Accuracy of administrative data for identification of patients with infective endocarditis

doi:10.1016/j.ijcard.2016.09.030

International Journal of Cardiology

Volume 224, 1 December 2016, Pages 162-164

https://doi.org/10.1016/j.ijcard.2016.09.030 Get rights and content

Abstract

Background

Infective endocarditis is associated with high morbidity and mortality rates that have plateaued over recent decades. Research to improve outcomes for these patients is limited by the rarity of this condition. Therefore, we sought to validate administrative database codes for the diagnosis of infective endocarditis.

Methods

We conducted a retrospective validation study of International Classification of Diseases (ICD-10-CM) codes for infective endocarditis against clinical Duke criteria (definite and probable) at a large acute care hospital between October 1, 2013 and June 30, 2015. To identify potential cases missed by ICD-10-CM codes, we also screened the hospital's valvular heart surgery database and the microbiology laboratory database (the latter for patients with bacteremia due to organisms commonly causing endocarditis).

Results

Using definite Duke criteria or probable criteria with clinical suspicion as the reference standard, the ICD-10-CM codes had a sensitivity (SN) of 0.90 (95% confidence interval (CI), 0.81–0.95), specificity (SP) of 1 (95% CI, 1–1), positive predictive value (PPV) of 0.78 (95% CI, 0.68–0.85) and negative predictive value (NPV) of 1 (95% CI, 1–1). Restricting the case definition to definite Duke criteria resulted in an increase in SN to 0.95 (95% CI, 0.86–0.99) and a decrease in PPV to 0.6 (95% CI, 0.49–0.69), with no change in specificity.

Conclusion

ICD-10-CM codes can accurately identify patients with infective endocarditis, and so administrative databases offer a potential means to study this infection over large jurisdictions, and thereby improve the prediction, diagnosis, treatment and prevention of this rare but serious infection.

Introduction

Infective endocarditis is a serious condition that continues to be associated with significant morbidity and mortality. Despite innovations in the detection and management of this disease, mortality has plateaued in recent decades; in-hospital mortality ranges from 15 to 30%, while five-year mortality is approximately 40% [1], [2], [3]. Non-lethal complications are common and can be debilitating, often necessitating prolonged, intensive and costly medical care [4], [5], [6].

These persistent challenges underscore the need for ongoing research of the epidemiology and outcomes of patients with infective endocarditis. Administrative databases represent a comprehensive and easily accessible source of patient information, and have been widely used to study this condition [6], [8], [9], [10], [11], [12], [13]. In particular, with recent changes in the guidelines for antibiotic prophylaxis to prevent infective endocarditis, administrative databases have been used to evaluate the impact of such changes on disease incidence at the population level [14], [15], [16]. However, the utility of such research depends on accurate application of diagnostic codes. We therefore sought to validate the accuracy of the International Classification of Diseases 10th revision (ICD-10-CM) codes for infective endocarditis against a clinical reference standard.

Section snippets

General study design and data sources

We conducted a retrospective validation study of administrative codes for infective endocarditis using medical records from Sunnybrook Health Sciences Centre (SHSC), a large, acute care teaching hospital in Toronto, Ontario, Canada. The study was approved by the SHSC Research Ethics Board. We identified all patients who were discharged from SHSC with ICD-10-CM codes for infective endocarditis in any diagnostic field between October 1, 2013 and June 30, 2015. I33 (acute and subacute infective

Results

From October 1, 2013 to June 30, 2015, 119 discharges from SHSC received an ICD-10-CM code for infective endocarditis (I33, I38, I39), representing 99 unique patients. During the same period, there were 35,172 unique patients admitted to a medical or surgical service. The microbiology database yielded 477 patients with one or more positive blood cultures for microorganisms that commonly cause infective endocarditis, of whom 10 were included in the study. The valve surgery database included 262

Discussion

Administrative databases are critical to clinical epidemiology research involving rare conditions such as infective endocarditis, because many important observational research questions can only be addressed with large sample sizes that are unlikely to be accrued in any one hospital. However, to utilize administrative databases for this purpose, the diagnostic codes must be sufficiently accurate to detect patients with the condition of interest. This study evaluated the diagnostic performance

Conflict of interest

The current study was unfunded. The authors have no relevant conflicts of interest to disclose.

Funding sources

Nick Daneman is supported by a clinician scientist salary award from the Canadian Institutes of Health Research. This research did not receive any specific grant from funding agencies in the public, commercial or not-for-profit sectors.

Disclosures

None.

Acknowledgements

We would like to acknowledge the Sunnybrook Health Sciences Centre Divisions of Infectious Diseases, Cardiology, Cardiac Surgery and Neurology, and the Department of Critical Care Medicine, for enabling our multi-disciplinary case conferencing quality improvement initiative for patients with infective endocarditis. This validation study was nested within this quality improvement initiative.

