Hospital Discharge Abstract Data Research Articles

Validating International Classification of Disease 10th revision algorithms for identifying influenza and respiratory syncytial virus hospitalizations

Objective: Routinely collected health administrative data can be used to efficiently assess disease burden in large populations, but it is important to evaluate the validity of these data. The objective of this study was to develop and validate International Classification of Disease 10PthP revision (ICD -10) algorithms that identify laboratory-confirmed influenza or laboratory-confirmed respiratory syncytial virus (RSV) hospitalizations using population-based health administrative data from Ontario, Canada.
 Study Design and Setting: Influenza and RSV laboratory data from the 2014-15 through to 2017-18 respiratory virus seasons were obtained from the Ontario Laboratories Information System (OLIS) and were linked to hospital discharge abstract data to generate influenza and RSV reference cohorts. These reference cohorts were used to assess the sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) of the ICD-10 algorithms. To minimize misclassification in future studies, we prioritized specificity and PPV in selecting top-performing algorithms.
 Results: 83,638 and 61,117 hospitalized patients were included in the influenza and RSV reference cohorts, respectively. The best influenza algorithm had a sensitivity of 73% (95% CI 72% to 74%), specificity of 99% (95% CI 99% to 99%), PPV of 94% (95% CI 94% to 95%), and NPV of 94% (95% CI 94% to 95%). The best RSV algorithm had a sensitivity of 69% (95% CI 68% to 70%), specificity of 99% (95% CI 99% to 99%), PPV of 91% (95% CI 90% to 91%) and NPV of 97% (95% CI 97% to 97%).
 Conclusion: We identified two highly specific algorithms that best ascertain patients hospitalized with influenza or RSV. These algorithms may be applied to hospitalized patients if data on laboratory tests are not available, and will thereby improve the power of future epidemiologic studies of influenza, RSV, and potentially other severe acute respiratory infections.

Validating International Classification of Disease 10th Revision algorithms for identifying influenza and respiratory syncytial virus hospitalizations

Routinely collected health administrative data can be used to efficiently assess disease burden in large populations, but it is important to evaluate the validity of these data. The objective of this study was to develop and validate International Classification of Disease 10th revision (ICD -10) algorithms that identify laboratory-confirmed influenza or laboratory-confirmed respiratory syncytial virus (RSV) hospitalizations using population-based health administrative data from Ontario, Canada. Influenza and RSV laboratory data from the 2014-15, 2015-16, 2016-17 and 2017-18 respiratory virus seasons were obtained from the Ontario Laboratories Information System (OLIS) and were linked to hospital discharge abstract data to generate influenza and RSV reference cohorts. These reference cohorts were used to assess the sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) of the ICD-10 algorithms. To minimize misclassification in future studies, we prioritized specificity and PPV in selecting top-performing algorithms. 83,638 and 61,117 hospitalized patients were included in the influenza and RSV reference cohorts, respectively. The best influenza algorithm had a sensitivity of 73% (95% CI 72% to 74%), specificity of 99% (95% CI 99% to 99%), PPV of 94% (95% CI 94% to 95%), and NPV of 94% (95% CI 94% to 95%). The best RSV algorithm had a sensitivity of 69% (95% CI 68% to 70%), specificity of 99% (95% CI 99% to 99%), PPV of 91% (95% CI 90% to 91%) and NPV of 97% (95% CI 97% to 97%). We identified two highly specific algorithms that best ascertain patients hospitalized with influenza or RSV. These algorithms may be applied to hospitalized patients if data on laboratory tests are not available, and will thereby improve the power of future epidemiologic studies of influenza, RSV, and potentially other severe acute respiratory infections.

