Accuracy of identification of patients with immune thrombocytopenic purpura through administrative records: a data validation study

Implementation of Primary Immune Thrombocytopenia Clinical Practice Guidelines for Management of Pregnancy

Background: Managing primary immune thrombocytopenia (ITP) in pregnancy is challenging. Providers must balance bleeding risk against medication toxicity. The evaluation of the implementation of pregnancy-specific ITP clinical guidelines has not been widely studied. The goal of this study was to describe the implementation of pregnancy-specific ITP guidelines at an academic health center. Methods: We conducted a retrospective chart review at the University of Oklahoma Health system from 2011 to 2020. Descriptive statistics were calculated to summarize the characteristics of the study population. Management, according to the clinical guidelines (American Society of Hematology; American College of Obstetricians and Gynecologists) was evaluated during pregnancy and during/for delivery. Results: A total of 85 pregnant persons with ITP were included. The majority (68%; 58/85) delivered vaginally. There were 0 maternal deaths and 2 infant deaths. No patients had major bleeding during pregnancy. Postpartum hemorrhage was experienced by 14%. The management of thrombocytopenia during pregnancy was 100% adherent to the strong recommendation for severe (n = 13) and mild (n = 11) thrombocytopenia. However, 18/50 (36%) asymptomatic persons with moderate thrombocytopenia received treatment despite the strong recommendation that treatment was unnecessary. Additionally, 8/21 (38%) persons with moderate thrombocytopenia received treatment to increase platelet counts for epidural anesthesia despite the guideline's suggestion that it was unnecessary. Conclusions: During pregnancy, patients with severe thrombocytopenia (i.e., most at risk of bleeding) received treatment. On the other hand, approximately 40% of pregnant persons with ITP received unnecessary treatment for moderate asymptomatic thrombocytopenia either during pregnancy or for an epidural. Utilizing clinical practice guidelines would reduce the overtreatment of pregnant persons which would reduce the potential side effects of therapy for the mother and infant.

A Qualitative Study of Physicians' Views on the Reuse of Electronic Health Record Data for Secondary Analysis

Electronic health records (EHRs) have become ubiquitous in clinical practice. Given the rich biomedical data captured for a large panel of patients, secondary analysis of these data for health research is also commonplace. Yet, there are many caveats to EHR data that the researchers must be aware of, such as the accuracy of and motive for documentation, and the reason for patients' visits to the clinic. The clinician-the author of the documentation-is thus central to the correct interpretation of EHR data for research purposes. In this study, I interviewed 11 physicians in various clinical specialties to bring attention to their view on the validity of research using EHR data. Qualitative, in-depth, one-on-one interviews were conducted with practicing physicians in inpatient and outpatient medicine. Content analysis using a data-driven, inductive approach to identify themes related to challenges and opportunities in the reuse of EHR data for secondary analysis generated seven themes. Themes that reflected challenges of EHRs for research included (1) audience, (2) accuracy of data, (3) availability of data, (4) documentation practices, and (5) representativeness. Themes that reflected opportunities of EHRs for research included (6) endorsement and (7) enablers. The greatest perceived barriers reflected the intended audience of the EHR, the interpretation and meaning of the data, and the quality of the data for research purposes. Physicians generally expressed more perceived challenges than opportunities in the reuse of EHR data for research purposes; however, they remained optimistic.

Can researchers trust ICD-10 coding of medical comorbidities in orthopaedic trauma patients?

Objectives

The 10th revision of the International Classification of Diseases (ICD-10) coding system may prove useful to orthopaedic trauma researchers to identify and document populations based on comorbidities. However, its use for research first necessitates determination of its reliability. The purpose of this study was to assess the reliability of electronic medical record (EMR) ICD-10 coding of nonorthopaedic diagnoses in orthopaedic trauma patients relative to the gold standard of prospective data collection.

Design

Nonexperimental cross-sectional study.

Setting

Level 1 Trauma Center.

Patients/Participants

Two hundred sixty-three orthopaedic trauma patients from 2 prior prospective studies from September 2018 to April 2022.

Intervention

Prospectively collected data were compared with EMR ICD-10 code abstraction for components of the Charlson Comorbidity Index (CCI), obesity, alcohol abuse, and tobacco use (retrospective data).

Main Outcome Measurements

Percent agreement and Cohen's kappa reliability.

Results

Percent agreement ranged from 86.7% to 96.9% for all CCI diagnoses and was as low as 72.6% for the diagnosis "overweight." Only 2 diagnoses, diabetes without end-organ damage (kappa = 0.794) and AIDS (kappa = 0.798) demonstrated Cohen's kappa values to indicate substantial agreement.

Conclusion

EMR diagnostic coding for medical comorbidities in orthopaedic trauma patients demonstrated variable reliability. Researchers may be able to rely on EMR coding to identify patients with diabetes without complications or AIDS. Chart review may still be necessary to confirm diagnoses. Low prevalence of most comorbidities led to high percentage agreement with low reliability.

Level of Evidence

Level 1 diagnostic.

Real-World Data Analysis of Patients Affected by Immune-Mediated Thrombotic Thrombocytopenic Purpura in Italy

Background: The therapeutic management of immune-mediated thrombotic thrombocytopenic purpura (iTTP) has recently benefited from the introduction of caplacizumab, an agent directed at the inhibition of platelet aggregation. This real-world analysis investigated the epidemiology and the demographic and clinical characteristics of iTTP patients in Italy before and after caplacizumab introduction in 2020. Methods: Hospitalized adults with iTTP were included using the administrative databases of healthcare entities covering 17 million residents. Epidemiological estimates of iTTP considered the 3-year period before and after caplacizumab introduction. After stratification by treatment with or without caplacizumab, iTTP patients were characterized for their baseline features. Results: The annual incidence before and after 2020 was estimated in the range of 4.3-5.8 cases/million and 3.6-4.6 cases/million, respectively. From 2018 to 2022, 393 patients with iTTP were included, and 42 of them were treated with caplacizumab. Caplacizumab-treated patients showed better clinical outcomes, with tendentially shorter hospital stays and lower mortality rates (no treated patients died at either 1 month or 3 months after caplacizumab treatment initiation, compared to 10.5% and 11.1% mortality rates at 1 and 3 months, respectively, of the untreated ones). Conclusions: These findings may suggest that caplacizumab advent provided clinical and survival benefits for patients with iTTP.

