Identifying patients with severe sepsis using administrative claims: patient-level validation of the angus implementation of the international consensus conference definition of severe sepsis

The Effect of Severe Sepsis and Septic Shock Management Bundle (SEP-1) Compliance and Implementation on Mortality Among Patients With Sepsis : A Systematic Review

BACKGROUND

The Centers for Medicare & Medicaid Services (CMS) Severe Sepsis and Septic Shock Management Bundle (SEP-1) is now included in the Hospital Value-Based Purchasing (VBP) Program.

PURPOSE

To assess the evidence supporting SEP-1 compliance or SEP-1 implementation in improving sepsis mortality.

DATA SOURCES

PubMed, Web of Science, EMBASE, CINAHL Complete, and Cochrane Library from inception to 26 November 2024.

STUDY SELECTION

Studies of adults with sepsis that included 3- or 6-hour sepsis bundles defined by SEP-1 specifications.

DATA EXTRACTION

Article screening, full-text review, data extraction, and risk-of-bias assessment were independently performed by 2 authors. Level of evidence was determined using GRADE (Grading of Recommendations Assessment, Development and Evaluation) criteria and National Quality Forum criteria.

DATA SYNTHESIS

A total of 4403 unique references were screened, and 17 studies were included. Twelve studies assessed the relationship between SEP-1 compliance and mortality; 5 showed statistically significant benefit, whereas 7 did not. Among studies showing benefit, 1 did not adjust for confounders, 1 found benefit only among patients with severe sepsis, 1 included only patients with septic shock, and 1 included only Medicare beneficiaries. Five studies assessed the relationship between SEP-1 implementation and sepsis mortality; only 1 showed significant benefit, but it did not adjust for mortality trends before SEP-1 implementation. All 17 studies were observational, and none had low risk of bias.

LIMITATIONS

The conclusions are limited by the underlying quality of the available studies, as all were observational. Because there was considerable methodologic heterogeneity among the included studies, a meta-analysis was not performed as the results could have been misleading.

CONCLUSION

This review found no moderate- or high-level evidence to support that compliance with or implementation of SEP-1 was associated with sepsis mortality. CMS should reconsider the addition of SEP-1 to the Hospital VBP Program.

PRIMARY FUNDING SOURCE

None. (PROSPERO: CRD42023482787).

Unsupervised clustering for sepsis identification in large-scale patient data: a model development and validation study

BACKGROUND

Sepsis is a major global health problem. However, it lacks a true reference standard for case identification, complicating epidemiologic surveillance. Consensus definitions have changed multiple times, clinicians struggle to identify sepsis at the bedside, and differing identification algorithms generate wide variation in incidence rates. The two current identification approaches use codes from administrative data, or electronic health record (EHR)-based algorithms such as the Center for Disease Control Adult Sepsis Event (ASE); both have limitations. Here our primary purpose is to report initial steps in developing a novel approach to identifying sepsis using unsupervised clustering methods. Secondarily, we report preliminary analysis of resulting clusters, using identification by ASE criteria as a familiar comparator.

METHODS

This retrospective cohort study used hospital administrative and EHR data on adults admitted to intensive care units (ICUs) at five Canadian medical centres (2015-2017), with split development and validation cohorts. After preprocessing 592 variables (demographics, encounter characteristics, diagnoses, medications, laboratory tests, and clinical management) and applying data reduction, we presented 55 principal components to eight different clustering algorithms. An automated elbow method determined the optimal number of clusters, and the optimal algorithm was selected based on clustering metrics for consistency, separation, distribution and stability. Cluster membership in the validation cohort was assigned using an XGBoost model trained to predict cluster membership in the development cohort. For cluster analysis, we prospectively subdivided clusters by their fractions meeting ASE criteria (≥ 50% ASE-majority clusters vs. ASE-minority clusters), and compared their characteristics.

RESULTS

There were 3660 patients in the development cohort and 3012 in the validation cohort, of which 21.5% (development) and 19.1% (validation) were ASE (+). The Robust and Sparse K-means Clustering (RSKC) method performed best. In the development cohort, it identified 48 clusters of hospitalizations; 11 ASE-majority clusters contained 22.4% of all patients but 77.8% of all ASE (+) patients. 34.9% of the 209 ASE (-) patients in the ASE-majority clusters met more liberal ASE criteria for sepsis. Findings were consistent in the validation cohort.

CONCLUSIONS

Unsupervised clustering applied to diverse, large-scale medical data offers a promising approach to the identification of sepsis phenotypes for epidemiological surveillance.

Postdischarge Mental Health Care and Emergency General Surgery Readmission for Patients With Serious Mental Illness

OBJECTIVE

To determine the association between postdischarge mental health care and odds of readmission after emergency general surgery (EGS) hospitalization for patients with serious mental illness (SMI).

BACKGROUND

A mental health visit (MHV) after medical hospitalization is associated with decreased readmissions for patients with SMI. The impact of a MHV after surgical hospitalization is unknown.

METHODS

Using Medicare claims, we performed a retrospective cohort study of hospitalized EGS patients with SMI aged above 65.5 (2016-2018). EGS included colorectal, general abdominal, hepatopancreatobiliary, hernia, intestinal obstruction, resuscitation, and upper gastrointestinal conditions. SMI was defined as schizophrenia spectrum, mood, or anxiety disorders. The exposure was MHV within 30 days of discharge. The primary outcome was 30-day readmission. Secondary outcomes included emergency department presentation and psychiatric admission. Inverse probability weighting was used to evaluated outcomes.

RESULTS

Of 88,092 analyzed patients, 11,755 (13.3%) had a MHV within 30 days of discharge. 23,696 (26.9%) of patients were managed operatively, 64,395 (73.1%) nonoperatively. After adjustment for potential confounders, patients with a postdischarge MHV had lower odds of acute care readmission than patients without a MHV in both operative (OR=0.60; 95% CI: 0.40-0.90) and nonoperative (OR=0.67; 95% CI: 0.53-0.84) cohorts. There was no association between postdischarge MHV and ED presentation or psychiatric admission in the operative or nonoperative groups.

CONCLUSIONS

Postdischarge MHV after EGS hospitalization was associated with decreased odds of readmission for patients with SMI managed operatively and nonoperatively. In older EGS patients with SMI, coordination of MHVs may be a mechanism to reduce readmission disparities.

Population Sepsis Incidence, Mortality, and Trends in Hong Kong Between 2009 and 2018 Using Clinical and Administrative Data

BACKGROUND

Sepsis surveillance using electronic health record (EHR)-based data may provide more accurate epidemiologic estimates than administrative data, but experience with this approach to estimate population-level sepsis burden is lacking.

METHODS

This was a retrospective cohort study including all adults admitted to publicly funded hospitals in Hong Kong between 2009 and 2018. Sepsis was defined as clinical evidence of presumed infection (clinical cultures and treatment with antibiotics) and concurrent acute organ dysfunction (≥2-point increase in baseline Sequential Organ Failure Assessment [SOFA] score). Trends in incidence, mortality, and case fatality risk (CFR) were modeled by exponential regression. Performance of the EHR-based definition was compared with 4 administrative definitions using 500 medical record reviews.

RESULTS

Among 13 540 945 hospital episodes during the study period, 484 541 (3.6%) had sepsis by EHR-based criteria with 22.4% CFR. In 2018, age- and sex-adjusted standardized sepsis incidence was 756 per 100 000 (relative change: +2.8%/y [95% CI: 2.0%-3.7%] between 2009 and 2018) and standardized sepsis mortality was 156 per 100 000 (relative change: +1.9%/y; 95% CI: .9%-2.8%). Despite decreasing CFR (relative change: -0.5%/y; 95% CI: -1.0%, -.1%), sepsis accounted for an increasing proportion of all deaths (relative change: +3.9%/y; 95% CI: 2.9%-4.8%). Medical record reviews demonstrated that the EHR-based definition more accurately identified sepsis than administrative definitions (area under the curve [AUC]: .91 vs .52-.55; P < .001).

CONCLUSIONS

An objective EHR-based surveillance definition demonstrated an increase in population-level standardized sepsis incidence and mortality in Hong Kong between 2009 and 2018 and was much more accurate than administrative definitions. These findings demonstrate the feasibility and advantages of an EHR-based approach for widescale sepsis surveillance.

Characteristics and Readmission Risks Following Sepsis Discharges to Home

OBJECTIVE

To examine the characteristics and risk factors associated with 30-day readmissions, including the impact of home health care (HHC), among older sepsis survivors transitioning from hospital to home.

RESEARCH DESIGN

Retrospective cohort study of the Medical Information Mart for Intensive Care (MIMIC)-IV data (2008-2019), using generalized estimating equations (GEE) models adjusting for patient sociodemographic and clinical characteristics.

SUBJECTS

Sepsis admission episodes with in-hospital stays, aged over 65, and discharged home with or without HHC were included.

MEASURES

The outcome was all-cause hospital readmission within 30 days following sepsis hospitalization. Covariates, including the primary predictor (HHC vs. Home discharges), were collected during hospital stays.

RESULTS

Among 9115 sepsis admissions involving 6822 patients discharged home (66.8% HHC, 33.2% Home), HHC patients, compared with those discharged without services, were older, had more comorbidities, longer hospital stays, more prior hospitalizations, more intensive care unit admissions, and higher rates of septic shock diagnoses. Despite higher illness severity in the HHC discharges, both groups had high 30-day readmission rates (30.2% HHC, 25.2% Home). GEE analyses revealed 14% higher odds of 30-day readmission for HHC discharges after adjusting for risk factors (aOR: 1.14; 95% CI: 1.02-1.27; P=0.02).

CONCLUSIONS

HHC discharges experienced higher 30-day readmission rates than those without, indicating the need for specialized care in HHC settings for sepsis survivors due to their complex health care needs. Attention to sepsis survivors, regardless of HHC receipt, is crucial given the high readmission rates in both groups. Further research is needed to optimize postacute care/interventions for older sepsis survivors.

National Analysis of Preexisting Immunosuppressive Conditions and Infection-Related Readmissions Among Sepsis Survivors

BACKGROUND

Recurrent episodes of infection and subsequent sepsis are a frequent cause of readmission after sepsis hospitalization. Although persistent immune dysregulation initiated during the sepsis episode may play a role, the impact of preexisting immune suppression (including HIV, organ transplantation, and cancer) and common chronic diseases associated with immune suppression (diabetes and chronic kidney disease) on the risk of recurrent infections after sepsis is unknown.

OBJECTIVES

To investigate the role of preexisting immune-suppressive conditions (PISCs) and other common chronic diseases on infection-related readmissions after a sepsis admission.

METHODS

Using the US 2015 Nationwide Readmissions Database, we identified a retrospective cohort of adult patients with an unplanned sepsis index admission from April to September, excluding patients with sepsis during the preceding 90 days. We followed all sepsis survivors for subsequent infection-related admissions for 90 days. We identified clinical conditions using International Classification of Diseases coding.

