Diagnostic accuracy of ICD code versus discharge summary-based query for endocarditis cohort identification

Validated administrative data based ICD-10 algorithms for chronic conditions: A systematic review

OBJECTIVE

This systematic review aimed to identify ICD-10 based validated algorithms for chronic conditions using health administrative data.

METHODS

A comprehensive systematic literature search using Ovid MEDLINE, Embase, PsycINFO, Web of Science and CINAHL was performed to identify studies, published between 1983 and May 2023, on validated algorithms for chronic conditions using administrative health data. Two reviewers independently screened titles and abstracts and reviewed full text of selected studies to complete data extraction. A third reviewer resolved conflicts arising at the screening or study selection stages. The primary outcome was validated studies of ICD-10 based algorithms with both sensitivity and PPV of ≥70 %. Studies with either sensitivity or PPV <70 % were included as secondary outcomes.

RESULTS

Overall, the search identified 1686 studies of which 54 met the inclusion criteria. Combining a previously published literature search, a total of 61 studies were included for data extraction. The study identified 40 chronic conditions with high validity and 22 conditions with moderate validity. The validated algorithms were based on administrative data from different countries including Canada, USA, Australia, Japan, France, South Korea, and Taiwan. The algorithms identified included several types of cancers, cardiovascular conditions, kidney diseases, gastrointestinal disorders, and peripheral vascular diseases, amongst others.

CONCLUSION

With ICD-10 prominently used across the world, this up-to-date systematic review can prove to be a helpful resource for research and surveillance initiatives using administrative health data for identifying chronic conditions.

Trends in epidemiology, surgical management, and prognosis of infective endocarditis during the XXI century in Spain: A population-based nationwide study

BACKGROUND

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.

METHODS

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.

RESULTS

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).

CONCLUSIONS

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.

Lead Extraction and Mortality Among Patients With Cardiac Implanted Electronic Device Infection

Importance

Complete hardware removal is a class I recommendation for cardiovascular implantable electronic device (CIED) infection, but practice patterns and outcomes remain unknown.

Objective

To quantify the number of Medicare patients with CIED infections who underwent implantation from 2006 to 2019 and lead extraction from 2007 to 2019 to analyze the outcomes in these patients in a nationwide clinical practice cohort.

Design, Setting, and Participants

This cohort study included fee-for-service Medicare Part D beneficiaries from January 1, 2006, to December 31, 2019, who had a de novo CIED implantation and a CIED infection more than 1 year after implantation. Data were analyzed from January 1, 2005, to December 31, 2019.

Exposure

A CIED infection, defined as (1) endocarditis or infection of a device implant and (2) documented antibiotic therapy.

Main Outcomes and Measures

The primary outcomes of interest were device infection, device extraction, and all-cause mortality. Time-varying multivariable Cox proportional hazards regression models were used to evaluate the association between extraction and survival.

Results

Among 1 065 549 patients (median age, 78.0 years [IQR, 72.0-84.0 years]; 50.9% male), mean (SD) follow-up was 4.6 (2.9) years after implantation. There were 11 304 patients (1.1%) with CIED infection (median age, 75.0 years [IQR, 67.0-82.0 years]); 60.1% were male, and 7724 (68.3%) had diabetes. A total of 2102 patients with CIED infection (18.6%) underwent extraction within 30 days of diagnosis. Infection occurred a mean (SD) of 3.7 (2.4) years after implantation, and 1-year survival was 68.3%. There was evidence of highly selective treatment, as most patients did not have extraction within 30 days of diagnosed infection (9202 [81.4%]), while 1511 (13.4%) had extraction within 6 days of diagnosis and 591 (5.2%) had extraction between days 7 and 30. Any extraction was associated with lower mortality compared with no extraction (adjusted hazard ratio [AHR], 0.82; 95% CI, 0.74-0.90; P < .001). Extraction within 6 days was associated with even lower risk of mortality (AHR, 0.69; 95% CI, 0.61-0.78; P < .001).

Conclusions and Relevance

In this study, a minority of patients with CIED infection underwent extraction. Extraction was associated with a lower risk of death compared with no extraction. The findings suggest a need to improve adherence to guideline-directed care among patients with CIED infection.