The Accuracy of Japanese Administrative Data in Identifying Acute Exacerbation of Idiopathic Pulmonary Fibrosis

BACKGROUND

This study aimed to develop criteria for identifying patients with acute exacerbation of idiopathic pulmonary fibrosis (AE-IPF) from Japanese administrative data and validate the pre-existing criteria.

METHODS

This retrospective, multi-center validation study was conducted at eight institutes in Japan to verify the diagnostic accuracy of the disease name for AE-IPF. We used the Japanese Diagnosis Procedure Combination data to identify patients with a disease name that could meet the diagnostic criteria for AE-IPF, who were admitted to the eight institutes from January 2016 to February 2019. As a reference standard, two respiratory physicians performed a chart review to determine whether the patients had a disease that met the diagnostic criteria for AE-IPF. Furthermore, two radiologists interpreted the chest computed tomography findings of cases considered AE-IPF and confirmed the diagnosis. We calculated the positive predictive value (PPV) for each disease name and its combination.

RESULTS

We included 830 patients; among them, 216 were diagnosed with AE-IPF through the chart review. We combined the groups of disease names and yielded two criteria: the criteria with a high PPV (0.72 [95% confidence interval 0.62 to 0.81]) and that with a slightly less PPV (0.61 [0.53 to 0.68]) but more true positives. Pre-existing criteria showed a PPV of 0.40 (0.31 to 0.49).

CONCLUSION

The criteria derived in this study for identifying AE-IPF from Japanese administrative data show a fair PPV. Although these criteria should be carefully interpreted according to the target population, our findings could be utilized in future database studies on AE-IPF.

Background incidence rates of hospitalisations and emergency department visits for thromboembolic and coagulation disorders in Ontario, Canada for COVID-19 vaccine safety assessment: a population-based retrospective observational study

OBJECTIVE

The objective of this study was to estimate background rates of selected thromboembolic and coagulation disorders in Ontario, Canada.

DESIGN

Population-based retrospective observational study using linked health administrative databases. Records of hospitalisations and emergency department visits were searched to identify cases using International Statistical Classification of Diseases and Related Health Problems, 10th Revision, Canada diagnostic codes.

PARTICIPANTS

All Ontario residents.

PRIMARY OUTCOME MEASURES

Incidence rates of ischaemic stroke, intracerebral haemorrhage, subarachnoid haemorrhage, deep vein thrombosis, pulmonary embolism, idiopathic thrombocytopaenia, disseminated intravascular coagulation and cerebral venous thrombosis during five prepandemic years (2015-2019) and 2020.

RESULTS

The average annual population was 14 million with 51% female. The mean annual rates per 100 000 population during 2015-2019 were 127.1 (95% CI 126.2 to 127.9) for ischaemic stroke, 22.0 (95% CI 21.6 to 22.3) for intracerebral haemorrhage, 9.4 (95% CI 9.2 to 9.7) for subarachnoid haemorrhage, 86.8 (95% CI 86.1 to 87.5) for deep vein thrombosis, 63.7 (95% CI 63.1 to 64.3) for pulmonary embolism, 6.1 (95% CI 5.9 to 6.3) for idiopathic thrombocytopaenia, 1.6 (95% CI 1.5 to 1.7) for disseminated intravascular coagulation, and 1.5 (95% CI 1.4 to 1.6) for cerebral venous thrombosis. Rates were lower in 2020 than during the prepandemic years for ischaemic stroke, deep vein thrombosis and idiopathic thrombocytopaenia. Rates were generally consistent over time, except for pulmonary embolism, which increased from 57.1 to 68.5 per 100 000 between 2015 and 2019. Rates were higher for females than males for subarachnoid haemorrhage, pulmonary embolism and cerebral venous thrombosis, and vice versa for ischaemic stroke and intracerebral haemorrhage. Rates increased with age for most of these conditions, but idiopathic thrombocytopaenia demonstrated a bimodal distribution with incidence peaks at 0-19 years and ≥60 years.

CONCLUSIONS

Our estimated background rates help contextualise observed events of these potential adverse events of special interest and to detect potential safety signals related to COVID-19 vaccines.

Thrombotic Events after COVID-19 Vaccination in the Over-50s: Results from a Population-Based Study in Italy

Several European countries suspended or changed recommendations for the use of Vaxzevria (AstraZeneca) for suspected adverse effects due to atypical blood-clotting. This research aims to identify a reference point towards the number of thrombotic events expected in the Italian population over 50 years of age who received Vaxzevria from 22 January to 12 April 2021. The venous thromboembolism (VT) and immune thrombocytopenia (ITP) event rates were estimated from a population-based cohort. The overall VT rate was 1.15 (95% CI 0.93–1.42) per 1000 person-years, and the ITP rate was 2.7 (95% CI 0.7–11) per 100,000 person-years. These figures translate into 83 and two expected events of VT and ITP, respectively, in the 15 days following the first administration of Vaxzevria. The number of thrombotic events reported from the Italian Medicines Agency does not appear to have increased beyond that expected in individuals over 50 years of age.

Risk factors for COVID-19 deaths among elderly nursing home Medicare beneficiaries in the pre-vaccine period

BACKGROUND

Evaluate pre-vaccine pandemic period COVID-19 death risk factors among nursing home (NH) residents.

METHODS

Retrospective cohort study covering Medicare fee-for-service beneficiaries ages ≥65 residing in U.S. NHs. We estimated adjusted hazard ratios (HRs) using multivariate Cox proportional hazards regressions.

RESULTS

Among 608,251 elderly NH residents, 57,398 (9.4%) died of COVID-related illness April 1 to December 22, 2020. About 46.9% (26,893) of these COVID-19 deaths occurred without prior COVID-19 hospitalizations. We observed a consistently increasing age trend for COVID-19 deaths. Racial/ethnic minorities generally shared a similarly high risk of NH COVID-19 deaths with Whites. NH facility characteristics including for-profit ownership and low health inspection ratings were associated with higher death risk. Resident characteristics, including male (HR 1.69), end-stage renal disease (HR 1.42), cognitive impairment (HR 1.34), and immunocompromised status (HR 1.20) were important death risk factors. Other individual-level characteristics were less predictive of death than they were in community-dwelling population.

