Biomedical subjects

Peter C Austin

Publications and source records attributed to Peter C Austin.

79 records · Page 5Linked to original sources

PubMed2002

Adverse effects of observational studies when examining adverse outcomes of drugs: case-control studies with low prevalence of exposure.

OBJECTIVES: The case-control study is commonly used to examine adverse drug events, in which prevalence of exposure in the source population is frequently very low. The objective of the current study was to examine the bias inherent in the odds ratio assessing the association between exposure and an adverse outcome when prevalence of exposure in the source population is extremely low. DESIGN: Monte Carlo simulations examined the effect of sample size, exposure prevalence, and magnitude of the underlying odds ratio on the bias of the estimated risk ratio, and the power to detect a non-zero risk ratio. RESULTS: Once the underlying odds ratio was at least four, the adverse effects of low prevalence of exposure was minimal. Studies with small sample sizes and low prevalence of exposure, coupled with small to moderate effect sizes, can result in biased estimates of association between exposure and disease status. With a sample size of 200 and an exposure prevalence of 0.5% in the control population, the bias in the estimated odds ratio can be as large as 115%. However, bias becomes negligible as sample size becomes large (n > or = 2000), even when prevalence of exposure is very low. Once the expected number of exposed controls is at least eight, the bias in the estimated odds ratio was no more than 5%. CONCLUSIONS: Studies with small sample sizes and low prevalence of exposure, coupled with small to moderate effect sizes can result in biased estimates of association between exposure status and adverse drug effects. However, bias becomes negligible as sample size becomes large.

Bias↗

PubMedDate not supplied

A comparison of methods for analyzing health-related quality-of-life measures.

OBJECTIVES: Self-reported health status is often measured using psychometric or utility indices that provide a score intended to summarize an individual's health. Measurements of health status can be subject to a ceiling effect. Frequently, researchers want to examine relationships between determinants of health and measures of health status. Regression methods that ignore the censoring in the health status measurement can produce biased coefficient estimates. The authors examine the performance of three different models for assessing the relationship between demographic characteristics and health status. METHODS: Three methods that allow one to analyze data subject to a ceiling effect are compared. The first model is the classic Tobit model. The second and third models are robust variants of the Tobit model: symmetrically trimmed least squares and censored least absolute deviations (Censored LAD) regression. These models were fit to data from the Canadian National Population Health Survey. The results are compared to three models that ignore the presence of a ceiling effect. RESULTS: The Censored LAD model produced coefficient estimates that tended to be shrunk toward 0, compared to the other two models. The three models produced conflicting evidence on the effect of gender on health status. Similarly, the rate of decay in health status with increasing age differed across the three models. The Censored LAD model produced results very similar to median regression. Furthermore, the censored LAD model had the lowest prediction error in an independent validation dataset. CONCLUSIONS: Our results highlight the need for careful consideration about how best to model variation in health status. Based upon our study, we recommend the use of Censored LAD regression.

Age Distribution↗

PubMedDate not supplied

Bayesian extensions of the Tobit model for analyzing measures of health status.

Self-reported health status is often measured using utility indices that provide a score intended to summarize an individual's health. Measurements of health status can be subject to a ceiling effect. Frequently, researchers want to examine relationships between determinants of health and measures of health status. In this article, Bayesian extensions of the classical Tobit model are used to study the relationship between health status and predictors of health. The author examined models where the conditional distribution of health status was either normal or lognormal, and allowed for both homoscedasticity and heteroscedasticity. Bayes factors were then used to compare the evidence for a given model against that for a competing model. The author found very strong evidence that the distribution of the Health Utilities Index, conditional on age, gender, income adequacy, and number of chronic conditions, was normal with nonuniform variance, compared to the competing models.

Bayes Theorem↗

PubMedDate not supplied

A comparison of Bayesian methods for profiling hospital performance.

There is a growing interest in the use of Bayesian methods for profiling institutional performance. In the literature, several studies have compared different frequentist methods for classifying hospitals as performance outliers. The purpose of this study was to compare 4 different Bayesian methods for classifying hospitals as outcomes outliers, using 30-day hospital-level mortality rates for a cohort of acute myocardial infarction patients as a test case. The 1st Bayesian method involved determining the probability that a hospital's mortality rare for an average patient exceeded a specified threshold. The 2nd method involved ranking hospitals according to their mortality rate for an average patient. The 3rd method involved determining the probability that a hospital's standardized mortality ratio exceeded a specified threshold. The 4th method involved ranking hospitals according to their standardized mortality ratio. In most of the scenarios examined, there was only marginal agreement between the different methods. In only 4 of 19 comparisons, was there good agreement between the different methods (0.40 < or = kappa < or = 0.75). Methods based on ranking institutions were relatively insensitive to differences between hospitals. These inconsistencies raise questions about the choice of methods for classifying hospital performance, and they suggest a need for urgent research into which methods are best able to discriminate between institutions and which are most meaningful to decision makers.

