Journal of Parenteral and Enteral Nutrition · 2019 · 16 citations · 37 references
AgricultureCardiologyNutritionClinical EpidemiologyNutritional EpidemiologyOutcomes ResearchMalnutritionMedical Nutrition TherapyPropensity Score MethodsAcute Myocardial Infarction DiagnosisCardiac Outcome AssessmentComorbidity AnalysisHealth Outcomes AnalysisPropensity InferencePropensity Scoring
BACKGROUND: Protein-energy malnutrition (PEM) diminishes amino acid and energy availability, impairing the body's healing capability after injury, such as in myocardial damage following acute myocardial infarction (AMI). AIMS: We sought to investigate the influence of PEM on clinical outcomes of AMI. METHODS: We identified records with a primary discharge diagnosis of AMI from the Nationwide Inpatient Sample (2012-2014), stratified by concomitant PEM. We matched PEM to no-PEM (1:1) using a greedy algorithm-based propensity methodology and estimated the impact of PEM on health outcomes (SAS 9.4). RESULTS: Of the 332,644 hospitalizations for AMI, 11,675 had concomitant PEM accounting for roughly $US 1.5 billion and over 119,792 hospital days. PEM was associated with older age (74.43- vs. 66.90-years; P < 0.0001), female sex (49.19% vs. 38.44%; P < 0.0001), black race (12.78% vs. 10.46%; P < 0.0001), and higher comorbidity burden (Deyo > 3: 32.77% vs. 16.69%; P < 0.0001). After propensity matching, PEM was associated with higher mortality (Adjusted odds ratio [AOR]: 1.59 [1.46-1.73]), cardiogenic shock (AOR: 2.26 [2.08-2.44]), discharge to secondary facilities (AOR: 2.21 [2.10-2.33]), charges ($135,500 [$131,956-139,139] vs. $81,084 [$79,241-82,970]), cardiac artery bypass surgery (AOR:1.81 [1.66-1.97]), intra-aortic balloon pump placement (AOR: 1.83 [1.65-2.04]) and longer length of stay (10.15- vs. 5.52-days). CONCLUSIONS: PEM is a predisposing factor for devastating clinical outcomes among AMI hospitalizations. Higher prevention, identification and management of PEM among high-risk individuals (older age, female sex, and black race) residing in the community are needed.
37
Comorbidity Measures for Use with Administrative Data
Anne Elixhauser, Claudia Steiner, D.R. Harris et al. · Medical Care · 1998 · 9.9K citations
Education, Health Services Administration, Health Assessment +16