CONCLUSIONS

Low NH health inspection ratings and private ownership contributed to COVID-19 death risks. Nearly half of NH COVID-19 deaths occurred without prior COVID-19 hospitalization and older residents were less likely to get hospitalized with COVID-19. No substantial differences were observed by race/ethnicity and socioeconomic status for NH COVID-19 deaths.

Natural History of Coronavirus Disease 2019: Risk Factors for Hospitalizations and Deaths Among >26 Million US Medicare Beneficiaries

BACKGROUND

The current study was performed to evaluate risk factors for severe coronavirus disease 2019 (COVID-19) outcomes among Medicare beneficiaries during the pandemic's early phase.

METHODS

In a retrospective cohort study covering Medicare fee-for-service beneficiaries, we separated out elderly residents in nursing homes (NHs) and those with end-stage renal disease (ESRD) from the primary study population of individuals age ≥65 years. Outcomes included COVID-19 hospital encounters and COVID-19-associated deaths. We estimated adjusted odds ratios (ORs) using logistic regression.

RESULTS

We analyzed 25 333 329 elderly non-NH beneficiaries without ESRD, 653 966 elderly NH residents, and 292 302 patients with ESRD. COVID-related death rates (per 10 000) were much higher among elderly NH residents (275.7) and patients with ESRD (60.8) than in the primary study population (5.0). Regression-adjusted clinical predictors of death among the primary population included immunocompromised status (OR, 1.43), frailty index conditions such as cognitive impairment (3.16), and other comorbid conditions, including congestive heart failure (1.30). Demographic-related risk factors included male sex (OR, 1.77), older age (3.09 for 80- vs 65-year-olds), Medicaid dual-eligibility status (2.17), and racial/ethnic minority. Compared with whites, ORs were higher for blacks (2.47), Hispanics (3.11), and Native Americans (5.82). Results for COVID-19 hospital encounters were consistent.

CONCLUSIONS

Frailty, comorbid conditions, and race/ethnicity were strong risk factors for COVID-19 hospitalization and death among the US elderly.

Second-line treatments and outcomes for immune thrombocytopenia: A retrospective study with electronic health records

BACKGROUND

Second-line treatment for immune thrombocytopenia (ITP) is not well reported for patients treated in real-world clinical settings.

OBJECTIVE

The purpose of this study was to compare outcomes of four second-line treatments for ITP.

PATIENTS/METHODS

Included adult patients had at least two medical records containing ITP diagnoses and second-line eltrombopag, romiplostim, rituximab, or splenectomy. Date of treatment initiation or splenectomy was set as index date, between July 1, 2008, and March 31, 2017. Patients had first-line corticosteroid or intravenous immune globulin treatment and continuous database activity from 6 months before to 12 months after index. Patient characteristics, treatment patterns, platelet counts, bleeding-related episodes (BREs), and thrombotic events (TEs) were compared by second-line treatment cohort.

RESULTS

The sample included 3332 patients (mean age, 60.5 years; 52.3% female): eltrombopag (5.8%), romiplostim (9.9%), rituximab (73.3%), and splenectomy (11.0%). Patients having splenectomy were younger, more likely female and commercially insured, and less likely to require a third line of treatment than medical regimen cohorts. Proportions of patients having treatment-free (≥180 days with no second-line index or rescue agent) periods varied significantly (P = .01) by regimen: 33% for eltrombopag, 23% for romiplostim, 26% for rituximab, and 17% for splenectomy. All regimens significantly improved platelet counts, while TE and BRE rates differed significantly (P = .03 and P = .01, respectively) when all treatment groups were compared.

CONCLUSIONS

Over an average 7-year follow-up, all second-line regimens improved platelet counts, but eltrombopag yielded the highest proportion of patients with completely treatment-free periods of at least 180 days.

Primary immune thrombocytopenia in US clinical practice: incidence and healthcare burden in first 12 months following diagnosis

Introduction: Primary immune thrombocytopenia (ITP), an autoimmune disorder characterized by low platelet count, can lead to serious bleeding events. Little is known about the current epidemiology of ITP in the US, and even less is known about the current healthcare burden of ITP, especially in the 12-month period following ITP diagnosis.Method: We used a retrospective cohort design and data from two US private healthcare claims databases (2010-2016) to identify persons with evidence of newly diagnosed ITP. We weighted estimates of the annual incidence of ITP by age and sex to reflect the US population, and summarized healthcare utilization and expenditures (2016 US$) during the first 12 months after ITP diagnosis ("follow-up period").Results: Annual incidence of ITP in the US was 6.1 per 100,000 persons, higher among females versus males (6.7 vs. 5.5), and highest among children aged 0-4 years (8.1) and adults aged ≥65 years (13.7). Patients with ITP averaged 0.33 (95% CI: 0.32-0.35) hospitalizations and 15.3 (15.1-15.6) ambulatory encounters during the follow-up period; mean total healthcare expenditures during this period were $21,290 (20,502-22,031). Hospitalizations were more common during the first 3 months following diagnosis, and were twice as frequent among children versus adults; expenditures for ambulatory encounters were substantially higher for adults versus children aged 0-4 years.Conclusions: Our findings suggest that nearly 20,000 children and adults are newly diagnosed with ITP each year in the US, substantially higher than previously reported. Among patients requiring formal medical care, the economic burden during the first 12 months following diagnosis is high, with estimated US expenditures totaling over $400 million.

Reliability of International Classification of Disease-9 Versus International Classification of Disease-10 Coding for Proximal Femur Fractures at a Level 1 Trauma Center

INTRODUCTION

The Centers for Medicare & Medicaid services proposed that transitioning from the 9th to the 10th revision of the International Classification of Disease (ICD) would provide better data for research. This study sought to determine the reliability of ICD-10 compared with ICD-9 for proximal femur fractures.

METHODS

Available imaging studies from 196 consecutively treated proximal femur fractures were retrospectively reviewed and assigned ICD codes by three physicians. Intercoder reliability (ICR) was calculated. Collectively, the physicians agreed on what should be the correct codes for each fracture, and this was compared with coding found in the medical and billing records.

RESULTS

No significant difference was observed in ICR for both ICD-9 and ICD-10 exact coding, which were both unreliable. Less specific coding improved ICR. ICD-9 general coding was better than ICD-10. Electronic medical record coding was unreliable. Billing codes were also unreliable, yet ICD-10 was better than ICD-9.

