The outcomes of mandibular third molar removal and non-removal: a study of patients' preferences using a multi-attribute method

The aim was to study patients' preferences about outcomes of mandibular third molar removal and nonremoval using multi-attribute utility (MAU) methodology. The study comprised three stages. Stage 1: Elicitation of domains, i.e. main areas of patients' lives which could be affected by third molar removal and non-removal. Stage 2a: Interdomain weighting was obtained by relative weighting of the domains elicited in Stage 1. Stage 2b: Intradomain weighting obtained by patients' designation of values for different health states of each domain. Stage 3: Rating of outcomes. The patients were asked to imagine experiencing a variety of outcomes of mandibular third molar removal and non-removal, described in 19 short vignettes. The numbers of patients interviewed for the three stages were 30, 78, and 55, respectively. Five domains were identified. The mean relative weightings were approximately equal for the domains ''Home and social life'' and ''General health and well-being'', followed in order of importance by ''Job and studies'' and ''Health and comfort of mouth, teeth and gums''. ''Your appearance'' received the lowest mean relative weighting. The vignette, which described the presence of a fluid-filled sac and suggested that this tooth must be removed, received the highest mean preference (least effect on patients' lives). The lowest mean preference (most effect on patients' lives) was generated by the vignette, which stated that the jaw was broken and that the teeth must be wired together for 6 weeks. We conclude that, from the patient's perspective, outcomes of non-removal were preferable to outcomes of mandibular third molar removal.