Development and validation of data quality rules in administrative health data using association rule mining

IntroductionData quality assessment is a challenging facet for researches using coded administrative health data. Our previous study had demonstrated the potentials of association rule mining to assess data quality. The objective of this study is to develop and validate a set of coding association rules for data quality assessment.
 Objectives and ApproachWe used the Canadian reabstracted hospital discharge abstract data (DAD) with clinical diagnosis coded in International Classification of Disease – 10th revision, Canada (ICD-10-CA) codes for rule development. The DAD data were divided into 5 age groups. Association rule mining were conducted on reabstracted DAD in each age group to extract ICD-10 coding association rules at the three and four digits levels. The rule strength was assessed using support and confidence. The rules will be reviewed by a panel of 5 physicians and 2 coding specialists to assess their appropriateness from clinical and coding perspectives using a modified Delphi rating
 ResultsIn total, 975 rules at the three digits level and 822 rules at the four digits level were learned from the data. Half of the rules were in the age group of ≥65 and no rules were found in the age group of 5 to 19. The interquartile range of rule confidences were 0.112 to 0.425 in the three digits level and 0.073 to 0.222 in the four digits level. Two-thirds of rules had the diagnosis codes related to endocrine and metabolic disorders and diseases of circulatory, respiratory and genitourinary systems. The panel review will be conducted in early April and will have the final set of rules available before the conference.
 Conclusion/ImplicationsThis study developed a set of validated ICD-10 coding association rules and creates a useful tool to cost-effectively assess data quality in routinely collected administrative health data.

Linking cohort data with administrative health data to develop a new hypertension prediction model to aid precision health approach

IntroductionHypertension is a common medical condition, affecting 1 in 5 Canadians, and is a major risk factor for heart attack, stroke, and kidney disease. Predicting the risk of developing incident hypertension may help to inform targeted preventive strategies.
 Objectives and ApproachIdentification of major risk factors and incorporation into a multivariable model for risk stratification may help to identify individuals who are at highest risk for developing incident hypertension and would potentially benefit most from intervention. The goal of the proposed research is to develop a robust hypertension prediction model for the general population using the Alberta Tomorrow Project (ATP) cohort data linked with Alberta’s administrative health data. ATP is Alberta's largest population health cohort, contains baseline data on socio-demographic characteristic, personal and family history of disease, medication use, lifestyle and health behavior, environmental exposures, physical measures and bio samples.
 ResultsAlberta’s administrative health data additionally provides information on health care utilization, enrollment, drugs, physician services, and hospital services. A prediction model for hypertension will be developed using logistic regression where information on candidate variables for the model will be gathered from ATP data and outcome (incident hypertension) will be ascertained from administrative health data (physicians/practitioner claim data and hospital discharge abstract data). Lacking follow-up information in current ATP data has laid the foundation of linking the two data sources through an anonymous unique person identifier (e.g. PHN) that will eventually provide follow-up information on ATP participants who are free of hypertension at baseline developed the disease as well as information on other potential variables.
 Conclusion/ImplicationsThe proposed prediction model will help to identify individuals at highest risk for developing hypertension and those who may benefit most from targeted healthy behavioral interventions and/or treatment. Such identification of high risk people may help prevent hypertension as well as the continuing costly cycle of managing hypertension and its complications.

Associations between provider and hospital volumes and postoperative mortality following total hip arthroplasty in New Brunswick: results from a provincial-level cohort study.

Several international studies have reported negative associations between hospital and/or provider volume and risk of postoperative death following total hip arthroplasty (THA). The only Canadian studies to report on this have been based in Ontario and have found no such association. We describe associations between postoperative deaths following THA and provider caseload volume, also adjusted for hospital volume, in a population-based cohort in New Brunswick. Our analyses are based on hospital discharge abstract data linked to vital statistics and to patient registry data. We considered all first known admissions for THA in New Brunswick between Jan. 1, 2007, and Dec. 31, 2013. Provider volume was defined as total THAs performed over the preceding 2 years. We fit logistic regression models to identify odds of dying within 30 and 90 days according to provider caseload volume adjusted for selected personal and contextual characteristics. About 7095 patients were admitted for THA in New Brunswick over the 7-year study period and 170 died within 30 days. We found no associations with provider volume and postoperative mortality in any of our models. Adjustment for contextual characteristics or hospital volume had no effects on this association. Our results suggest that patients admitted for hip replacements in New Brunswick can expect to have similar risk of death regardless of whether they are admitted to see a provider with high or low THA volumes and of whether they are admitted to the province's larger or smaller hospitals.

Hospital safety among neurologic patients: A population-based cohort study of adverse events.