DISCUSSION

ICD-9 and ICD-10 lack reliability in coding proximal femur fractures. ICD-10 results in data that are no more reliable than those found with ICD-9.

LEVEL OF EVIDENCE

Level I diagnostic.

Multimorbidity states associated with higher mortality rates in organ dysfunction and sepsis: a data-driven analysis in critical care

BACKGROUND

Sepsis remains a complex medical problem and a major challenge in healthcare. Diagnostics and outcome predictions are focused on physiological parameters with less consideration given to patients' medical background. Given the aging population, not only are diseases becoming increasingly prevalent but occur more frequently in combinations ("multimorbidity"). We hypothesized the existence of patient subgroups in critical care with distinct multimorbidity states. We further hypothesize that certain multimorbidity states associate with higher rates of organ failure, sepsis, and mortality co-occurring with these clinical problems.

METHODS

We analyzed 36,390 patients from the open source Medical Information Mart for Intensive Care III (MIMIC III) dataset. Morbidities were defined based on Elixhauser categories, a well-established scheme distinguishing 30 classes of chronic diseases. We used latent class analysis to identify distinct patient subgroups based on demographics, admission type, and morbidity compositions and compared the prevalence of organ dysfunction, sepsis, and inpatient mortality for each subgroup.

RESULTS

We identified six clinically distinct multimorbidity subgroups labeled based on their dominant Elixhauser disease classes. The "cardiopulmonary" and "cardiac" subgroups consisted of older patients with a high prevalence of cardiopulmonary conditions and constituted 6.1% and 26.4% of study cohort respectively. The "young" subgroup included 23.5% of the cohort composed of young and healthy patients. The "hepatic/addiction" subgroup, constituting 9.8% of the cohort, consisted of middle-aged patients (mean age of 52.25, 95% CI 51.85-52.65) with the high rates of depression (20.1%), alcohol abuse (47.75%), drug abuse (18.2%), and liver failure (67%). The "complicated diabetics" and "uncomplicated diabetics" subgroups constituted 9.4% and 24.8% of the study cohort respectively. The complicated diabetics subgroup demonstrated higher rates of end-organ complications (88.3% prevalence of renal failure). Rates of organ dysfunction and sepsis ranged 19.6-69% and 12.5-46.7% respectively in the six subgroups. Mortality co-occurring with organ dysfunction and sepsis ranges was 8.4-23.8% and 11.7-27.4% respectively. These adverse outcomes were most prevalent in the hepatic/addiction subgroup.

CONCLUSION

We identify distinct multimorbidity states that associate with relatively higher prevalence of organ dysfunction, sepsis, and co-occurring mortality. The findings promote the incorporation of multimorbidity in healthcare models and the shift away from the current single-disease paradigm in clinical practice, training, and trial design.

Rate of bleeding-related episodes in elderly patients with primary immune thrombocytopenia: a retrospective cohort study

OBJECTIVE

Immune thrombocytopenia (ITP) is characterized by low platelet counts and a tendency toward increased bleeding and bruising. We aimed to describe bleeding frequency and use of rescue ITP therapy to treat or prevent bleeding in elderly ITP patients in a real-world setting.

METHODS

Using Medicare 20% sample data, 2007-2012, we identified elderly (ages ≥67 years) Medicare fee-for-service enrollees diagnosed with primary ITP between 1 January 2009 and 30 September 2012. Bleeding-related episodes (BREs) were defined as ≥1 bleeding event or use of ITP therapies commonly considered for rescue or emergency therapy. BRE rates were examined for the cohort overall, by time since ITP onset, and by splenectomy status. Patients were followed from ITP onset until the earliest of death, disenrollment from fee-for-service coverage, or 31 December 2012.

RESULTS

We identified 3007 elderly patients diagnosed with primary ITP (mean [SD] age: 79.6 [7.5] years; 55% female); 2178 (72%) experienced at least one BRE (8867 BREs); 92 (3%) underwent splenectomy. Nearly half of BREs were defined by rescue therapy use alone. The overall rate was 1.72 BREs per patient-year (95% CI; 1.68-1.75); rates were higher during the first 3 months after ITP onset and after splenectomy.

CONCLUSION

Elderly ITP patients experienced about two BREs per patient-year after ITP onset. Most patients experienced at least one BRE. These real-world results demonstrate the importance of examining both bleeding and use of rescue or emergency ITP therapy in the assessment of disease burden in elderly patients with ITP.

Appropriateness of ICD-coded Diagnostic Inpatient Hospital Discharge Data for Medical Practice Assessment

Objectives: We performed a systematic review to investigate the quality of diagnostic hospital discharge data (DHDD) in order to gain insight in the usefulness of these data for medical practice assessment. We investigated the methods used to evaluate data quality, factors that determine data quality and its consequences for medical practice assessment.

Methods: We selected studies in which both completeness (or sensitivity: SENS) and correctness (or positive predictive value: PPV) were measured. We used the random-effects model to calculate mean SENS and PPV and to explore the effect of a number of covariates.

Results: The 101 included studies were very heterogeneous. We distinguished six typical study designs. We found a mean SENS of 0.67 (95%CI: 0.62– 0.73) and PPV of 0.76 (95%CI: 0.73– 0.79); SENS was significantly lower for comorbidity and complication studies than for some single disease studies. PPV was significantly higher for Scandinavian countries than for other countries. Recoding compared to re-abstracting of the medical record as a gold standard gave a significantly lower PPV. Diagnostic data were considered appropriate by the authors of the studies for quality of care purposes when both SENS and PPV were at least 0.85. Only 13% of the studies fulfilled this criterion.

Conclusions: Variability in quality of care between settings can easily be overshadowed by variability in data quality. However, the use of DHDD by physicians to evaluate their own medical practice may be useful. But only if physicians are willing to critically interpret the meaning of the information for their medical practice assessment.

Reliable assessment of the incidence of childhood autoimmune hemolytic anemia

BACKGROUND

Childhood autoimmune hemolytic anemia (AIHA) is a rare and severe disease characterized by hemolysis and positive direct antiglobulin test (DAT). Few epidemiologic indicators are available for the pediatric population. The objective of our study was to reliably estimate the number of AIHA cases in the French Aquitaine region and the incidence of AIHA in patients under 18 years old.

