Early impact of the ICD-10-CM transition on selected health outcomes in 13 electronic health care databases in the United States

Translation of a Claims-Based Frailty Index From the International Classification of Diseases, Ninth Revision, Clinical Modification to the Tenth Revision

The Faurot frailty index (FFI) is a validated algorithm that uses enrollment and International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM)-based billing information from Medicare claims data as a proxy for frailty. In October 2015, the US health-care system transitioned from the ICD-9-CM to the International Classification of Diseases, Tenth Revision, Clinical Modification (ICD-10-CM). Applying the Centers for Medicare and Medicaid Services General Equivalence Mappings, we translated diagnosis-based frailty indicator codes from the ICD-9-CM to the ICD-10-CM, followed by manual review. We used interrupted time-series analysis of Medicare data to assess the comparability of the pre- and posttransition FFI scores. In cohorts of beneficiaries enrolled in January 2015-2017 with 8-month frailty look-back periods, we estimated associations between the FFI and 1-year risk of aging-related outcomes (mortality, hospitalization, and admission to a skilled nursing facility). Updated indicators had similar prevalences as pretransition definitions. The median FFI scores and interquartile ranges (IQRs) for the predicted probability of frailty were similar before and after the International Classification of Diseases transition (pretransition: median, 0.034 (IQR, 0.02-0.07); posttransition: median, 0.038 (IQR, 0.02-0.09)). The updated FFI was associated with increased risks of mortality, hospitalization, and skilled nursing facility admission, similar to findings from the ICD-9-CM era. Studies of medical interventions in older adults using administrative claims should use validated indices, like the FFI, to mitigate confounding or assess effect-measure modification by frailty.

Adverse drug reactions in hospitals: population estimates for Portugal and the ICD-9-CM to ICD-10-CM crosswalk

BACKGROUND

Adverse drug reactions (ADR), both preventable and non-preventable, are frequent and pose a significant burden. This study aimed to produce up-to-date estimates for ADR rates in hospitals, in Portugal, from 2010 to 2018. In addition, it explores possible pitfalls when crosswalking between ICD-9-CM and ICD-10-CM code sets for ADR identification.

METHODS

The Portuguese Hospital Morbidity Database was used to identify hospital episodes (outpatient or inpatient) with at least one ICD code of ADR. Since the study period spanned from 2010 to 2018, both ICD-9-CM and ICD-10-CM codes based on previously published studies were used to define episodes. This was an exploratory study, and descriptive statistics were used to provide ADR rates and summarise episode features for the full period (2010-2018) as well as for the ICD-9-CM (2010-2016) and ICD -10-CM (2017-2018) eras.

RESULTS

Between 2010 and 2018, ADR occurred in 162,985 hospital episodes, corresponding to 1.00% of the total number of episodes during the same period. Higher rates were seen in the oldest age groups. In the same period, the mean annual rate of episodes related to ADR was 174.2/100,000 population. The episode rate (per 100,000 population) was generally higher in males, except in young adults (aged '15-20', '25-30' and '30-35' years), although the overall frequency of ADR in hospital episodes was higher in females.

CONCLUSIONS

Despite the ICD-10-CM transition, administrative health data in Portugal remain a feasible source for producing up-to-date estimates on ADR in hospitals. There is a need for future research to identify target recipients for preventive interventions and improve medication safety practices in Portugal.

Sex differences in temporal trends in main and secondary pulmonary embolism diagnosis and case fatality rates: 2003-2019

AIMS

There is controversy regarding the incidence and outcomes of pulmonary embolism (PE) according to sex. Our aim was to address sex differences in temporal trends in main and secondary hospital PE diagnoses, management and case fatality rates (CFR).

METHODS AND RESULTS

Retrospective analysis of Spain´s National Healthcare System hospital database, years 2003-2019, for patients ≥18 years with main or secondary PE diagnosis. Trends by sex in hospital diagnosis, use of procedures, and CFRs were analysed by joinpoint and Poisson regression models. Of 339 469 PE diagnoses, 52% were in women. Sixty-five percent were main diagnosis, 35.2% secondary. Total annual diagnoses and frequentation rates increased similarly in men and women: average annual percent change (AAPC): 2.0% (95% CI, 1.3-2.6; P < 0.005). Secondary PEs were more common in men (37.8% vs. 32.9%, P < 0.001). Men showed greater comorbidity than women (Charlson index 2.22 ± 0.01 vs. 1.74 ± 0.01, P < 0.001), particularly cancer in the secondary diagnosis group (40.9% vs. 31.6%, P < 0.001). CFRs for PE as main diagnosis were comparable and decreased in parallel in men (from 13.8% in 2003 to 7.3% in 2019) and women (from 13.1% in 2003 to 6.9% in 2019). However, for PE as secondary diagnosis, CFRs remained higher (P < 0.001) in men (from 42.5% in 2003 to 26.2% in 2019) than women (from 34.4% in 2003 to 22.8% in 2019).

CONCLUSION

PE hospital diagnosis increased significantly between 2003 and 2019 in men and women for both main and secondary diagnosis. Although in-hospital CFR decreased one third still remains very high, especially in men with secondary PE diagnosis.

The Impact of Principal Diagnosis on Readmission Risk among Patients Hospitalized for Community-Acquired Pneumonia

Coding variation distorts performance/outcome statistics not eliminated by risk adjustment. Among 1596 community-acquired pneumonia patients hospitalized from 1998 to 2012 identified using an evidence-based algorithm, the authors measured the association of principal diagnosis (PD) with 30-day readmission, stratified by Pneumonia Severity Index risk class. The 152 readmitted patients were more ill (Pneumonia Severity Index class V 38.8% versus 25.8%) and less likely to have a pneumonia PD (52.6% versus 69.9%). Among patients with PDs of pneumonia, respiratory failure, sepsis, and aspiration, mortality/readmission rates were 3.9/8.5%, 28.8/14.0%, 24.7/19.6%, and 9.0/15.0%, respectively. The nonpneumonia PDs were associated with a greater risk of adjusted 30-day readmission: respiratory failure odds ratio (OR) 1.89 (95% confidence interval [CI], 1.13-3.15), sepsis OR 2.54 (95% CI, 1.52-4.26), and possibly aspiration OR 1.73 (95% CI, 0.88-3.41). With increasing use of alternative PDs among pneumonia patients, quality reporting must account for variations in condition coding practices. Rigorous risk adjustment does not eliminate the need for accurate, consistent case definition in producing valid quality measures.

Impact of the Transition from ICD-9-CM to ICD-10-CM on the Rates of Severe Maternal Morbidity in Arkansas: An Analysis of Claims Data

Background:

Severe maternal morbidity (SMM) is considered as a near miss for maternal death, therefore it is crucial to identify and prevent SMM. Medical insurance claims can be used to identify SMM. There was a national transition from the International Classification of Diseases, 9th Revision, Clinical Modification (ICD-9-CM) to International Classification of Diseases, 10th Revision, Clinical Modification/Procedure Coding System (ICD-10-CM/PCS) in October 2015.

Objective:

This study investigates the impact of transition from ICD-9-CM to ICD-10-CM on the rates of SMM in the state of Arkansas using birth certificates linked with insurance claims data in the Arkansas All-Payer Claims Database (APCD).

Study Design:

Birth certificates between January 1, 2013, and December 31, 2017, were linked to insurance claims data from the APCD. SMM was defined using the algorithm provided by the Centers for Disease Control and Prevention, using ICD-9 codes for births before October 1, 2015, and ICD-10-CM codes for births on or after October 1, 2015.

Results:

The incidence of SMM increased after transition to the ICD-10-CM system in Arkansas. The relatively higher rate of SMM in ICD-10-CM versus ICD-9-CM was greater in magnitude on the delivery day and throughout the 42-day postpartum period (odds ratio [OR]: 1.30; 95% confidence interval [CI]: 1.20–1.42) compared with the rate on the day of delivery (OR: 1.20; 95% CI: 1.06–1.36). When excluding blood transfusions, the higher rate of SMM during the ICD-10 era was even greater both in the delivery day and 42-day postpartum period (OR: 1.66; 95% CI: 1.49–1.85) and on the day of delivery (OR: 1.58; 95% CI: 1.31–1.90).

Background rates of five thrombosis with thrombocytopenia syndromes of special interest for COVID-19 vaccine safety surveillance: Incidence between 2017 and 2019 and patient profiles from 38.6 million people in six European countries

Background

Thrombosis with thrombocytopenia syndrome (TTS) has been reported among individuals vaccinated with adenovirus‐vectored COVID‐19 vaccines. In this study, we describe the background incidence of non‐vaccine induced TTS in six European countries.

