A pseudo-likelihood method for estimating misclassification probabilities in competing-risks settings when true-event data are partially observed.

Outcome misclassification occurs frequently in binary-outcome studies and can result in biased estimation of quantities such as the incidence, prevalence, cause-specific hazards, cumulative incidence functions, and so forth. A number of remedies have been proposed to address the potential misclassification of the outcomes in such data. The majority of these remedies lie in the estimation of misclassification probabilities, which are in turn used to adjust analyses for outcome misclassification. A number of authors advocate using a gold-standard procedure on a sample internal to the study to learn about the extent of the misclassification. With this type of internal validation, the problem of quantifying the misclassification also becomes a missing data problem as, by design, the true outcomes are only ascertained on a subset of the entire study sample. Although, the process of estimating misclassification probabilities appears simple conceptually, the estimation methods proposed so far have several methodological and practical shortcomings. Most methods rely on missing outcome data to be missing completely at random (MCAR), a rather stringent assumption which is unlikely to hold in practice. Some of the existing methods also tend to be computationally-intensive. To address these issues, we propose a computationally-efficient, easy-to-implement, pseudo-likelihood estimator of the misclassification probabilities under a missing at random (MAR) assumption, in studies with an available internal-validation sample. We present the estimator through the lens of studies with competing-risks outcomes, though the estimator extends beyond this setting. We describe the consistency and asymptotic distributional properties of the resulting estimator, and derive a closed-form estimator of its variance. The finite-sample performance of this estimator is evaluated via simulations. Using data from a real-world study with competing-risks outcomes, we illustrate how the proposed method can be used to estimate misclassification probabilities. We also show how the estimated misclassification probabilities can be used in an external study to adjust for possible misclassification bias when modeling cumulative incidence functions.

Hospital outcomes in non-surgical patients identified at risk for OSA.

BACKGROUND: In-hospital respiratory outcomes of non-surgical patients with undiagnosed obstructive sleep apnea (OSA), particularly those with significant comorbidities are not well defined. Undiagnosed and untreated OSA may be associated with increased cardiopulmonary morbidity. STUDY OBJECTIVES: Evaluate respiratory failure outcomes in patients identified as at-risk for OSA by the Berlin Questionnaire (BQ). METHODS: This was a retrospective study conducted using electronic health records at a large health system. The BQ was administered at admission to screen for OSA to medical-service patients under the age of 80 years old meeting the following health system criteria: (1) BMI greater than 30; (2) any of the following comorbid diagnoses: hypertension, heart failure, acute coronary syndrome, pulmonary hypertension, arrhythmia, cerebrovascular event/stroke, or diabetes. Patients with known OSA or undergoing surgery were excluded. Patients were classified as high-risk or low-risk for OSA based on the BQ score as follows: low-risk (0 or 1 category with a positive score on the BQ); high-risk (2 or more categories with a positive score on BQ). The primary outcome was respiratory failure during index hospital stay defined by any of the following: orders for conventional ventilation or intubation; at least two instances of oxygen saturation less than 88% by pulse oximetry; at least two instances of respiratory rate over 30 breaths per minute; and any orders placed for non-invasive mechanical ventilation without a previous diagnosis of sleep apnea. Propensity scores were used to control for patient characteristics. RESULTS: Records of 15,253 patients were assessed. There were no significant differences in the composite outcome of respiratory failure by risk of OSA (high risk: 11%, low risk: 10%, p = 0.55). When respiratory failure was defined as need for ventilation, more patients in the low-risk group experienced invasive mechanical ventilation (high-risk: 1.8% vs. low-risk: 2.3%, p = 0.041). Mortality was decreased in patients at high-risk for OSA (0.86%) vs. low risk for OSA (1.53%, p < 0.001). CONCLUSIONS: Further prospective studies are needed to understand the contribution of undiagnosed OSA to in-hospital respiratory outcomes.

Neurodevelopment in Young Children Born to HIV-Infected Mothers: A Meta-analysis.

CONTEXT: HIV-infected (HIV+) children have worse neurodevelopmental outcomes compared with HIV-uninfected children. However, little is known regarding the differences in neurodevelopment between young HIV+ children, HIV-exposed but uninfected (HEU) children, and HIV-unexposed and uninfected (HUU) children. OBJECTIVE: To systematically review and meta-analyze data on neurodevelopmental performance between young HIV+, HEU, and HUU children. DATA SOURCES: We systematically searched the following electronic bibliographic databases: Ovid Medline, Embase, PsycINFO, Education Resources Information Center, and the Cochrane Database of Systematic Reviews. STUDY SELECTION: Studies were selected on the basis of defined inclusion criteria. Titles, abstracts, and full texts were assessed by 2 independent reviewers. DATA EXTRACTION: Data were extracted by 2 independent reviewers and cross-checked by 2 additional reviewers. RESULTS: Forty-five studies were identified for inclusion in the systematic review, and of these, 11 were included in the meta-analysis on the basis of availability of Bayley Scales of Infant and Toddler Development scores. Within the meta-analysis, when compared with their HUU peers, HIV+ and HEU children had lower cognitive and motor scores. HIV+ and HEU children with antiretroviral (ARV) exposure had lower cognitive and motor scores compared with those without ARV exposure. LIMITATIONS: We were unable to control adequately for intravenous drug use, geographic location, or quality of the assessment independently. CONCLUSIONS: Both HIV+ and HEU children had worse developmental outcomes compared with HUU children. HIV+ and HEU children with ARV exposure also had worse developmental outcomes compared with those without exposure; however, these results should be interpreted with caution. More research is needed to identify the impact of ARV exposure on young children.