NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.

School of Health and Related Research (ScHARR), University of Sheffield. Clinical Guidelines for Type 2 Diabetes: Prevention and Management of Foot Problems [Internet]. Sheffield (UK): University of Sheffield; 2003. (NICE Clinical Guidelines, No. 10.)

  • This publication is provided for historical reference only and the information may be out of date.

This publication is provided for historical reference only and the information may be out of date.

Cover of Clinical Guidelines for Type 2 Diabetes

Clinical Guidelines for Type 2 Diabetes: Prevention and Management of Foot Problems [Internet].

Show details

4Foot care: general management aproach

General management approach


Effective care involves a partnership between patients and professionals and all decision making should be shared. (D)

The role that any informal carers of the person with diabetes has in providing care and receiving information to allow them to fulfil this role should be discussed with the person with diabetes, and any decisions about this should be that of the person with diabetes. (D)

Arrange recall and annual review as part of ongoing care. (A)

Health care professionals and other personnel involved in the assessment of diabetic feet should receive adequate training. (D)

As part of annual review, trained personnel should examine patients’ feet to detect risk factors for ulceration. (A)

To improve knowledge, encourage beneficial self-care and minimise inadvertent self-harm, healthcare professionals should discuss and agree with patients a management plan that includes appropriate foot care education.* (B)

Extra vigilance should be used for people who are older (over 70 years of age), have had diabetes for a long time, have poor vision, have poor footwear, smoke, are socially deprived or live alone. (C)

Health care professionals may need to discuss, agree and make special arrangements for people who are housebound or living in care or nursing homes to ensure equality of access to foot care assessments and treatments. (D)

Evidence statements

Where people with diabetes and general practices are willing participants and when an organised system of recall and review is in place which specifies components of foot care, shared care arrangements between hospital and general practice provide levels of surveillance as good as hospital diabetes clinic attendance alone. (Ib)

An agreed plan of management between people with diabetes and health care professionals is important to achieve adequate levels of care. (Ib)

Prompt and appropriate information exchange between primary and secondary care, and with patients, improves communication and quality of care. (IV)


Shared care has been defined as “the joint participation of hospital consultants and general practitioners in the planned delivery of care for patients with a chronic condition, informed by enhanced information exchange over and above routine discharge and referral notices” (Hickman et al, 1994). Studies of the consequences of organising care have used several models: e.g. devolving care of certain patients to primary care until a problem occurs, or retaining annual visits to a hospital clinic with interim care in general practice. Details of the workings of a number of shared care schemes in the UK have been reviewed (Greenhalgh, 1994).


Randomised trials comparing the effectiveness of general practice or hospital care are the subject of a Cochrane review (Griffin and Kinmonth, 1997). This review identified five trials from Australia or the British Isles evaluating Type 2 diabetes or both Type 2 and Type 1 diabetes patients (Porter, 1982; Hayes et al, 1984; Hoskins et al, 1993; Hurwitz et al, 1993; DICE, 1994).

Only one new randomised controlled study was identified as published in the past four years. Donohoe et al (2000) examined the impact of an integrated care model for use in general practices and compared against current care plus an unrelated educational intervention in five control practices. The integrated care model comprised an organisational framework centred on the primary care-based annual review and examination of the diabetic foot. It incorporated practice visits, education of the whole primary care team on the recognition, examination and management of the diabetic foot, referral criteria and responsibilities of professionals. Foot care leaflets outlining the patients’ roles and responsibilities in the integrated care arrangements were also provided for dissemination to patients by the practices and the Semmes Weinstein monofilament was introduced to practice staff. Randomisation was at practice level and the five practices in the intervention group received regular reinforcement visits over a six-month period. In addition a further separate education programme was provided for chiropodists in the intervention group. Outcomes were changes over a six-month period in patient attitude to, and knowledge of, foot care. Whilst knowledge and attitude improved in both groups, significant differences between the group only occurred in the attitude scores (change 3.18, 95% confidence interval 1.29, 5.07) in favour of the intervention group. Improvements in the intervention group were also seen in the health professionals’ knowledge (p<0.001), producing a change between groups of 13.46 (95% CI 8.30, 18.62), and in appropriate referrals to specialised foot clinics (p=0.05).

In two British studies (Hurwitz et al, 1993; DICE, 1994) which featured regular prompting of patients and/or doctors, and reviews with specified components, overall mortality and glycaemic control were at least as good in general practice as with hospital outpatient care while losses to follow-up were significantly lower. One of these studies found that primary care patients were more likely to be referred to a chiropodist than hospital attenders (DICE, 1994) whilst the other found no significant difference (Hurwitz et al, 1993). The consequent benefit, in foot-health terms, of the increased review and referral in these studies is not known.

An Australian study (Hoskins et al, 1993) demonstrated that similar levels of follow-up could be achieved either with properly structured shared care or hospital care, but loss to follow-up in both groups was considerable, limiting the usefulness of the endpoint comparisons.

The early British studies (Porter, 1982; Hayes et al, 1984) differ from the others in that no automated system of recall was provided to prompt either patients or GPs to initiate a consultation. The first study (Porter, 1982) was published as a letter and although outcomes appear equivalent, insufficient detail is presented for an adequate assessment. In the second study (Hayes et al, 1984) nearly all local GP practices agreed to participate and there was a relatively long period of follow-up (5 years) compared to other studies. This study suggests poorer follow-up and overall mortality for patients receiving solely general practice rather than hospital care.