Methods

Electronic medical records from France, the Netherlands, Italy, Germany, Spain, and the United Kingdom informed the study. Incidence rates of cerebral venous sinus thrombosis (CVST), splanchnic vein thrombosis (SVT), deep vein thrombosis (DVT), pulmonary embolism (PE), and myocardial infarction or ischemic stroke, all with concurrent thrombocytopenia, were estimated among the general population of persons in a database between 2017 and 2019. A range of additional potential adverse events of special interest for COVID‐19 vaccinations were also studied in a similar manner.

Findings

A total of 38 611 617 individuals were included. Background rates ranged from 1.0 (95% CI: 0.7–1.4) to 8.5 (7.4–9.9) per 100 000 person‐years for DVT with thrombocytopenia, from 0.5 (0.3–0.6) to 20.8 (18.9–22.8) for PE with thrombocytopenia, from 0.1 (0.0–0.1) to 2.5 (2.2–2.7) for SVT with thrombocytopenia, and from 1.0 (0.8–1.2) to 43.4 (40.7–46.3) for myocardial infarction or ischemic stroke with thrombocytopenia. CVST with thrombocytopenia was only identified in one database, with incidence rate of 0.1 (0.1–0.2) per 100 000 person‐years. The incidence of non‐vaccine induced TTS increased with age, and was typically greater among those with more comorbidities and greater medication use than the general population. It was also more often seen in men than women. A large proportion of those affected were seen to have been taking antithrombotic and anticoagulant therapies prior to their event.

Interpretation

Although rates vary across databases, non‐vaccine induced TTS has consistently been seen to be a very rare event among the general population. While still remaining very rare, rates were typically higher among older individuals, and those affected were also seen to generally be male and have more comorbidities and greater medication use than the general population.

Cardiovascular Effectiveness of Sodium-Glucose Cotransporter 2 Inhibitors and Glucagon-Like Peptide-1 Receptor Agonists in Older Patients in Routine Clinical Care With or Without History of Atherosclerotic Cardiovascular Diseases or Heart Failure

Background Randomized trials demonstrate the cardioprotective effects of sodium-glucose cotransporter 2 inhibitors (SGLT2i) and glucagon-like peptide-1 receptor agonists (GLP-1RA). We evaluated their relative cardiovascular effectiveness in routine care populations with a broad spectrum of atherosclerotic cardiovascular diseases (CVDs) or heart failure (HF). Methods and Results We identified Medicare beneficiaries from 2013 to 2017, aged >65 years, initiating SGLT2i (n=24 747) or GLP-1RA (n=22 596) after a 1-year baseline. On the basis of diagnoses during baseline, we classified patients into: (1) no HF or CVD, (2) HF but no CVD, (3) no HF but CVD, and (4) both HF and CVD. We identified hospitalized HF and atherosclerotic CVD outcomes from drug initiation until treatment changes, death, or disenrollment. We estimated propensity score-weighted 2-year risk ratios (RRs) and risk differences, accounting for measured confounding, informative censoring, and competing risk. In patients with no CVD or HF, SGLT2i reduced the hospitalized HF risk compared with GLP-1RA (propensity score-weighted RR, 0.65; 95% CI, 0.43-0.96). The association was strongest in those who had HF but no CVD (RR, 0.48; 95% CI, 0.25-0.85). The combined myocardial infarction, stroke, and mortality outcome risk was slightly higher for SGLT2i compared with GLP-1RA in those without CVD or HF (RR, 1.31; 95% CI, 1.09-1.56). The association was favorable toward SGLT2i in subgroups with a history of HF. Conclusions SGLT2i reduced the cardiovascular risk versus GLP-1RA in patients with a history of HF but no CVD. Atherosclerotic CVD events were less frequent with GLP-1RA in those without prior CVD or HF.

Inpatient Diagnosis of Delirium and Encephalopathy: Coding Trends 2011-2018

OBJECTIVES

Ten medical societies have called for scientific literature to integrate research on delirium and encephalopathy, while physicians continually debate how to accurately document diagnoses of acute confusional states. To promote this integration, we evaluated trends in diagnoses of delirium and encephalopathy among hospitalized adults and physician specialties, incorporating transitions to the Diagnostic and Statistical Manual of Mental Disorders-5 (DSM-5) and the International Classification of Disease-10 (ICD-10).

METHODS

Using 2011-2018 IBM MarketScan datasets, we identified delirium/encephalopathy patients ≥ 18 years using ICD-9/10 codes among hospitalized patients. We identified physician specialties associated with the hospitalization and comorbidities within one year prior to the diagnosis of delirium or encephalopathy. Log-binomial models were used to evaluate diagnostic trends, adjusting for age, gender, insurance and comorbidities.

RESULTS

We identified 10,509 hospitalized patients with diagnosis of delirium and 94,438 with encephalopathy between 2011-2018. Although the number of patients with either diagnosis increased over time, increased use of delirium diagnosis was less than it was for encephalopathy compared to 2011 after adjusting for covariates (ARR 0.45; 95% CI 0.43 to 0.48). During the 8 years, neurologists and internists increased their use of both diagnoses, whereas psychiatrists only increased their use of delirium. Family practice physicians and nurse practitioners presented no significant change in either diagnosis for this timeframe.

CONCLUSIONS

Our results suggest that refined diagnostic codes and criteria may alter trends among clinicians in diagnosing delirium and/or encephalopathy. Additional diagnostic clarity may be necessary to support refined diagnoses among family practice physicians and nurse practitioners.

Developing real-world evidence from real-world data: Transforming raw data into analytical datasets

Development of evidence-based practice requires practice-based evidence, which can be acquired through analysis of real-world data from electronic health records (EHRs). The EHR contains volumes of information about patients-physical measurements, diagnoses, exposures, and markers of health behavior-that can be used to create algorithms for risk stratification or to gain insight into associations between exposures, interventions, and outcomes. But to transform real-world data into reliable real-world evidence, one must not only choose the correct analytical methods but also have an understanding of the quality, detail, provenance, and organization of the underlying source data and address the differences in these characteristics across sites when conducting analyses that span institutions. This manuscript explores the idiosyncrasies inherent in the capture, formatting, and standardization of EHR data and discusses the clinical domain and informatics competencies required to transform the raw clinical, real-world data into high-quality, fit-for-purpose analytical data sets used to generate real-world evidence.

Impact of ICD-9 to ICD-10 Coding Transition on Prevalence Trends in Neurology

Objective

To determine the effect of International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) to International Classification of Diseases, 10th Revision, Clinical Modification (ICD-10-CM) coding transition on the point prevalence and longitudinal trends of 16 neurologic diagnoses.

Methods

We used 2014-2017 data from the National Inpatient Sample to identify hospitalizations with one of 16 common neurologic diagnoses. We used published ICD-9-CM codes to identify hospitalizations from January 1, 2014, to September 30, 2015, and used the Agency for Healthcare Research and Quality's MapIt tool to convert them to equivalent ICD-10-CM codes for October 1, 2015-December 31, 2017. We compared the prevalence of each diagnosis before vs after the ICD coding transition using logistic regression and used interrupted time series regression to model the longitudinal change in disease prevalence across time.

Results

The average monthly prevalence of subarachnoid hemorrhage was stable before the coding transition (average monthly increase of 4.32 admissions, 99.7% confidence interval [CI]: -8.38 to 17.01) but increased after the coding transition (average monthly increase of 24.32 admissions, 99.7% CI: 15.71-32.93). Otherwise, there were no significant differences in the longitudinal rate of change in disease prevalence over time between ICD-9-CM and ICD-10-CM. Six of 16 neurologic diagnoses (37.5%) experienced significant changes in cross-sectional prevalence during the coding transition, most notably for status epilepticus (odds ratio 0.30, 99.7% CI: 0.26-0.34).

Conclusions

The transition from ICD-9-CM to ICD-10-CM coding affects prevalence estimates for status epilepticus and other neurologic disorders, a potential source of bias for future longitudinal neurologic studies. Studies should limit to 1 coding system or use interrupted time series models to adjust for changes in coding patterns until new neurology-specific ICD-9 to ICD-10 conversion maps can be developed.

Validation of a Claims-based Algorithm to Identify Pregestational Diabetes Among Pregnant Women in the United States

BACKGROUND

Identifying pregestational diabetes in pregnant women using administrative claims databases is important for studies of the safety of antidiabetic treatment in pregnancy, but limited data are available on the validity of case-identifying algorithms. The purpose of this study was to evaluate the validity of an administrative claims-based algorithm to identify pregestational diabetes.

