More randomised controlled trials on frequent attendance

We appreciate the comments that Smits¹ et al made on our article.² From this reading we deduce that they would agree with its main findings: the way in which frequent attendance is defined has an impact on the factors associated with it and their discriminative power, and the use of the top decile cut-off seems to be more recommended than the top quartile.

They introduce an interesting idea that is clearly relevant to this discussion. We should focus on those frequent attenders that persist over time, as there is a significant proportion of those who left their status after 2 or 3 years. No doubt this is a reasonable and pragmatic approach. Unfortunately, no randomised control trials that show there is some kind of GP intervention to reduce these visits of persistent frequent attenders have been published. However, a randomised control trial of a successful GP intervention with frequent attenders in primary care was published in the BJGP.³ Although further randomised controlled trials are necessary, this comprehensive GP intervention with frequent attenders resulted in a significant and relevant reduction in their consultations. In fact total visits of frequent attenders of the intervention group were reduced by nearly 40%, while in the control group there was virtually no change. Moreover, this effect was found for frequent attenders of only 1 year (‘short frequent attenders’).

Therefore, given the evidence available so far, we cannot conclude persistent frequent attending is of more importance and clinical usefulness than short-term frequent attending, but rather the opposite.

We believe it would be more interesting to concentrate scientific efforts to determine whether that or other interventions are effective in reducing frequent attendance and if it is achieved by cost-effectiveness and cost-utility.