Validity, reliability, and responsiveness of a self-reported foot and ankle score (SEFAS).

Sir—We read with concern the article by Coster, Karlsson et al. (Acta Orthopaedica 2012; 83 (2): 197-203). This article presented a self-administered questionnaire, originally devised by individuals at the New Zealand National Joint Registry (Hosman et al. 2007), intended to be used for assessing outcomes following ankle replacement surgery.

From the brief description of the methods, the development of this questionnaire appears to have involved no patients with foot or ankle problems. Indeed the only patients who were involved, were those chiefly with osteoarthritis of the hip, undergoing hip replacement! This is because the instrument comprises a majority of items copied directly from the Oxford hip score (Dawson et al. 1996) with the word ‘hip’ simply changed to ‘ankle’. This is a curious method of devising a condition-specific measure that is very different from the methods most usually adopted by people well versed in psychometric methods, and which are the methods now recommended by the FDA (US Department of Health and Human Services Food and DrugAsministration 2009). These include the recommendation that instrument item generation is incomplete without involving (i.e. interviews or focus groups) patients with the condition of interest. Without this, it is difficult to claim that a patient-reported measure represents the patient’s perspective and its content validity is undermined.

The paper mentioned a few other self-administered measures (although not ‘patient-reported’ in the strict sense of the term i.e. devised, from the outset, with patients) that have been used for foot or ankle treatment evaluation, none of which have been developed with patients undergoing foot surgery and validated in the surgical setting. We were also particularly perplexed by the lack of any mention of the Manchester-Oxford Foot Questionnaire (MOXFQ) (Dawson et al. 2006, 2007) a Patient-Reported Outcome Measure that have been subjected to rigorous testing (Dawson et al. 2011, 2012). This measure is increasingly being used by European specialists in foot surgery (Marinozzi et al. 2009, Harrison et al. 2007, Maher and Kilmartin 2010a) and is gradually being translated into other languages.

Readers of the journal may wish to use the measure, or at least be aware that such an instrument exists. For a copy of the MOXFQ questionnaire, and permission to use it please email: healthoutcomes@isis.ox.ac.uk

Sir—Thank you for your interest in our recent paper on the SEFAS score (Cöster et al 2012).

The New Zealand National Joint Register (NJR) has used a self-reported questionnaire for assessing pain and function following ankle replacement surgery since the year 2000 although the questionnaire was not published until 2007 (Hosman et al. 2007). NJR has never claimed that it is an Oxford questionnaire, but has always stated in annual reports that it was developed based on the Oxford hip questionnaire, but not validated. This is also stated in the methodology section of the paper by Hosman et al. (2007). In 8 of the 12 questions the word “the hip” was changed to “the ankle” whereas the remaining 4 questions were new foot and /or ankle-related questions.

When we translated and culturally adapted the questionnaire used by NJR, and now also the Swedish Ankle Register (www.swedankle.se), we did not initially involve patients with foot or ankle problems. However, the validation process is currently on-going and patients with various foot and ankle related problems are now involved. To improve the content and face validity we have used focus groups with patients and experts on the field.

KOOS is a well-known knee-specific score translated to several languages (www.koos.nu). When questionnaires were developed for ankle and hip, they also changed the word “knee” to “foot” (FAOS) “or “hip” (HOOS) and after that they involved patients with conditions of interest. The SEFAS is developed in the same way.

While preparing our manuscript and during the publication process we were aware of the Manchester-Oxford-Foot Questionnaire (MOXFQ) in the context for hallux valgus (Dawson et al. 2006, 2007). However, at that time we found no publication on MOXFQ in the context for ankle.