To examine the frequency and type of adverse events (AEs) experienced by neurologic patients in hospital. This population-based, retrospective cohort study used hospital discharge abstract data for children and adults admitted to hospital from 2009 to 2015 with 1 of 9 neurologic conditions (Alzheimer disease and related dementia, brain tumor, epilepsy, motor neuron disease, multiple sclerosis, parkinsonism/Parkinson disease, spinal cord injury, traumatic brain injury, and stroke). Neurologic conditions were identified with ICD-10-CA codes. Eighteen AEs were examined with ICD-10-CA codes. The proportion of AEs was calculated, and regression analysis was used to examine factors and outcomes associated with AEs (age, sex, comorbidity, length of stay, and mortality). The overall proportion of admissions associated with an AE among those with a neurologic condition was 11 per 100 admissions. Those with a spinal cord injury had the highest proportion of AEs (39.4 per 100 admissions). The most common AEs were infections and respiratory complications (32.0% and 16.7%, respectively). Age and the presence of comorbidities were associated with higher odds of an AE, while readmission was associated with lower odds of an AE. Having an AE was associated with increased length of stay and higher odds of mortality. This study demonstrates that neurologic patients have a high proportion of AEs in hospital. The findings provide information on the quality and safety of care for people with neurologic conditions in hospital, which can help inform future quality improvement initiatives.

Does Higher Spending Improve Survival Outcomes for Myocardial Infarction? Examining the Cost-Outcomes Relationship Using Time-Varying Covariates.

Previous patient-level acute myocardial infarction (AMI) research has found higher hospital spending to be associated with improved survival; however, survivor-treatment selection bias traditionally has been overlooked. The purpose of this study was to examine the AMI cost-outcome relationship, taking into account this form of bias. Hospital Discharge Abstract data tracked costs for AMI hospitalizations. Ontario Vital Statistics data tracked patient mortality. A standard Cox survival model was compared to an extended Cox model using hospital costs as a time-varying covariate to examine the impact of cost on 1-year survival in a cohort of 30,939 first-time AMI patients in Ontario, Canada, from 2007 to 2010. Higher patient-level AMI spending decreased the hazard of dying (Standard Model: log-cost hazard ratio: 0.513, 95 percent CI: 0.479-0.549; Extended Model: log-cost hazard ratio: 0.700, 95 percent CI: 0.645-0.758); however, the protective effect was overestimated by 62 percent when survivor-treatment bias was overlooked. In the extended model, a 10 percent increase in spending was associated with a 3.6 percent decrease in hazard of death. The findings of this study suggest that if survivor-treatment bias is overlooked, future research may materially overstate the protective effect of patient-level spending on outcomes.

Coding of obesity in administrative hospital discharge abstract data: accuracy and impact for future research studies

BackgroundObesity is a pervasive problem and a popular subject of academic assessment. The ability to take advantage of existing data, such as administrative databases, to study obesity is appealing. The objective of our study was to assess the validity of obesity coding in an administrative database and compare the association between obesity and outcomes in an administrative database versus registry.MethodsThis study was conducted using a coronary catheterization registry and an administrative database (Discharge Database (DAD)). A Body Mass Index (BMI) ≥30 kg/m2 within the registry defined obesity. In the DAD obesity was defined by diagnosis codes E65 – E68 (ICD-10). The sensitivity, specificity, negative predictive value (NPV) and positive predictive value (PPV) of an obesity diagnosis in the DAD was determined using obesity diagnosis in the registry as the referent. The association between obesity and outcomes was assessed.ResultsThe study population of 17380 subjects was largely male (68.8%) with a mean BMI of 27.0 kg/m2. Obesity prevalence was lower in the DAD than registry (2.4% vs. 20.3%). A diagnosis of obesity in the DAD had a sensitivity 7.75%, specificity 98.98%, NPV 80.84% and PPV 65.94%. Obesity was associated with decreased risk of death or re-hospitalization, though non-significantly within the DAD. Obesity was significantly associated with an increased risk of cardiac procedure in both databases.ConclusionsOverall, obesity was poorly coded in the DAD. However, when coded, it was coded accurately. Administrative databases are not an optimal datasource for obesity prevalence and incidence surveillance but could be used to define obese cohorts for follow-up.

Validity of AHRQ patient safety indicators derived from ICD-10 hospital discharge abstract data (chart review study)

ObjectiveTo assess if the Agency for Healthcare Research and Quality patient safety indictors (PSIs) could be used for case findings in the International Classification of Disease 10th revision (ICD-10) hospital...