References (20)

B. Bikdeli et al.
Trends in hospitalization rates and outcomes of endocarditis among Medicare beneficiaries
J. Am. Coll. Cardiol.
(2013)
S. Pant et al.
Trends in infective endocarditis incidence, microbiology, and valve replacement in the United States from 2000 to 2011
J. Am. Coll. Cardiol.
(2015)
M.J. Dayer et al.
Incidence of infective endocarditis in England, 2000–13: a secular trend, interrupted time-series analysis
Lancet
(2015)
S. Schneeweiss et al.
Veteran's affairs hospital discharge databases coded serious bacterial infections accurately
J. Clin. Epidemiol.
(2007)
E.I. Benchimol et al.
Development and use of reporting guidelines for assessing the quality of validation studies of health administrative data
J. Clin. Epidemiol.
(2011)
B. Hoen et al.
Clinical practice. Infective endocarditis
N. Engl. J. Med.
(2013)
C. Selton-Suty et al.
Preeminence of Staphylococcus aureus in infective endocarditis: a 1-year population-based survey
Clin. Infect. Dis.
(2012)
D.R. Murdoch et al.
Clinical presentation, etiology, and outcome of infective endocarditis in the 21st century: the International Collaboration on Endocarditis-Prospective Cohort Study
Arch. Intern. Med.
(2009)
G. Karth et al.
Complicated infective endocarditis necessitating ICU admission: clinical course and prognosis
Crit. Care
(2002)
P.B. Francioli et al.
Outpatient treatment of infective endocarditis
Clin. Microbiol. Infect.
(1998)

There are more references available in the full text version of this article.

Cited by (43)