METHODS

Using a cohort of pregnant women nested within the Medicaid Analytic Extract (MAX) database, we developed an algorithm to identify pregestational type 1 and type 2 diabetes, distinct from gestational diabetes. Within a single large healthcare system in the Boston area, we identified women who delivered an infant between 2000 and 2010 and were covered by Medicaid, and linked their electronic health records to their Medicaid claims within MAX. Medical records were reviewed by two physicians blinded to the algorithm classification to confirm or rule out pregestational diabetes, with disagreements resolved by discussion. We calculated positive predictive values with 95% confidence intervals using the medical record as the reference standard.

RESULTS

We identified 49 pregnancies classified by the claims-based algorithm as pregestational diabetes that were linked to the electronic health records and had records available for review. The PPV for any pregestational diabetes was 92% [95% confidence interval (CI) 82%, 97%], type 2 diabetes 87% (68%, 95%), and type 1 diabetes 57% (37%, 75%).

CONCLUSIONS

The claims-based algorithm for pregestational diabetes and type 2 diabetes performed well; however, the PPV was low for type 1 diabetes.

Claims-based measures of prescription opioid utilization: A practical guide for researchers

BACKGROUND

Given the increased attention to the opioid epidemic and the role of inappropriate prescribing, there has been a marked increase in the number of studies using claims data to study opioid use and policies designed to curb misuse. Our objective is to review the medical literature for recent studies that use claims data to construct opioid use measures and to develop a guide for researchers using these measures.

METHODS

We searched for articles relating to opioid use measured in health insurance claims data using a defined set of search terms for the years 2014-2020. Original research articles based in the United States that used claims-based measures of opioid utilization were included and information on the study population and measures of any opioid use, quantity of opioid use, new opioid use, chronic opioid use, multiple providers, and overlapping prescriptions was abstracted.

RESULTS

A total of 164 articles met inclusion criteria. Any opioid use was the most commonly included measure, defined by 85 studies. This was followed by quantity of opioids (68 studies), chronic opioid use (53 studies), overlapping prescriptions (28 studies), and multiple providers (8 studies). Each measure contained multiple, distinct definitions with considerable variation in how each was operationalized.

CONCLUSIONS

Claims-based opioid utilization measures are commonly used in research, but definitions vary significantly from study to study. Researchers should carefully consider which opioid utilization measures and definitions are most appropriate for their study and recognize how different definitions may influence study results.

Impact of the ICD-9-CM to ICD-10-CM transition on the incidence of severe maternal morbidity among delivery hospitalizations in the United States

BACKGROUND

Surveillance of maternal mortality and severe maternal morbidity is important to identify temporal trends, evaluate the impact of clinical practice changes or interventions, and monitor quality of care. A common source for severe maternal morbidity surveillance is hospital discharge data. On October 1, 2015, all hospitals in the United States transitioned from the International Classification of Diseases, Ninth Revision, Clinical Modification to the International Classification of Diseases, Tenth Revision, Clinical Modification coding for diagnoses and procedures.

OBJECTIVE

This study aimed to evaluate the impact of the transition from the International Classification of Diseases, Ninth Revision, Clinical Modification to the International Classification of Diseases, Tenth Revision, Clinical Modification coding systems on the incidence of severe maternal morbidity in the United States in hospital discharge data.

STUDY DESIGN

Using data from the National Inpatient Sample, obstetrical deliveries between January 1, 2012, and December 31, 2017, were identified using a validated case definition. Severe maternal morbidity was defined using the International Classification of Diseases, Ninth Revision, Clinical Modification (January 1, 2012, to September 30, 2015) and the International Classification of Diseases, Tenth Revision, Clinical Modification (October 1, 2015, to December 31, 2017) codes provided by the Centers for Disease Control and Prevention. An interrupted time series and segmented regression analysis was used to assess the impact of the transition from the International Classification of Diseases, Ninth Revision, Clinical Modification to the International Classification of Diseases, Tenth Revision, Clinical Modification coding on the incidence of severe maternal morbidity per 1000 obstetrical deliveries.

RESULTS

From 22,751,941 deliveries, the incidence of severe maternal morbidity in the International Classification of Diseases, Ninth Revision, Clinical Modification coding era was 19.04 per 1000 obstetrical deliveries and decreased to 17.39 per 1000 obstetrical deliveries in the International Classification of Diseases, Tenth Revision, Clinical Modification coding era (P<.001). The transition to International Classification of Diseases, Tenth Revision, Clinical Modification coding led to an immediate decrease in the incidence of severe maternal morbidity (-2.26 cases of 1000 obstetrical deliveries) (P<.001). When blood products transfusion was removed from the case definition, the magnitude of the decrease in the incidence of SMM was much smaller (-0.60 cases/1000 obstetric deliveries), but still significant (P<.001).

CONCLUSION

After the transition to the International Classification of Diseases, Tenth Revision, Clinical Modification coding for health diagnoses and procedures in the United States, there was an abrupt statistically significant and clinically meaningful decrease in the incidence of severe maternal morbidity in hospital discharge data. Changes in the underlying health of the obstetrical population are unlikely to explain the sudden change in severe maternal morbidity. Although much work has been done to validate the International Classification of Diseases, Ninth Revision, Clinical Modification codes for severe maternal morbidity, it is critical that validation studies be undertaken to validate the International Classification of Diseases, Tenth Revision, Clinical Modification codes for severe maternal morbidity to permit ongoing surveillance, quality improvement, and research activities that rely on hospital discharge data.

Capturing Intravenous Thrombolysis for Acute Stroke at the ICD-9 to ICD-10 Transition: Case Volume Discontinuity in the United States National Inpatient Sample

Background

Transition from International Classification of Diseases (ICD) Ninth and Tenth Revisions (ICD‐9 and ICD‐10) for hospital discharge data was mandated for US hospitals on October 1, 2015. We examined the volume of patients receiving thrombolysis in ischemic stroke (IS) identified using ICD codes within this transition period in the 2015 to 2016 National Inpatient Sample, a weighted 20% sample of all inpatient US hospital discharges.

Methods and Results

During the ICD‐10 period, 2 case identification strategies were used. Codes for IS were combined with: (1) only the ICD‐10 code for thrombolytic given into a peripheral vein and (2) all new ICD‐10 codes mapped to the ICD‐9 code for all thrombolysis. On visual inspection there was an obvious discontinuity in the volume of patients with IS treated with IV thrombolysis corresponding to 3 time periods: ICD‐9 (study period 1), transition (period 2), and ICD‐10 (period 3). With Strategy 1, analysis using a linear spline with 2 knots shows that the volume of patients with IS treated with IV thrombolysis was significantly different between study periods 1 and 2 (slope difference −1880, 95% CI −2834 to −928, P=0.005), and periods 2 to 3 (slope difference 1980, 95% CI 1207–2754, P = 0.002). With Strategy 2, volumes did not change significantly between periods 1 to 2, though there was a significant difference between periods 2 and 3 (slope difference 719, 95% CI 91–1347, P=0.034).

Conclusions

The significant discontinuity in thrombolysis volumes for IS during the transition period for ICD‐9 to ICD‐10 coding suggests that more rigorous validation of US administrative data during this time period may be necessary for research, resource planning, and quality assurance.

Effect of ICD-9-CM to ICD-10-CM coding system transition on identification of common conditions: An interrupted time series analysis

PURPOSE

To evaluate the effect of diagnostic coding system transition on the identification of common conditions recorded in Taiwan's national claims database.

METHODS

Using the National Health Insurance Research Database, we estimated the 3-month prevalence of recorded diagnosis of 32 conditions based on the ICD-9-CM codes in 2014-2015 and the ICD-10-CM codes in 2016-2017. Two algorithms were assessed for ICD-10-CM: validated ICD-10 codes in the literature and codes translated from ICD-9-CM using an established mapping algorithm. We used segmented regression analysis on time-series data to examine changes in the 3-month prevalence (both level and trend) before and after the ICD-10-CM implementation.

RESULTS

Significant changes in the level were found in 19 and 11 conditions when using the ICD-10 codes from the literature and mapping algorithm, respectively. The conditions with inconsistent levels by both of the algorithms were valvular heart disease, peripheral vascular disease, mild liver disease, moderate to severe liver disease, metastatic cancer, rheumatoid arthritis and collagen vascular diseases, coagulopathy, blood loss anemia, deficiency anemia, alcohol abuse, and psychosis. Nine conditions had significant changes in the trend when using the ICD-10 codes from the literature or mapping algorithm.

CONCLUSIONS

Less than half of the 32 conditions studied had a smooth transition between the ICD-9-CM and ICD-10-CM coding systems. Researchers should pay attention to the conditions where the coding definitions result in inconsistent time series estimates.