RESULTS

We identified 649 029 unique unplanned sepsis admissions over 6 months; 189 604 (29.2%) had sepsis with PISC, and 459 425 (70.8%) had sepsis without PISC. Overall, sepsis survivors were older (median age, 70 years), and 145 156 (22.4%) experienced at least 1 infection-related readmission within 90 days. The incidence of infection-related readmission among sepsis survivors with PISC was 26.1%, whereas it was 20.8% for those without PISC. The excess risk of infection-related readmissions attributed to PISC was 5.3%, whereas the excess risk attributed to diabetes and chronic kidney disease was 3.7% and 4.7%, respectively. The background risk attributed to new-onset sepsis among participants with none of these conditions was 16.3%. Multivariable regression analysis adjusting for age, gender, and acute illness factors indicated that odds of infection-related readmission were strongly associated with PISC (odds ratio 1.30; 95% confidence interval, 1.29-1.32), closely followed by chronic kidney disease (1.28 [1.27-1.32]) and diabetes (1.17 [1.16-1.19]).

CONCLUSION

The risk of subsequent infection is higher among patients with PISC, although chronic kidney disease and diabetes are also important risk factors for subsequent infection and sepsis readmissions.

Sepsis survivors readmitted within 30 days: outcomes of a single-center retrospective study

OBJECTIVE

To investigate a cohort of sepsis survivors readmitted within 30 days postdischarge, explore the one-year mortality rate based on different causes of readmission and identify factors associated with increased one-year mortality risk among all sepsis survivors readmitted within this timeframe.

METHODS

This was a single-center retrospective cohort study involving adult sepsis survivors who were readmitted within 30 days of discharge. Patients were categorized into 3 groups based on the cause of readmission: same-source infectious readmission, different-source infectious readmission, and noninfectious readmission. The outcome of interest was all-cause one-year mortality. Cox proportional hazard analysis was performed to compare factors associated with one-year mortality.

RESULTS

Of the 1,666 patients admitted with sepsis, 243 (14.5%) were readmitted within 30 days. Readmissions were due to same-source infections (40.7%), different-source infections (21.4%), or noninfectious causes (37.9%). All-cause one-year mortality was 46.9%, with no difference between the groups. Age (HR 1.02; 95%CI: 1.003 - 1.04; p = 0.01), Sequential Organ Failure Assessment score (HR 1.1; 95%CI: 1.02 - 1.18; p = 0.01), discharge to a care facility during index admission (HR 1.6; 95%CI: 1.04 - 2.40; p = 0.03), and malignancy (HR 2.3; 95%CI: 1.5 - 3.7; p < 0.001) were associated with one-year mortality.

CONCLUSION

Thirty-day readmission in sepsis survivors was common and was associated with a 46.9% one-year mortality rate regardless of readmission cause. Quality improvement patient safety initiatives based on local institutional factors may allow for targeted interventions to improve sepsis survivor outcomes.

Adjudication of Codes for Identifying Sepsis in Hospital Administrative Data by Expert Consensus

OBJECTIVES

Refine the administrative data definition of sepsis in hospitalized patients, including less severe cases.

DESIGN AND SETTING

For each of 1928 infection and 108 organ dysfunction codes used in Canadian hospital abstracts, experts reached consensus on the likelihood that it could relate to sepsis. We developed a new algorithm, called AlgorithmL, that requires at least one infection and one organ dysfunction code adjudicated as likely or very likely to be related to sepsis. AlgorithmL was compared with four previously described algorithms, regarding included codes, population-based incidence, and hospital mortality rates-separately for ICU and non-ICU cohorts in a large Canadian city. We also compared sepsis identification from these code-based algorithms with the Centers for Disease Control's Adult Sepsis Event (ASE) definition.

SUBJECTS

Among Calgary's adult population of 1.033 million there were 61,632 eligible hospitalizations.

INTERVENTIONS

None.

MEASUREMENTS AND MAIN RESULTS

AlgorithmL includes 720 infection codes and 50 organ dysfunction codes. Comparison algorithms varied from 42-941 infection codes to 2-36 organ codes. There was substantial nonoverlap of codes in AlgorithmL vs. the comparators. Annual sepsis incidence rates (per 100,000 population) based on AlgorithmL were 91 in the ICU and 291 in the non-ICU cohort. Incidences based on comparators ranged from 28-77 for ICU to 11-266 for non-ICU cohorts. Hospital sepsis mortality rates based on AlgorithmL were 24% in ICU and 17% in non-ICU cohorts; based on comparators, they ranged 27-38% in the ICU cohort and 18-47% for the non-ICU cohort. Of AlgorithmL-identified cases, 41% met the ASE criteria, compared with 42-82% for the comparator algorithms.

CONCLUSIONS

Compared with other code-based algorithms, AlgorithmL includes more infection and organ dysfunction codes. AlgorithmL incidence rates are higher; hospital mortality rates are lower. AlgorithmL may more fully encompass the full range of sepsis severity.

Association of Area Deprivation Index With Mortality in Critically Ill Adults With COVID-19

BACKGROUND

Various social determinants of health have been established as significant risk factors for COVID-19 transmission, prevalence, incidence, and mortality. Area deprivation index (ADI, a composite score made up of educational, housing, and poverty markers) is an accepted multidimensional social determinants of health measure. Little is known about how structural social determinants of health before hospitalization, including ADI, may affect mortality related to COVID-19 in critically ill patients.

OBJECTIVES

To examine the association of ADI with intensive care unit (ICU) mortality in patients with COVID-19 and compare its predictive power with that of clinical factors.

METHODS

This was a retrospective cohort study of critically ill adults with COVID-19 in 3 hospitals within a single health system. Multivariable logistic regression models (adjusted for demographic and clinical variables) were used to examine the association of ADI with ICU mortality.

RESULTS

Data from 1784 patients hospitalized from 2020 to 2022 were analyzed. In multivariable models, no association was found between national ADI and ICU mortality. Notable factors associated with ICU mortality included treatment year, age, van Walraven weighted score, invasive mechanical ventilation, and body mass index.

CONCLUSION

In this study, clinical factors were more predictive of mortality than ADI and other social determinants of health. The influence of ADI may be most relevant before hospital admission. These findings could serve as a foundation for shaping targeted public health strategies and hospital interventions, enhancing care delivery, and potentially contributing to better outcomes in future pandemics.

Healthcare Use and Expenditures in Rural Survivors of Hospitalization for Sepsis

OBJECTIVES

Sepsis survivors have greater healthcare use than those surviving hospitalizations for other reasons, yet factors associated with greater healthcare use in this population remain ill-defined. Rural Americans are older, have more chronic illnesses, and face unique barriers to healthcare access, which could affect postsepsis healthcare use. Therefore, we compared healthcare use and expenditures among rural and urban sepsis survivors. We hypothesized that rural survivors would have greater healthcare use and expenditures.

DESIGN, SETTING, AND PATIENTS

To test this hypothesis, we used data from 106,189 adult survivors of a sepsis hospitalization included in the IBM MarketScan Commercial Claims and Encounters database and Medicare Supplemental database between 2013 and 2018.

INTERVENTIONS

None.

MEASUREMENTS AND MAIN RESULTS

We identified hospitalizations for severe sepsis and septic shock using the International Classification of Diseases , 9th Edition (ICD-9) or 1CD-10 codes. We used Metropolitan Statistical Area classifications to categorize rurality. We measured emergency department (ED) visits, inpatient hospitalizations, skilled nursing facility admissions, primary care visits, physical therapy visits, occupational therapy visits, and home healthcare visits for the year following sepsis hospitalizations. We calculated the total expenditures for each of these categories. We compared outcomes between rural and urban patients using multivariable regression and adjusted for covariates. After adjusting for age, sex, comorbidities, admission type, insurance type, U.S. Census Bureau region, employment status, and sepsis severity, those living in rural areas had 17% greater odds of having an ED visit (odds ratio [OR] 1.17; 95% CI, 1.13-1.22; p < 0.001), 9% lower odds of having a primary care visit (OR 0.91; 95% CI, 0.87-0.94; p < 0.001), and 12% lower odds of receiving home healthcare (OR 0.88; 95% CI, 0.84-0.93; p < 0.001). Despite higher levels of ED use and equivalent levels of hospital readmissions, expenditures in these areas were 14% (OR 0.86; 95% CI, 0.80-0.91; p < 0.001) and 9% (OR 0.91; 95% CI, 0.87-0.96; p < 0.001) lower among rural survivors, respectively, suggesting these services may be used for lower-acuity conditions.

CONCLUSIONS

In this large cohort study, we report important differences in healthcare use and expenditures between rural and urban sepsis survivors. Future research and policy work is needed to understand how best to optimize sepsis survivorship across the urban-rural continuum.

Ceftazidime/avibactam alone or in combination with an aminoglycoside for treatment of carbapenem-resistant Enterobacterales infections: A retrospective cohort study

BACKGROUND

Ceftazidime/avibactam is one of the preferred treatment options for carbapenem-resistant Enterobacterales (CRE). However, the benefit of combining ceftazidime/avibactam with another antibiotic remains unclear.

OBJECTIVES

To identify variables associated with treatment failure during the use of ceftazidime/avibactam for CRE infections and assess the effect of combining an aminoglycoside with ceftazidime/avibactam.

METHODS

This was a retrospective cohort study of patients with a positive CRE culture treated with ceftazidime/avibactam between 2015 and 2021 in 134 Veterans Affairs (VA) facilities. The primary outcome was 30-day mortality and the secondary outcome was in-hospital mortality. A subanalysis in patients who received an aminoglycoside was also performed.

RESULTS

A total of 303 patients were included. The overall 30-day and in-hospital mortality rates were 12.5% and 24.1%, respectively. Age (aOR 1.052, 95% CI 1.013-1.093), presence in the ICU (aOR 2.704, 95% CI 1.071-6.830), and receipt of an aminoglycoside prior to initiation of ceftazidime/avibactam (aOR 4.512, 95% CI 1.797-11.327) were independently associated with 30-day mortality. In the subgroup of patients that received an aminoglycoside (n = 77), their use in combination with ceftazidime/avibactam had a 30-day mortality aOR of 0.321 (95% CI, 0.089-1.155).

CONCLUSION

In veterans treated with ceftazidime/avibactam for CRE infections, increased age, receipt of an empiric aminoglycoside, and presence in the ICU at the time of index culture were associated with higher 30-day mortality. Among patients who received an aminoglycoside, their use in combination with ceftazidime/avibactam trended toward protectiveness of 30-day mortality, suggesting a potential role for this combination to treat CRE infections in patients who are more severely ill.

Chronic kidney disease and risk of bloodstream infections and sepsis: a 17-year follow-up of the population-based Trøndelag Health Study in Norway

PURPOSE

Bloodstream infections (BSI) and sepsis are important causes of hospitalization, loss of health, and death globally. Targetable risk factors need to be identified to improve prevention and treatment. In this study, we aimed to evaluate the association of chronic kidney disease (CKD) and risk of and mortality from BSI and sepsis in the general population during a 22-year period.