Case Definitions for Acute Myocardial Infarction in Administrative Databases and Their Impact on In‐Hospital Mortality Rates

To identify validated ICD-9-CM/ICD-10 coded case definitions for acute myocardial infarction (AMI). Ovid Medline (1950-2010) was searched to identify studies that validated acute myocardial infarction (AMI) case definitions. Hospital discharge abstract data and chart data were linked to validate identified AMI definitions. Systematic literature review, chart review, and administrative data analysis. Data on sensitivity/specificity/positive and negative predictive values (PPV and NPV) were extracted from previous studies to identify validated case definitions for AMI. These case definitions were validated in administrative data through chart review and applied to hospital discharge data to assess in-hospital mortality. Of the eight ICD-9-CM definitions validated in the literature, use of ICD-9-CM code 410 to define AMI had the highest sensitivity (94 percent) and specificity (99 percent). In our data, ICD-9-CM/ICD-10 codes 410/I21-I22 in all available coding fields had high sensitivity (83.3 percent/82.8 percent) and PPV (82.8 percent/82.2 percent). The in-hospital mortality among AMI patients identified using this case definition was 7.6 percent in ICD-9-CM data and 6.6 percent in ICD-10 data. We recommend that ICD-9-CM 410 or ICD-10 I21-I22 in the primary diagnosis coding field should be used to define AMI. The use of a consistent validated case definition would improve comparability across studies.

Application of a national administrative case definition for the identification of pre-existing diabetes mellitus in pregnancy

Accurate ascertainment of pregnant women with pre-existing diabetes allows for the comprehensive surveillance of maternal and neonatal outcomes associated with this chronic disease. To determine the accuracy of case definitions for pre-existing diabetes mellitus when applied to a pregnant population, a cohort of women who were pregnant in Nova Scotia, Canada, between 1991 and 2003 was obtained from a population-based provincial perinatal database, the Nova Scotia Atlee Perinatal Database (NSAPD). Person-level data from administrative databases using hospital discharge abstract data and outpatient physician services data were linked to this cohort. Various algorithms for defining diabetes mellitus from the administrative data, including the algorithm suggested by the National Diabetes Surveillance System (NDSS), were compared to a reference standard definition from the NSAPD. Validation of the NDSS case definition applied to this pregnant population demonstrated a sensitivity of 87% and a positive predictive value (PPV) of 66.4%. Use of ICD-9 and ICD-10 diagnostic codes among hospitalizations with diabetes mellitus in pregnancy showed important increases in sensitivity and PPV, especially for those pregnancies delivered in tertiary centres. In this population, pregnancy-related administrative data from the hospitalization database alone appear to be a more accurate data source for identifying pre-existing diabetes than applying the NDSS case definition, particularly when pregnant women are delivered in a tertiary hospital. Although the NDSS definition of diabetes performs reasonably well compared to a reference standard definition of diabetes, using this definition for evaluating maternal and perinatal outcomes associated with diabetes in pregnancy will result in a certain degree of misclassification and, therefore, biased estimates of outcomes.

Using hospital discharge abstract data to identify incident breast cancer cases and assess quality of care

This study assesses the feasibility of using hospital discharge data to identify women with incident breast cancer, and it evaluates the use of these data to assess compliance with an indicator of quality of care. The study’s goals are to: 1) compare breast cancer case-finding algorithms based upon ICD-9 CM codes found in hospital discharge data with cancer registry data and 2) compare the utility of these algorithms for use with a selected quality indicator.

Development and Validation of the Excess Mortality Ratio–adjusted Injury Severity Score Using the International Classification of Diseases 10th Edition