Acute myocardial infarction: Development and application of an ICD-10-CM-based algorithm to a large U.S. healthcare claims-based database

Background

Healthcare administrative claims data hold value for monitoring drug safety and assessing drug effectiveness. The U.S. Food and Drug Administration Biologics Effectiveness and Safety Initiative (BEST) is expanding its analytical capacity by developing claims-based definitions—referred to as algorithms—for populations and outcomes of interest. Acute myocardial infarction (AMI) was of interest due to its potential association with select biologics and the lack of an externally validated International Classification of Diseases, 10^th Revision, Clinical Modification (ICD-10-CM) algorithm.

Objective

Develop and apply an ICD-10-CM-based algorithm in a U.S. administrative claims database to identify and characterize AMI populations.

Methods

A comprehensive literature review was conducted to identify validated AMI algorithms. Building on prior published methodology and consistent application of ICD-9-CM codes, an ICD-10-CM algorithm was developed via forward-backward mapping using General Equivalence Mappings and refined with clinical input. An AMI population was then identified in the IBM^® MarketScan^® Research Databases and characterized using descriptive statistics.

Results and discussion

Between 2014–2017, 2.83–3.16 individuals/1,000 enrollees/year received ≥1 AMI diagnosis in any healthcare setting. The 2015 transition to ICD-10-CM did not result in a substantial change in the proportion of patients identified. Average patient age at first AMI diagnosis was 64.9 years, and 61.4% of individuals were male. Unspecified chest pain, hypertension, and coronary atherosclerosis of native coronary vessel/artery were most commonly reported within one day of AMI diagnosis. Electrocardiograms were the most common medical procedure and beta-blockers were the most commonly ordered cardiac medication in the one day before to 14 days following AMI diagnosis. The mean length of inpatient stay was 5.6 days (median 3 days; standard deviation 7.9 days). Findings from this ICD-10-CM-based AMI study were internally consistent with ICD-9-CM-based findings and externally consistent with ICD-9-CM-based studies, suggesting that this algorithm is ready for validation in future studies.

The Impact of Transitioning From International Classification of Diseases, Ninth Revision to International Classification of Diseases, Tenth Revision on Reported Complication Rates Following Primary Total Knee Arthroplasty

BACKGROUND

In 2015, the healthcare system transitioned from International Classification of Diseases, Ninth Revision (ICD-9) coding to the Tenth Revision (ICD-10). We sought to determine the effect of this change on the reported incidence of complications following total knee arthroplasty (TKA).

METHODS

The Humana administrative claims database was queried from 2 years prior to October 1, 2015 (ICD-9 cohort) and for 1 year after this date (ICD-10 cohort) to identify all TKA procedures. Complications occurring within 6 months of surgery were captured using the respective coding systems. Incidence of each complication was compared between cohorts using risk ratios (RR) and 95% confidence intervals.

RESULTS

There were 19,009 TKAs in the ICD-10 cohort and 38,172 TKAs in the ICD-9 cohort. The incidence of each complication analyzed was significantly higher in the ICD-9 cohort relative to the ICD-10 cohort. Periprosthetic joint infection occurred in 1.9% vs 1.3% (RR 1.5, 1.3-1.9), loosening in 0.3% vs 0.1% (RR 2.7, 1.8-4.9), periprosthetic fracture in 0.3% vs 0.1% (RR 3.0, 1.6-4.5), and other mechanical complications in 0.7% vs 0.4% (RR 2.0, 1.5-2.5) (P < .05 for all).

CONCLUSION

The transition from ICD-9 to ICD-10 coding has altered the reported incidence of complications following TKA. These results are likely due to the added complexity of ICD-10 which is joint and laterality specific. It is important to understand the differences between coding systems as this data is used for quality initiatives, risk adjustment models, and clinical research. Thoughtful methodology will be necessary when ICD-9 and ICD-10 data are being analyzed simultaneously.

Using and improving distributed data networks to generate actionable evidence: the case of real-world outcomes in the Food and Drug Administration's Sentinel system

The US Food and Drug Administration (FDA) Sentinel System uses a distributed data network, a common data model, curated real-world data, and distributed analytic tools to generate evidence for FDA decision-making. Sentinel system needs include analytic flexibility, transparency, and reproducibility while protecting patient privacy. Based on over a decade of experience, a critical system limitation is the inability to identify enough medical conditions of interest in observational data to a satisfactory level of accuracy. Improving the system's ability to use computable phenotypes will require an "all of the above" approach that improves use of electronic health data while incorporating the growing array of complementary electronic health record data sources. FDA recently funded a Sentinel System Innovation Center and a Community Building and Outreach Center that will provide a platform for collaboration across disciplines to promote better use of real-world data for decision-making.

Health outcomes coding trends in the US Food and Drug Administration's Sentinel System during transition to International Classification of Diseases-10 coding system: A brief review

Background and purpose

The transition from International Classification of Diseases, 9th revision, clinical modification (ICD‐9‐CM) to ICD‐10‐CM poses a challenge to epidemiologic studies that use diagnostic codes to identify health outcomes and covariates. We evaluated coding trends in health outcomes in the US Food and Drug Administration's Sentinel System during the transition.

Methods

We reviewed all health outcomes coding trends reports on the Sentinel website through November 30, 2019 and analyzed trends in incidence and prevalence across the ICD‐9‐CM and ICD‐10‐CM eras by visual inspection.

Results

We identified 78 unique health outcomes (22 acute, 32 chronic, and 24 acute or chronic) and 140 time‐series graphs of incidence and prevalence. The reports also included code lists and code mapping methods used. Of the 140 graphs reviewed, 81 (57.9%) showed consistent trends across the ICD‐9‐CM and ICD‐10‐CM eras, while 51 (36.4%) and 8 (5.7%) graphs showed inconsistent and uncertain trends, respectively. Chronic HOIs and acute/chronic HOIs had higher proportions of consistent trends in prevalence definitions (83.9% and 78.3%, respectively) than acute HOIs (28.6%). For incidence, 55.6% of acute HOIs showed consistent trends, while 41.2% of chronic HOIs and 39.3% of acute/chronic HOIs showed consistency.

Conclusions

Researchers using ICD‐10‐CM algorithms obtained by standardized mappings from ICD‐9‐CM algorithms should assess the mapping performance before use. The Sentinel reports provide a valuable resource for researchers who need to develop and assess mapping strategies. The reports could benefit from additional information about the algorithm selection process and additional details on monthly incidence and prevalence rates.

Key points

We reviewed health outcomes coding trends reports on the US FDA Sentinel website through November 30, 2019 and analyzed trends in incidence and prevalence across the International Classification of Diseases, 9th revision, Clinical Modification (ICD‐9‐CM) and ICD‐10‐CM eras by code mapping method and the type of health outcomes of interest (acute, chronic, acute or chronic).

More than a third of the 140 time‐series graphs of incidence and prevalence of health outcomes showed inconsistent or uncertain trends. Consistency in trends varied by code mapping method, type of health outcomes of interest, and whether the measurement was incidence or prevalence.

Studies using ICD‐9‐CM‐based algorithms mapped to ICD‐10‐CM codes need to assess the performance of the mappings and conduct manual refinement of the algorithms as needed before using them.

Real-world safety of palbociclib in breast cancer patients in the United States: a new user cohort study

Background

There is limited real-world safety information on palbociclib for treatment of advanced stage HR+/HER2- breast cancer.

Methods

We conducted a cohort study of breast cancer patients initiating palbociclib and fulvestrant from February 2015 to September 2017 using the HealthCore Integrated Research Database (HIRD), a longitudinal claims database of commercial health plan members in the United States.

The historical comparator cohort comprised patients initiating fulvestrant monotherapy from January 2011 to January 2015. Propensity score matching and Cox regression were used to estimate hazard ratios for various safety events. For acute liver injury (ALI), additional analyses and medical record validation were conducted.

Results

There were 2445 patients who initiated palbociclib including 566 new users of palbociclib-fulvestrant, and 2316 historical new users of fulvestrant monotherapy. Compared to these historical new users of fulvestrant monotherapy, new users of palbociclib-fulvestrant had a greater than 2-fold elevated risk for neutropenia, leukopenia, thrombocytopenia, stomatitis and mucositis, and ALI. Incidence of anemia and QT prolongation were more weakly associated, and incidences of serious infections and pulmonary embolism were similar between groups after propensity score matching. After adjustment for additional ALI risk factors, the elevated risk of ALI in new users of palbociclib-fulvestrant persisted (e.g. primary ALI algorithm hazard ratio (HR) = 3.0, 95% confidence interval (CI) = 1.1–8.4).

