Suggestions regarding adaptation of the STRICTA guidelines for reporting acupuncture practice and research.

Dear Editor, We are writing to share some thoughts regarding our use of the STandards for Reporting Interventions in Clinical Trials of Acupuncture (STRICTA) guidelines in two recent reviews of acupuncture for plantar heel pain. 3 STRICTA was, of course, designed as a guide for the reporting of acupuncture and similar interventions. We needed to evaluate reporting from particularly diverse sources for review purposes, so we made a number of adaptations to the criteria (detailed below). We felt that some of the original items did not capture enough of the data that we were interested in, so we subdivided them to create some additional items. STRICTA item 1c addresses the ‘extent to which treatment was varied’. We found this ambiguous; as well as varying between individual patients or practitioners, treatments may vary between successive sessions. To make this distinction, we subdivided this item into 1ci (individualisation) and 1cii (follow-up variation). Item 4b (‘setting and context’) includes instructions to practitioners and information for patients. We felt that additional aspects of the setting were important, so we expanded it to create 4bi, ‘instructions to patients’ (such as being told to walk about during needling), and added 4bii (clinic context) and 4biii (country). Item 5 (‘practitioner background’) focuses on acupuncture qualifications, experience, and affiliations. We were also interested to know the other professional backgrounds of the acupuncture providers, so we subdivided this into 5i (practitioner background in acupuncture) and 5ii (primary profession). In addition, we wanted to quantify the data, and therefore developed a scoring system similar to that used by Prady et al. We allocated one point for each item (or sub-item). In some cases a fractional score was used to indicate partial fulfilment. Because some items did not apply to all sources (eg, 6a and 6b for non-comparative studies), we made the scoring proportionate and presented the total achieved as a percentage of the available points. In most cases scoring was straightforward but issues did arise with some, as follows. Item 1a (‘style of acupuncture’) was rarely stated clearly. It was sometimes apparent from the context (eg, journal of publication, or clinic address); however, we only scored the point when it was made explicit. Items 2d (‘responses sought’) and 4a (‘other interventions’) were difficult to score. If either item was mentioned there was no problem but, if not, this could either have meant that no response was required (or no other interventions were used) or that it was poorly reported—we had no way of knowing which. Therefore we decided to score 0 and to add a ‘? n/a’ note. Another related aspect is the potential for responses indicating the patient’s tolerance or comfort level. A focus on patient-centred care was interesting to us but is not, we suspect, what the STRICTA team originally had in mind. Item 2g (‘needle type’) requires four items for a full record —material, diameter, length and make. We scored 1⁄4 point for each. Table 1 shows the scores achieved by each of the STRICTA items across the 25 studies included in our linked review. Table 2 shows distribution of scores in relation to year and country of origin of source. Details of the sources and analysis are available in the paper and its online supplementary files. Our studies were not designed to evaluate STRICTA so we offer impressions rather than definitive conclusions. However, we identified