Original article
Selecting a patient characteristics index for the prediction of medical outcomes using administrative claims data

https://doi.org/10.1016/0895-4356(94)00202-2Get rights and content

Abstract

Recently, there has been a great deal of discussion regarding the use of administrative databases to study outcomes of medical care. A major issue in this discussion is how to classify patients in terms of characteristics such as disease-severity, comorbidities, resource needs, stability, etc. Different indices have been developed in an attempt to provide a common classification scheme in terms of these characteristics. In this paper, we examine the utility of four indices in the prediction of length of stay and 30-day mortality for patients undergoing total knee replacement surgery between 1985 and 1989. The indices that we compare are the Deyo-adapted Charlson Index, the Relative Intensity Score derived from Patient Management Categories (PMCs), the Patient Severity Level derived from PMCs, and the number of diagnoses (up to five) listed in the Medicare claims data. The first three of these indices represent measures of comorbidity, resource use, and severity of illness, respectively. The number of diagnoses is likely to capture aspects of each of these characteristics. We find that all of the indices improve (in terms of model fit) over the baseline (no index) models of length of stay and mortality, and the Relative Intensity Score and Patient Severity Level result in the greatest improvement in measures of model fit. We found, however, that these two indices have a non-monotonic relationship with length of stay and mortality. The Deyo-adapted Charlson Index performed least well of the four indices in terms of explanatory ability. The number of diagnoses performed well, and does not suffer from problems associated with miscoding on claims data.

References (18)

There are more references available in the full text version of this article.

Cited by (121)

  • The impact of operative time and hypothermia in acute burn surgery

    2017, Burns
    Citation Excerpt :

    Patients excluded from analysis were those admitted for reconstructive surgery, conditions other than burn injury, or if only part of their acute burn management was undertaken at the centre, and those for whom documentation was either absent or incomplete. Demographic data collected included sex, age, total body surface area, presence of inhalation injury, substance abuse (alcohol, cigarette, and other recreational drug use), comorbidities [18], operative details, and post-operative complications. Operative length was defined a priori as the total duration in the operating room (from time of entry to exit).

  • Evaluation of three co-morbidity measures to predict mortality in patients undergoing total joint arthroplasty

    2016, Osteoarthritis and Cartilage
    Citation Excerpt :

    However, in total joint arthroplasty patients there is no consensus as to which co-morbidity measure is the best in regards to mortality prediction. In a sample of almost 250,000 US Medicare surgical patients between 1985 and 1989, Melfi et al. reported that the unweighted co-morbidity score at the time of surgery was the best performer (c = 0.82) in predicting short term death in comparison to the Charlson weighted score (c = 0.65), which performed very poorly in comparison to the other measures (c = 0.65–0.82)48. More recently, when Mnatzaganian et al. evaluated some of these measures in a smaller cohort of 819 Australian men, they reported that the Charlson weighted score (c = 0.88) performed better in the 1 year mortality prediction than the Elixhauser conditions (c = 0.86), but similarly to our results the differences were small between the co-morbidity measures (c = 0.76–0.88)46.

  • Effect of Minor Liver Function Test Abnormalities and Values Within the Normal Range on Survival in Heart Failure

    2015, American Journal of Cardiology
    Citation Excerpt :

    Only the first complete set of LFTs was used for the present analysis. Co-morbid conditions were defined using the ICD-9 diagnostic coding for the Charlson co-morbidity criteria.12 Additional co-morbidities were defined as follows: hypertension (ICD-9 401 to 405) and ischemic heart disease (ICD-9 410 to 414).

View all citing articles on Scopus
View full text