Conclusions

This real-world study found increased risks of several adverse events identified in clinical trials, including neutropenia, leukopenia, and thrombocytopenia, but no increased risk of serious infections or pulmonary embolism when comparing new users of palbociclib-fulvestrant to fulvestrant monotherapy. We observed an increased risk of ALI, extending clinical trial findings of significant imbalances in grade 3/4 elevations of alanine aminotransferase (ALT).

Supplementary Information

The online version contains supplementary material available at 10.1186/s12885-021-07790-z.

Measuring complications of serious pediatric emergencies using ICD-10

OBJECTIVE

To create definitions for complications for 16 serious pediatric conditions using the International Classification of Diseases, 10th Revision, Clinical Modification or Procedure Coding System (ICD-10-CM/PCS), and to assess whether complication rates are similar to those measured with ICD-9-CM/PCS.

DATA SOURCES

The Healthcare Cost and Utilization Project State Emergency Department and Inpatient Databases from five states between 2014 and 2017 were used to identify cases and assess complication rates. Incidences were calculated using population counts from the 5-year American Community Survey.

DATA COLLECTION/EXTRACTION METHODS

Patients were identified by the presence of a diagnosis code for one of the 16 serious conditions. Only the first encounter for a given condition by a patient was included. Encounters resulting in transfer were excluded as the presence of complications was unknown.

STUDY DESIGN

We defined complications using data elements routinely available in administrative databases including ICD-10-CM/PCS codes. The definitions were adapted from ICD-9-CM/PCS using general equivalence mappings and refined using consensus opinion. We included 16 serious conditions: appendicitis, bacterial meningitis, compartment syndrome, new-onset diabetic ketoacidosis (DKA), ectopic pregnancy, empyema, encephalitis, intussusception, mastoiditis, myocarditis, orbital cellulitis, ovarian torsion, sepsis, septic arthritis, stroke, and testicular torsion. Using data from children under 18 years, we compared incidences and complication rates across the ICD-10-CM/PCS transition for each condition using interrupted time series.

PRINCIPAL FINDINGS

There were 61 314 ED visits for a serious condition; the most common was appendicitis (n = 37 493). Incidence rates for each condition were not significantly different across the ICD-10-CM/PCS transition for 13/16 conditions. Three differed: empyema (increased 42%), orbital cellulitis (increased 60%), and sepsis (increased 26%). Complication rates were not significantly different for each condition across the ICD-10-CM/PCS transition, except appendicitis (odds ratio 0.62, 95% CI 0.57-0.68), DKA (OR 3.79, 95% CI 1.92-7.50), and orbital cellulitis (OR 0.53, 95% CI 0.30-0.95).

CONCLUSIONS

For most conditions, incidences and complication rates were similar before and after the transition to ICD-10-CM/PCS codes, suggesting our system identifies complications of conditions in administrative data similarly using ICD-9-CM/PCS and ICD-10-CM/PCS codes. This system may be applied to screen for cases with complications and in health services research.

A pitfall of using general equivalence mappings to estimate national trends of surgical utilization for pediatric patients

BACKGROUND

General equivalence mappings (GEMs) were developed to facilitate a transition from International Classification of Diseases, Ninth Revision (ICD-9) to ICD, Tenth Revision (ICD-10). Validation of GEMs is suggested as coding errors have been reported for adult populations. The purpose of this study was to illustrate limitations of the GEMs for pediatric surgical procedures.

METHODS

Using the 2014 to 2016 National Inpatient Sample, we evaluated all patients undergoing inguinal hernia repair. ICD-9 codes for the repair were independently classified as laparoscopic or open approach by two surgeons. Conversions of the ICD-9 to ICD-10 codes were compared between the GEMs strategy and surgeons' manual mapping. National trends were compared for overall, adult, and pediatric populations.

RESULTS

We found significant inconsistencies in the proportion of laparoscopic inguinal hernia repair based on mapping strategies employed. For adults, the comparison of the proportions in 2016 was 17.79% (GEMs) versus 21.44% (Manual). In pediatric population, the contrast was 0.45% (GEMs) versus 17.75% (Manual), and no laparoscopic repair cases were found using GEMs in the last quarter of 2015.

CONCLUSION

Some conversions of ICD-9 and ICD-10 using the current GEMs are not valid for certain populations and procedures. Clinical validation of coding conversions is essential.

LEVEL OF EVIDENCE

Level V.

Risk of Stroke and Bleeding in Atrial Fibrillation Treated with Apixaban Compared with Warfarin

BACKGROUND

A previous FDA study reported a favorable benefit risk for apixaban compared with warfarin for stroke prevention in older non-valvular atrial fibrillation (NVAF) patients (≥ 65 years). However, it remains unclear whether this favorable benefit risk persists in other populations including younger users. We examined if a similar benefit risk was observed in the Sentinel System and if it varied by age group.

OBJECTIVE

To examine the risk of ischemic stroke, gastrointestinal (GI) bleeding, and intracranial hemorrhage (ICH) in apixaban users compared with warfarin users in Sentinel Distributed Database (SDD).

DESIGN AND PARTICIPANTS

A retrospective new user cohort study was conducted among patients, 21 years and older initiating apixaban and warfarin for NVAF, between December 28, 2012, and June 30, 2018, in the SDD.

MAIN MEASURES

Cox proportional hazard regression was used to estimate the hazard ratios (HR) and 95% confidence intervals (95% CI) for each outcome (ischemic stroke, GI bleeding, and ICH) in propensity score matched apixaban users compared with the warfarin users. Subgroup analyses by age (21-64, 65-74, and 75+ years) were conducted.

KEY RESULTS

After matching, 55.3% and 58.4% (n = 55,038) of the apixaban and warfarin users were included in the main analysis. GI bleeding was the most common outcome. The HR (95% CI) for GI bleeding, ICH, and ischemic stroke in apixaban users compared with warfarin users were 0.57 (0.50-0.66), 0.53 (0.40-0.70), and 0.56 (0.45-0.71) respectively. The reduced risk of these outcomes in apixaban compared with warfarin users persisted across age groups.

CONCLUSION

In NVAF patients of all ages initiating either apixaban or warfarin for stroke prevention in the Sentinel System, apixaban was associated with a decreased risk of GI bleeding, ICH, and ischemic stroke compared with warfarin. Among patients less than 65 years of age, apixaban use was associated with a decreased risk of GI bleeding and ischemic stroke.

Strategy for reliable identification of ischaemic stroke, thrombolytics and thrombectomy in large administrative databases

Background

Administrative data are frequently used in stroke research. Ensuring accurate identification of patients who had an ischaemic stroke, and those receiving thrombolysis and endovascular thrombectomy (EVT) is critical to ensure representativeness and generalisability. We examined differences in patient samples based on mode of identification, and propose a strategy for future patient and procedure identification in large administrative databases.

Methods

We used non-public administrative data from the state of California to identify all patients who had an ischaemic stroke discharged from an emergency department (ED) or inpatient hospitalisation from 2010 to 2017 based on International Classification of Disease (ICD-9) (2010–2015), ICD-10 (2015–2017) and Medicare Severity-Diagnosis-related Group (MS-DRG) discharge codes. We identified patients with interhospital transfers, patients receiving thrombolytics and patients treated with EVT based on ICD, Current Procedural Terminology (CPT) and MS-DRG codes. We determined what proportion of these transfers and procedures would have been identified with ICD versus MS-DRG discharge codes.

Results

Of 365 099 ischaemic stroke encounters, most (87.70%) had both a stroke-related ICD-9 or ICD-10 code and stroke-related MS-DRG code; 12.28% had only an ICD-9 or ICD-10 code and 0.02% had only an MS-DRG code. Nearly all transfers (99.99%) were identified using ICD codes. We identified 32 433 thrombolytic-treated patients (8.9% of total) using ICD, CPT and MS-DRG codes; the combination of ICD and CPT codes identified nearly all (98%). We identified 7691 patients treated with EVT (2.1% of total) using ICD and MS-DRG codes; both MS-DRG and ICD-9/ICD-10 codes were necessary because ICD codes alone missed 13.2% of EVTs. CPT codes only pertain to outpatient/ED patients and are not useful for EVT identification.

Conclusions

ICD-9/ICD-10 diagnosis codes capture nearly all ischaemic stroke encounters and transfers, while the combination of ICD-9/ICD-10 and CPT codes are adequate for identifying thrombolytic treatment in administrative datasets. However, MS-DRG codes are necessary in addition to ICD codes for identifying EVT, likely due to favourable reimbursement for EVT-related MS-DRG codes incentivising accurate coding.