Bayes Theorem↗

PubMedDate not supplied

The use of fixed- and random-effects models for classifying hospitals as mortality outliers: a Monte Carlo assessment.

BACKGROUND: There is an increasing movement towards the release of hospital "report-cards. "However, there is a paucity of research into the abilities of the different methods to correctly classify hospitals as performance outliers. OBJECTIVE: To examine the ability of risk-adjusted mortality rates computed using conventional logistic regression and random-effects logistic regression models to correctly identify hospitals that have higher than acceptable mortality. RESEARCH DESIGN: Monte Carlo simulations. MEASURES: Sensitivity, specificity, and positive predictive value of a classification as a high-outlier for identifying hospitals with higher than acceptable mortality rates. RESULTS: When the distribution of hospital specific log-odds of death was normal, random-effects models had greater specificity and positive predictive value than fixed-effects models. However, fixed-effects models had greater sensitivity than random-effects models. CONCLUSIONS: Researchers and policy makers need to carefully consider the balance between false positives and false negatives when choosing statistical models for determining which hospitals have higher than acceptable mortality in performance profiling.

Hospital Mortality↗

PubMedDate not supplied

Optimal statistical decisions for hospital report cards.

PURPOSE: Hospital report cards provide information designed to help patients and providers to make decisions. The purpose of this study was to place the design of hospital report cards into a decision-theoretic framework. The authors' objectives were 2-fold: 1st, to determine what the choice of significance level implies about the relative value of the different types of misclassifications that can arise. Second, to determine optimal significance levels for specific cost functions describing the relative costs associated with different types of misclassifications. METHODS: Using a previously published theoretical model for hospital mortality, the authors computed false positive (i.e., falsely classified as providing poor-quality care) and false negative (falsely classified as providing good-quality care) rates. First, they determined the cost functions for false negatives and false positives that are implicitly associated with the use of significance levels of 0.05 and 0.01 for identifying hospitals with higher than average mortality. Second, they determined the levels of statistical significance that should be chosen to minimize predefined cost functions, thus minimizing costs associated with misclassifying hospitals. RESULTS: The lower the statistical significance level required for identifying hospitals with higher than average mortality, the lower the implicit cost of false negatives compared to false positives. For a given significance level, the greater the number of patients treated at each hospital or the greater the proportion of truly poorly performing hospitals, the lower the value of the implicit cost incurred by a false negative compared to that for a false positive. For cost functions that put a high relative penalty on false negatives compared to false positives, the use of significance levels of 0.05 or 0.01 does not result in optimal decisions across expected number of patients treated at each hospital or proportions of truly poor-quality care. CONCLUSIONS: Hospital report cards that use significance levels of either 0.05 or 0.01 to identify hospitals that have statistically significantly higher than average mortality make implicit assumptions about cost functions, and the values of the optimal cost function vary across scenarios.

Decision Support Techniques↗

PubMedDate not supplied

The impact of unmeasured clinical variables on the accuracy of hospital report cards: a Monte Carlo study.

PURPOSE: Hospital report cards are commonly produced using administrative data. The objective of this study was to determine the impact of unmeasured clinical data on the accuracy of hospitals' report cards. METHODS: Monte Carlo simulations were based on both administrative and detailed clinical data for patients hospitalized with an acute myocardial infarction in Ontario, Canada. Data were simulated such that the true performance of each hospital was known. Both clinical and administrative risk scores were randomly generated for each patient. The ability of hospital report cards to correctly identify hospitals that truly had higher than acceptable mortality was compared when both clinical and administrative data were used and when only administrative data were used. By using Monte Carlo simulations, we were able to incrementally increase the divergence between the 2 risk scores. RESULTS: In a wide range of settings, sensitivity and specificity of hospital report cards was only negligibly greater when both administrative and clinical data were used compared to when only administrative data were used. CONCLUSIONS: Unmeasured clinical data have at most a minor impact on the accuracy of cardiac hospital report cards.

Hospital Administration↗