METHODS

We conducted a prospective cohort study among participants in the population-based Norwegian HUNT Study, where 68,438 participated. The median follow-up time was 17.4 years. The exposures were estimated glomerular filtration rate (eGFR) and albumin-creatinine ratio (ACR) in urine. The outcomes were hazard ratios (HR) of hospital admission or death due to BSI or sepsis. The associations were adjusted for age, sex, diabetes, obesity, systolic blood pressure, smoking status, and cardiovascular disease.

RESULTS

Participants with eGFR < 30 ml/min/1.732 had HR 3.35 for BSI (95% confidence intervals (CI) 2.12-5.3) and HR 2.94 for sepsis (95% CI 1.82-4.8) compared to normal eGFR (≥ 90 ml/min/1.732). HRs of death from BSI and sepsis were 4.2 (95% CI 1.71-10.4) and 4.1 (95% CI 1.88-8.9), respectively. Participants with severely increased albuminuria (ACR > 30 mg/mmol) had HR 3.60 for BSI (95% CI 2.30-5.6) and 3.14 for sepsis (95% CI 1.94-5.1) compared to normal albumin excretion (ACR < 3 mg/mmol). HRs of death were 2.67 (95% CI 0.82-8.7) and 2.16 (95% CI 0.78-6.0), respectively.

CONCLUSION

In this large population-based cohort study, CKD was clearly associated with an increased risk of BSI and sepsis and related death.

Antipsychotic medications and severe sepsis in schizophrenia: A nested case-control study

BACKGROUND

Sepsis constitutes a condition that involves life-threatening organ dysfunction induced by severe infection. This nested case-control study investigated risk factors for severe sepsis and whether antipsychotic use is associated with severe sepsis risk in patients with schizophrenia, a topic that has not been comprehensively explored in previous studies.

METHODS

We selected 39,432 patients with schizophrenia aged between 15 and 65 years from Taiwan's Psychiatric Inpatient Medical Claims database for the period 2000-2012. The case group comprised patients with severe sepsis after their first psychiatric admission (n = 1382). The case and control groups were randomly matched (1:4) by age, sex and first psychiatric admission (year) and finally comprised 1382 and 5528 individuals, respectively. We employed multivariable conditional logistic regression to identify (1) risk factors (physical illnesses and nonpsychiatric medications) and (2) antipsychotic-severe sepsis associations.

RESULTS

Higher numbers of psychiatric admissions and physical illnesses such as delirium, cerebrovascular disease and cancer were significantly associated with a higher risk of severe sepsis. Furthermore, severe sepsis was associated with the use of antithrombotic agents, systemic corticosteroids and agents targeting the renin-angiotensin system. Clozapine (adjusted risk ratio = 1.65) and quetiapine (adjusted risk ratio = 1.59) use were associated with an increased risk of severe sepsis. The use of more than one antipsychotic drug could further increase this risk.

CONCLUSION

Several physical illnesses and nonpsychiatric medications increase the risk of severe sepsis in patients with schizophrenia. Specifically, clozapine or quetiapine use significantly increased the risk of severe sepsis in these patients.

Development and Validation of a Machine Learning Model for Early Detection of Untreated Infection

BACKGROUND

Early diagnostic uncertainty for infection causes delays in antibiotic administration in infected patients and unnecessary antibiotic administration in noninfected patients.

OBJECTIVE

To develop a machine learning model for the early detection of untreated infection (eDENTIFI), with the presence of infection determined by clinician chart review.

DERIVATION COHORT

Three thousand three hundred fifty-seven adult patients hospitalized between 2006 and 2018 at two health systems in Illinois, United States.

VALIDATION COHORT

We validated in 1632 patients in a third Illinois health system using area under the receiver operating characteristic curve (AUC).

PREDICTION MODEL

Using a longitudinal discrete-time format, we trained a gradient boosted machine model to predict untreated infection in the next 6 hours using routinely available patient demographics, vital signs, and laboratory results.

RESULTS

eDENTIFI had an AUC of 0.80 (95% CI, 0.79-0.81) in the validation cohort and outperformed the systemic inflammatory response syndrome criteria with an AUC of 0.64 (95% CI, 0.64-0.65; p < 0.001). The most important features were body mass index, age, temperature, and heart rate. Using a threshold with a 47.6% sensitivity, eDENTIFI detected infection a median 2.0 hours (interquartile range, 0.9-5.2 hr) before antimicrobial administration, with a negative predictive value of 93.6%. Antibiotic administration guided by eDENTIFI could have decreased unnecessary IV antibiotic administration in noninfected patients by 10.8% absolute or 46.4% relative percentage points compared with clinicians.

CONCLUSION

eDENTIFI could both decrease the time to antimicrobial administration in infected patients and unnecessary antibiotic administration in noninfected patients. Further prospective validation is needed.

Trends in Empiric Broad-Spectrum Antibiotic Use for Suspected Community-Onset Sepsis in US Hospitals

Importance

Little is known about the degree to which suspected sepsis drives broad-spectrum antibiotic use in hospitals, what proportion of antibiotic courses are unnecessarily broad in retrospect, and whether these patterns are changing over time.

Objective

To describe trends in empiric broad-spectrum antibiotic use for suspected community-onset sepsis.

Design, Setting, and Participants

This cross-sectional study used clinical data from adults admitted to 241 US hospitals in the PINC AI Healthcare Database. Eligible participants were aged 18 years or more and were admitted between 2017 and 2021 with suspected community-onset sepsis, defined by a blood culture draw, lactate measurement, and intravenous antibiotic administration on admission.

Exposures

Empiric anti-methicillin-resistant Staphylococcus aureus (MRSA) and/or antipseudomonal β-lactam agent use.

Main Outcomes and Measures

Annual rates of empiric anti-MRSA and/or antipseudomonal β-lactam agent use and the proportion that were likely unnecessary in retrospect based on the absence of β-lactam resistant gram-positive or ceftriaxone-resistant gram-negative pathogens from clinical cultures obtained through hospital day 4. Annual trends were calculated using mixed-effects logistic regression models, adjusting for patient and hospital characteristics.

Results

Among 6 272 538 hospitalizations (median [IQR] age, 66 [53-78] years; 443 465 male [49.6%]; 106 095 Black [11.9%], 65 763 Hispanic [7.4%], 653 907 White [73.1%]), 894 724 (14.3%) had suspected community-onset sepsis, of whom 582 585 (65.1%) received either empiric anti-MRSA (379 987 [42.5%]) or antipseudomonal β-lactam therapy (513 811 [57.4%]); 311 213 (34.8%) received both. Patients with suspected community-onset sepsis accounted for 1 573 673 of 3 141 300 (50.1%) of total inpatient anti-MRSA antibiotic days and 2 569 518 of 5 211 745 (49.3%) of total antipseudomonal β-lactam days. Between 2017 and 2021, the proportion of patients with suspected sepsis administered anti-MRSA or antipseudomonal therapy increased from 63.0% (82 731 of 131 275 patients) to 66.7% (101 003 of 151 435 patients) (adjusted OR [aOR] per year, 1.03; 95% CI, 1.03-1.04). However, resistant organisms were isolated in only 65 434 cases (7.3%) (30 617 gram-positive [3.4%], 38 844 gram-negative [4.3%]) and the proportion of patients who had any resistant organism decreased from 9.6% to 7.3% (aOR per year, 0.87; 95% CI, 0.87-0.88). Most patients with suspected sepsis treated with empiric anti-MRSA and/or antipseudomonal therapy had no resistant organisms (527 356 of 582 585 patients [90.5%]); this proportion increased from 88.0% in 2017 to 91.6% in 2021 (aOR per year, 1.12; 95% CI, 1.11-1.13).

Conclusions and Relevance

In this cross-sectional study of adults admitted to 241 US hospitals, empiric broad-spectrum antibiotic use for suspected community-onset sepsis accounted for half of all anti-MRSA or antipseudomonal therapy; the use of these types of antibiotics increased between 2017 and 2021 despite resistant organisms being isolated in less than 10% of patients treated with broad-spectrum agents.

The rates of septicemia in older adults with prostate cancer treated with abiraterone or enzalutamide: A population-based study

INTRODUCTION

Prostate cancer (PCa) is the most common non-cutaneous tumor among American men. Androgen receptor signaling inhibitors such as abiraterone and enzalutamide have been approved for similar disease states among patients with advanced PCa. Existing data suggest using steroids is associated with an increased risk of infection. Because abiraterone is usually prescribed with prednisone, we sought to compare the risk of septicemia in patients using abiraterone vs. enzalutamide.

MATERIALS AND METHODS

We utilized the SEER-Medicare-linked data and used negative binomial regression models to compare the changes in the rates of septicemia-related hospitalizations six months pre- and post-abiraterone and enzalutamide initiation.

RESULTS

We found that the incidence of septicemia-related hospitalizations increased 2.77 fold within six months of initiating abiraterone (incidence rate ratio [IRR]: 2.77, 95% confidence interval [CI]: 2.17-3.53) 1.97 fold within six months of starting enzalutamide (IRR: 1.97, 95% CI: 1.43-2.72). However, the difference in the changes did not reach statistical significance (interaction IRR: 0.71, 95% CI: 0.48-1.06).

DISCUSSION

The findings suggest that both abiraterone and enzalutamide are associated with an increased risk of septicemia-related hospitalizations. However, the difference in the increase of septicemia risk following the two treatments did not reach statistical significance. Further studies are warranted to understand the mechanisms at play.

Characteristics and Outcomes of Obstetric Sepsis by Three Sets of Clinical Diagnostic Criteria: A Retrospective Study

OBJECTIVE

 Pregnancy-related mortality in the United States is the highest of all developed nations with a reported rate of 17 deaths per 100,000 live births in 2014 to 2017. Sepsis-related mortality is a major component of pregnancy-related mortality. Similar to nonpregnancy-related sepsis, the criteria for pregnancy-related sepsis are evolving. The purposes of this study were to compare three criteria for sepsis (Sepsis-2, Sepsis-3, California Maternal Quality Care Collaborative [CMQCC]) with one another and to determine patient outcomes using those three sets of criteria.

STUDY DESIGN

 Using the electronic medical record, we obtained granular data on all patients at University of Michigan Medical Center from July 10, 2009 to September 4, 2019 with suspected sepsis (blood cultures and administration of antibiotics) during pregnancy until the 42nd postpartum day. Agreement between the three criteria were assessed with kappa and shown by a Venn diagram. Groups were compared using standardized differences and chi square, rank sum, or independent t-tests.

RESULTS

 Of the 228 patients having sepsis by any criteria, 191 (83%) patients met the criteria for Sepsis-2, 131 (57%) for Sepsis-3, and 62 (27%) met criteria according to CMQCC. Agreement between the three criteria ranged from kappa = 0.13 (95% confidence interval [CI]: 0.09, 0.18) to kappa = 0.31 (95% CI: 0.23, 0.39). Patients who met CMQCC criteria tended to have more comorbidities and higher APACHE II (Acute Physiology And Chronic Health Evaluation) scores. Mortality (by 90 days) among the groups was low with 10 (4%) patients dying. Patients meeting criteria for CMQCC sepsis had higher mortality than the non-CMQCC patients with sepsis (10 vs. 2%, standardized difference = 0.31, p = 0.027).