PROCEDURE

In this retrospective study, the capture-recapture method and log-linear model were used for the period 2000-2008 in the Aquitaine region from the following three data sources for the diagnosis of AIHA: the OBS'CEREVANCE database cohort, positive DAT collected from the regional blood bank database, and the French medico-economic information system.

RESULTS

A list of 281 different patients was obtained after cross-matching the three databases; 44 AIHA cases were identified in the period 2000-2008; and the total number of cases was estimated to be 48 (95% confidence interval [CI]: 45-55). The calculated incidence of the disease was 0.81/100,000 children under 18 years old per year (95% CI 0.76-0.92).

CONCLUSION

Accurate methods are required for estimating the incidence of AIHA in children. Capture-recapture analysis corrects underreporting and provides optimal completeness. This study highlights a possible under diagnosis of this potentially severe disease in various pediatric settings. AIHA incidence may now be compared with the incidences of other hematological diseases and used for clinical or research purposes.

Incidence of Clinician-Diagnosed Lyme Disease, United States, 2005-2010

Extrapolation from a large medical claims database suggests that 329,000 cases occur annually.

National surveillance provides important information about Lyme disease (LD) but is subject to underreporting and variations in practice. Information is limited about the national epidemiology of LD from other sources. Retrospective analysis of a nationwide health insurance claims database identified patients from 2005–2010 with clinician-diagnosed LD using International Classification of Diseases, Ninth Revision, Clinical Modification, codes and antimicrobial drug prescriptions. Of 103,647,966 person-years, 985 inpatient admissions and 44,445 outpatient LD diagnoses were identified. Epidemiologic patterns were similar to US surveillance data overall. Outpatient incidence was highest among boys 5–9 years of age and persons of both sexes 60–64 years of age. On the basis of extrapolation to the US population and application of correction factors for coding, we estimate that annual incidence is 106.6 cases/100,000 persons and that ≈329,000 (95% credible interval 296,000–376,000) LD cases occur annually. LD is a major US public health problem that causes substantial use of health care resources.

Management of immune thrombocytopenia in adults: a population-based analysis of the French hospital discharge database from 2009 to 2012

The present study describes the current clinical practice and hospital management of adults with immune thrombocytopenia (ITP) between 2009 and 2012 in France, based on the national discharge hospital database. Adult ITP patients were managed almost exclusively in public hospitals. A relatively stable number of patients, around 3200 per year, were hospitalized for ITP annually over the 4-year period, about two-thirds of whom were newly-diagnosed ITP. Re-hospitalizations tended to decrease over the study period. Intravenous immunoglobulin administration, concerning half of ITP hospitalized patients, and rituximab administration were stable over time, whereas a slight decrease of splenectomies was observed.

Are risk evaluation and mitigation strategies associated with less off-label use of medications? The case of immune thrombocytopenia

Using data from a large commercial health insurer, we studied prescribing of romiplostim (Nplate) and eltrombopag (Promacta), two drugs for primary immune thrombocytopenia (ITP) for which risk evaluation and mitigation strategies (REMS) with elements to assure safe use were initially imposed and then removed. We identified 103 and 117 new users of romiplostim and eltrombopag, respectively. Use was almost exclusively for FDA-approved indications ("on-label") while the REMS with elements to assure safe use were in place. After these elements were lifted, off-label use of eltrombopag among patients with hepatitis C virus (HCV), a subsequently approved indication, increased. The ratio of incidence rate ratios of off-label/HCV to on-label initiation of eltrombopag between the two time periods was significant (13.41; P < 0.001). Our finding of an association with reduced off-label prescribing suggests that REMS with elements to assure safe use can help promote patient safety but may also prevent promising off-label drug uses.

Epidemiology of incident immune thrombocytopenia: a nationwide population-based study in France

Incidence of ITP was 2.9/100 000 person-years with age, seasonal, and regional variations; in adults, 18% were secondary. Severe (gastrointestinal or central nervous system) bleeding at ITP onset was rare (<1%); the risk increased with age.

Agreement of Medicaid claims and electronic health records for assessing preventive care quality among adults

To compare the agreement of electronic health record (EHR) data versus Medicaid claims data in documenting adult preventive care. Insurance claims are commonly used to measure care quality. EHR data could serve this purpose, but little information exists about how this source compares in service documentation. For 13 101 Medicaid-insured adult patients attending 43 Oregon community health centers, we compared documentation of 11 preventive services, based on EHR versus Medicaid claims data. Documentation was comparable for most services. Agreement was highest for influenza vaccination (κ =  0.77; 95% CI 0.75 to 0.79), cholesterol screening (κ = 0.80; 95% CI 0.79 to 0.81), and cervical cancer screening (κ = 0.71; 95% CI 0.70 to 0.73), and lowest on services commonly referred out of primary care clinics and those that usually do not generate claims. EHRs show promise for use in quality reporting. Strategies to maximize data capture in EHRs are needed to optimize the use of EHR data for service documentation.

Minimizing signal detection time in postmarket sequential analysis: balancing positive predictive value and sensitivity

PURPOSE

Outcome misclassification in retrospective epidemiologic analyses has been well-studied, but little is known about such misclassification with respect to sequential statistical analysis during surveillance of medical product-associated risks, a planned capability of the US Food and Drug Administration's Sentinel System.

METHODS

Using a vaccine example, we model and simulate sequential database surveillance in an observational data network using a variety of outcome detection algorithms. We consider how these algorithms, as characterized by sensitivity and positive predictive value, impact the length of surveillance and timeliness of safety signal detection. We show investigators/users of these networks how they can perform preparatory study design calculations that consider outcome misclassification in sequential database surveillance.

RESULTS

Non-differential outcome misclassification generates longer surveillance times and less timely safety signal detection as compared with the case of no misclassification. Inclusive algorithms characterized by high sensitivity but low positive predictive value outperform more narrow algorithms when detecting rare outcomes. This decision calculus may change considerably if medical chart validation procedures were required.

CONCLUSIONS

These findings raise important questions regarding the design of observational data networks used for pharmacovigilance. Specifically, there are tradeoffs involved when choosing to populate such networks with component databases that are large as compared with smaller integrated delivery system databases that can more easily access laboratory or clinical data and perform medical chart validation.

Combining structured and unstructured data to identify a cohort of ICU patients who received dialysis

OBJECTIVE

To develop a generalizable method for identifying patient cohorts from electronic health record (EHR) data-in this case, patients having dialysis-that uses simple information retrieval (IR) tools.

