Marina Yogendran

Using multiple data features improved the validity of osteoporosis case ascertainment from administrative databases

OBJECTIVES The aim was to construct and validate algorithms for osteoporosis case ascertainment from administrative databases and to estimate the population prevalence of osteoporosis for these algorithms. STUDY DESIGN AND SETTING Artificial neural networks, classification trees, and logistic regression were applied to hospital, physician, and pharmacy data from Manitoba, Canada. Discriminative performance and calibration (i.e., error) were compared for algorithms defined from different sets of diagnosis, prescription drug, comorbidity, and demographic variables. Algorithms were validated against a regional bone mineral density testing program. RESULTS Discriminative performance and calibration were poorer and sensitivity was generally lower for algorithms based on diagnosis codes alone than for algorithms based on an expanded set of data features that included osteoporosis prescriptions and age. Validation measures were similar for neural networks and classification trees, but prevalence estimates were lower for the former model. CONCLUSION Multiple features of administrative data generally resulted in improved sensitivity of osteoporosis case-detection algorithm without loss of specificity. However, prevalence estimates using an expanded set of features were still slightly lower than estimates from a population-based study with primary data collection. The classification methods developed in this study can be extended to other chronic diseases for which there may be multiple markers in administrative data.

The Association Between Corticosteroid Use and Development of Fractures Among IBD Patients in a Population-Based Database

OBJECTIVE:Because the rate of fracture among patients with inflammatory bowel disease (IBD) is only slightly higher than that in the general population, it is important to define high-risk groups worthy of diagnostic evaluation or prophylactic interventions. Corticosteroid use has been considered in other diseases to be a risk for fracture, although not all studies in IBD are concordant on this point. We aimed to determine whether patients with IBD drawn from a population-based database who sustain fractures are more likely to have been using corticosteroids than a matched group of IBD patients who did not fracture.METHODS:We extracted from our population-based University of Manitoba Inflammatory Bowel Disease Epidemiology Database the number of patients with a new diagnosis of fracture between the years 1997–2000. From within our Inflammatory Bowel Disease Epidemiology Database, we extracted a control group of IBD patients who did not develop fractures matched to the case group who did by age, gender, diagnosis, year of diagnosis, and geographic area of residence. We linked our cohorts with Manitoba Healths Drug Program Information Network to study corticosteroid use within 2 yr before fracture diagnosis. The Drug Program Information Network is a population-based database, established in 1995, which records all prescription drugs.RESULTS:Fractures were identified in 13 patients with Crohns disease and in 28 patients with ulcerative colitis. The control group included 103 Crohns disease and 173 ulcerative colitis patients who did not fracture. In Crohns disease, for the group who fractured compared with the controls who did not fracture, corticosteroid use before fracture was evident in seven (54%) compared with 21 (22%) who did not fracture (χ2 = 4.45, df = 1, p = 0.035). In ulcerative colitis, for the group who fractured compared with the controls who did not fracture, corticosteroid use before fracture was evident in five (18%) compared with 37 (21%) who did not fracture (χ2 = 0.031, df = 1, p = 0.861). Fracture cases were more likely to be exposed to oral corticosteroids (OR = 1.75; 95% CI = 0.82–3.75), but this result is not significant. Regarding corticosteroid dosing among the 12 patients with IBD who fractured and used corticosteroids, the mean total days supply was 314 days ± 236 days compared with 258 days ± 278 days in those who did not fracture (p = 0.16). The prescribed daily dose among corticosteroid users was comparable for those who fractured versus those who did not fracture (18 mg/day vs 21 mg/day, p = 0.90).CONCLUSIONS:Patients who require corticosteroids in Crohns disease should be considered at risk for fracture. Further research is required to delineate after how much corticosteroid use are subjects at risk and/or after what duration of active disease.

Population-based trends in osteoporosis management after new initiations of long-term systemic glucocorticoids (1998-2008).