CONCLUSION

 The agreement among Sepsis-2, Sepsis-3, and CMQCC diagnostic criteria is weak. CMQCC criteria identifies patients with sepsis at higher risk of death.

KEY POINTS

· Agreements (kappa) between the three criteria are poor.. · CMQCC sepsis patients have more comorbidities.. · Endometritis was the most common cause of sepsis.. · CMQCC mortality was 10%; non-CMQCC mortality was 2%..

Understanding the biases to sepsis surveillance and quality assurance caused by inaccurate coding in administrative health data

PURPOSE

Timely and accurate data on the epidemiology of sepsis are essential to inform policy decisions and research priorities. We aimed to investigate the validity of inpatient administrative health data (IAHD) for surveillance and quality assurance of sepsis care.

METHODS

We conducted a retrospective validation study in a disproportional stratified random sample of 10,334 inpatient cases of age ≥ 15 years treated in 2015-2017 in ten German hospitals. The accuracy of coding of sepsis and risk factors for mortality in IAHD was assessed compared to reference standard diagnoses obtained by a chart review. Hospital-level risk-adjusted mortality of sepsis as calculated from IAHD information was compared to mortality calculated from chart review information.

RESULTS

ICD-coding of sepsis in IAHD showed high positive predictive value (76.9-85.7% depending on sepsis definition), but low sensitivity (26.8-38%), which led to an underestimation of sepsis incidence (1.4% vs. 3.3% for severe sepsis-1). Not naming sepsis in the chart was strongly associated with under-coding of sepsis. The frequency of correctly naming sepsis and ICD-coding of sepsis varied strongly between hospitals (range of sensitivity of naming: 29-71.7%, of ICD-diagnosis: 10.7-58.5%). Risk-adjusted mortality of sepsis per hospital calculated from coding in IAHD showed no substantial correlation to reference standard risk-adjusted mortality (r = 0.09).

CONCLUSION

Due to the under-coding of sepsis in IAHD, previous epidemiological studies underestimated the burden of sepsis in Germany. There is a large variability between hospitals in accuracy of diagnosing and coding of sepsis. Therefore, IAHD alone is not suited to assess quality of sepsis care.

Costs of Care for Operative and Nonoperative Management of Emergency General Surgery Conditions

OBJECTIVE

Many emergency general surgery (EGS) conditions can be managed operatively or nonoperatively, with outcomes that vary by diagnosis. We hypothesized that operative management would lead to higher in-hospital costs but to cost savings over time.

BACKGROUND

EGS conditions account for $28 billion in health care costs in the United States annually. Compared with scheduled surgery, patients who undergo emergency surgery are at increased risk of complications, readmissions, and death, with accompanying costs of care that are up to 50% higher than elective surgery. Our prior work demonstrated that operative management had variable impacts on clinical outcomes depending on the EGS condition.

METHODS

This was a nationwide, retrospective study using fee-for-service Medicare claims data. We included patients 65.5 years of age or older with a principal diagnosis for an EGS condition 7/1/2015-6/30/2018. EGS conditions were categorized as: colorectal, general abdominal, hepatopancreaticobiliary (HPB), intestinal obstruction, and upper gastrointestinal. We used near-far matching with a preference-based instrumental variable to adjust for confounding and selection bias. Outcomes included Medicare payments for the index hospitalization and at 30, 90, and 180 days.

RESULTS

Of 507,677 patients, 30.6% received an operation. For HPB conditions, costs for operative management were initially higher but became equivalent at 90 and 180 days. For all others, operative management was associated with higher inpatient costs, which persisted, though narrowed, over time. Out-of-pocket costs were nearly equivalent for operative and nonoperative management.

CONCLUSIONS

Compared with nonoperative management, costs were higher or equivalent for operative management of EGS conditions through 180 days, which could impact decision-making for clinicians, patients, and health systems in situations where clinical outcomes are similar.

Using a theory-based, customized video game as an educational tool to improve physicians' trauma triage decisions: study protocol for a randomized cluster trial

BACKGROUND

Transfer of severely injured patients to trauma centers, either directly from the field or after evaluation at non-trauma centers, reduces preventable morbidity and mortality. Failure to transfer these patients appropriately (i.e., under-triage) remains common, and occurs in part because physicians at non-trauma centers make diagnostic errors when evaluating the severity of patients' injuries. We developed Night Shift, a theory-based adventure video game, to recalibrate physician heuristics (intuitive judgments) in trauma triage and established its efficacy in the laboratory. We plan a type 1 hybrid effectiveness-implementation trial to determine whether the game changes physician triage decisions in real-life and hypothesize that it will reduce the proportion of patients under-triaged.

METHODS

We will recruit 800 physicians who work in the emergency departments (EDs) of non-trauma centers in the US and will randomize them to the game (intervention) or to usual education and training (control). We will ask those in the intervention group to play Night Shift for 2 h within 2 weeks of enrollment and again for 20 min at quarterly intervals. Those in the control group will receive only usual education (i.e., nothing supplemental). We will then assess physicians' triage practices for older, severely injured adults in the 1-year following enrollment, using Medicare claims, and will compare under-triage (primary outcome), 30-day mortality and re-admissions, functional independence, and over-triage between the two groups. We will evaluate contextual factors influencing reach, adoption, implementation, and maintenance with interviews of a subset of trial participants (n = 20) and of other key decision makers (e.g., patients, first responders, administrators [n = 100]).

DISCUSSION

The results of the trial will inform future efforts to improve the implementation of clinical practice guidelines in trauma triage and will provide deeper understanding of effective strategies to reduce diagnostic errors during time-sensitive decision making.

TRIAL REGISTRATION

ClinicalTrials.gov; NCT06063434 . Registered 26 September 2023.

Sepsis in Latvia-Incidence, Outcomes, and Healthcare Utilization: A Retrospective, Observational Study

This study explores the incidence, outcomes, and healthcare resource utilization concerning sepsis in Latvia's adult population. Using a merged database from the National Health Service and the Latvian Centre for Disease Prevention and Control, sepsis-related hospitalizations were analyzed from 2015-2020. Findings revealed a 53.1% surge in sepsis cases from 2015-2018 with subsequent stabilization. This spike was more prominent among elderly patients. The age/sex adjusted case fatality rate rose from 34.7% in 2015 to 40.5% in 2020. Of the 7764 sepsis survivors, the one-year mortality rate was 12% compared to 2.2% in a reference group of 20,686 patients with infections but no further signs of sepsis. Sepsis survivors also incurred higher healthcare costs, driven by longer rehospitalizations and increased pharmaceutical needs, though they accessed outpatient services less frequently than the reference group. These findings underscore the growing detection of sepsis in Latvia, with survivors facing poorer outcomes and suggesting the need for enhanced post-sepsis outpatient care.

Accuracy of International Classification of Disease Coding Methods to Estimate Sepsis Epidemiology: A Scoping Review

PURPOSE

To provide an overview of various sepsis International Classification of Diseases (ICD) coding methods and their diagnostic accuracy.

METHODS

We undertook a systematic scoping review between 1991 and 2020 (search terms: sepsis, coding, and epidemiology) to include studies reporting the accuracy of a sepsis ICD coding method. Studies were grouped by ICD coding method, number of diagnostic accuracy parameters, ICD version, reference standard, design, country, setting, type of dataset and sepsis definition. ICD coding methods were categorised as explicit or implicit, with the explicit methods further divided into wide and narrow groups. Descriptive statistics were used to present data.

RESULTS

We analysed 17 studies, of which 16 (94.1%) used retrospective medical chart review as the reference standard for clinical sepsis, and eight (47.1%) used hospital administrative data to identify sepsis. There were 53 assessments of various ICD coding methods, with 32 (60.4%) of them being explicit and 21 (39.6%) implicit methods. The coding methods had a median sensitivity of <75% but a median specificity of >85%. However, a wide variation was noted in the diagnostic accuracy parameters of all ICD coding methods. Most of the studies showed high methodological quality.

CONCLUSION

None of the current ICD coding methods is optimal for identifying sepsis.

An acute care surgeon's dilemma: Operative vs. non-operative management of emergency general surgery conditions in patients with recent colorectal cancer treatment

BACKGROUND

This comparative effectiveness study examined outcomes of operative vs. non-operative management for emergency general surgery (EGS) conditions in patients with recent cancer treatment (RT).

METHODS

Medicare beneficiaries with a history of colorectal cancer hospitalized for an EGS condition (2016-2018) were identified. RT was defined as chemotherapy/radiation within 3 months prior to admission. Instrumental variable analysis assessed the impact of management on mortality and readmissions among survivors (30d, 60d, and 90d), for patients in whom there was clinical equipoise regarding optimal management strategy.

RESULTS

Of 26,097 patients, 13% had undergone RT. In both the RT and non-RT groups, the optimal management strategy was uncertain in 14%. Operative management conferred increased risk of mortality but not readmission in patients with RT compared to those without (90d mortality:+43%, p ​= ​0.03; 90d readmission:+7.1%, p ​= ​0.776).

CONCLUSIONS

In patients with RT for whom there is clinical equipoise regarding EGS management, operative intervention increases risk of mortality.

Challenges of assessing the burden of sepsis

BACKGROUND

Sepsis is one of the most frequent causes of death worldwide, but the recording of population-based epidemiology is challenging, which is why reliable data on sepsis incidence and mortality are only available in a few, mostly highly-resourced countries.

OBJECTIVE

The aim of this narrative review is to provide an overview of sepsis epidemiology worldwide and in Germany based on current literature, to identify challenges in this research area, and to give an outlook on future developments.

MATERIALS AND METHODS

Selective literature review. PubMed and Google Scholar were searched for current literature. The results were processed narratively.

RESULTS

Based on modeling studies or meta-analyses of prospective studies, global annual sepsis incidence was found to be 276-678/100,000 persons. Case fatality ranged from 22.5 to 26.7%. However, current data sources have several limitations, as administrative data of selected individual countries-mostly with high income-were used as their basis. In these administrative data, sepsis is captured with limited validity. Prospective studies using clinical data often have limited comparability or lack population reference.

CONCLUSION

There is a lack of reliable data sources and definitions to monitor the epidemiology of sepsis and collect reliable global estimates. Increased policy efforts and new scientific approaches are needed to improve our understanding of sepsis epidemiology, identify vulnerable populations, and develop and target effective interventions.

Trends in mortality after a sepsis hospitalization: a nationwide prospective registry study from 2008 to 2021

BACKGROUND

Few studies have reported on mortality beyond one year after sepsis. We aim to describe trends in short- and long-term mortality among patients admitted with sepsis, and to describe the association between clinical characteristics and mortality for improved monitoring, treatment and prognosis.