METHODS

We used the coded data and clinical notes from the 24,506 adult patients in the Multiparameter Intelligent Monitoring in Intensive Care database to identify patients who had dialysis. We used SQL queries to search the procedure, diagnosis, and coded nursing observations tables based on ICD-9 and local codes. We used a domain-specific search engine to find clinical notes containing terms related to dialysis. We manually validated the available records for a 10% random sample of patients who potentially had dialysis and a random sample of 200 patients who were not identified as having dialysis based on any of the sources.

RESULTS

We identified 1844 patients that potentially had dialysis: 1481 from the three coded sources and 1624 from the clinical notes. Precision for identifying dialysis patients based on available data was estimated to be 78.4% (95% CI 71.9% to 84.2%) and recall was 100% (95% CI 86% to 100%).

CONCLUSIONS

Combining structured EHR data with information from clinical notes using simple queries increases the utility of both types of data for cohort identification. Patients identified by more than one source are more likely to meet the inclusion criteria; however, including patients found in any of the sources increases recall. This method is attractive because it is available to researchers with access to EHR data and off-the-shelf IR tools.

Determining a definite diagnosis of primary immune thrombocytopenia by medical record review

The objective of this study is to establish a method to identify patients with primary immune thrombocytopenia (ITP) utilizing administrative data from diverse data sources that would be appropriate for epidemiologic studies of ITP, regardless of patients' age and source of health care. Medical records of the Oklahoma University Medical Center, 1995-2004, were reviewed to document the accuracy of the administrative code ICD-9-CM 287.3 for identifying children and adults with ITP, using novel, explicit levels of evidence to identify patients with a definite diagnosis. The proportion of patients diagnosed by hematologists compared to non-hematologists and the proportion of patients diagnosed as outpatients compared to inpatients were determined. For children, age <16 years, 323 outpatient medical records were reviewed; 225 adult outpatient medical records were reviewed. The positive predictive value for the administrative code for identifying patients with a definite diagnosis of ITP by a hematologist was 0.72 in children and 0.69 in adults. In 98% of children and 92% of adults seen as outpatients, the definite diagnosis of ITP was established by a hematologist. One hundred eighteen child and 141 adult inpatient medical records were reviewed. In 95% of children and 83% of adults, the definite diagnosis of ITP by a hematologist was established as an outpatient. This study confirmed the previously reported positive predictive value for the administrative code for identifying patients with ITP. Additionally, it was determined that analysis of hematologists' outpatient administrative codes identified most children and adults with ITP. Am. J. Hematol. 2012. © 2012 Wiley Periodicals, Inc.

Prevalence of primary immune thrombocytopenia in Oklahoma

To determine the prevalence of immune thrombocytopenia (ITP) in Oklahoma regardless of age, clinical characteristics, insurance status, and source of health care. Patients with ITP were identified by the administrative code ICD-9-CM 287.3 in Oklahoma hematologists' offices for a 2-year period, 2003-2004. Prevalence was estimated separately for children (<16 years old) and adults because of their distinct clinical characteristics. Oklahoma census data for 2000 was used as the denominator. Eighty-seven (94%) of 93 eligible Oklahoma hematologists participated; 620 patients with ITP were identified. The average annual prevalences were as follows: 8.1 (95% CI: 6.7-9.5) per 100,000 children, 12.1 (95% CI: 11.1-13.0) per 100,000 adults, and 11.2 (95% CI: 10.4-12.0) per 100,000 population. Among children and adults less than age 70 years, the prevalence was greater among women. Among adults aged 70 years and older, the prevalence was greater among men. The highest prevalence of ITP was among men age 80 years and older. These data document for the first time the prevalence of ITP regardless of age, clinical characteristics, insurance status, and source of health care. The methodology developed for this prevalence analysis may be adaptable for epidemiologic studies of other uncommon disorders which lack specific diagnostic criteria and are treated primarily by medical specialists. Am. J. Hematol. 2012. © 2012 Wiley Periodicals, Inc.

Comorbidities in patients with persistent or chronic immune thrombocytopenia

There is a paucity of epidemiological data on the risk of comorbidities in adults with persistent or chronic immune thrombocytopenia (ITP). In this study, we compared the rates of cataracts, diabetes, renal failure, vascular events, lymphoma, and leukemia among patients with and without persistent or chronic ITP. Using administrative data, adult patients with medical claims for ITP from January, 2000 through September, 2006 were identified. An age- and gender-matched comparison cohort without evidence of ITP was randomly selected. The incidence rate ratio (IRR) of each comorbidity among ITP patients relative to the comparison group was estimated using Poisson regression, adjusting for baseline covariates. A total of 3,131 patients with persistent or chronic ITP were identified, and 9,392 were selected for the comparison cohort. The adjusted IRRs were as follows: diabetes 1.73 (95% CI 1.36-2.20), renal failure 2.05 (95% CI 1.67-2.51), any vascular event 1.70 (95% CI 1.41-2.05), lymphoma 5.91 (95% CI 2.61-13.37), leukemia 19.83 (95% CI 5.84-67.34), and mortality 4.21 (95% CI 3.06-5.79). There was no increased risk for cataract or myocardial infarction in the ITP cohort. Patients with persistent or chronic ITP are at increased risk for several comorbidities including hematologic malignancies, relative to a matched comparison cohort.

Cost and mortality associated with hospitalizations in patients with immune thrombocytopenic purpura

Immune thrombocytopenic purpura (ITP) is associated with low platelet counts and, consequently, a high risk of adverse events leading to hospitalization. However, there are few data on the clinical and economic burden of hospitalizations for ITP. The Nationwide Inpatient Sample (NIS) database of discharges, a stratified 20% sample of all United States (US) community hospitals across all payers, was used to evaluate discharges in ITP patients. We developed nationally representative numbers of discharges in ITP patients from 2003 to 2006 based on diagnosis codes. Using appropriate weights for each NIS discharge, we created national estimates of average cost, length of stay, and in-hospital mortality for specific groups of ITP-related hospitalizations. Approximately 129,000 discharges occurred between 2003 and 2006 in ITP patients. The average cost associated with all discharges in 2008 dollars was 16,476, with a 6.4-day length of stay and in-hospital mortality of 3.8%. In contrast, the average cost of all hospitalizations in the US population during the same period was 10,039, the average length of stay was 4.8 days, and in-hospital mortality was 2.5%. Mortality risk was higher for ITP patients than for the standard US population adjusted for age and gender, with a relative mortality ratio of 1.5 (95% CI: 1.4-1.6). On the basis of a nationally representative sample of US discharge records from 2003 to 2006, hospitalization with ITP represents an economically and clinically important event. ITP was associated with higher costs, longer stays, and more in-hospital deaths on average than all other hospitalized patients combined.