Rates of Vascular Events in Patients With Migraine: A MarketScan® Database Retrospective Cohort Study

Objective

To estimate the baseline rates of vascular events among people with migraine.

Background

Several novel medications that target the calcitonin gene‐related peptide (CGRP) pathway are approved to treat people with migraine. Given that the CGRP pathway also plays a role in maintaining cardiovascular homeostasis, determining the baseline rates of vascular events among people with migraine will help inform the safety of these novel medications.

Methods

In this retrospective cohort study, patients 18‐ to 64‐year‐old patients with migraine were identified from the MarketScan^® database (January 2013‐December 2017) and were categorized into 4 vascular risk categories: migraine with aura; and high, medium, and low vascular risk. Event rates (per 1000 person‐years [PY]) for 19 vascular events were estimated overall, by risk category, and by baseline characteristics.

Results

Among 1,195,696 patients with migraine, 4.8% (57,853/1,195,696) had migraine with aura, and 2.8% (33,949/1,195,696), 15.5% (184,782/1,195,696), and 77.9% (931,059/1,195,696) were at high, medium, and low risk of vascular events, respectively. Rates of ischemic stroke (per 1000 PY) were 5.1 (95% confidence interval [CI]: 5.0, 5.2) overall, 8.6 (95% CI: 8.1, 9.1) for patients with migraine aura, 47.2 (95% CI: 45.3, 49.0) in the high‐risk group, 9.4 (95% CI: 9.1, 9.7) in the medium‐risk group, and 2.9 (95% CI: 2.9, 3.0) in the low‐risk group. Rates of acute myocardial infarction (per 1000 PY) were 1.8 (95% CI: 1.8, 1.9) overall, 1.9 (95% CI: 1.7, 2.2) for patients with migraine aura, 14.0 (95% CI: 13.0, 14.9) in the high‐risk group, 3.9 (95% CI: 3.7, 4.1) in the medium‐risk group, and 1.1 (95% CI: 1.0, 1.1) in the low‐risk group. High‐risk patients had the highest rates of each of 19 evaluated vascular events, and rates were higher for men, older age groups, and those with higher comorbidity scores, medication usage, and medical utilization.

Conclusion

Our findings provide recent rates of vascular disease in patients with migraine. In the future, this information will be useful to help inform clinical risk:benefit decision making when assessing the use of therapies such as CGRP antagonists for migraine.

Impact of the Transition from ICD-9-CM to ICD-10-CM on the Identification of Pregnancy Episodes in US Health Insurance Claims Data

Background

Before October 2015, pregnancy cohorts assembled from US health insurance claims have relied on medical encounters with International Classification of Diseases-ninth revision-clinical modification (ICD-9-CM) codes. We aimed to extend existing pregnancy identification algorithms into the ICD-10-CM era and evaluate performance.

Methods

We used national private insurance claims data (2005-2018) to develop and test a pregnancy identification algorithm. We considered validated ICD-9-CM diagnosis and procedure codes that identify medical encounters for live birth, stillbirth, ectopic pregnancy, abortions, and prenatal screening to identify pregnancies. We then mapped these codes to the ICD-10-CM system using general equivalent mapping tools and reconciled outputs with literature and expert opinion. Both versions were applied to the respective coding period to identify pregnancies. We required 45 weeks of health plan enrollment from estimated conception to ensure the capture of all pregnancy endpoints.

Results

We identified 7,060,675 pregnancy episodes, of which 50.1% met insurance enrollment requirements. Live-born deliveries comprised the majority (76.5%) of episodes, followed by abortions (20.3%). The annual prevalence for all pregnancy types was stable across the ICD transition period except for postterm pregnancies, which increased from 0.5% to 3.4%. We observed that ICD codes indicating gestational age were available for 86.8% of live-born deliveries in the ICD-10 era compared to 23.5% in the ICD-9 era. Patterns of prenatal tests remained stable across the transition period.

Conclusion

Translation of existing ICD-9-CM pregnancy algorithms into ICD-10-CM codes provided reasonable consistency in identifying pregnancy episodes across the ICD transition period. New codes for gestational age can potentially improve the precision of conception estimates and minimize measurement biases.

Using Electronic Health Record and Administrative Data to Analyze Maternal and Neonatal Delivery Complications

BACKGROUND

Obstetric quality of care measures have largely focused on severe maternal morbidity (SMM), with little consensus about measures of less severe but more prevalent delivery and neonatal complications. This study analyzes risk-adjusted maternal and neonatal outcomes using both ICD-10 coding and electronic health record (EHR) data.

METHODS

Complication rates at seven health system hospitals from January 2016 to August 2019 were analyzed. EHR data and ICD-10 codes were used to identify the incidence of SMM as well as other route-specific maternal and neonatal complications. Researchers tested the association of maternal sociodemographic and clinical risk markers with the likelihood of maternal and neonatal complications using multiple logistic and Poisson regression.

RESULTS

Among 42,681 deliveries, the SMM rate was 1.3%, and other complication rates were 12.9% for vaginal and 19.7% for cesarean deliveries. The neonatal complication rate was 20.2%. Risk factors for all complications included multiple gestation and hypertensive disorders of pregnancy. Risk factors for SMM included nulliparity, cesarean delivery, and preexisting conditions; risks for neonatal complications included academic medical center admission, cesarean delivery, higher maternal body mass index, and preterm birth. There were significant racial disparities in maternal and neonatal outcomes.

CONCLUSION

This study is among the first to combine EHR and administrative discharge data to describe a wide range of maternal and neonatal birth outcomes, including associations with established risk factors. Although SMM was rare, route-specific and neonatal complications were much more common and may offer a better focus for obstetric quality improvement efforts.

Pneumonia Hospitalization Coding Changes Associated With Transition From the 9th to 10th Revision of International Classification of Diseases

Objectives:

To evaluate the impact of International Classification of Disease, 10th revision, Clinical Modification (ICD-10-CM) implementation on pneumonia hospitalizations rates, which had declined following pneumococcal conjugate vaccine introduction for infants in 2000.

Methods:

We randomly selected records from a single hospital 1 year before (n = 500) and after (n = 500) October 2015 implementation of ICD-10-CM coding. We used a validated ICD-9-CM algorithm and translation of that algorithm to ICD-10-CM to identify pneumonia hospitalizations pre- and post-implementation, respectively. We recoded ICD-10-CM records to ICD-9-CM and vice versa. We calculated sensitivity and positive predictive value (PPV) of the ICD-10-CM algorithm using ICD-9-CM coding as the reference. We used sensitivity and PPV values to calculate an adjustment factor to apply to ICD-10 era rates to enable comparison with ICD-9-CM rates. We reviewed primary diagnoses of charts not meeting the pneumonia definition when recoded.

Results:

Sensitivity and PPV of the ICD-10-CM algorithm were 94% and 92%, respectively, for young children and 74% and 79% for older adults. The estimated adjustment factor for ICD-10-CM period rates was −2.09% (95% credible region [CR], −7.71% to +3.0%) for children and +6.76% (95% CR, −3.06% to +16.7%) for older adults. We identified a change in coding adult charts that met the ICD-9-CM pneumonia definition that led to recoding in ICD-10-CM as chronic obstructive pulmonary disease (COPD) exacerbation.

Conclusions:

The ICD-10-CM algorithm derived from a validated ICD-9-CM algorithm should not introduce substantial bias for evaluating pneumonia trends in children. However, changes in coding of pneumonia associated with COPD in adults warrant further study.

Secular trends of stroke incidence and mortality in China, 1990 to 2016: The Global Burden of Disease Study 2016

BACKGROUND

The impact of socioeconomic developments on stroke incidence and mortality must be understood to target prevention strategies appropriately. We assessed the secular trends in stroke incidence and mortality in China based on data from the Global Burden of Disease Study 2016.

METHODS

Trends of stroke incidence and mortality of China was described in different categories of age, sex and stroke type using the GBD study database. Also a comparative study was conducted between China and Japan, U.S. to find reasonable references for development. Secular trends in incidence and mortality (per 100,000 population) were assessed for stroke, including ischemic and hemorrhagic stroke from 1990 to 2016. Population pyramid was used to illustrate changes in age- and sex-specific incidence and mortality rates.

RESULTS

During the study period, stroke incidence in China increased from 204.52 to 403.08 and mortality increased from 122.09 to 130.94; the corresponding age-standardized rates changed from 335.63 to 353.70 and from 231.28 to 132.84, respectively. Among those aged 15-49 and 50-69 years, the incidence rates of ischemic stroke and hemorrhagic stroke both tended to increase, whereas the mortality rates tended to decline in all age groups. The incidence and mortality were highest among those aged ≥70 years. Compared with the U.S. and Japan, age-standardized rates of incidence and mortality were higher in China.