METHODS

Patients ≥ 18 years admitted to all Norwegian hospitals (2008-2021) with a first sepsis episode were identified using Norwegian Patient Registry and International Classification of Diseases 10th Revision codes. Sepsis was classified as implicit (known infection site plus organ dysfunction), explicit (unknown infection site), or COVID-19-related sepsis. The outcome was all-cause mortality. We describe age-standardized 30-day, 90-day, 1-, 5- and 10-year mortality for each admission year and estimated the annual percentage change with 95% confidence interval (CI). The association between clinical characteristics and all-cause mortality is reported as hazard ratios (HRs) adjusted for age, sex and calendar year in Cox regression.

RESULTS

The study included 222,832 patients, of whom 127,059 (57.1%) had implicit, 92,928 (41.7%) had explicit, and 2,845 (1.3%) had COVID-19-related sepsis (data from 2020 and 2021). Trends in overall age-standardized 30-day, 90-day, 1- and 5-year mortality decreased by 0.29 (95% CI - 0.39 to - 0.19), 0.43 (95% CI - 0.56 to - 0.29), 0.61 (95% CI - 0.73 to - 0.49) and 0.66 (95% CI - 0.84 to - 0.48) percent per year, respectively. The decrease was observed for all infections sites but was largest among patients with respiratory tract infections. Implicit, explicit and COVID-19-related sepsis had largely similar overall mortality, with explicit sepsis having an adjusted HR of 0.980 (95% CI 0.969 to 0.991) and COVID-19-related sepsis an adjusted HR of 0.916 (95% CI 0.836 to 1.003) compared to implicit sepsis. Patients with respiratory tract infections have somewhat higher mortality than those with other infection sites. Number of comorbidities was positively associated with mortality, but mortality varied considerably between different comorbidities. Similarly, number of acute organ dysfunctions was strongly associated with mortality, whereas the risk varied for each type of organ dysfunction.

CONCLUSION

Overall mortality has declined over the past 14 years among patients with a first sepsis admission. Comorbidity, site of infection, and acute organ dysfunction are patient characteristics that are associated with mortality. This could inform health care workers and raise the awareness toward subgroups of patients that needs particular attention to improve long-term mortality.

Primary Care Physician Follow-Up and 30-Day Readmission After Emergency General Surgery Admissions

Importance

The benefit of primary care physician (PCP) follow-up as a potential means to reduce readmissions in hospitalized patients has been found in other medical conditions and among patients receiving high-risk surgery. However, little is known about the implications of PCP follow-up for patients with an emergency general surgery (EGS) condition.

Objective

To evaluate the association between PCP follow-up and 30-day readmission rates after hospital discharge for an EGS condition.

Design, Setting, and Participants

This cohort study used data from the Centers for Medicare & Medicaid Services Master Beneficiary Summary File, Inpatient, Carrier (Part B), and Durable Medical Equipment files for beneficiaries aged 66 years or older who were hospitalized with an EGS condition that was managed operatively or nonoperatively between September 1, 2016, and November 30, 2018. Eligible patients were enrolled in Medicare fee-for-service, admitted through the emergency department with a primary diagnosis of an EGS condition, and received a general surgery consultation during the admission. Data were analyzed between July 11, 2022, and June 5, 2023.

Exposure

Follow-up with a PCP within 30 days after hospital discharge for the index admission.

Main Outcomes and Measures

The primary outcome was readmission within 30 days after discharge for the index admission. An inverse probability weighted regression model was used to estimate the risk-adjusted association of PCP follow-up with 30-day readmission. The secondary outcome was readmission within 30 days after discharge stratified by treatment type (operative vs nonoperative treatment) during their index admission.

Results

The study included 345 360 Medicare beneficiaries (mean [SD] age, 74.4 [12.0] years; 187 804 females [54.4%]) hospitalized with an EGS condition. Of these, 156 820 patients (45.4%) had a follow-up PCP visit, 108 544 (31.4%) received operative treatment during their index admission, and 236 816 (68.6%) received nonoperative treatment. Overall, 58 253 of 332 874 patients (17.5%) were readmitted within 30 days after discharge for the index admission. After risk adjustment and propensity weighting, patients who had PCP follow-up had 67% lower odds of readmission (adjusted odds ratio [AOR], 0.33; 95% CI, 0.31-0.36) compared with patients without PCP follow-up. After stratifying by treatment type, patients who were treated operatively during their index admission and had subsequent PCP follow-up within 30 days after discharge had 79% reduced odds of readmission (AOR, 0.21; 95% CI, 0.18-0.25); a similar association was seen among patients who were treated nonoperatively (AOR, 0.36; 95% CI, 0.34-0.39). Infectious conditions, heart failure, acute kidney failure, and chronic kidney disease were among the most frequent diagnoses prompting readmission overall and among operative and nonoperative treatment groups.

Conclusions and Relevance

In this cohort study, follow-up with a PCP within 30 days after discharge for an EGS condition was associated with a significant reduction in the adjusted odds of 30-day readmission. This association was similar for patients who received operative care or nonoperative care during their index admission. In patients aged 66 years or older with an EGS condition, primary care coordination after discharge may be an important tool to reduce readmissions.

Emergency Surgery, Multimorbidity and Hospital-Free Days: A Retrospective Observational Study

INTRODUCTION

Analyzing hospital-free days (HFDs) offers a patient-centered approach to health services research. We hypothesized that, within emergency general surgery (EGS), multimorbidity would be associated with fewer HFDs, whether patients were managed operatively or nonoperatively.

METHODS

EGS patients were identified using national Medicare claims data (2015-2018). Patients were classified as multimorbid based on the presence of a Qualifying Comorbidity Set and stratified by treatment: operative (received surgery within 48 h of index admission) and nonoperative. HFDs were calculated through 180 d, beginning on the day of index admission, as days alive and spent outside of a hospital, an Emergency Department, or a long-term acute care facility. Univariate comparisons were performed using Kruskal-Wallis tests and risk-adjusted HFDs were compared between multimorbid and nonmultimorbid patients using multivariable zero-inflated negative binomial regression models.

RESULTS

Among 174,891 operative patients, 45.5% were multimorbid. Among 398,756 nonoperative patients, 59.2% were multimorbid. Multimorbid patients had fewer median HFDs than nonmultimorbid patients among operative and nonoperative cohorts (P < 0.001). At 6 mo, among operative patients, multimorbid patients had 6.5 fewer HFDs (P < 0.001), and among nonoperative patients, multimorbid patients had 7.9 fewer HFDs (P < 0.001). When length of stay was included as a covariate, nonoperative multimorbid patients still had 7.9 fewer HFDs than nonoperative, nonmultimorbid patients (P < 0.001).

CONCLUSIONS

HFDs offer a patient-centered, composite outcome for claims-based analyses. For EGS patients, multimorbidity was associated with less time alive and out of the hospital, especially when patients were managed nonoperatively.

Association between hospital onset of infection and outcomes in sepsis patients - A propensity score matched cohort study based on health claims data in Germany

BACKGROUND

Hospital-acquired infections are a common source of sepsis. Hospital onset of sepsis was found to be associated with higher acute mortality and hospital costs, yet its impact on long-term patient-relevant outcomes and costs is unknown.

OBJECTIVE

We aimed to assess the association between sepsis origin and acute and long-term outcomes based on a nationwide population-based cohort of sepsis patients in Germany.

METHODS

This retrospective cohort study used nationwide health claims data from 23 million health insurance beneficiaries. Sepsis patients with hospital-acquired infections (HAI) were identified by ICD-10-codes in a cohort of adult patients with hospital-treated sepsis between 2013 and 2014. Cases without these ICD-10-codes were considered as sepsis cases with community-acquired infection (CAI) and were matched with HAI sepsis patients by propensity score matching. Outcomes included in-hospital/12-month mortality and costs, as well as readmissions and nursing care dependency until 12 months postsepsis.

RESULTS

We matched 33,110 HAI sepsis patients with 28,614 CAI sepsis patients and 22,234 HAI sepsis hospital survivors with 19,364 CAI sepsis hospital survivors. HAI sepsis patients had a higher hospital mortality than CAI sepsis patients (32.8% vs. 25.4%, RR 1.3, p < .001). Similarly, 12-months postacute mortality was higher (37.2% vs. 30.1%, RR=1.2, p < .001). Hospital and 12-month health care costs were 178% and 22% higher in HAI patients than in CAI patients, respectively. Twelve months postsepsis, HAI sepsis survivors were more often newly dependent on nursing care (33.4% vs. 24.0%, RR=1.4, p < .001) and experienced 5% more hospital readmissions (mean number of readmissions: 2.1 vs. 2.0, p < .001).

CONCLUSIONS

HAI sepsis patients face an increased risk of adverse outcomes both during the acute sepsis episode and in the long-term. Measures to prevent HAI and its progression into sepsis may be an opportunity to mitigate the burden of long-term impairments and costs of sepsis, e.g., by early detection of HAI progressing into sepsis, particularly in normal wards; adequate sepsis management and adherence to sepsis bundles in hospital-acquired sepsis; and an improved infection prevention and control.

The Conditional Effects of Multimorbidity on Operative Versus Nonoperative Management of Emergency General Surgery Conditions: A Retrospective Observational Study Using an Instrumental Variable Analysis

OBJECTIVE

To understand how multimorbidity impacts operative versus nonoperative management of emergency general surgery (EGS) conditions.

BACKGROUND

EGS is a heterogenous field, encompassing operative and nonoperative treatment options. Decision-making is particularly complex for older patients with multimorbidity.

METHODS

Using an instrumental variable approach with near-far matching, this national, retrospective observational cohort study of Medicare beneficiaries examines the conditional effects of multimorbidity, defined using qualifying comorbidity sets, on operative versus nonoperative management of EGS conditions.

RESULTS

Of 507,667 patients with EGS conditions, 155,493 (30.6%) received an operation. Overall, 278,836 (54.9%) were multimorbid. After adjustment, multimorbidity significantly increased the risk of in-hospital mortality associated with operative management for general abdominal patients (+9.8%; P = 0.002) and upper gastrointestinal patients (+19.9%, P < 0.001) and the risk of 30-day mortality (+27.7%, P < 0.001) and nonroutine discharge (+21.8%, P = 0.007) associated with operative management for upper gastrointestinal patients. Regardless of multimorbidity status, operative management was associated with a higher risk of in-hospital mortality among colorectal patients (multimorbid: + 12%, P < 0.001; nonmultimorbid: +4%, P = 0.003), higher risk of nonroutine discharge among colorectal (multimorbid: +42.3%, P < 0.001; nonmultimorbid: +55.1%, P < 0.001) and intestinal obstruction patients (multimorbid: +14.6%, P = 0.001; nonmultimorbid: +14.8%, P = 0.001), and lower risk of nonroutine discharge (multimorbid: -11.5%, P < 0.001; nonmultimorbid: -11.9%, P < 0.001) and 30-day readmissions (multimorbid: -8.2%, P = 0.002; nonmultimorbid: -9.7%, P < 0.001) among hepatobiliary patients.

CONCLUSIONS

The effects of multimorbidity on operative versus nonoperative management varied by EGS condition category. Physicians and patients should have honest conversations about the expected risks and benefits of treatment options, and future investigations should aim to understand the optimal management of multimorbid EGS patients.