Trends in epidemiology, surgical management, and prognosis of infective endocarditis during the XXI century in Spain: A population-based nationwide study
2024, Journal of Infection and Public Health
Few population-based studies have evaluated the epidemiology of infective endocarditis (IE). Changes in population demographics and guidelines on IE may have affected both the incidence and outcomes of IE. Therefore, the aim of our study is to provide contemporary population-based epidemiological data of IE in Spain.
Retrospective nationwide observational study using data from the Spanish National Health System Discharge Database. We included all patients hospitalized with IE from January 2000 to December 2019.
A total of 64,550 IE episodes were included. The incidence of IE rose from 5.25 cases/100,000 person-year in 2000 to 7.21 in 2019, with a 2% annual percentage change (95% CI 1.3–2.6). IE incidence was higher among those aged 85 or older (43.5 cases/100.000 person-years). Trends across the study period varied with sex and age.
Patients with IE were progressively older (63.9 years in 2000–2004 to 70.0 in 2015–2019, p < 0.001) and had more frequent comorbidities and predispositions, including, previous valvular prosthesis (12.1% vs 20.9%, p < 0.001). After adjustment, a progressive reduction in mortality was noted including in 2015–2019 compared to 2010–2014 (adjusted odds ratio 0.93, 95% confident interval 0.88–0.99, p = 0.023)., which was associated with more frequent cardiac surgery in recent years (15.1% in 2010–2014 vs 19.9% in 2015–2019).
In Spain, the incidence of IE has increased during the XXI century, with a more pronounced increase in elderly individuals. Adjusted-mortality decreased over the years, which could be related to a higher percentage of surgery. Our results highlight the changing epidemiology of IE.
Updated estimates for the burden of chronic limb-threatening ischemia in the Medicare population
2023, Journal of Vascular Surgery
Estimates of chronic limb-threatening ischemia (CLTI) based on diagnosis codes of the International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) suggest a prevalence of 0.23%-0.32% and incidence of 0.20%-0.26% among Medicare patients. ICD-10-CM includes 144 CLTI diagnosis codes, allowing improved specificity in identifying affected patients. We sought to use ICD-10-CM diagnosis codes to determine the prevalence of CLTI among Medicare patients and describe the patient cohort affected by this condition.
Using two years of data from Centers for Medicare and Medicaid Services, we identified all patients that had at least one CLTI diagnosis code to determine prevalence and incidence rates. Sensitivity analyses were performed to compare our methodology to prior publications and quantify the extent of missed diagnoses. The number and type of vascular procedures that occurred after diagnosis were tabulated. A cohort of patients with two or more CLTI diagnosis codes were then identified for further descriptive analysis. Associations between patient demographics and survival were analyzed using Cox proportional hazards models.
Over 65 million patients were enrolled in Medicare in 2017 to 2018. Of these, 480,227 had diagnosis of CLTI, with a corresponding to a 1-year incidence of 0.33% and a 2-year prevalence of 0.74%. Patients underwent an average of 43.6 vascular procedures per 100 person-years. Sensitivity analyses identified 89,805 additional patients that had a diagnosis code of peripheral arterial disease who underwent revascularization or amputation. Patients with CLTI were predominantly male (56.2%), white (76.4%), and qualified for Medicare due to age (64.0%). Thirty-seven percent were dual-eligible. One-year survival was 77.7%, significantly lower than estimated actuarial survival adjusted for age, sex, and race (95.1%; P < .001). Cox proportional hazards models demonstrate significantly increased mortality for men vs women (hazard ratio, 1.07; 95% confidence interval, 1.04-1.10; P < .001), but no association between race and overall survival (hazard ratio, 0.99; 95% confidence interval, 0.98-1.01; P = .83).
Using ICD-10-CM diagnosis codes, we demonstrated slightly higher incidence and prevalence of CLTI than in published literature, reflecting our more complete methodology. Sensitivity analyses suggest that increased complexity of the highly specific ICD-10-CM coding may diminish capture of CLTI. Inclusion of patients with non-CLTI peripheral arterial disease diagnoses produces moderate increases in incidence and prevalence at the cost of decreased specificity in identifying patients with CLTI. Medicare patients with CLTI are older, and more commonly male, black, and dual eligible compared with the general Medicare population. Observed mid-term survival for patients with CLTI is significantly lower than actuarial estimates, confirming the importance of focused efforts on identifying and aligning goals of care in this complex patient population.
Opioid Use Disorder Increases Readmissions After Cardiac Surgery: A Call to Action
2022, Annals of Thoracic Surgery
Owing to the opioid epidemic, more cardiac surgery patients present with opioid use disorder (OUD). A better understanding of national readmissions among these patients is necessary to improve outcomes and optimize resource utilization. We sought to examine the effect of OUD on readmission after cardiac surgery.
Of 555 394 cardiac surgery patients from 2016 to 2017 in the Nationwide Readmissions Database, 6082 (1.1%) presented with OUD. These patients were assessed at 30, 90, and 180 days after discharge. The OUD patients and non-OUD patients were propensity score matched for patient- and procedure-level characteristics. Kaplan-Meier curves were compared using the log rank test.
First-time readmissions were significantly higher among patients with OUD (30 days 19.7% vs 15.7%, P = .04; 90 days 31.8% vs 24.2%, P < .0001; and 180 days 42.3% vs 30.6%, P < .0001). There was a trend toward higher reoperation by 180 days, with 90% of those being isolated valve surgery. By 180 days, significantly more OUD patients had three or more readmissions (7.8% vs 4.5%) compared with non-OUD patients. Yet, only 2.4% of OUD patients received any counseling or treatment for substance abuse during the index admission. The most common readmitting diagnosis was infection (55% vs 41%, P < .0001) including endocarditis, prosthetic infections, and skin or subcutaneous infections. Respiratory failure, opioid overdose, and acute pain were also more common among patients with OUD.
Cardiac surgery patients with OUD have multiple readmissions but are rarely provided adequate addiction management during their index admission. Greater emphasis on multidisciplinary management is necessary to limit costs and morbidity associated with readmission or reoperation.
Antibiotic Prophylaxis Against Infective Endocarditis Before Invasive Dental Procedures
