To the Editor
Patients are frequently re-hospitalized in the 90 days after severe sepsis1. Little is known, however, about the reasons for this utilization and whether it can be reduced. We sought to determine the most common readmission diagnoses after severe sepsis, the extent to which readmissions may be potentially preventable by post-hospitalization ambulatory care, and whether the pattern of readmission diagnoses differs compared to that of other acute medical conditions.
Methods
We studied participants in the nationally representative U.S. Health and Retirement Study, a multistage probability sample of households with adults ages 50 years and older, linked to Medicare claims (1998-2010)2. We identified hospitalizations with severe sepsis using a validated approach that requires ICD-9-CM codes for both infection and acute organ dysfunction3,4. We matched hospitalizations for severe sepsis to hospitalizations for 15 common acute medical conditions (Table 1 legend) one-to-one by age, gender, post-discharge comorbidity burden (Charlson Comorbidity Index), pre-hospitalization functional disability (limitations of activities and instrumental activities of daily living), and length of hospitalization using coarsened exact matching5.
Table 1.
Top 10 Readmission Diagnoses After Severe Sepsis Hospitalization
| Rank | Diagnosis | N (%) (95%CI) of Survivors with a 90-Day Readmission for this Diagnosis | ||
|---|---|---|---|---|
| Severe Sepsis Survivors (N=2,617) | Survivors of Matched Hospitalizations for Other Acute Medical Conditions (N=2,617) | McNemar P | ||
| 1 | Sepsis | 167 (6.4%) (5.4%-7.3%) | 73 (2.8%) (2.2%-3.4%) | <0.001 |
| 2 | Congestive Heart Failure | 144 (5.5%) (4.6%-6.4%) | 204 (7.8%) (6.8%-8.8%) | 0.001 |
| 3 | Pneumonia | 92 (3.5%) (2.8%-4.2%) | 85 (3.3%) (2.6%-3.9%) | 0.58 |
| 4 | Acute Renal Failure | 87 (3.3%) (2.6%-4.0%) | 30 (1.2%) (0.7%-1.6%) | <0.001 |
| 5 | Rehabilitation | 74 (2.8%) (2.2%-3.5%) | 120 (4.6%) (3.8%-5.4%) | 0.001 |
| 6 | Respiratory Failure | 65 (2.5%) (1.9%-3.1%) | 38 (1.5%) (1.0%-1.9%) | 0.007 |
| 7 | Complication of Device, Implant, or Graft | 52 (2.0%) (1.5%-2.5%) | 59 (2.3%) (1.7%-2.8%) | 0.50 |
| 8 | COPD Exacerbation | 49 (1.9%) (1.4%-2.4%) | 41 (1.6%) (1.1%-2.0%) | 0.40 |
| 9 | Aspiration Pneumonitis | 47 (1.8%) (1.3%-2.3%) | 31 (1.2%) (0.8%-1.6%) | 0.06 |
| 10 | Urinary Tract Infection | 44 (1.7%) (1.2%-2.2%) | 47 (1.8%) (1.3%-2.3%) | 0.75 |
The top 10 readmission diagnoses accounted for 51.5% of all readmissions in the 90 days after severe sepsis hospitalization. Principal diagnoses for the matched hospitalizations for other acute medical conditions were the 15 most common acute hospitalization non-sepsis diagnoses in our cohort: heart failure; pneumonia; cardiac arrhythmia; COPD exacerbation; acute myocardial infarction; acute cerebrovascular disease; complication of a device, implant, or graft; chest pain; fluid or electrolyte disorder; urinary tract infection; hip fracture; gastrointestinal hemorrhage; complication or surgical or medical care; syncope; and diabetes with complication.
We measured the rate and 95%CI of 90-day readmissions. Using Healthcare Cost & Utilization Project's Clinical Classification Software, we determined the most common readmission diagnoses. To gauge what proportion of re-hospitalizations may be potentially preventable, we measured ambulatory care sensitive conditions (ACSCs)—diagnoses for which effective outpatient care can reduce hospitalization rates6. We used ACSCs identified by Agency for Healthcare Quality & Research6, and an expanded definition also including sepsis, skin/soft tissue infection, acute renal failure, and aspiration pneumonitis, all of which could plausibly be prevented or treated early to avoid re-hospitalization.
We compared readmission rates using McNemar chi-squared tests with significance at p<0.001 (two-sided) given multiple comparisons. The University of Michigan IRB approved this study; patients provided oral informed consent at enrollment and for Medicare linkage.
Results
We identified 3,494 severe sepsis hospitalizations, of which 2,843(81.4%) survived to discharge. Of these, 2,617(92.1%) were matched to hospitalizations for other acute medical conditions. The cohort's mean age was 78.9±8.9 years, 57.3% were female, and they had some pre-existing functional disability [median=1 limitation (IQR:0-4)]. At discharge, patients had moderate comorbidity burden [median Charlson Index=6 (IQR:3-8)]. Median(IQR) hospitalization length was 7(4-11) days. Age, gender, comorbidity burden, functional status, and hospitalization length did not differ between severe sepsis and matched acute medical conditions, p>0.05 for each.