Association of Established Primary Care Use With Postoperative Mortality Following Emergency General Surgery Procedures

Importance

Sixty-five million individuals in the US live in primary care shortage areas with nearly one-third of Medicare patients in need of a primary care health care professional. Periodic health examinations and preventive care visits have demonstrated a benefit for surgical patients; however, the impact of primary care health care professional shortages on adverse outcomes from surgery is largely unknown.

Objective

To determine if preoperative primary care utilization is associated with postoperative mortality following an emergency general surgery (EGS) operation among Black and White older adults.

Design, Setting, and Participants

This was a retrospective cohort study that took place at US hospitals with an emergency department. Participants were Medicare patients aged 66 years or older who were admitted from the emergency department for an EGS condition between July 1, 2015, and June 30, 2018, and underwent an operation on hospital day 0, 1, or 2. The analysis was performed during December 2022. Patients were classified into 1 of 5 EGS condition categories based on principal diagnosis codes; colorectal, general abdominal, hepatopancreatobiliary, intestinal obstruction, or upper gastrointestinal. Mixed-effects multivariable logistic regression was used in the risk-adjusted models. An interaction term model was used to measure effect modification by race.

Exposure

Primary care utilization in the year prior to presentation for an EGS operation.

Main Outcome and Measures

In-hospital, 30-day, 60-day, 90-day, and 180-day mortality.

Results

A total of 102 384 patients (mean age, 73.8 [SD, 11.5] years) were included in the study. Of those, 8559 were Black (8.4%) and 93 825 were White (91.6%). A total of 88 340 patients (86.3%) had seen a primary care physician in the year prior to their index hospitalization. After risk adjustment, patients with primary care exposure had 19% lower odds of in-hospital mortality than patients without primary care exposure (odds ratio [OR], 0.81; 95% CI, 0.72-0.92). At 30 days patients with primary care exposure had 27% lower odds of mortality (OR, 0.73; 95% CI, 0.67-0.80). This remained relatively stable at 60 days (OR, 0.75; 95% CI, 0.69-0.81), 90 days (OR, 0.74; 95% CI, 0.69-0.81), and 180 days (OR, 0.75; 95% CI, 0.70-0.81). None of the interactions between race and primary care physician exposure for mortality at any time interval were significantly different.

Conclusions and Relevance

In this observational study of Black and White Medicare patients, primary care utilization had no impact on in-hospital mortality for Black patients, but was associated with decreased mortality for White patients. Primary care utilization was associated with decreased mortality for both Black and White patients at 30, 60, 90 and 180 days.

National Cohort Study of Resource Utilization in Older Adults With Emergency General Surgery Conditions

INTRODUCTION

Prior studies have sought to describe Emergency General Surgery (EGS) burden, but a detailed description of resource utilization for both operative and nonoperative management of EGS conditions has not been undertaken.

METHODS

Patient and hospital characteristics were extracted from Medicare data, 2015-2018. Operations, nonsurgical procedures, and other resources (i.e., radiology) were defined using Current Procedural Terminology codes.

RESULTS

One million eight hundred two thousand five hundred forty-five patients were included in the cohort. The mean age was 74.7 y and the most common diagnoses were upper gastrointestinal. The majority of hospitals were metropolitan (75.1%). Therapeutic radiology services were available in 78.4% of hospitals and operating rooms or endoscopy suites were available in 92.5% of hospitals. There was variability in resource utilization across EGS subconditions, with hepatobiliary (26.4%) and obstruction (23.9%) patients most frequently undergoing operation.

CONCLUSIONS

Treatment of EGS diseases in older adults involves several interventional resources. Changes in EGS models, acute care surgery training, and interhospital care coordination may be beneficial to the treatment of EGS patients.

Conditional Effects of Race on Operative and Nonoperative Outcomes of Emergency General Surgery Conditions

INTRODUCTION

Many emergency general surgery (EGS) conditions can be managed both operatively or nonoperatively; however, it is unknown whether the decision to operate affects Black and White patients differentially.

METHODS

We identified a nationwide cohort of Black and White Medicare beneficiaries, hospitalized for common EGS conditions from July 2015 to June 2018. Using near-far matching to adjust for measurable confounding and an instrumental variable analysis to control for selection bias associated with treatment assignment, we compare outcomes of operative and nonoperative management in a stratified population of Black and White patients. Outcomes included in-hospital mortality, 30-day mortality, nonroutine discharge, and 30-day readmissions. An interaction test based on a t test was used to determine the conditional effects of operative versus nonoperative management between Black and White patients.

RESULTS

A total of 556,087 patients met inclusion criteria, of which 59,519 (10.7%) were Black and 496,568 (89.3%) were White. Overall, 165,932 (29.8%) patients had an operation and 390,155 (70.2%) were managed nonoperatively. Significant outcome differences were seen between operative and nonoperative management for some conditions; however, no significant differences were seen for the conditional effect of race on outcomes.

CONCLUSIONS

The decision to manage an EGS patient operatively versus nonoperatively has varying effects on surgical outcomes. These effects vary by EGS condition. There were no significant conditional effects of race on the outcomes of operative versus nonoperative management among universally insured older adults hospitalized with EGS conditions.

Long-term temporal trends in incidence rate and case fatality of sepsis and COVID-19-related sepsis in Norwegian hospitals, 2008-2021: a nationwide registry study

OBJECTIVES

To estimate temporal trends in incidence rate (IR) and case fatality during a 14-year period from 2008 to 2021, and to assess possible shifts in these trends during the COVID-19 pandemic.

SETTING

All Norwegian hospitals 2008-2021.

PARTICIPANTS

317 705 patients ≥18 year with a sepsis International Classification of Diseases 10th revision code retrieved from The Norwegian Patient Registry.

PRIMARY AND SECONDARY MEASURES

Annual age-standardised IRs with 95% CIs. Poisson regression was used to estimate changes in IRs across time, and logistic regression was used to estimate ORs for in-hospital death.

RESULTS

Among 12 619 803 adult hospitalisations, a total of 317 705 (2.5%) hospitalisations in 222 832 (70.0%) unique patients met the sepsis criteria. The overall age-standardised IR of a first sepsis admission was 246/100 000 (95% CI 245 to 247), whereas the age-standardised IR of all sepsis admissions was 352/100 000 (95% CI 351 to 354). In the period 2009-2019, the annual IR for a first sepsis episode was stable (IR ratio (IRR) per year, 0.999; 95% CI 0.994 to 1.004), whereas for recurrent sepsis the IR increased (annual IRR, 1.048; 95% CI 1.037 to 1.059). During the COVID-19 pandemic, the IRR for a first sepsis was 0.877 (95% CI 0.829 to 0.927) in 2020 and 0.929 (95% CI 0.870 to 0.992) in 2021, and for all sepsis it was 0.870 (95% CI 0.810 to 0.935) in 2020 and 0.908 (95% CI 0.840 to 0.980) in 2021, compared with the previous 11-year period. Case fatality among first sepsis admissions declined in the period 2009-2019 (annual OR 0.954 (95% CI 0.950 to 0.958)), whereas case fatality increased during the COVID-19 pandemic in 2020 (OR 1.061 (95% CI 1.001 to 1.124) and in 2021 (OR 1.164 (95% CI 1.098 to 1.233)).

CONCLUSION

The overall IR of sepsis increased from 2009 to 2019, due to an increasing IR of recurrent sepsis, and indicates that sepsis awareness with updated guidelines and education must continue.

Effect of Serious Mental Illness on Surgical Consultation and Operative Management of Older Adults with Acute Biliary Disease: A Nationwide Study

BACKGROUND

Mental illness is associated with worse outcomes after emergency general surgery. To understand how preoperative processes of care may influence disparate outcomes, we examined rates of surgical consultation, treatment, and operative approach between older adults with and without serious mental illness (SMI).

STUDY DESIGN

We performed a nationwide, retrospective cohort study of Medicare beneficiaries aged 65.5 years or more hospitalized via the emergency department for acute cholecystitis or biliary colic. SMI was defined as schizophrenia spectrum, mood, and/or anxiety disorders. The primary outcome was surgical consultation. Secondary outcomes included operative treatment and surgical approach (laparoscopic vs open). Multivariable logistic regression was used to examine outcomes with adjustment for potential confounders related to patient demographics, comorbidities, and rates of imaging.

RESULTS

Of 85,943 included older adults, 19,549 (22.7%) had SMI. Before adjustment, patients with SMI had lower rates of surgical consultation (78.6% vs 80.2%, p < 0.001) and operative treatment (68.2% vs 71.7%, p < 0.001), but no significant difference regarding laparoscopic approach (92.0% vs 92.1%, p = 0.805). In multivariable regression models with adjustment for confounders, there was no difference in odds of receiving a surgical consultation (odds ratio 0.98 [95% CI 0.93 to 1.03]) or undergoing operative treatment (odds ratio 0.98 [95% CI 0.93 to 1.03]) for patients with SMI compared with those without SMI.

CONCLUSIONS

Older adults with SMI had similar odds of receiving surgical consultation and operative treatment as those without SMI. As such, differences in processes of care that result in SMI-related disparities likely occur before or after the point of surgical consultation in this universally insured patient population.

Operative and Nonoperative Outcomes of Emergency General Surgery Conditions: An Observational Study Using a Novel Instrumental Variable

OBJECTIVE

To determine the effect of operative versus nonoperative management of emergency general surgery conditions on short-term and long-term outcomes.

BACKGROUND

Many emergency general surgery conditions can be managed either operatively or nonoperatively, but high-quality evidence to guide management decisions is scarce.

METHODS

We included 507,677 Medicare patients treated for an emergency general surgery condition between July 1, 2015, and June 30, 2018. Operative management was compared with nonoperative management using a preference-based instrumental variable analysis and near-far matching to minimize selection bias and unmeasured confounding. Outcomes were mortality, complications, and readmissions.

RESULTS

For hepatopancreaticobiliary conditions, operative management was associated with lower risk of mortality at 30 days [-2.6% (95% confidence interval: -4.0, -1.3)], 90 days [-4.7% (-6.50, -2.8)], and 180 days [-6.4% (-8.5, -4.2)]. Among 56,582 intestinal obstruction patients, operative management was associated with a higher risk of inpatient mortality [2.8% (0.7, 4.9)] but no significant difference thereafter. For upper gastrointestinal conditions, operative management was associated with a 9.7% higher risk of in-hospital mortality (6.4, 13.1), which increased over time. There was a 6.9% higher risk of inpatient mortality (3.6, 10.2) with operative management for colorectal conditions, which increased over time. For general abdominal conditions, operative management was associated with 12.2% increased risk of inpatient mortality (8.7, 15.8). This effect was attenuated at 30 days [8.5% (3.8, 13.2)] and nonsignificant thereafter.

CONCLUSIONS

The effect of operative emergency general surgery management varied across conditions and over time. For colorectal and upper gastrointestinal conditions, outcomes are superior with nonoperative management, whereas surgery is favored for patients with hepatopancreaticobiliary conditions. For obstructions and general abdominal conditions, results were equivalent overall. These findings may support patients, clinicians, and families making these challenging decisions.

