Evidence for current recommendations concerning the management of foot health for people with chronic long-term conditions: a systematic review

Search criteria for the systematic review were identified using the “PICO” statement. Search terms were defined by three experts in podiatry (CB, AR, AB) and the final list agreed through a consensus meeting. Electronic databases (Medline, Embase, Cinahl, Web of Science, SCOPUS and The Cochrane Library) were searched for English language publications and translations reporting recommendations for the management of foot health for people with chronic conditions between 2000 and 2016. The year 2000 was considered an appropriate point to commence the literature search, given the establishment of degree education in podiatry from 1993, and the subsequent development of postgraduate and post-doctoral qualifications and research skills within the profession.

Searches were performed using the following text word terms and MeSH headings: foot, chronic, chronic disease, long term care, neurodegenerative diseases, multiple sclerosis, arthritis, lung diseases, emphysema, diabetes mellitus, hypertension, cerebrovascular disorders, myocardial ischemia, heart failure, colonic diseases, human immunodeficiency virus, osteoporosis, fibromyalgia, neoplasms, mental disorders, depression, asthma, chronic obstructive pulmonary disease, thyroid diseases, hyperlipidaemia, psoriatic arthritis, gout, ankylosing spondylitis, colitis, scleroderma, practice guidelines, algorithms.

Hand searches of bibliographic references identified additional publications. Grey literature was included based on an initial search using the terms foot, feet and podiatry and rerun in conjunction with the terms guideline, recommendation and management to ensure that the search had captured all relevant sources. Publications identified through the grey literature and hand searches were not restricted by date. The systematic literature search was facilitated by a medical librarian at the Bodleian library, University of Oxford. The full search strategy can be seen in Additional file 1: Appendix A.

A publication was included if it met all the following criteria:

Three independent expert reviewers (CB, AB & KE) assessed all the included publications (July 2016) for quality using the AGREE II checklist [27]. The AGREE II checklist was developed to address issues of variability in the quality of practice guidelines and is used internationally as a means of assessing the standard of methodology used in the approach to guideline development. The checklist evaluates the quality of 23-items across six domains that relate to aspects of the guideline development through assessment of six domains including, scope and purpose, involvement of stakeholders, developmental rigour, presentation, applicability and editorial independence. An overall score of 7 reflects that the guideline strongly agrees with most of the AGREE II criteria, while an overall score of 1 shows a weak agreement. Whilst the consortium for the development of the AGREE II criteria set the domain scores, they did not set minimum domain scores for high or low quality guidelines. Instead they advise that the decision should be made by the user and guided by the context in which AGREE II is being used. Therefore, through consensus of all authors, we set threshold scores at high quality (scores 6–7), mid-level quality and partially meeting the AGREE II criteria (scores 4–5), low methodological quality (scores 1–3). The scores for all included papers can be seen in Additional file 2: Appendix B.

Data from the included publications was synthesised via template analysis, employing a thematic organisation and structure. Template analysis is a process by which textual data is organised and analysed according to themes, allowing unstructured textual material to be categorised into meaningful datasets. Indexing sections of text relevant to a theme also permits the development of codes which emerge from the data [28]. Predetermined themes related to foot care and / or podiatry services were established from synthesis of current literature by experts in the field of podiatry (CB, AB) [29, 30] (Table 1). Each theme was assigned a code which when noted within a publication was readily entered into the data extraction sheet. A theme could be present singularly or all codes could exist within the same publication.

Where stated, the level of evidence supporting the recommendations related to foot health management was extracted and interpreted according to the GRADE (Grading of Recommendations Assessment, Development and Evaluation) system [31]. The GRADE system classifies the quality of evidence as high, moderate, low, and very low (Table 2). For example, evidence based on randomised controlled trials is ranked as high quality evidence and expert opinion, case series is ranked as the lowest level of evidence [31]. The reasoning for this more rigorous assessment was that, in some cases, while the overall publication quality may have been low or average based on the AGREE II criteria [27], some evidence underpinning the publication may in fact be of high quality or vice-versa.

Of the 166 included publications, 74 (45%) were from the grey literature, 65 (39%) were from electronic databases and 27 (16%) were from hand searches.

Quality assessment of the 166 included publications was carried out using the AGREE II checklist (Fig. 6). We found that the methodological quality was spread across the range of AGREE II scores. Sixty-two publications fell into the category of high quality (scores 6–7); of these the majority were related to diabetes and the diabetic foot (15.7%), and published in the USA (4.8%). Comparatively, 51 publications were considered to be of mid-level quality and partially met the AGREE II criteria (scores 4–5). Forty-six publications were considered to be of low methodological quality (scores 1–3); of these, 31 (18.7%) came from the electronic literature, 14 (8.4%) from the hand searched literature and one (0.6%) from the grey literature.

Synthesis of the themes related to foot health management show that recommendations are made for all themes, but not in all included publications (Fig. 7). The most common recommendation is for preventive care or assessments (e.g. ABI tests), followed by clinical interventions such as foot orthoses, foot ulcer care and foot health education. Comparatively, there is a noticeably lower number of publications containing recommendations for core podiatry and podiatric surgery. Areas with the lowest numbers of recommendations involved themes such as antifungal and pain medication.

Examining patterns across the 166 publications would suggest that the majority of supporting evidence for foot health management is included within UK guidelines, recommendations or standards of care, followed by the US, Europe and Australia. This is not surprising as our search was limited to English only publications, however of note, within these countries podiatry is a recognised allied health discipline [32, 33]. Podiatry within the UK and Australia has been building a research capacity over the past decade [34] and it is possible that the trend in year on year increases in publications containing supporting evidence for foot health management related to other chronic conditions is linked to this.

