Understanding the basis of treatment choices for varicose veins: a model for decision making with the repertory grid technique.

OBJECTIVES

To use the repertory grid technique as a method for identifying and rating the criteria that clinicians use to make a choice between the different treatment options for patients with a common condition such as varicose veins.

DESIGN

The "expert panel" consensus method for rating the appropriateness of clinical procedures was modified with an existing psychometric method, the repertory grid technique. To identify the criteria used to decide about treatment, the panel members compared and contrasted a range of nine "treatment prototypes". They were then required to rate each criterion for its relevance to each treatment prototype.

SETTING

The panel was selected from different geographical locations in the South Western Regional Health Authority.

SUBJECTS

The expert panel was composed of six vascular surgeons, three from teaching and three from non-teaching hospitals; two general practitioners who were also clinical assistants in vascular surgery; and one honorary senior lecturer in general practice.

MAIN MEASURES

Decision making criteria were categorised according to their content. Their frequency of replication was noted-that is, how many clinicians used the same criterion. Computer analysis of the rating scores for the nine panel members identified the relative importance of each treatment criterion for each treatment option.

RESULTS

161 criteria for the treatment of varicose veins were elicited from the nine participants. These criteria were wide ranging, from clinical indications (48% of those used), to social (32%), and organisational factors (20%). Clinical indications were more likely to be used when deciding about surgery as a high priority, whereas social and organisational criteria were more likely to be applied in decisions about surgery as a low priority, day case surgery, and cosmetic surgery.

CONCLUSIONS

The repertory grid technique proved to be effective in modelling decision making for a condition such as varicose veins: its use enabled both the identification of the wide range of criteria underlying the decision to treat and the exploration of the relative importance of these criteria in relation to several treatment options. Its potential as a method for reducing variation in clinical decision making and thus improving distribution of high quality care lies in its ability to pinpoint dilemmas of decision making rather than as the basis for drawing up guidelines to regulate decision making practice.