Clinical, contextual and hospital-level factors associated with escalation and de-escalation of empiric Gram-negative antibiotics among US inpatients

Background

Empiric Gram-negative antibiotics are frequently changed in response to new information. To inform antibiotic stewardship, we sought to identify predictors of antibiotic changes using information knowable before microbiological test results.

Methods

We performed a retrospective cohort study. Survival-time models were used to evaluate clinical factors associated with antibiotic escalation and de-escalation (defined as an increase or decrease, respectively, in the spectrum or number of Gram-negative antibiotics within 5 days of initiation). Spectrum was categorized as narrow, broad, extended or protected. Tjur's D statistic was used to estimate the discriminatory power of groups of variables.

Results

In 2019, 2 751 969 patients received empiric Gram-negative antibiotics at 920 study hospitals. Antibiotic escalation occurred in 6.5%, and 49.2% underwent de-escalation; 8.8% were changed to an equivalent regimen. Escalation was more likely when empiric antibiotics were narrow-spectrum (HR 19.0 relative to protected; 95% CI: 17.9-20.1), broad-spectrum (HR 10.3; 95% CI: 9.78-10.9) or extended-spectrum (HR 3.49; 95% CI: 3.30-3.69). Patients with sepsis present on admission (HR 1.94; 95% CI: 1.91-1.96) and urinary tract infection present on admission (HR 1.36; 95% CI: 1.35-1.38) were more likely to undergo antibiotic escalation than patients without these syndromes. De-escalation was more likely with combination therapy (HR 2.62 per additional agent; 95% CI: 2.61-2.63) or narrow-spectrum empiric antibiotics (HR 1.67 relative to protected; 95% CI: 1.65-1.69). Choice of empiric regimen accounted for 51% and 74% of the explained variation in antibiotic escalation and de-escalation, respectively.

Conclusions

Empiric Gram-negative antibiotics are frequently de-escalated early in hospitalization, whereas escalation is infrequent. Changes are primarily driven by choice of empiric therapy and presence of infectious syndromes.

Return to work after sepsis-a German population-based health claims study

Background

Long-term impairments after sepsis can impede the return to work in survivors. We aimed to describe rates of return to work 6 and 12 months postsepsis.

Methods

This retrospective, population-based cohort study was based on health claims data of the German AOK health insurance of 23.0 million beneficiaries. We included 12-months survivors after hospital-treated sepsis in 2013/2014, who were ≤60 years at the time of the admission and were working in the year presepsis. We assessed the prevalence of return to work (RTW), persistent inability to work and early retirement.

Results

Among 7,370 working age sepsis survivors, 69.2% returned to work at 6 months postsepsis, while 22.8% were on sick leave and 8.0% retired early. At 12 months postsepsis, the RTW rate increased to 76.9%, whereas 9.8% were still on sick leave and 13.3% retired early. Survivors who returned to work had a mean of 70 (SD 93) sick leave days in the 12 months presepsis (median 28 days, IQR 108 days).

Conclusion

One out of four working age sepsis survivors does not resume work in the year postsepsis. Specific rehabilitation and targeted aftercare may be opportunities to reduce barriers to RTW after sepsis.

Comparative Effectiveness of Fludrocortisone and Hydrocortisone vs Hydrocortisone Alone Among Patients With Septic Shock

Importance

Patients with septic shock may benefit from the initiation of corticosteroids. However, the comparative effectiveness of the 2 most studied corticosteroid regimens (hydrocortisone with fludrocortisone vs hydrocortisone alone) is unclear.

Objective

To compare the effectiveness of adding fludrocortisone to hydrocortisone vs hydrocortisone alone among patients with septic shock using target trial emulation.

Design, Setting, and Participants

This retrospective cohort study from 2016 to 2020 used the enhanced claims-based Premier Healthcare Database, which included approximately 25% of US hospitalizations. Participants were adult patients hospitalized with septic shock and receiving norepinephrine who began hydrocortisone treatment. Data analysis was performed from May 2022 to December 2022.

Exposure

Addition of fludrocortisone on the same calendar day that hydrocortisone treatment was initiated vs use of hydrocortisone alone.

Main Outcome and Measures

Composite of hospital death or discharge to hospice. Adjusted risk differences were calculated using doubly robust targeted maximum likelihood estimation.

Results

Analyses included 88 275 patients, 2280 who began treatment with hydrocortisone-fludrocortisone (median [IQR] age, 64 [54-73] years; 1041 female; 1239 male) and 85 995 (median [IQR] age, 67 [57-76] years; 42 136 female; 43 859 male) who began treatment with hydrocortisone alone. The primary composite outcome of death in hospital or discharge to hospice occurred among 1076 (47.2%) patients treated with hydrocortisone-fludrocortisone vs 43 669 (50.8%) treated with hydrocortisone alone (adjusted absolute risk difference, -3.7%; 95% CI, -4.2% to -3.1%; P < .001).

Conclusions and Relevance

In this comparative effectiveness cohort study among adult patients with septic shock who began hydrocortisone treatment, the addition of fludrocortisone was superior to hydrocortisone alone.

National incidence and mortality of hospitalized sepsis in China

BACKGROUND

Sepsis is a leading cause of preventable death around the world. Population-based estimation of sepsis incidence is lacking in China. In this study, we aimed to estimate the population-based incidence and geographic variation of hospitalized sepsis in China.

METHODS

We retrospectively identified hospitalized sepsis from the nationwide National Data Center for Medical Service (NDCMS) and the National Mortality Surveillance System (NMSS) by ICD-10 codes for the period from 2017 to 2019. In-hospital sepsis case fatality and mortality rate were calculated to extrapolate the national incidence of hospitalized sepsis. The geographic distribution of hospitalized sepsis incidence was examined using Global Moran's Index.

RESULTS

We identified 9,455,279 patients with 10,682,625 implicit-coded sepsis admissions in NDCMS and 806,728 sepsis-related deaths in NMSS. We estimated that the annual standardized incidence of hospitalized sepsis was 328.25 (95% CI 315.41-341.09), 359.26 (95% CI 345.4-373.12) and 421.85 (95% CI 406.65-437.05) cases per 100,000 in 2017, 2018 and 2019, respectively. We observed 8.7% of the incidences occurred among neonates less than 1 year old, 11.7% among children aged 1-9 years, and 57.5% among elderly older than 65 years. Significant spatial autocorrelation for incidence of hospitalized sepsis was observed across China (Moran's Index 0.42, p = 0.001; 0.45, p = 0.001; 0.26, p = 0.011 for 2017, 2018, 2019, respectively). Higher number of hospital bed supply and higher disposable income per capita were significantly associated with a higher incidence of hospitalized sepsis.

CONCLUSION

Our study showed a greater burden of sepsis hospitalizations than previous estimated. The geographical disparities suggested more efforts were needed in prevention of sepsis.

Cardiovascular Events Among Survivors of Sepsis Hospitalization: A Retrospective Cohort Analysis

Background Sepsis is associated with an elevated risk of late cardiovascular events among hospital survivors. Methods and Results We included OptumLabs Data Warehouse patients from 2009 to 2019 who survived a medical/nonsurgical hospitalization lasting at least 2 nights. The association between sepsis during hospitalization, based on explicit and implicit discharge International Classification of Diseases, Ninth Revision (ICD-9)/Tenth Revision (ICD-10) diagnosis codes, with subsequent death and rehospitalization was analyzed using Kaplan-Meier survival analysis and multivariable Cox proportional-hazards models. The study population included 2 258 464 survivors of nonsurgical hospitalization (5 396 051 total patient-years of follow-up). A total of 808 673 (35.8%) patients had a sepsis hospitalization, including implicit sepsis only in 448 644, explicit sepsis only in 124 841, and both in 235 188. Patients with sepsis during hospitalization had an elevated risk of all-cause mortality (adjusted hazard ratio [HR], 1.27 [95% CI, 1.25-1.28]; P<0.001), all-cause rehospitalization (adjusted HR, 1.38 [95% CI, 1.37-1.39]; P<0.001), and cardiovascular hospitalization (adjusted HR, 1.43 [95% CI, 1.41-1.44]; P<0.001), especially heart failure hospitalization (adjusted HR, 1.51 [95% CI, 1.49-1.53]). Patients with implicit sepsis had higher risk than those with explicit sepsis. A sensitivity analysis using the first hospitalization yielded concordant results for cardiovascular hospitalization (adjusted HR, 1.78 [95% CI, 1.76-1.78]; P<0.001), as did a propensity-weighted analysis (adjusted HR, 1.52 [95% CI, 1.50-1.54]; P<0.001). Conclusions Survivors of sepsis hospitalization are at elevated risk of early and late post-discharge death as well as cardiovascular and non-cardiovascular rehospitalization. This hazard spans the spectrum of cardiovascular events and may suggest that sepsis is an important cardiovascular risk factor.

Defining Posttraumatic Sepsis for Population-Level Research

IMPORTANCE

Multiple classification methods are used to identify sepsis from existing data. In the trauma population, it is unknown how administrative methods compare with clinical criteria for sepsis classification.

OBJECTIVES

To characterize the agreement between 3 approaches to sepsis classification among critically ill patients with trauma and compare the sepsis-associated risk of adverse outcomes when each method was used to define sepsis.

DESIGN, SETTING, AND PARTICIPANTS

This retrospective cohort study used data collected between January 1, 2012, and December 31, 2020, from patients aged 16 years or older with traumatic injury, admitted to the intensive care unit of a single-institution level 1 trauma center and requiring invasive mechanical ventilation for at least 3 days. Statistical analysis was conducted from August 1, 2021, to March 31, 2022.

EXPOSURE

Hospital-acquired sepsis, as classified by 3 methods: a novel automated clinical method based on data from the electronic health record, the National Trauma Data Bank (NTDB), and explicit and implicit medical billing codes.

MAIN OUTCOMES AND MEASURES

The primary outcomes were chronic critical illness and in-hospital mortality. Secondary outcomes included number of days in an intensive care unit, number of days receiving mechanical ventilation, discharge to a skilled nursing or long-term care facility, and discharge to home without assistance.

RESULTS

Of 3194 patients meeting inclusion criteria, the median age was 49 years (IQR, 31-64 years), 2380 (74%) were male, and 2826 (88%) sustained severe blunt injury (median Injury Severity Score, 29 [IQR, 21-38]). Sepsis was identified in 747 patients (23%) meeting automated clinical criteria, 118 (4%) meeting NTDB criteria, and 529 (17%) using medical billing codes. The Light κ value for 3-way agreement was 0.16 (95% CI, 0.14-0.19). The adjusted relative risk of chronic critical illness was 9.9 (95% CI, 8.0-12.3) for sepsis identified by automated clinical criteria, 5.0 (95% CI, 3.4-7.3) for sepsis identified by the NTDB, and 4.5 (95% CI, 3.6-5.6) for sepsis identified using medical billing codes. The adjusted relative risk for in-hospital mortality was 1.3 (95% CI, 1.0-1.6) for sepsis identified by automated clinical criteria, 2.7 (95% CI, 1.7-4.3) for sepsis identified by the NTDB, and 1.0 (95% CI, 0.7-1.2) for sepsis identified using medical billing codes.