Guidelines, standards of care and recommendations play an important role in synthesising evidence for health care and health policy information however according to the AGREE II consortium [27] the potential benefits are only as good as the quality of the guidelines themselves. We found 62 publications that fell into the category of high quality, 51 were considered to be of mid-level quality and 46 were considered low methodological quality; 7 publications were not given an AGREE II score as the content did not address a sufficient number of the domains being assessed. We found that the methods for guideline development markedly differed according to the stakeholders involved. The high-quality guidelines were largely from government agencies and national professional bodies such as the National Institute for Health and Care Excellence (NICE), the Scottish Intercollegiate Guidelines Network (SIGN), and the National Institutes of Health (NIH), whereas the low-quality guidelines tended to be developed by smaller local or regional health providers and often were single authored publications. This is consistent with other studies that have investigated guidelines specifically for management of foot and ankle problems in rheumatoid arthritis [21] and guidelines for foot screening in diabetes [22].

From our review, we also established that it was rare that a podiatrist or foot care specialist was included as a stakeholder in the higher scoring publications. This is of concern as the recommendations contained within guidelines are defined by the scope of the stakeholders. Stakeholders advise guideline developers in searching for, selecting, critiquing and combining data [24]. Without engagement with podiatrists or foot care specialists it is possible that evidence included in these publications may have incomplete, ill-informed or biased evaluations which in turn can lead to inappropriate recommendations [24]. If podiatrists and foot care specialists are to become involved either as leaders of guideline development or as members of the stakeholder group it is imperative that the methodological approach follows the robust principles, such as those outlined by the AGREE II consortium [27]. The validity of the guideline recommendations should be open to being judged in a rigorous and reproducible way.

The strength of recommendations within guidelines, recommendations or standards of care is informed by the quality of the research evidence on which they are based [24, 25]. Following an assessment of the overall quality of the included publications, a more in-depth analysis was carried out that specifically considered the levels of evidence that underpinned the strength of their recommendations concerning foot care, foot health management and/or podiatry. Disappointingly, we found a high number of publications providing a recommendation in the context of a narrative but without an indication of the strength or quality of the underlying evidence for foot care / podiatry. We also found a high number of publications that indicated that the recommendations provided are based on expert opinion or best practice.

Where a level of evidence was given, this tended to be in the higher scoring publications for methodological quality and related to GRADE (Grading of Recommendations Assessment, Development and Evaluation) system [31] as strong-moderate quality. Therefore, some evidence for foot care, foot health management and /or podiatry exists that is underpinned by a limited number of systematic reviews and randomised controlled trials such that those recommendations could apply to most patients in most circumstances [31]. Evidence for preventive care or foot assessments (vascular, neurological, musculoskeletal, dermatological) features in the most publications, followed by foot health education. There is moderate evidence for preventive care advancements or foot assessments (diagnostic ultrasound, ABPIs, advanced training or skills) and clinical interventions (provision of foot orthoses, insoles, nail surgery, injection therapies, physical therapies). Evidence for core podiatry (nail, corn and callus care) is rare and no publications included evidence for podiatric surgery.

According to GRADE, for those themes with emerging good quality evidence, further research, if performed, is likely to have an important impact on confidence in the estimate of the effect and may change the estimate [31]. As the body of work continues to grow, it is thus likely to have an important impact on the evidence for ‘podiatry’ however it is clear that considerable work is still required for key areas that will do well to have a targeted and strategic approach.

The findings of this systematic review need to be considered in the context of several limitations. First the scrutiny for evidence for ‘podiatry’ was determined a priori by the authors as being most accessible from guidelines as they usually incorporate research evidence into their clinical recommendations. In this instance it would have been inappropriate to select publications by single interventions due to the wide range of interventions related to management of foot health / podiatry in long term conditions and the multi-professional nature involved in the interventions. As only English-language publications were included in the review this may have resulted in additional relevant, high-quality evidence from non-English publications being omitted.

The hand searching process also procured a number of publications not identified from the electronic database search. This may be due to the grey literature and hand searched references not being restricted by date as had the electronic searches. This highlights the differences in approach required in searching the electronic and grey literature, and the challenges inherent in designing an inclusive search strategy. In view of this, the grey literature search in particular may not have captured all relevant articles; however, every attempt was made to minimise the potential of ‘missed’ literature by developing the search strategy with input from experts in the field of podiatry and foot health care, and guided by a specialist in designing literature search strategies.

The reader needs also be aware that while the GRADE system is widely adopted by medical researchers, it does have its limitations. Often manuscripts retrieved from the grey literature, such as observational studies, narratives and clinical expert opinion, tend to be classified as being low methodological quality; however in the absence of any other more suitable system it was decided to use GRADE as “the extensive user testing done on the GRADE system could be one factor that has contributed to its popularity among guideline developers” [35].

Lastly, quality of evidence was determined by methodological assessment of the publications in accordance with the AGREE II consortium and interpretation of cited evidence [27] through the Cochrane’s recommended approach for grading the quality of evidence (GRADE) [31]. We found difficulty in conducting these assessments on all publications due to the high proportion of narrative reviews and publications based on consensus /expert opinion that were included. Often, there was circularity of evidence and recommendations in such publications tended to have very little new evidence included. We attempted to reduce the biases that may have been introduced due to the subjective nature in these assessments by ensuring all reviewers met three times during the assessment period to discuss controversies and arrive at consensus for the final scores. We also ensured consistency in scoring by allocating publications for assessment and scoring between the three reviewers (KE, CB, AB). CB and AB independently scored 83 publications and KE independently scored 40 of each of AB and CBs publications.