.1%)	18 (94.7%)	< 0.01
Q51: Hospital holds lectures (e.g., grand rounds) or an annual meeting focused on sepsis	3 (15.9%)	3 (15.9%)	16 (84.2%)	< 0.01
Q52: Hospital includes sepsis recognition and treatment in annual nursing competencies	5 (26.3%)	10 (52.6%)	18 (94.7%)	< 0.01

Abbreviations: IT, information technology; CMO, Chief Medical Officer; CNO, Chief Nursing Officer; QI, quality improvement; ED, emergency department; IP, inpatient; ICU, intensive care unit; HAI, healthcare-associated infection; HO, hospital-onset

Method 1: All responses=Yes; Method 2: ≥ 0.70 responses=Yes; Method 3: Any responses=Yes

DISCUSSION

Our findings indicate that reports of hospital SPCE frequently varied by respondents within each hospital. This variation may limit the SPCE Assessment Tool’s utility to monitor the composition of hospital sepsis programs and its relationship with effective sepsis outcomes or differentiate performance among hospitals.⁵ We recommend hospitals convene a multidisciplinary respondent committee to promote a collective, comprehensive survey response and mitigate misclassification or undesirable variation. If multiple responses are elicited from within hospitals, strategies to aggregate responses should be standardized to overcome the significant classification differences produced by alternative approaches.

Limitations of our analysis include issues that are common to survey research like low response, contextual limitations of survey questions, and potential for response or recall biases. Additionally, findings from participant sites may not generalize to all geographic regions and hospital settings. Finally, we considered role type in assessing characteristics associated with response variation, but other factors may be important to explore.

In conclusion, we found wide response variation to the CDC SPCE Assessment Tool, with evidence of between- and within-hospital differences. To best monitor sepsis program composition and management practices, hospitals should strive to obtain one singular survey response that synthesizes multiple stakeholders and perspectives. Future research is needed to examine the association between SPCE and hospital outcomes.

Funding

Research reported in this publication was supported by the Agency for Healthcare Research and Quality under award number R01HS029656. The content is solely the responsibility of the authors and does not necessarily represent the official views of the Agency for Healthcare Research and Quality.

Data Availability

The datasets analyzed during the current study are available from the corresponding author on reasonable request.

Declarations:

Ethics Approval:

The study was approved by the University of Michigan Institutional Review Board (IRB#00241324) on February 22, 2024.

Conflict of Interest:

Dr. Kowalkowski reports receiving grant support from the National Institutes of Health, Agency for Healthcare Research and Quality, and Duke Endowment outside of the submitted work. Dr. Birken reports receiving grant support from the National Institutes of Health, Patient Centered Outcomes Research Institute, and Duke Endowment outside of the submitted work. Dr. Miller reports receiving grant support from the National Institutes of Health outside of the submitted work. Dr. Palakshappa reports receiving grant support from the National Institutes of Health outside of the submitted work. Dr. Taylor reports receiving grant support from the National Institutes of Health outside the submitted work. All other authors report no potential conflicts.