Review: use of electronic medical records for health outcomes research: a literature review

This review assessed the use of electronic medical record (EMR) systems in outcomes research. We systematically searched PubMed to identify articles published from January 2000 to January 2007 involving EMR use for outpatient-based outcomes research in the United States. EMR-based outcomes research studies (n = 126) have increased sixfold since 2000. Although chronic conditions were most common, EMRs were also used to study less common diseases, highlighting the EMRs' flexibility to examine large cohorts as well as identify patients with rare diseases. Traditional multi-variate modeling techniques were the most commonly used technique to address confounding and potential selection bias. Data validation was a component in a quarter of studies, and many evaluated the EMR's ability to achieve similar results previously achieved using other data sources. Investigators using EMR data should aim for consistent terminology, focus on adequately describing their methods, and consider appropriate statistical methods to control for confounding and treatment-selection bias.

The incidence of idiopathic thrombocytopenic purpura among adults: a population-based study and literature review

Published data on the epidemiology of idiopathic thrombocytopenic purpura (ITP) among adults are very limited. We conducted a study of ITP incidence using the General Practice Research Database in the United Kingdom. From 1992 to 2005, there were 840 cases of ITP among adults considering 21 749 623 person-years (PYs) of follow-up, for a crude incidence of 3.9 per 100 000 PYs [95% confidence interval (CI): 3.6, 4.1]. The incidence was higher among women [4.5 per 100 000 PYs (95% CI: 4.2, 4.9)] than men [3.2 per 100 000 PYs (95% CI: 2.8, 3.5)]. Among both women and men, incidence was higher at older ages and in later study years. In a systematic review of previously published literature, incidence of ITP among adults ranged from 1.6 to 2.68 per 100 000 persons per year; prevalence ranged from 9.5 to 23.6 per 100 000 persons. In order to improve the understanding of the disease burden of ITP, future studies should include a clearly defined definition of ITP and focus on well-described source populations that are geographically and ethnically diverse.

Intraobserver and interobserver agreement of International Classification of Diseases, Ninth Revision codes in classifying shoulder instability

HYPOTHESIS

The purpose of this study was to investigate the intraobserver and interobserver reliability of the International Classification of Diseases, Ninth Revision (ICD-9) system when applied to the classification of shoulder instability.

MATERIALS AND METHODS

From December 2005 through February 2007, 50 patients with shoulder instability were evaluated and classified by an attending physician using one of the 16 ICD-9 codes for shoulder instability. Patients were reassessed after two weeks by the original physician and two additional shoulder specialists. 42 patients completed the study. These data were then analyzed to assess intraobserver and interobserver reliability.

RESULTS

Intraobserver agreement for ICD-9 codes was 50% (kappa=0.25, fair). Interobserver agreement was 23% (kappa=0.002, poor).

DISCUSSION

The ICD-9 coding system is the recognized standard for classifying disease states and is used for large epidemiologic studies. The poor agreement demonstrated in this study suggests that the ICD-9 coding system has poor agreement and as such is not a precise method to classify shoulder instability.

CONCLUSION

Shoulder instability cannot reliably be classified using the ICD-9 coding system. Until a more reliable system is developed, epidemiologic studies of shoulder instability that use ICD-9 codes may be difficult to interpret.

LEVEL OF EVIDENCE

Level 1; Testing a previously developed classification system.

Searching for a needle in a haystack: use of ICD-9-CM codes in drug-induced liver injury

OBJECTIVES

The aim of our study was to compare three search strategies using a computerized administrative database to identify cases of idiosyncratic drug-induced liver injury (DILI) due to amoxicillin/clavulanic acid, phenytoin, valproic acid, and isoniazid.

METHODS

In search 1, electronic medical records from patients seen between 1994 and 2004 with an ICD-9-CM code of acute liver injury were identified and cross-searched for the specific drug names in the dictation text. In search 2, all patients with an ICD-9-CM code of drug poisoning/overdose due to one of the four study drugs were identified. In search 3, patients with a poisoning code as well as an acute liver injury code were identified.

RESULTS

Review of the records from the 7,395 search 1 patients yielded 51 DILI cases (0.7%). In contrast, the 566 search 2 patients yielded only three DILI cases (0.5%). Finally, search 3 provided the greatest specificity but a low rate of detection with only two patients (3.9%) having DILI due to one of the four drugs.

CONCLUSION

Acute liver injury ICD-9-CM codes combined with a text search of the dictated medical record yielded the greatest number of DILI cases but was less specific than crossing acute liver injury and poisoning codes. Use of ICD-9-CM codes to identify rare adverse events like DILI remains problematic and highlights the need for prospective surveillance networks.

Simulations showed that validation of database-derived diagnostic criteria based on a small subsample reduced bias

OBJECTIVE

To evaluate alternative approaches to correct for bias due to inaccurate diagnostic criteria in database studies of associations.

STUDY DESIGN AND SETTINGS

A simulation study of a hypothetical cohort of 10,000 subjects selected based on database-derived diagnostic criteria with positive predictive value (PPV) of either 53% or 80%. Analyses focus on the putative association between a drug and the time to a negative outcome. The association is confounded for "false positive" subjects, where the drug acts as a marker for unobserved frailty. First, we estimate the conventional multivariable Cox's Model 1. We then assume having in-depth evaluation of a fraction of subjects, which permits estimating the probabilities of having the disease for all subjects in the cohort. Alternative correction methods use the estimated probability as a confounder (Model 2), a modifier of the drug effect (Model 3), or an importance weight (Model 4).

RESULTS

With a PPV of 53%, Models 1 and 2 induced about 50% underestimation bias for the drug effect. Interaction-based Model 3 yielded the least biased estimates (25% bias), whereas weighting by probability (Model 4) resulted in slightly more biased (33%), but more stable estimates.

CONCLUSION

Proposed methods help reducing bias due to sample contamination.