This study aimed to develop and validate a new method for measuring injury severity, the excess mortality ratio-adjusted Injury Severity Score (EMR-ISS), using the International Classification of Diseases 10th Edition (ICD-10). An injury severity grade similar to the Abbreviated Injury Scale (AIS) was converted from the ICD-10 codes on the basis of quintiles of the EMR for each ICD-10 code. Like the New Injury Severity Score (NISS), the EMR-ISS was calculated from three maximum severity grades using data from the Korean National Injury Database. The EMR-ISS was then validated using the Hosmer-Lemeshow goodness-of-fit chi-square (HL chi-square, with lower values preferable), the area under the receiver operating characteristic curve (AUC-ROC), and the Pearson correlation coefficient to compare it with the International Classification of Diseases 9th Edition-based Injury Severity Score (ICISS). Nationwide hospital discharge abstract data (DAD) from stratified-sample general hospitals (n = 150) in 2004 were used for an external validation. The total number of study subjects was 29,282,531, with five subgroups of particular interest identified for further study: traumatic brain injury (TBI, n = 3,768,670), traumatic chest injury (TCI,n = 1,169,828), poisoning (n = 251,565), burns (n = 869,020), and DAD (n = 26,374). The HL chi-square was lower for EMR-ISS than for ICISS in all groups: 42,410.8 versus 55,721.9 in total injury, 7,139.6 versus 20,653.9 in TBI, 6,603.3 versus 4,531.8 in TCI, 2,741.2 versus 9,112.0 in poisoning, 764.4 versus 4,532.1 in burns, and 28.1 versus 49.4 in DAD. The AUC-ROC for death was greater for EMR-ISS than for ICISS: 0.920 versus 0.728 in total injury, 0.907 versus 0.898 in TBI, 0.675 versus 0.799 in TCI, 0.857 versus 0.900 in poisoning, 0.735 versus 0.682 in burns, and 0.850 versus 0.876 in DAD. The Pearson correlation coefficient between the two scores was )0.68 in total injury, )0.76 in TBI, )0.86 in TCI, )0.69 in poisoning,)0.58 in burns, and )0.75 in DAD. The EMR-ISS showed better calibration and discrimination power for prediction of death than the ICISS in most injury groups. The EMR-ISS appears to be a feasible tool for passive injury surveillance of large data sets, such as insurance data sets or community injury registries containing diagnosis codes. Additional further studies for external validation on prospectively collected data sets should be considered.

Hospital discharge abstract data on comorbidity improved the prediction of death among patients hospitalized with aspiration pneumonia

Objective To use diagnoses reported as present at admission in California hospital discharge abstract data to identify categories of comorbid disease and conditions related to aspiration pneumonia and to assess their association with hospital mortality. Study Design and Setting The study population included all persons hospitalized in California from 1996 through 1999, with a principal diagnosis of aspiration pneumonia. Present at admission diagnoses representing comorbid diseases were separated from conditions closely related to aspiration pneumonia by a physician panel through a computer supported Delphi process. Multivariable logistic regression was used to assess the probability of hospital death after adjusting for these patient characteristics. The statistical performance of this method was compared to the performance of two independent methods for measuring comorbid disease. The practical significance of differences in statistical performance was assessed by comparing the estimated effects of age, race, and ethnicity after adjustments using each method. Results Mortality risk adjustment using present at admission diagnoses resulted in substantially better statistical performance and in different measurements of the adjusted effects of age, race, and ethnicity. Conclusion Reporting present at admission diagnoses in hospital discharge data yields meaningful improvements in hospital mortality risk adjustment.

The importance of comorbidities in explaining differences in patient costs.

The authors examine to what extent comorbidities contribute to differences in patient hospital costs. The medical record data for this study were collected from 15 metropolitan Boston hospital for 4,439 patients admitted mostly in 1985 for one of eight common conditions. Massachusetts hospital discharge abstract data for 1985 and 1993 also were used. Comorbidities were identified from the medical record for the 15-hospital data set and from discharge abstracts for all cases. Stepwise regression models were used to develop comorbidity scores. Across all conditions, the medical record-based comorbidity score increased the R2 value from .42 in a model with diagnosis-related groups alone to .50. In condition-specific analyses, including the comorbidity score increased the R2 by more than 50% in six of eight conditions, and was more important than several other dimensions of severity in explaining condition-specific costs. When comorbidities were determined from discharge abstract data rather than medical records, only approximately half as much comorbidity was found. Also, there was much less explanatory power: the all-condition R2 only went from .42 to .44. However, a comorbidity score developed from statewide hospital discharge abstract data was more useful in explaining variations in charges in the eight condition-specific analyses conducted on patients 65 years and older. Comorbidities, particularly when determined from the medical record, are important determinants of patient costs.

