Calendar | For Your Patients | PHA Main Site | Contact Us | About Us | Not a registered user? Sign up here.

Resource Library

Determining Clinical Utility of the CAMPHOR Questionnaire

Cherylanne Glassner

S. Watson

Mardi Gomberg-Maitland


  Sign in to add a review

Leave a Comment

Conference: 2012 International PHA Conference and Scientific Sessions

Release Date: 06.22.2012

Presentation Type: Abstracts

Glassner C, Watson S, Gomberg­Maitland M.

University of Chicago, USA

BACKGROUND: In recent years, the Cambridge Pulmonary Hypertension Outcome Review (CAMPHOR) has become accepted as a reliable method of patient assessment. The University of Chicago Pulmonary Hypertension (PH) Program administers these questionnaires to its patients in order to measure quality of life and health status. The potential for the CAMPHOR to be used as a clinical endpoint in PH is promising. In order to serve as an endpoint, the CAMPHOR must provide an appropriate indication of current condition and disease progression.

METHODS: We performed a retrospective review of CAMPHOR data from October 1, 2008 to October 12, 2011, collecting scores by category and total, as well as the date of questionnaire administration. Other clinical measures of PH, clinical worsening (CW) events (hospitalization and death), and basic demographic information were also collected. Logistic regression and Cox regression models were used to correlate CAMPHOR score (baseline or net change) with log odds of CW and hazard rate (i.e. risk) of CW, respectively. An ROC curve was used to determine how well CAMPHOR could distinguish between patient functional class (FC) I/II and III/IV.

RESULTS: Analysis was performed on a subset of 43 patients who had CAMPHOR scores and clinician assessed FC at two time points in 2009. Change in score components failed to significantly correlate with log odds and hazard rate of CW. However, in models of the baseline scores, we found evidence of a statistically significant relationship between log odds of CW and activities score (p=0.020). Cox regression results similarly indicate that there are significant associations between the risk of CW and baseline symptoms (HR 0.914, p=0.046) and activities (HR 1.2, p=<0.001) scores. Upon evaluation of ROC, change in CAMPHOR score did not differentiate FC, whereas in follow up, symptoms score was most predictive of FC (Symptoms AUC 0.8; Activities AUC 0.76, Quality of Life AUC 0.74; Total AUC 0.78).

CONCLUSION: Use of CAMPHOR the clinical setting can be helpful for predicting probability and risk of hospitalization and death. The symptoms score can help the clinician assess patient FC.