OBJECTIVES Our objective was to describe changes in glucocorticoid-induced osteoporosis (GIOP) preventive care from 1998-2008 including rates and correlates of bone mineral density (BMD) testing and osteoporosis treatment in new long-term glucocorticoid initiations. METHODS A population-based study of adults aged 20 yr or older in Manitoba, Canada, was conducted using linked healthcare databases. Subjects with new long-term (≥90 d) systemic glucocorticoid initiations were identified within each fiscal year. High-quality GIOP preventive care was defined by the composite of BMD testing or osteoporosis treatment within 6 months of starting glucocorticoids. For each initiation, we identified sociodemographic and clinical characteristics, prednisone dose equivalents, and prescriber specialty. Multivariable Poisson regression models were used to calculate adjusted incidence rate ratios (aIRR). RESULTS We studied 17,736 new long-term glucocorticoid initiations; one third were at least 10 mg prednisone daily, and most (64%) were prescribed by general practitioners. Overall, 6-month rates of BMD testing were 6%, osteoporosis treatment 22%, and the composite of testing or treatment 25%. From 1998-2008, there were modest increases in BMD testing (from 4 to 6%), osteoporosis treatment (from 15 to 24%), and testing or treatment [from 17 to 27%; aIRR = 1.51; 95% confidence interval (CI) = 1.40-1.63]. High-quality GIOP preventive care varied significantly by age (16% for those <50 yr vs. 27% for those ≥70 yr; aIRR = 0.57; 95% CI = 0.52-0.63), sex (13% for men vs. 34% for women; aIRR = 0.40; 95% CI = 0.37-0.43), and prescriber (23% general practice vs. 44% rheumatology; aIRR = 0.56; 95% CI = 0.52-0.60). CONCLUSIONS Quality of GIOP preventive care has improved but remains suboptimal with only one quarter of those starting long-term glucocorticoids receiving BMD testing or osteoporosis treatment. Interventions to improve GIOP prevention, especially targeting younger patients, men, and nonspecialists, are needed.

Rates of readmission and death associated with leaving hospital against medical advice: a population-based study

Background: Leaving hospital against medical advice may have adverse consequences. Previous studies have been limited by evaluating specific types of patients, small sample sizes and incomplete determination of outcomes. We hypothesized that leaving hospital against medical advice would be associated with increases in subsequent readmission and death. Methods: In a population-based analysis involving all adults admitted to hospital and discharged alive in Manitoba from Apr. 1, 1990, to Feb. 28, 2009, we evaluated all-cause 90-day mortality and 30-day hospital readmission. We used multivariable regression, adjusted for age, sex, socioeconomic status, year of hospital admission, patient comorbidities, hospital diagnosis, past frequency of admission to hospital, having previously left hospital against medical advice and data clustering (patients with multiple admissions). For readmission, we assessed both between-person and within-person effects of leaving hospital against medical advice. Results: Leaving against medical advice occurred in 21 417 of 1 916 104 index hospital admissions (1.1%), and was associated with higher adjusted rates of 90-day mortality (odds ratio [OR] 2.51, 95% confidence interval [CI] 2.18–2.89), and 30-day hospital readmission (within-person OR 2.10, CI 1.99–2.21; between-person OR 3.04, CI 2.79–3.30). In our additional analyses, elevated rates of readmission and death associated with leaving against medical advice were manifest within 1 week and persisted for at least 180 days after discharge. Interpretation: Adults who left the hospital against medical advice had higher rates of hospital readmission and death. The persistence of these effects suggests that they are not solely a result of incomplete treatment of acute illness. Interventions aimed at reducing these effects may need to include longitudinal interventions extending beyond admission to hospital.

The accuracy of administrative data for identifying the presence and timing of admission to intensive care units in a Canadian province.

Background:A prerequisite for using administrative data to study the care of critically ill patients in intensive care units (ICUs) is that it accurately identifies such care. Only limited data exist on this subject. Objective:To assess the accuracy of administrative data in the Canadian province of Manitoba for identifying the existence, number, and timing of admissions to adult ICUs. Research Design:For the period 1999 to 2008, we compared information about ICU care from Manitoba hospital abstracts, with the criterion standard of a clinical ICU database that includes all admissions to adult ICUs in its largest city of Winnipeg. Comparisons were made before and after a national change in administrative data requirements that mandated specific data elements identifying the existence and timing of ICU care. Results:In both time intervals, hospital abstracts were extremely accurate in identifying the presence of ICU care, with positive predictive values exceeding 98% and negative predictive values exceeding 99%. Administrative data correctly identified the number of separate ICU admissions for 93% of ICU-containing hospitalizations; inaccuracy increased with more ICU stays per hospitalization. Hospital abstracts were highly accurate for identifying the timing of ICU care, but only for hospitalizations containing a single ICU admission. Conclusions:Under current national-reporting requirements, hospital administrative data in Canada can be used to accurately identify and quantify ICU care. The high accuracy of Manitoba administrative data under the previous reporting standards, which lacked standardized coding elements specific to ICU care, may not be generalizable to other Canadian jurisdictions.