CONCLUSIONS

Although the incidence of stroke has increased in China, overall mortality has decreased. A priority of stroke prevention and control strategies will transition from reducing mortality to controlling the incidence in at-risk populations.

How much can we trust electronic health record data?

Trust in EHR data is becoming increasingly important as a greater share of clinical and health services research use EHR data. We discuss reasons for distrust and acknowledge limitations. Researchers continue to use EHR data because of strengths including greater clinical detail than sources like administrative billing claims. Further, many limitations are addressable with existing methods including data quality checks and common data frameworks. We discuss how to build greater trust in the use of EHR data for research, including additional transparency and research priority areas that will both enhance existing strengths of the EHR and mitigate its limitations.

Pharmacoepidemiologic Screening of Potential Oral Anticoagulant Drug Interactions Leading to Thromboembolic Events

Drug-drug interactions (DDIs) with oral anticoagulants may lead to under-anticoagulation and increased risk of thromboembolism. Although warfarin is susceptible to numerous DDIs, few studies have examined DDIs resulting in thromboembolism or those involving direct-acting oral anticoagulants (DOACs). We aimed to identify medications that increase the rate of hospitalization for thromboembolic events when taken concomitantly with oral anticoagulants. We conducted a high-throughput pharmacoepidemiologic screening study using Optum Clinformatics Data Mart, 2000-2016. We performed self-controlled case series studies among adult users of oral anticoagulants (warfarin, dabigatran, rivaroxaban, apixaban, and edoxaban) with at least one hospitalization for a thromboembolic event. Among eligible patients, we identified all oral medications frequently co-prescribed with oral anticoagulants as potential interacting precipitants. Conditional Poisson regression was used to estimate rate ratios comparing precipitant exposed vs. unexposed time for each anticoagulant-precipitant pair. To minimize within-person confounding by indication for the precipitant, we used pravastatin as a negative control object drug. Multiple estimation was adjusted using semi-Bayes shrinkage. We screened 1,622 oral anticoagulant-precipitant drug pairs and identified 226 (14%) drug pairs associated with statistically significantly elevated risk of thromboembolism. Using pravastatin as the negative control object drug, this list was reduced to 69 potential DDI signals for thromboembolism, 33 (48%) of which were not documented in the DDI knowledge databases Lexicomp and/or Micromedex. There were more DDI signals associated with warfarin than DOACs. This study reproduced several previously documented oral anticoagulant DDIs and identified potential DDI signals that deserve to be examined in future etiologic studies.

Age-Period-Cohort Analysis of Type-Specific Stroke Morbidity and Mortality in China

BACKGROUND

Stroke has become the leading cause of death in China. This study aimed to assess the age-period-cohort (APC) effects on the long-term trends of type-specific stroke morbidity and mortality in China between 1993 and 2017.Methods and Results:The data were obtained from the Global Burden of Disease 2017 (GBD 2017) and were analyzed with the age-period-cohort framework. The net drifts of mortality were below 0 (hemorrhagic stroke [HS]: males: -1.620%, females: -3.531%; ischemic stroke [IS]: males: -1.041%, females: -3.002%), and the local drift values were below 0 in all age groups and for both genders. The net drifts of HS incidence were below 0 (males: -1.412%, females: -2.688%), while those of IS were above 0 (males: 1.425%, females: 1.117%). Period effect of mortality showed similar monotonic downward patterns for both genders, with a faster decrease for females than for males. Period effect of incidence showed a declined trend of incidence for HS, but an elevated trend for IS in both genders. After controlling for age and period effects, cohort effects on incidence found a monotonic decline trend for HS, while for IS, an elevated trend was found at first to peak during the 1950-1970 s, then declined steadily afterwards. Cohort effects on mortality showed a monotonic declined trend.

CONCLUSIONS

By using Age-Period-Cohort (APC) analysis, a disparity between HS and IS was identified. Different prevention and control strategies should be used depending on the subtypes of stroke.

Translating claims-based CHA2 DS2 -VaSc and HAS-BLED to ICD-10-CM: Impacts of mapping strategies

PURPOSE

The CHA₂ DS₂ -VaSc and HAS-BLED risk scores are commonly used in the studies of oral anticoagulants (OACs). The best ways to map these scores to the International Classification of Diseases, 10th Revision, Clinical Modification (ICD-10-CM) codes is unclear, as is how they perform in various types of OAC users. We aimed to assess the distributions of CHA₂ DS₂ -VaSc and HAS-BLED scores and C-statistics for outcome prediction in the ICD-10-CM era using different mapping strategies.

METHODS

We compared the distributions of CHA₂ DS₂ -VaSc and HAS-BLED scores from various mapping strategies in atrial fibrillation patients before, during, and after ICD-10-CM transition. We estimated the C-statistics predicting the 90-day risk of hospitalized stroke (for CHA₂ DS₂ -VaSc) or hospitalized bleeding (for HAS-BLED) in patients identified at least 6 months after the ICD-10-CM transition, overall and by anticoagulant type.

RESULTS

Forward-backward mapping produced higher CHA₂ DS₂ -VaSc and HAS-BLED scores in the ICD-10-CM era compared to the ICD-9-CM era: the mean difference was 0.074 (95% confidence interval 0.064-0.085) for CHA₂ DS₂ -VaSc and 0.055 (0.048-0.062) for HAS-BLED. Both scores had higher C-statistics in patients taking no OACs (0.697 [0.677-0.717] for CHA₂ DS₂ -VaSc; 0.719 [0.702-0.737] for HAS-BLED) or direct OACs (0.695 [0.654-0.735] for CHA₂ DS₂ -VaSc; 0.700 [0.673-0.728] for HAS-BLED) than those taking warfarin (0.655 [0.613-0.697] for CHA₂ DS₂ -VaSc; 0.663 [0.6320.695] for HAS-BLED).

CONCLUSIONS

Existing mapping strategies generally preserved the distributions of CHA₂ DS₂ -VaSc and HAS-BLED scores after ICD-10-CM transition. Both scores performed better in patients on no OACs or direct OACs than patients on warfarin.

Mapping from the International Classification of Diseases (ICD) 9th to 10th Revision for Research in Biologics and Biosimilars Using Administrative Healthcare Data

PURPOSE

The Centers for Medicare and Medicaid Services (CMS) mandated the transition from ICD-9 to ICD-10 codes on October 1, 2015. Postmarketing surveillance of newly marketed drugs, including novel biologics and biosimilars, requires a robust approach to convert ICD-9 to ICD-10 codes for study variables. We examined three mapping methods for health conditions (HCs) of interest to the Biologics and Biosimilars Collective Intelligence Consortium (BBCIC) and compared their prevalence.

METHODS

Using CMS General Equivalence Mappings, we applied forward-backward mapping (FBM) to 108 HCs and secondary mapping (SM) and tertiary mapping (TM) to seven preselected HCs. A physician reviewed the mapped ICD-10 codes. The prevalence of the 108 HCs defined by ICD-9 versus ICD-10 codes was examined in BBCIC's distributed research network (September 1, 2012 to March 31, 2018). We visually assessed prevalence trends of these HCs and applied a threshold of 20% level change in ICD-9 versus ICD-10 prevalence.

RESULTS

Nearly four times more ICD-10 codes were mapped by SM and TM than FBM, but most were irrelevant or nonspecific. For conditions like myocardial infarction, SM or TM did not generate additional ICD-10 codes. Through visual inspection, one-fifth of the HCs had inconsistent ICD-9 versus ICD-10 prevalence trends. 13% of HCs had a level change greater than +/-20%.

CONCLUSION

FBM is generally the most efficient way to convert ICD-9 to ICD-10 codes, yet manual review of converted ICD-10 codes is recommended even for FBM. The lack of existing guidance to compare the performance of ICD-9 with ICD-10 codes led to challenges in empirically determining the quality of conversions.

Incidence of statin use in older adults with and without cardiovascular disease and diabetes mellitus, January 2008- March 2018

Background

Data from randomized controlled trials and observational studies on older adults who take statins for primary prevention of atherosclerotic cardiovascular disease are limited. To determine the incidence of statin use in older adults with and without cardiovascular disease (CVD) and/or diabetes (DM), we conducted a descriptive observational study.

Methods

The cohort consisted of health plan members in the NIH Collaboratory Distributed Research Network aged >75 years who had continuous drug and medical benefits for ≥183 days during the study period, January 1, 2008- March 31, 2018. We defined DM and CVD using diagnosis codes, and identified statins using dispensing data. Statin use was considered incident if a member had no evidence of statin exposure in the claims during the previous 183 days, and the use was considered long-term if statins were supplied for ≥180 days. Incidence rates were reported among members with and without CVD and/or diabetes, and stratified by year, sex, and age group.