1,115(42.7%) severe sepsis survivors were re-hospitalized within 90 days. The 10 most common readmission diagnoses following severe sepsis included several ACSCs: heart failure, pneumonia, COPD exacerbation, and urinary infection (Table 1). Collectively, ACSCs accounted for 22.2%(95%CI: 20.3%-24.5%) of 90-day readmissions. Using the expanded definition, ACSCs accounted for 41.6%(95%CI: 39.1%-44.1%) of 90-day readmissions after severe sepsis.
Patterns of readmission differed between survivors of severe sepsis and matched acute medical conditions (Table 1, Figure 1); rates of readmission for sepsis and renal failure were higher and accounted for a greater proportion of the total readmissions after severe sepsis. Readmissions for a primary diagnosis of infection (sepsis, pneumonia, urinary tract, and skin/soft tissue infection) occurred in 11.9%(95%CI: 10.6%-13.1%) of severe sepsis survivors, compared to 8.0%(95%CI: 7.0%-9.1%) of matched acute medical conditions, p<0.001. Readmissions for ACSCs were more common after severe sepsis versus matched acute conditions [21.6%(95%CI: 20.0%-23.2%) versus 19.1% (95%CI: 17.7%-20.7%), p=0.022] and accounted for a greater proportion of all 90-day readmissions after severe sepsis [41.6% (95%CI: 39.2%-44.1%) versus 37.1%(95%CI: 34.8%-39.5%) of readmissions, p=0.009].
Figure 1. Total and Potentially Preventable 90-Day Readmissions among Survivors of Severe Sepsis and Matched Hospitalizations for Acute Medical Conditions.
Potentially preventable readmission diagnoses include pneumonia, hypertension, dehydration, asthma, urinary tract infection, chronic obstructive pulmonary disease exacerbation, perforated appendix, diabetes, angina, congestive heart failure, sepsis, acute renal failure, skin/soft tissue infection, and aspiration pneumonitis. 95% CI for readmission proportions are depicted in the shaded areas.
Discussion
Hospitalizations in the 90 days after severe sepsis are common, and 42% occurred for diagnoses that could potentially be prevented or treated early to avoid hospitalization compared to 37% after matched acute medical conditions. A limitation of the present study is that we inferred potential preventability of re-hospitalizations by measuring readmissions for ACSCs. Nonetheless, the high prevalence and great concentration of specific diagnoses during this early post-discharge period suggests that further study is warranted of the feasibility and potential benefit of post-discharge interventions tailored to patients’ personalized risk for a limited number of common conditions.
Acknowledgements
The Health and Retirement Study is funded by the National Institute on Aging and performed at the Institute for Social Research, University of Michigan. We appreciate the expert programming of Ryan McCammon, MS, and Vanessa Dickerman, MS, at the University of Michigan. They were not compensated for their contributions besides salary. Dr. Prescott has had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Funding: This work was supported by grants T32 HL007749, R01 AG0030155, R21 AG044752, and U01 AG09740 from the National Institutes of Health and IIR 11-109 from the Department of Veterans Affairs Health Services Research & Development Service. The views expressed in this article are those of the authors and do not necessarily reflect the position or policy of the Department of Veterans Affairs or the US government.
Role of Funders: The funders had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.
Footnotes
Author Contributions: HCP designed the study, analyzed the data, interpreted the data, and drafted the manuscript. KML acquired the data, interpreted the data, and revised the manuscript critically for intellectual content. TJI interpreted the data and revised the manuscript critically for intellectual content.
Declarations: The authors received funding from the National Institutes of Health and the Department of Veterans Affairs. The authors have no other conflicts of interest.
References
- 1.Prescott HC, Langa KM, Liu V, Escobar GJ, Iwashyna TJ. Increased 1-year healthcare use in survivors of severe sepsis. Am. J. Respir. Crit. Care Med. 2014 Jul 1;190(1):62–69. doi: 10.1164/rccm.201403-0471OC. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Sonnega A, Faul JD, Ofstedal MB, Langa KM, Phillips JWR, Weir DR. Cohort Profile: the Health and Retirement Study (HRS). Int. J. Epidemiol. 2014 doi: 10.1093/ije/dyu067. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Angus DC, Linde-Zwirble WT, Lidicker J, Clermont G, Carcillo J, Pinsky MR. Epidemiology of severe sepsis in the United States: analysis of incidence, outcome, and associated costs of care. Crit. Care Med. 2001 Jul;29(7):1303–1310. doi: 10.1097/00003246-200107000-00002. [DOI] [PubMed] [Google Scholar]
- 4.Iwashyna TJ, Odden A, Rohde J, et al. Identifying patients with severe sepsis using administrative claims: patient-level validation of the angus implementation of the international consensus conference definition of severe sepsis. Med. Care. 2014 Jun;52(6):e39–43. doi: 10.1097/MLR.0b013e318268ac86. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Blackwell Matthew, Iacus Stefano, King Gary, Porro Giuseppe. cem: Coarsened Exact Mzatching in Stata. The Stata Journal. 2009;9(4):524–546. [Google Scholar]
- 6.Agency for Healthcare Research and Quality [September 30, 2014];Guide to Prevention Quality Indicators. Updated March, 2007. Available at: http://www.qualityindicators.ahrq.gov/Downloads/Modules/PQI/V31/pqi_guide_v31.pdf.