2022, Journal of the American College of Cardiology
Antibiotic prophylaxis (AP) before invasive dental procedures (IDPs) is recommended to prevent infective endocarditis (IE) in those at high IE risk, but there are sparse data supporting a link between IDPs and IE or AP efficacy in IE prevention.
The purpose of this study was to investigate any association between IDPs and IE, and the effectiveness of AP in reducing this.
We performed a case-crossover analysis and cohort study of the association between IDPs and IE, and AP efficacy, in 7,951,972 U.S. subjects with employer-provided Commercial/Medicare-Supplemental coverage.
Time course studies showed that IE was most likely to occur within 4 weeks of an IDP. For those at high IE risk, case-crossover analysis demonstrated a significant temporal association between IE and IDPs in the preceding 4 weeks (OR: 2.00; 95% CI: 1.59-2.52; P = 0.002). This relationship was strongest for dental extractions (OR: 11.08; 95% CI: 7.34-16.74; P < 0.0001) and oral-surgical procedures (OR: 50.77; 95% CI: 20.79-123.98; P < 0.0001). AP was associated with a significant reduction in IE incidence following IDP (OR: 0.49; 95% CI: 0.29-0.85; P = 0.01). The cohort study confirmed the associations between IE and extractions or oral surgical procedures in those at high IE risk and the effect of AP in reducing these associations (extractions: OR: 0.13; 95% CI: 0.03-0.34; P < 0.0001; oral surgical procedures: OR: 0.09; 95% CI: 0.01-0.35; P = 0.002).
We demonstrated a significant temporal association between IDPs (particularly extractions and oral-surgical procedures) and subsequent IE in high–IE-risk individuals, and a significant association between AP use and reduced IE incidence following these procedures. These data support the American Heart Association, and other, recommendations that those at high IE risk should receive AP before IDP.
Sex differences in health resource utilization, costs and mortality during hospitalization for infective endocarditis in the United States
2021, American Heart Journal Plus: Cardiology Research and Practice
Few studies have assessed the association between sex and outcomes among patients with infective endocarditis. The aim of the study was to better understand the association between biologic sex, clinical outcomes and surgical treatment patterns among a contemporary cohort of patients admitted to hospital with infective endocarditis.
We used the National Inpatient Sample dataset from the Health Care Utilization Project to identify adult patients admitted for infective endocarditis between January and December 2016. We compared outcomes between men and women including inpatient hospital mortality, direct hospital costs, length of stay, and inpatient surgical treatment patterns. Multivariable analyses were performed with adjustment for age, socioeconomic status, and comorbidity burden.
Among 18,702 patients with infective endocarditis, there were 8730 (46.7%) women and 1753 (8.4%) in-hospital deaths. In multivariable analysis, female sex was associated with a trend toward lower in-hospital mortality (adjusted odds ratio (OR) 0.90; 95% confidence interval (CI) 0.80 to 1.01, p = 0.06). Additionally, female sex was associated with significantly shorter hospital length of stay (−0.5 days; 95% CI −0.88 to −0.12, p = 0.009) and lower hospital costs (−$3035; 95% CI −$4277 to −$1792; p < 0.001). Notably, women were less likely to undergo surgical intervention (adjusted OR 0.59; 95% CI 0.52 to 0.67, p < 0.001).
In a contemporary, nationally representative cohort of patients admitted for IE in the United States, there were sex-specific differences in management and in-hospital outcomes. Possible sex-based bias in treatment patterns and access to inpatient surgical intervention for infective endocarditis warrants further study.
Clinical and Economic Burden of Hospitalizations for Infective Endocarditis in the United States
2020, Mayo Clinic Proceedings
To assess contemporary trends in the incidence, characteristics, and outcomes of hospital admissions for infective endocarditis (IE) in the United States.
Patients ≥18 years admitted with IE between January 1, 2003, and December 31, 2016, were identified in the National Inpatient Sample. We assessed the annual incidence, clinical characteristics, morbidity, mortality, and cost of IE-related hospitalizations.
The incidence of IE-related hospitalizations increased from 34,488 (15.9; 95% confidence interval [CI], 15.73, 16.06) per 100,000 adults) in 2003 to 54,405 (21.8; 95% CI, 21.60-21.97) per 100,000 adults) in 2016 (P<.001). The prevalence of patients below 30 years of age, and those who inject drugs, increased from 7.3% to 14.5% and from 4.8% to 15.1%, respectively (P<.001). The annual volume of valve surgery for IE increased from 4049 in 2003 to 6460 in 2016 (P<.001), but the ratio of valve surgery to IE-hospitalizations did not decrease (11.7% in 2003; 11.8% in 2016). There was also a temporal increase in risk-adjusted rates of stroke (8.0% to 13.2%), septic shock (5.4% to 16.3%), and mechanical ventilation (7.7% to 16.5%; P<.001). However, risk-adjusted mortality decreased from 14.4% to 9.8% (P<.001). Median length-of-stay and mean inflation-adjusted cost decreased from 11 to 10 days and from $45,810±$61,787 to $43,020±$55,244, respectively, (P<.001). Nonetheless, the expenditure on IE hospitalizations increased ($1.58 billion in 2003 to $2.34 billion in 2016; P<.001).
There is a substantial recent rise in endocarditis hospitalizations in the United States. Although the adjusted in-hospital mortality of endocarditis and the cost of admission decreased over time, the overall expenditure on in-hospital care for endocarditis increased.

View all citing articles on Scopus

¹: This author takes responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation.

²: Contributed equally to supervision of this study and shares senior authorship.

View full text

Accuracy of administrative data for identification of patients with infective endocarditis

Abstract

Background

Methods

Results

Conclusion

Introduction

Section snippets

General study design and data sources

Results

Discussion

Conflict of interest

Funding sources

Disclosures

Acknowledgements

J. Am. Coll. Cardiol.

J. Am. Coll. Cardiol.

Lancet

J. Clin. Epidemiol.

J. Clin. Epidemiol.

Clinical practice. Infective endocarditis

N. Engl. J. Med.

Preeminence of Staphylococcus aureus in infective endocarditis: a 1-year population-based survey

Clin. Infect. Dis.

Clinical presentation, etiology, and outcome of infective endocarditis in the 21st century: the International Collaboration on Endocarditis-Prospective Cohort Study

Arch. Intern. Med.

Complicated infective endocarditis necessitating ICU admission: clinical course and prognosis

Crit. Care

Outpatient treatment of infective endocarditis

Clin. Microbiol. Infect.