A Comparison of Obstetric Interventions and Outcomes Between Black and White Patients at an Urban Tertiary Medical Center

The objective of the study is to evaluate whether rates of selected labor and delivery interventions and severe maternal morbidity (SMM) differ between Black and White pregnant patients. This retrospective observational cohort study included all Black or White pregnant patients who delivered at the University of Chicago Medical Center between January 2015 and December 2019. Data queried included demographic information, antepartum complications, preterm interventions, labor and delivery events, and neonatal outcomes. SMM was a composite outcome, including intensive care unit admission, blood transfusion, hysterectomy, eclampsia, cardiac arrest, or death. In total, 10,885 parturients (9001 Black and 1884 White) and 11,211 neonates (9254 born to Black and 1957 to White patients) were included in the study. Black patients were more likely to have preterm labor (3.51% vs. 1.86%, p = 0.0002) and no prenatal care (17.83% vs. 4.05%, p < 0.0001). There was no significant difference in the administration of magnesium sulfate for fetal neuroprotection (Black 44.78% vs. White 49.32%, p = 0.48) or antenatal corticosteroids (Black 67.83% vs. White 71.98%, p = 0.28) among those with preterm delivery. There was no significant difference in SMM (Black 2.24% vs. White 2.44%, p = 0.60), and SMM rates decreased over time (OR 0.79 per year, 95% CI: 0.72-0.87, p < 0.0001) for all patients. Black patients had more pregnancy complications, but their complications were addressed with similar rates of obstetrical interventions. In a high-resource setting, there was no difference in rates of SMM when compared to White patients.

Deep Learning Analysis of Polish Electronic Health Records for Diagnosis Prediction in Patients with Cardiovascular Diseases

Electronic health records naturally contain most of the medical information in the form of doctor’s notes as unstructured or semi-structured texts. Current deep learning text analysis approaches allow researchers to reveal the inner semantics of text information and even identify hidden consequences that can offer extra decision support to doctors. In the presented article, we offer a new automated analysis of Polish summary texts of patient hospitalizations. The presented models were found to be able to predict the final diagnosis with almost 70% accuracy based just on the patient’s medical history (only 132 words on average), with possible accuracy increases when adding further sentences from hospitalization results; even one sentence was found to improve the results by 4%, and the best accuracy of 78% was achieved with five extra sentences. In addition to detailed descriptions of the data and methodology, we present an evaluation of the analysis using more than 50,000 Polish cardiology patient texts and dive into a detailed error analysis of the approach. The results indicate that the deep analysis of just the medical history summary can suggest the direction of diagnosis with a high probability that can be further increased just by supplementing the records with further examination results.

Correlates of 90-day Mortality Among People Who Do and Do Not Inject Drugs with Infective Endocarditis in Seattle, Washington

Background

Infective endocarditis (IE) remains highly morbid, but few studies have evaluated factors associated with IE mortality. We examined correlates of 90-day mortality among people who inject drugs (PWID) and people who do not inject drugs (non-PWID).

Methods

We queried the electronic medical record for cases of IE among adults ≥18 years of age at 2 academic medical centers in Seattle, Washington, from 1 January 2014 to 31 July 2019. Cases were reviewed to confirm a diagnosis of IE and drug use status. Deaths were confirmed through the Washington State death index. Descriptive statistics were used to characterize IE in PWID and non-PWID. Kaplan-Meier log-rank tests and Cox proportional hazard models were used to assess correlates of 90-day mortality.

Results

We identified 507 patients with IE, 213 (42%) of whom were PWID. Sixteen percent of patients died within 90 days of admission, including 14% of PWID and 17% of non-PWID (P = .50). In a multivariable Cox proportional hazard model, injection drug use was associated with a higher mortality within the first 14 days of admission (adjusted hazard ratio [aHR], 2.33 [95% confidence interval {CI}, 1.16–4.65], P = .02); however, there was no association between injection drug use and mortality between 15 and 90 days of admission (aHR, 0.63 [95% CI, .31–1.30], P = .21).

Conclusions

Overall 90-day mortality did not differ between PWID and non-PWID with IE, although PWID experienced a higher risk of death within 14 days of admission. These findings suggest that early IE diagnosis and treatment among PWID is critical to improving outcomes.