Results

Among 757,569 eligible members, 109,306 older adults initiated statins and 54,624 became long-term users. Health plan members with CVD had the highest incidence of statin use (143.9 initiators per 1,000 member-years for CVD & DM; 114.5 initiators per 1,000 member-years for CVD & No DM). Among health plan members without CVD, those with DM had rates of statin use that were over two times higher than members without DM (76.1 versus 34.5 initiators per 1,000 member-years, respectively). Statin initiation remained steady throughout 2008–2016, was slightly higher in males, and declined with increasing age.

Conclusion

Incidence of statin use varied by CVD and DM comorbidity, and was lowest among those without CVD. These results highlight the potential clinical equipoise to conduct large pragmatic clinical trials to generate evidence that could be used to inform future blood cholesterol guidelines.

ICD-10-CM extension with ICD-9 diagnosis codes to support integrated access to clinical legacy data

INTRODUCTION

ICD is currently the most widely used terminology to code diagnosis and procedures. The transition from ICD-9-CM to ICD-10-CM became effective on October 1, 2015 in US and many other countries. Projects that use this codification for research purposes, requires advanced methods to exploit data with both versions of ICD. Although the General Equivalence Mappings (GEMs), provided by the Centers for Medicare and Medicaid Services, might help to overcome these challenges, their direct use as translation mappings is not possible, mostly due to the further specificity of ICD-10-CM concepts.

OBJECTIVE

We propose a methodology to generate an extended version of ICD-10-CM with selected ICD-9-CM diagnosis codes.

METHODS

The extension was generated using the GEMs relations between concepts of both terminologies and the hierarchical relations of ICD-10-CM.

RESULTS

This extended ICD-10-CM, together with modifications to the mapping of ICD-9-CM concepts that were not inserted, allows the generation of an improved translation of legacy data, raising the number of 1-to-1 correspondences by +13.81%.

CONCLUSION

The extended ICD-10-CM enables the accurate integration of ICD-9-CM and ICD-10-CM diagnosis data into a single terminology. With such analysis of data possible without having to specify both ICD-9-CM and ICD-10-CM separately for each query.

The impact of the ICD-9-CM to ICD-10-CM transition on the prevalence of birth defects among infant hospitalizations in the United States

BACKGROUND

Many public health surveillance programs utilize hospital discharge data in their estimation of disease prevalence. These databases commonly use the International Classification of Diseases (ICD) coding scheme, which transitioned from the ICD-9 clinical modification (ICD-9-CM) to ICD-10-CM on October 1, 2015. This study examined this transition's impact on the prevalence of major birth defects among infant hospitalizations.

METHODS

Using data from the Agency for Health Care Research and Quality-sponsored National Inpatient Sample, hospitalizations during the first year of life with a discharge date between January 1, 2012 and December 31, 2016 were used to estimate the monthly national hospital prevalence of 46 birth defects for the ICD-9-CM and ICD-10-CM timeframes separately. Survey-weighted Poisson regression was used to estimate 95% confidence intervals for each hospital prevalence. Interrupted time series framework and corresponding segmented regression was used to estimate the immediate change in monthly hospital prevalence following the ICD-9-CM to ICD-10-CM transition.

RESULTS

Between 2012 and 2016, over 21 million inpatient hospitalizations occurred during the first year of life. Among the 46 defects studied, statistically significant decreases in the immediate hospital prevalence of five defects and significant increases in the immediate hospital prevalence of eight defects were observed after the ICD-10-CM transition.

CONCLUSIONS

Changes in prevalence were expected based on changes to ICD-10-CM. Observed changes for some conditions may result from variation in monthly hospital prevalence or initial unfamiliarity of coders with ICD-10-CM. These findings may help birth defects surveillance programs evaluate and interpret changes in their data related to the ICD-10-CM transition.

Heterogeneity introduced by EHR system implementation in a de-identified data resource from 100 non-affiliated organizations

Objectives

Electronic health record (EHR) data aggregated from multiple, non-affiliated, sources provide an important resource for biomedical research, including digital phenotyping. Unlike work with EHR data from a single organization, aggregate EHR data introduces a number of analysis challenges.

Materials and Methods

We used the Cerner Health Facts data, a de-identified aggregate EHR data resource populated by data from 100 independent health systems, to investigate the impact of EHR implementation factors on the aggregate data. These included use of ancillary modules, data continuity, International Classification of Disease (ICD) version and prompts for clinical documentation.

Results and Discussion

Health Facts includes six categories of data from ancillary modules. We found of the 664 facilities in Health Facts, 49 use all six categories while 88 facilities were not using any. We evaluated data contribution over time and found considerable variation at the health system and facility levels. We analyzed the transition from ICD-9 to ICD-10 and found that some organizations completed the shift in 2014 while others remained on ICD-9 in 2017, well after the 2015 deadline. We investigated the utilization of “discharge disposition” to document death and found inconsistent use of this field. We evaluated clinical events used to document travel status implemented in response to Ebola, height and smoking history. Smoking history documentation increased dramatically after Meaningful Use, but dropped in some organizations. These observations highlight the need for any research involving aggregate EHR data to consider implementation factors that contribute to variability in the data before attributing gaps to “missing data.”

Impact of ICD-10-CM Transition on Mental Health Diagnoses Recording

Objective:

This study examines the impact of the transition from ICD-9-CM to ICD-10-CM diagnosis coding on the recording of mental health disorders in electronic health records (EHRs) and claims data in ten large health systems. We present rates of these diagnoses across two years spanning the October 2015 transition.

Methods:

Mental health diagnoses were identified from claims and EHR data at ten health care systems in the Mental Health Research Network (MHRN). Corresponding ICD-9-CM and ICD-10-CM codes were compiled and monthly rates of people receiving these diagnoses were calculated for one year before and after the coding transition.

Results:

For seven of eight diagnostic categories, monthly rates were comparable during the year before and the year after the ICD-10-CM transition. In the remaining category, psychosis excluding schizophrenia spectrum disorders, aggregate monthly rates of decreased markedly with the ICD-10-CM transition, from 48 to 33 per 100,000. We propose that the change is due to features of General Equivalence Mappings (GEMS) embedded in the EHR.

Conclusions:

For most mental health conditions, the transition to ICD-10-CM appears to have had minimal impact. The decrease seen for psychosis diagnoses in these health systems is likely due to changes associated with EHR implementation of ICD-10-CM coding rather than an actual change in disease prevalence. It is important to consider the impact of the ICD-10-CM transition for all diagnostic criteria used in research studies, quality measurement, and financial analysis during this interval.

Assessing the impact of the new ICD-10-CM coding system on pharmacoepidemiologic studies-An application to the known association between angiotensin-converting enzyme inhibitors and angioedema

PURPOSE

To replicate the well-established association between angiotensin-converting enzyme inhibitors versus beta blockers and angioedema in the International Classification of Diseases, 10th Revision, Clinical Modification (ICD-10-CM) era.

METHODS

We conducted a retrospective, inception cohort study in a large insurance database formatted to the Sentinel Common Data Model. We defined study periods spanning the ICD-9-CM era only, ICD-10-CM era only, and ICD-9-CM and ICD-10-CM era and conducted simple-forward mapping (SFM), simple-backward mapping (SBM), and forward-backward mapping (FBM) referencing the General Equivalence Mappings to translate the outcome (angioedema) and covariates from ICD-9-CM to ICD-10-CM. We performed propensity score (PS)-matched and PS-stratified Cox proportional hazards regression to estimate hazard ratios (HRs) and 95% confidence intervals (CIs).

RESULTS

In the ICD-9-CM and ICD-10-CM eras spanning April 1 to September 30 of 2015 and 2016, there were 152 017 and 145 232 angiotensin-converting enzyme inhibitor initiators and 115 073 and 116 652 beta-blocker initiators, respectively. The PS-matched HR was 4.19 (95% CI, 2.82-6.23) in the ICD-9-CM era, 4.37 (2.92-6.52) in the ICD-10-CM era using SFM, and 4.64 (3.05-7.07) in the ICD-10-CM era using SBM and FBM. The PS-matched HRs from the mixed ICD-9-CM and ICD-10-CM eras ranged from 3.91 (2.69-5.68) to 4.35 (3.33-5.70).

CONCLUSION

The adjusted HRs across different diagnostic coding eras and the use of SFM versus SBM and FBM produced numerically different but clinically similar results. Additional investigations as ICD-10-CM data accumulate are warranted.