Epidemiology of critically ill patients in intensive care units: a population-based observational study

IntroductionEpidemiologic assessment of critically ill people in Intensive Care Units (ICUs) is needed to ensure the health care system can meet current and future needs. However, few such studies have been published.MethodsPopulation-based analysis of all adult ICU care in the Canadian province of Manitoba, 1999 to 2007, using administrative data. We calculated age-adjusted rates and trends of ICU care, overall and subdivided by age, sex and income.ResultsIn 2007, Manitoba had a population of 1.2 million, 118 ICU beds in 21 ICUs, for 9.8 beds per 100,000 population. Approximately 0.72% of men and 0.47% of women were admitted to ICUs yearly. The age-adjusted, male:female rate ratio was 1.75 (95% CI 1.64 to 1.88). Mean age was 64.5 ± 16.4 years. Rates rose rapidly after age 40, peaked at age 75 to 80, and declined for the oldest age groups. Rates were higher among residents of lower income areas, for example declining from 7.9 to 4.4 per 100,000 population from the poorest to the wealthiest income quintiles (p <0.0001). Rates of ICU admission slowly declined over time, while cumulative yearly ICU bed-days slowly rose; changes were age-dependent, with faster declines in admission rates with older age. There was a high rate of recidivism; 16% of ICU patients had received ICU care previously.ConclusionsThese temporal trends in ICU admission rates and cumulative bed-days used have significant implications for health system planning. The differences by age, sex and socioeconomic status, and the high rate of recidivism require further research to clarify their causes, and to devise strategies for reducing critical illness in high-risk groups.

Comparing administrative and survey data for ascertaining cases of irritable bowel syndrome: a population-based investigation

BackgroundAdministrative and survey data are two key data sources for population-based research about chronic disease. The objectives of this methodological paper are to: (1) estimate agreement between the two data sources for irritable bowel syndrome (IBS) and compare the results to those for inflammatory bowel disease (IBD); (2) compare the frequency of IBS-related diagnoses in administrative data for survey respondents with and without self-reported IBS, and (3) estimate IBS prevalence from both sources.MethodsThis retrospective cohort study used linked administrative and health survey data for 5,134 adults from the province of Manitoba, Canada. Diagnoses in hospital and physician administrative data were investigated for respondents with self-reported IBS, IBD, and no bowel disorder. Agreement between survey and administrative data was estimated using the κ statistic. The χ2 statistic tested the association between the frequency of IBS-related diagnoses and self-reported IBS. Crude, sex-specific, and age-specific IBS prevalence estimates were calculated from both sources.ResultsOverall, 3.0% of the cohort had self-reported IBS, 0.8% had self-reported IBD, and 95.3% reported no bowel disorder. Agreement was poor to fair for IBS and substantially higher for IBD. The most frequent IBS-related diagnoses among the cohort were anxiety disorders (34.4%), symptoms of the abdomen and pelvis (26.9%), and diverticulitis of the intestine (10.6%). Crude IBS prevalence estimates from both sources were lower than those reported previously.ConclusionsPoor agreement between administrative and survey data for IBS may account for differences in the results of health services and outcomes research using these sources. Further research is needed to identify the optimal method(s) to ascertain IBS cases in both data sources.

A population-based analysis of leaving the hospital against medical advice: incidence and associated variables

BackgroundPrior studies of patients leaving hospital against medical advice (AMA) have been limited by not being population-based or assessing only one type of patient.MethodsWe used administrative data at the Manitoba Centre for Health Policy to evaluate all adult residents of Manitoba, Canada discharged alive from acute care hospitals between April 1, 1990 and February 28, 2009. We identified the rate of leaving AMA, and used multivariable logistic regression to identify socio-demographic and diagnostic variables associated with leaving AMA.ResultsOf 1 916 104 live hospital discharges, 21 417 (1.11%) ended with the patient leaving AMA. The cohort contained 610 187 individuals, of whom 12 588 (2.06%) left AMA once and another 2 986 (0.49%) left AMA more than once. The proportion of AMA discharges did not change over time. Alcohol and drug abuse was the diagnostic group with the highest proportion of AMA discharges, at 11.71%. Having left AMA previously had the strongest association with leaving AMA (odds ratio 170, 95% confidence interval 156–185). Leaving AMA was more common among men, those with lower average household incomes, histories of alcohol or drug abuse or HIV/AIDS. Major surgical procedures were associated with a much lower chance of leaving the hospital AMA.ConclusionsThe rate of leaving hospital AMA did not systematically change over time, but did vary based on patient and illness characteristics. Having left AMA in the past was highly predictive of subsequent AMA events.

