Objectives The objective of this study was to develop and validate a decision algorithm that replicates expert opinion regarding the level of disease activity in patients with acromegaly. Methods A panel of key experts in endocrinology and outcomes research provided input as to what parameters are critical to assess disease activity. The top five parameters where then subsequently defined at three levels of severity. A final set of 243 unique health states were developed. A group of 21 expert endocrinologists from 5 European countries and Canada was recruited to participate in a discrete choice experiment. Each participant was presented with 10 common scenarios and 42 additional scenarios selected at random and asked to indicate whether the hypothetical patient was stable, having mild disease activity needing further evaluation, or having significant disease activity requiring clinical action. Concordance in decisions based on the 10 common scenarios was tested using Fleiss Kappa and a set of linear prediction models tested to predict outcome. Results The five parameters included: 1) IGF-I level (SDS score); 2) tumor status (change on MRI); 3) comorbidities (number and severity); 4) signs and symptoms (Patient Acromegaly Symptom Questionnaire score); and 5) health-related quality of life (scored on a disease specific measure). The Inter-rater reliability was adequate (Kappa = 0.52). Based on a CART analysis, the decision model of best fit was disjunctive with IGF-1 and Tumor status primarily predicting the health status. The remaining three parameters were instructive but not deterministic. Conclusions Biochemical and tumor status are the primary predictors of disease activity, with the patient’s perceived disease state playing a secondary role. This may highlight the need for a more patient-centered approach to acromegaly disease management. To enable clinical use of the model, a disease specific tool named ACRODAT (ACROmegaly Disease Activity Tool) is currently in further development.
Development of a Prediction Model of Disease Activity in Support of Clinical Practice - the Acrodat Experience
GHIGO, Ezio;
2015-01-01
Abstract
Objectives The objective of this study was to develop and validate a decision algorithm that replicates expert opinion regarding the level of disease activity in patients with acromegaly. Methods A panel of key experts in endocrinology and outcomes research provided input as to what parameters are critical to assess disease activity. The top five parameters where then subsequently defined at three levels of severity. A final set of 243 unique health states were developed. A group of 21 expert endocrinologists from 5 European countries and Canada was recruited to participate in a discrete choice experiment. Each participant was presented with 10 common scenarios and 42 additional scenarios selected at random and asked to indicate whether the hypothetical patient was stable, having mild disease activity needing further evaluation, or having significant disease activity requiring clinical action. Concordance in decisions based on the 10 common scenarios was tested using Fleiss Kappa and a set of linear prediction models tested to predict outcome. Results The five parameters included: 1) IGF-I level (SDS score); 2) tumor status (change on MRI); 3) comorbidities (number and severity); 4) signs and symptoms (Patient Acromegaly Symptom Questionnaire score); and 5) health-related quality of life (scored on a disease specific measure). The Inter-rater reliability was adequate (Kappa = 0.52). Based on a CART analysis, the decision model of best fit was disjunctive with IGF-1 and Tumor status primarily predicting the health status. The remaining three parameters were instructive but not deterministic. Conclusions Biochemical and tumor status are the primary predictors of disease activity, with the patient’s perceived disease state playing a secondary role. This may highlight the need for a more patient-centered approach to acromegaly disease management. To enable clinical use of the model, a disease specific tool named ACRODAT (ACROmegaly Disease Activity Tool) is currently in further development.
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