Agreement between caregiver reported healthcare utilization and administrative data for children with asthma

In asthma, healthcare utilization is frequently an outcome measure and can come from several sources. Asthma-related hospitalizations, emergency department (ED) visits, oral steroid bursts, and outpatient visits were compared between caregiver report and administrative data over 2 years. The difference between sources (caregiver minus administrative) was as follows: hospitalizations = -0.02 (95% limits of agreement, -0.66 to 0.61), ED visits = 0.18 (-1.16 to 1.52), steroid bursts = 0.26 (-3.98 to 4.49), and outpatient visits = 0.29 (-6.10 to 6.64). The percent of individuals with disagreement between sources was hospitalizations = 6.1%; ED visits = 20.2%; steroid bursts = 34.3%; and outpatient visits = 83.6%. The data sources resulted in similar estimates on the population level; however, there were pronounced differences for outpatient visits on an individual level. Importantly, the individual level disagreement between the data sources could negatively affect the perceived quality of care provided by a physician and reduce their compensation in a pay-for-performance system when physicians are rated using administrative data, yet they provide treatment based on patient-reported information.

Prevalence of immune thrombocytopenia: analyses of administrative data

BACKGROUND

The prevalence of immune thrombocytopenic purpura (ITP) in the USA is unknown. The paucity of data makes clinical trial design and resource allocation challenging.

OBJECTIVES

We aimed to quantify the prevalence of ITP in one state and to report on utilization of resources.

METHODS

The Maryland Health Care Commission supplied utilization data on all privately insured Maryland residents in 2002. We identified patients having two claims, separated by at least 30 days, for International Classification of Diseases, Ninth Revision, Clinical Modification code 287.3 (expected to be predominantly ITP). We excluded patients with concurrent diagnoses that made ITP unlikely. In sensitivity analyses, we varied the required visit interval between 14 and 180 days. We quantified ITP prevalence, resource utilization, and prevalence of concurrent autoimmune illnesses.

RESULTS

The age-adjusted prevalence of ITP was 9.5 per 100,000 persons (10.5 per 100,000 when requiring a minimum 14-day interval and 4.5 per 100,000 with a 180-day interval). There was a predominance of males in childhood and of females in the middle-adult years, with an overall prevalence rate ratio of 1.9 for females to males. Twenty per cent of these patients were hospitalized, but emergency department use was rare, as was splenectomy. A concurrent diagnosis of multiple sclerosis was 25 times more prevalent than anticipated.

CONCLUSIONS

We conclude that the prevalence of ITP in one populous state in the USA is comparable with that which has been reported in Europe. The suggested co-occurrence of ITP and multiple sclerosis in children merits further investigation.

Short-term prediction of mortality in patients with systemic lupus erythematosus: classification of outcomes using random forests

OBJECTIVE

To identify demographic and clinical characteristics that classify patients with systemic lupus erythematosus (SLE) at risk for in-hospital mortality.

METHODS

Patients hospitalized in California from 1996 to 2000 with a principal diagnosis of SLE (N = 3,839) were identified from a state hospitalization database. As candidate predictors of mortality, we used patient demographic characteristics; the presence or absence of 40 different clinical conditions listed among the discharge diagnoses; and 2 summary indexes derived from the discharge diagnoses, the Charlson Index and the SLE Comorbidity Index. Predictors of patients at increased risk of mortality were identified and validated using random forests, a statistical procedure that is a generalization of single classification trees. Random forests use bootstrapped samples of patients and randomly selected subsets of predictors to create individual classification trees, and this process is repeated to generate multiple trees (a forest). Classification is then done by majority vote across all trees.

RESULTS

Of the 3,839 patients, 109 died during hospitalization. Selecting from all available predictors, the random forests had excellent predictive accuracy for classification of death. The mean classification error rate, averaged over 10 forests of 500 trees each, was 11.9%. The most important predictors were the Charlson Index, respiratory failure, SLE Comorbidity Index, age, sepsis, nephritis, and thrombocytopenia.

CONCLUSION

Information on clinical diagnoses can be used to accurately predict mortality among hospitalized patients with SLE. Random forests represent a useful technique to identify the most important predictors from a larger (often much larger) number and to validate the classification.

Accuracy of coding for possible warfarin complications in hospital discharge abstracts

BACKGROUND

Hospital discharge abstracts could be used to identify complications of warfarin if coding for bleeding and thromboembolic events are accurate.

OBJECTIVES

To measure the accuracy of International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9CM) codes for bleeding and thromboembolic diagnoses.

SETTING

University affiliated, tertiary care hospital in Ottawa, Canada.

PATIENTS

A random sample of patients discharged between September 1999 and September 2000 with an ICD-9-CM code indicating a bleeding or thromboembolic diagnosis.

METHODS

Gold-standard coding was determined by a trained chart abstractor using explicit standard diagnostic criteria for bleeding, major bleeding, and acute thromboembolism. The abstractor was blinded to the original coding. We calculated the sensitivity, specificity, positive, and negative predictive values of the original ICD-9CM codes for bleeding or thromboembolism diagnoses.

RESULTS

We reviewed 616 medical records. 361 patients (59%) had a code indicating a bleeding diagnosis, 291 patients (47%) had a code indicating a thromboembolic diagnosis and 36 patients (6%) had a code indicating both. According to the gold standard criteria, 352 patients experienced bleeding, 333 experienced major bleeding, and 188 experienced an acute thromboembolism. For bleeding, the ICD-9CM codes had the following sensitivity, specificity, positive and negative predictive values [95% CI]: 93% [90-96], 88% [83-91], 91% [88-94], and 91% [87-94], respectively. For major bleeding, the ICD-9CM codes had the following sensitivity, specificity, positive and negative predictive values: 94% [91-96], 83% [78-87], 87% [83-90], and 92% [88-95], respectively. For thromboembolism, the ICD-9CM codes had the following sensitivity, specificity, positive and negative predictive values: 97% [94-99], 74% [70-79], 62% [57-68], and 98% [96-99], respectively. By selecting a sub-group of ICD-9CM codes for thromboembolism, the positive predictive value increased to 87%.

CONCLUSION

In our centre, the discharge abstract could be used to identify and exclude patients hospitalized with a major bleed or thromboembolism. If coding quality for bleeding is similar in other hospitals, these ICD-9-CM diagnostic codes could be used to study population-based warfarin-associated hemorrhagic complications using administrative databases.