CONCLUSIONS AND RELEVANCE

In this cohort study of critically ill patients with trauma, administrative methods misclassified sepsis and underestimated the incidence and severity of sepsis compared with an automated clinical method using data from the electronic health record. This study suggests that an automated approach to sepsis classification consistent with Third International Consensus Definitions for Sepsis and Septic Shock (Sepsis-3) clinical criteria is feasible and may improve existing approaches to health services and population-based research in this population.

Artificial and human intelligence for early identification of neonatal sepsis

Artificial intelligence may have a role in the early detection of sepsis in neonates. Machine learning can identify patterns that predict high or increasing risk for clinical deterioration from a sepsis-like illness. In developing this potential addition to NICU care, careful consideration should be given to the data and methods used to develop, validate, and evaluate prediction models. When an AI system alerts clinicians to a change in a patient's condition that warrants a bedside evaluation, human intelligence and experience come into play to determine an appropriate course of action: evaluate and treat or wait and watch closely. With intelligently developed, validated, and implemented AI sepsis systems, both clinicians and patients stand to benefit. IMPACT: This narrative review highlights the application of AI in neonatal sepsis prediction. It describes issues in clinical prediction model development specific to this population. This article reviews the methods, considerations, and literature on neonatal sepsis model development and validation. Challenges of AI technology and potential barriers to using sepsis AI systems in the NICU are discussed.

Outcomes Associated With Rural Emergency Department Provider-to-Provider Telehealth for Sepsis Care: A Multicenter Cohort Study

STUDY OBJECTIVE

To test the hypothesis that provider-to-provider tele-emergency department care is associated with more 28-day hospital-free days and improved Surviving Sepsis Campaign (SSC) guideline adherence in rural emergency departments (EDs).

METHODS

Multicenter (n=23), propensity-matched, cohort study using medical records of patients with sepsis from rural hospitals in an established, on-demand, rural video tele-ED network in the upper Midwest between August 2016 and June 2019. The primary outcome was 28-day hospital-free days, with secondary outcomes of 28-day inhospital mortality and SSC guideline adherence.

RESULTS

A total of 1,191 patients were included in the analysis, with tele-ED used for 326 (27%). Tele-ED cases were more likely to be transferred to another hospital (88% versus 8%, difference 79%, 95% confidence interval [CI] 75% to 83%). After matching and regression adjustment, tele-ED cases did not have more 28-day hospital-free days (difference 0.07 days more for tele-ED, 95% CI -0.04 to 0.17) or 28-day inhospital mortality (adjusted odds ratio [aOR] 0.51, 95% CI 0.16 to 1.60). Adherence with both the SSC 3-hour bundle (aOR 0.59, 95% CI 0.28 to 1.22) and complete bundle (aOR 0.45, 95% CI 0.02 to 11.60) were similar. An a priori-defined subgroup of patients treated by advanced practice providers suggested that the mortality was lower in the cohort with tele-ED use (aOR 0.11, 95% CI 0.02 to 0.73) despite no significant difference in complete SSC bundle adherence (aOR 2.88, 95% CI 0.52 to 15.86).

CONCLUSION

Rural emergency department patients treated with provider-to-provider tele-ED care in a mature network appear to have similar clinical outcomes to those treated without.

Analyzing Impact of Multimorbidity on Long-Term Outcomes after Emergency General Surgery: A Retrospective Observational Cohort Study

BACKGROUND

Little is known about the impact of multimorbidity on long-term outcomes for older emergency general surgery patients.

STUDY DESIGN

Medicare beneficiaries, age 65 and older, who underwent operative management of an emergency general surgery condition were identified using Centers for Medicare & Medicaid claims data. Patients were classified as multimorbid based on the presence of a Qualifying Comorbidity Set (a specific combination of comorbid conditions known to be associated with increased risk of in-hospital mortality in the general surgery setting) and compared with those without multimorbidity. Risk-adjusted outcomes through 180 days after discharge from index hospitalization were calculated using linear and logistic regressions.

RESULTS

Of 174,891 included patients, 45.5% were identified as multimorbid. Multimorbid patients had higher rates of mortality during index hospitalization (5.9% vs 0.7%, odds ratio [OR] 3.05, p < 0.001) and through 6 months (17.1% vs 3.4%, OR 2.33, p < 0.001) after discharge. Multimorbid patients experienced higher rates of readmission at 1 month (22.9% vs 11.4%, OR 1.48, p < 0.001) and 6 months (38.2% vs 21.2%, OR 1.48, p < 0.001) after discharge, lower rates of discharge to home (42.5% vs 74.2%, OR 0.52, p < 0.001), higher rates of discharge to rehabilitation/nursing facility (28.3% vs 11.3%, OR 1.62, p < 0.001), greater than double the use of home oxygen, walker, wheelchair, bedside commode, and hospital bed (p < 0.001), longer length of index hospitalization (1.33 additional in-patient days, p < 0.001), and higher costs through 6 months ($5,162 additional, p < 0.001).

CONCLUSIONS

Older, multimorbid patients experience worse outcomes, including survival and independent function, after emergency general surgery than nonmultimorbid patients through 6 months after discharge from index hospitalization. This information is important for setting recovery expectations for high-risk patients to improve shared decision-making.

Identifying infected patients using semi-supervised and transfer learning

Objectives

Early identification of infection improves outcomes, but developing models for early identification requires determining infection status with manual chart review, limiting sample size. Therefore, we aimed to compare semi-supervised and transfer learning algorithms with algorithms based solely on manual chart review for identifying infection in hospitalized patients.

Materials and Methods

This multicenter retrospective study of admissions to 6 hospitals included “gold-standard” labels of infection from manual chart review and “silver-standard” labels from nonchart-reviewed patients using the Sepsis-3 infection criteria based on antibiotic and culture orders. “Gold-standard” labeled admissions were randomly allocated to training (70%) and testing (30%) datasets. Using patient characteristics, vital signs, and laboratory data from the first 24 hours of admission, we derived deep learning and non-deep learning models using transfer learning and semi-supervised methods. Performance was compared in the gold-standard test set using discrimination and calibration metrics.

Results

The study comprised 432 965 admissions, of which 2724 underwent chart review. In the test set, deep learning and non-deep learning approaches had similar discrimination (area under the receiver operating characteristic curve of 0.82). Semi-supervised and transfer learning approaches did not improve discrimination over models fit using only silver- or gold-standard data. Transfer learning had the best calibration (unreliability index P value: .997, Brier score: 0.173), followed by self-learning gradient boosted machine (P value: .67, Brier score: 0.170).

Discussion

Deep learning and non-deep learning models performed similarly for identifying infection, as did models developed using Sepsis-3 and manual chart review labels.

Conclusion

In a multicenter study of almost 3000 chart-reviewed patients, semi-supervised and transfer learning models showed similar performance for model discrimination as baseline XGBoost, while transfer learning improved calibration.

Incidence, recurring admissions and mortality of severe bacterial infections and sepsis over a 22-year period in the population-based HUNT study

Purpose

Severe bacterial infections are important causes of hospitalization and loss of health worldwide. In this study we aim to characterize the total burden, recurrence and severity of bacterial infections in the general population during a 22-year period.

Methods

We investigated hospitalizations due to bacterial infection from eight different foci in the prospective population-based Trøndelag Health Study (the HUNT Study), where all inhabitants aged ≥ 20 in a Norwegian county were invited to participate. Enrollment was between 1995 and 1997, and between 2006 and 2008, and follow-up ended in February 2017. All hospitalizations, positive blood cultures, emigrations and deaths in the follow-up period were captured through registry linkage.

Results

A total of 79,393 (69.5% and 54.1% of the invited population) people were included, of which 42,237 (53%) were women and mean age was 48.5 years. There were 37,298 hospitalizations due to infection, affecting 15,496 (22% of all included) individuals. The median time of follow-up was 20 years (25^th percentile 9.5–75^th percentile 20.8). Pneumonia and urinary tract infections were the two dominating foci with incidence rates of 639 and 550 per 100,000 per year, respectively, and with increasing incidence with age. The proportion of recurring admissions ranged from 10.0% (central nervous system) to 30.0% (pneumonia), whilst the proportion with a positive blood culture ranged from 4.7% (skin- and soft tissue infection) to 40.9% (central nervous system). The 30-day mortality varied between 3.2% (skin- and soft tissue infection) and 20.8% (endocarditis).

Conclusions

In this population-based cohort, we observed a great variation in the incidence, positive blood culture rate, recurrence and mortality between common infectious diseases. These results may help guide policy to reduce the infectious disease burden in the population.

Prevalence and Outcomes of Previously Healthy Adults Among Patients Hospitalized With Community-Onset Sepsis

BACKGROUND

Devastating cases of sepsis in previously healthy patients have received widespread attention and have helped to catalyze state and national mandates to improve sepsis detection and care. However, it is unclear what proportion of patients hospitalized with sepsis previously were healthy and how their outcomes compare with those of patients with comorbidities.

RESEARCH QUESTION

Among adults hospitalized with community-onset sepsis, how many previously were healthy and how do their outcomes compare with those of patients with comorbidities?

STUDY DESIGN AND METHODS

We retrospectively identified all adults with community-onset sepsis hospitalized in 373 US hospitals from 2009 through 2015 using clinical indicators of presumed infection and organ dysfunction (Centers for Disease Control and Prevention's Adult Sepsis Event criteria). Comorbidities were identified using International Classification of Diseases, Ninth Revision, Clinical Modification codes. We applied generalized linear mixed models to measure the associations between the presence or absence of comorbidities and short-term mortality (in-hospital death or discharge to hospice), adjusting for severity of illness on admission.

RESULTS

Of 6,715,286 hospitalized patients, 337,983 (5.0%) were hospitalized with community-onset sepsis. Most patients with sepsis (329,052 [97.4%]) had received a diagnosis of at least one comorbidity; only 2.6% previously were healthy. Patients with sepsis who previously were healthy were younger than those with comorbidities (mean age, 58.0 ± 19.8 years vs 67.0 ± 16.5 years), were less likely to require ICU care on admission (37.9% vs 50.5%), and were more likely to be discharged home (57.9% vs 45.6%), rather than to subacute facilities (16.3% vs 30.8%), but showed higher short-term mortality rates (22.8% vs 20.8%; P < .001 for all). The association between previously healthy status and higher short-term mortality persisted after risk adjustment (adjusted OR, 1.99; 95% CI, 1.87-2.13).

INTERPRETATION

The vast majority of patients hospitalized with community-onset sepsis harbor pre-existing comorbidities. However, previously healthy patients may be more likely to die when they seek treatment at the hospital with sepsis compared with patients with comorbidities. These findings underscore the importance of early sepsis recognition and treatment for all patients.