Constructing episodes of inpatient care: data infrastructure for population-based research.

BackgroundDatabases used to study the care of patients in hospitals and Intensive Care Units (ICUs) typically contain a separate entry for each segment of hospital or ICU care. However, it is not uncommon for patients to be transferred between hospitals and/or ICUs, and when transfers occur it is necessary to combine individual entries to accurately reconstruct the complete episodes of hospital and ICU care. Failure to do so can lead to erroneous lengths-of-stay, and rates of admissions, readmissions, and death.MethodsThis study used a clinical ICU database and administrative hospital abstracts for the adult population of Manitoba, Canada from 2000–2008. We compared five methods for identifying patient transfers and constructing hospital episodes, and the ICU episodes contained within them. Method 1 ignored transfers. Methods 2–5 considered the time gap between successive entries (≤1 day vs. ≤2 days), with or without use of data fields indicating inter-hospital transfer. For the five methods we compared the resulting number and lengths of hospital and ICU episodes.ResultsDuring the study period, 48,551 hospital abstracts contained 53,246 ICU records. For Method 1 these were also the number of hospital and ICU episodes, respectively. Methods 2–5 gave remarkably similar results, with transfers included in approximately 25% of ICU-containing hospital episodes, and 10% of ICU episodes. Comparison with Method 1 showed that failure to account for such transfers resulted in overestimating the number of episodes by 7-10%, and underestimating mean or median lengths-of-stay by 9-30%.ConclusionsIn Manitoba is it not uncommon for critically ill patients to be transferred between hospitals and between ICUs. Failure to account for transfers resulted in inaccurate assessment of parameters relevant to researchers, clinicians, and policy-makers. The details of the method used to identify transfers, at least among the variations tested, made relatively little difference. In addition, we showed that these methods for constructing episodes of hospital and ICU care can be implemented in a large, complex dataset.

Major Osteoporotic to Hip Fracture Ratios in Canadian Men and Women With Swedish Comparisons: A Population‐Based Analysis

Fracture Risk Assessment (FRAX) tools are calibrated from country‐specific fracture epidemiology. Although hip fracture data are usually available, data on non‐hip fractures for most countries are often lacking. In such cases, rates are often estimated by assuming similar non‐hip to hip fracture ratios from historical (1987 to 1996) Swedish data. Evidence that countries share similar fracture ratios is limited. Using data from Manitoba, Canada (2000 to 2007, population 1.2 million), we identified 21,850 incident major osteoporotic fractures (MOF) in men and women aged >50 years. Population‐based age‐ and sex‐specific ratios of clinical vertebral, forearm, and humerus fractures to hip fractures were calculated, along with odds ratios (ORs) and 95% confidence intervals (CIs). All ratios showed decreasing trends with increasing age for both men and women. Men and women showed similar vertebral/hip fracture ratios (all p > 0.1, with ORs 0.86 to 1.25). Forearm/hip and humerus/hip fracture ratios were significantly lower among men than women (forearm/hip ratio: p < 0.01 for all age groups, with ORs 0.29 to 0.53; humerus/hip ratio: p < 0.05 for all age groups [except 80 to 84 years] with ORs 0.46 to 0.86). Ratios for any MOF/hip fracture were also significantly lower among men than women in all but two subgroups (p < 0.05 for all age groups [except 80 to 84 and 90+ years] with ORs 0.48 to 0.87). Swedish vertebral/hip fracture ratios were similar to the Canadian fracture ratios (within 7%) but significantly lower for other sites (men and women: 46% and 35% lower for forearm/hip ratios, 19% and 15% lower for humerus/hip ratios, and 19% and 23% lower for any MOF/hip ratios). These differences have implications for updating and calibrating FRAX tools, fracture risk estimation, and intervention rates. Moreover, wherever possible, it is important that countries try to collect accurate non‐hip fracture data.