Outcome of hospitalized injured patients after institution of a trauma system in an urban area.

To determine if risk of death for hospitalized injured patients changes when an urban trauma system is implemented. An analysis of the risk of death in hospitalized injured patients in 1984 and 1985 (pretrauma system), 1986 and 1987 (early trauma system), and 1990 and 1991 (established trauma system) using hospital discharge abstract data. A total of 18 acute care hospitals in the four-county area encompassing Portland, Ore. A cohort of 70,350 hospitalized patients with at least one discharge diagnosis indicating injury. Death during hospitalization. After the trauma system was established, 77% of patients in the region with an Injury Severity Score (ISS) of 16 or greater were admitted to level I trauma centers. More than 72% of patients with an ISS less than 16 were hospitalized in nontrauma centers. Risk of death for injured patients hospitalized at level I trauma centers declined after the trauma system was established (odds ratio, 0.65; 95% confidence interval, 0.51 to 0.81). Patients who died in trauma centers after institution of the trauma system were younger and had more severe injuries, and the majority died within 1 day of admission, whereas patients who died in nontrauma centers died a median of 5 days after admission. Establishment of a trauma system shifted the more seriously injured patients to level I trauma centers, where there was a significant reduction in the adjusted death rate.

The reliability of racial classifications in hospital discharge abstract data.

Many studies demonstrating an association between race and the use of medical services have used hospital discharge abstract data. The quality of the measures of race in such data sources has heretofore been unexplored. Hospital discharge abstract data from New York State were used to identify 767 cardiac patients who had been admitted to a hospital twice. Racial classifications during the two admissions were concordant 93.7% of the time. Kappa was .89 for Blacks, .72 for Whites, and .43 for all other racial groups. Evidence suggests that the misclassification of race in hospital discharge abstract data is nondifferential; racial discrepancies in access to medical services are thus probably even greater than those previously reported.

Toward a Comprehensive Emergency Medical System for Children

Systems management of life-threatening injuries in children and adults is now accepted as state-of-the-art by those who care for trauma victims in the United States and Canada. A few regional trauma systems for adults have had several decades of experience and have recently served as models for inclusion of pediatric trauma.1 In certain instances, notably the state of Pennsylvania, an emergency medical services (EMS) system has come into being with fully integrated adult and children's components. That the National Pediatric Trauma Registry includes more than 12 000 children is indicative of the significant problem of trauma in childhood; the Registry has provided a necessary base for statistical analysis of injury severity and long-term rehabilitation needs.2 Since 1985, several federally funded state demonstration grants for EMS for children (EMSC) have attempted to establish guidelines for patient care and to suggest methods of ongoing monitoring of the effectiveness of these systems, surveillance of quality, and review of patient outcome. A statewide designated pediatric trauma center for Maryland located in The Johns Hopkins Children's Center has been functional for 12 years.3 Data are now available for objective evaluation of the effectiveness and impact of this regional pediatric trauma program. The level of compliance within Maryland's regionalized pediatric trauma system from 1979 through 1986 was recently examined using hospital discharge abstract data routinely recorded for all discharges from 58 acute care hospitals in the state of Maryland.4 Compliance with regionalization was measured by examining (1) the proportion of patients with injuries of varying injury severity scores5,6 who were treated at each of three levels of care (statewide pediatric trauma center, regional trauma center, and community hospital) and (2) the proportion of in-hospital deaths occurring at each level of care.

Regional differences in surgical care based upon uniform physician and hospital discharge abstract data.

Population based surgical rates for various common surgical procedures were analyzed on a regional basis by examining select uniform hospital discharge abstract data from Wisconsin hospitals. The surgical asbracts of nearly 64,000 procedures were compared to the supply of physicians and showed a significant variation in the rates of common procedures even within rather large planning districts. In general, the volume of surgery correlated with the supply of surgeons. Exceptions were noted; for example, primary appendectomy, tonsillectomy and adenoidectomy (T & A), and inguinal herniorrhaphy did not correlate with the supply of surgeons, but did correlate with the supply of general practitioners. Further, T & A had a strong negative correlation to the supply of ear, nose and throat specialists. Information of this type has significance both for Professional Standard Review Organizations (PSROs) and Health Planning Agencies (HPAs). Further work will be necessary to define optimal surgical rates.