The Potential of High‐Dimensional Propensity Scores in Health Services Research: An Exemplary Study on the Quality of Care for Elective Percutaneous Coronary Interventions
Sie können Bookmarks mittels Listen verwalten, loggen Sie sich dafür bitte in Ihr SLUB Benutzerkonto ein.
Medientyp:
E-Artikel
Titel:
The Potential of High‐Dimensional Propensity Scores in Health Services Research: An Exemplary Study on the Quality of Care for Elective Percutaneous Coronary Interventions
Beschreibung:
<jats:sec><jats:title>Objective</jats:title><jats:p>Evaluating the potential of the high‐dimensional propensity score (<jats:styled-content style="fixed-case">HDPS</jats:styled-content>) to control for residual confounding in studies analyzing quality of care based on administrative health insurance data.</jats:p></jats:sec><jats:sec><jats:title>Data Source</jats:title><jats:p>Secondary data from 2004 to 2009 from three German statutory health insurance providers.</jats:p></jats:sec><jats:sec><jats:title>Study Design</jats:title><jats:p>We conducted a retrospective cohort study in patients with elective percutaneous coronary interventions (<jats:styled-content style="fixed-case">PCI</jats:styled-content>s) and compared the mortality risk between the in‐ and outpatient setting using Cox regression. Adjustment for predefined confounders was performed using conventional propensity score (<jats:styled-content style="fixed-case">PS</jats:styled-content>) techniques. Further, an <jats:styled-content style="fixed-case">HDPS</jats:styled-content> was calculated based on predefined and empirically selected confounders from the database.</jats:p></jats:sec><jats:sec><jats:title>Principal Findings</jats:title><jats:p>Conventional <jats:styled-content style="fixed-case">PS</jats:styled-content> methods showed a decreased mortality risk for outpatient compared to inpatient <jats:styled-content style="fixed-case">PCI</jats:styled-content>s, while trimming of patients with nonoverlap in the <jats:styled-content style="fixed-case">HDPS</jats:styled-content> distribution and weighting resulted in a comparable risk. Most comorbidities were less prevalent in the <jats:styled-content style="fixed-case">HDPS</jats:styled-content>‐trimmed population compared to the original one.</jats:p></jats:sec><jats:sec><jats:title>Conclusion</jats:title><jats:p>The <jats:styled-content style="fixed-case">HDPS</jats:styled-content> methodology may reduce residual confounding by rendering the studied cohort more comparable through restriction. However, results cannot be generalized for the entire study population. To provide unbiased results, full assessment of all unmeasured confounders from proxy information in the database would be necessary.</jats:p></jats:sec>