There are a number of potential problems in trying to interpret and generalise the findings of these trials. The two recent British studies enrolled self-selecting local practices. Patients randomised were self-selecting, stabilised, had no major medical complications, and were already attending hospital. The follow-up in these two trials was two years. The relative success of shared care in the later trials may be due to the implementation of structured recall and review but also due to the use of (enthusiastic) volunteer practices and patients, to relatively short follow-up periods and the trial context. A district-wide audit of diabetes care of patients discharged into primary care (where most practices had registers and recall systems) suggested an erratic and generally poor standard of supervision (Dunn et al, 1996). Other studies have questioned the care received by patients in the community (Day et al, 1987; Wilkes et al, 1980).

The unit of randomisation and analysis in each of the trials is the patient. However, some GPs will have provided care for more than one patient randomised to the same intervention introducing a form of clustering, potentially reducing degrees of freedom and leading to spurious accuracy. Additionally, GPs may have seen patients randomised to both shared care and hospital care introducing contamination. Given these concerns statistical pooling of data, at the level of the patient, by meta-analysis is not appropriate.

None of the trials reported components of foot surveillance or foot complications in any detail. The optimal period for routine surveillance of emergent foot problems is unclear but the trials featuring structured recall and review used periods of 3 to 6 months. Neither do the trials provide guidance on the relative merits of diabetes mini-clinics in general practice as opposed to care provided in routine GP consultations. One trial featured 2 mini-clinics and 1 routine care practice in the shared care group but did not show stratified findings (DICE, 1994).

A review (Hampson et al, 1996) suggested that at discharge to general practice the exchange of information between specialists and both GPs and patients may be unhelpfully delayed and inadequate in content. Structured notes are advocated, containing details of diagnosis, examinations and procedures, history and progress, admission and discharge treatments, a management plan and problem list. Similarly specialists require, when handling GP referrals, information including the reason for referral, past medical history, summary of the complaint and symptoms, GPs clinical findings and investigations, diagnosis, drug (and other treatment) details, allergies and patient expectations. Ultimately centralised computer databases of patient records could provide a solution, but the importance of structured communication appears an important facet of collaborative shared care.

There are a number of ways in which the care of people with diabetes overall and for their foot problems can be managed in a general practice. Smith et al (1998), in a retrospective study, examined the impact upon care of the use of an electronic management system in a sub-speciality diabetes clinic in the USA compared with providers in the same clinic using paper medical records only. The electronic system is designed to allow entry of clinical information in real time and automatic entry of laboratory data. It also provides predetermined option responses and a paper report at the end of the clinical consultation. Eighty two patients were randomly selected from a billing database of patients with diabetes, either Type, 39 of whom were cared for by providers using the electronic management system and 43 had paper records only. Each medical record was reviewed comprehensively by a trained nurse for data parameters documented in the previous 12 months. Baseline characteristics did not differ between the two groups of patients, but the number of foot examinations were significantly increased (p<0.001) in those patients whose providers used the electronic system, along with other performance indictors in the management of diabetes.

The Diabetes NSF and Diabetes Information Strategy are recent policy initiatives that may have a significant impact upon care for people with diabetes in terms of information, record and registers.


No evidence was found to specify the precise components, or method of transmission, of information, but the guideline development group felt it was good practice when organising services to give careful consideration to communication between primary care and specialist services.

A multi-disciplinary group was set up, under the aegis of the British Diabetic Association, to report on the implementation of the St Vincent Declaration guidelines with respect to diabetic feet (Edmonds et al, 1996). Among its recommendations was specific consideration of the role of coordinated multi-disciplinary teams in providing care. This detailed the respective roles of the chiropodist, diabetologist, general practitioner, nurse, orthotist, radiologist and surgeon arranged variously in hospital, community care and vascular service teams. There is no formal comparative evidence to indicate if any optimal arrangement of health care professionals in the care of diabetic feet exists. However, the guideline development group felt it is good practice to ensure that a multi-disciplinary team of professionals are available and co-ordinated to promptly provide the full range of appropriate services to patients.

Patient education


Structured patient education should be made available to all people with diabetes at the time of initial diagnosis, and then as required on an ongoing basis, based on a formal, regular assessment of need. [NICE 2003]

Offer patient education on an ongoing basis.* (A)

Use different patient education approaches until optimal methods appear to be identified in terms of desired outcomes. (B)

Evidence statements

Educational interventions can improve foot care knowledge and behaviour in the short term (up to 18 months). (Ib)

There is insufficient evidence currently available to recommend a specific type of education or provide guidance on the setting for, or frequency of, sessions. (NICE, Technology Appraisal 60)


Educational strategies seek to address the perception that whilst relatively simple actions by the patient and physician may minimise foot complications, often these simple procedures are not systematically applied. This may happen because patients are unaware, are embarrassed or do not understand the importance of basic care. It may be because specialist services (in foot care: podiatry, orthotics or orthopaedics) are unavailable or have long waiting lists. Finally, the health care system may not be organised in such a way that regular examination by, and education from, health care professionals does occur (Cohen, 1983). Recent studies are a response to evidence from surveys which suggest that people with diabetes may not have sufficient skills or knowledge to properly manage the disease (Thomson et al, 1992; Masson et al, 1989; Collier, 1971). The common elements of patient education are foot hygiene, awareness of fungal infections, and actions required for cutaneous injuries.


Educational interventions varied in their form, setting, length of follow-up (range 6 months to 7 years) and size of study population (between 50 and 530 randomised). Studies consistently reported relative improvements in knowledge about foot care and behaviour in intervention groups, although the importance of these changes is hard to assess. No other consistent patterns are present in study methods or findings, and it is necessary to interpret the findings of studies individually.

A systematic review by Valk et al (2002) assessed the evidence from randomised controlled trials that used educational programmes for the prevention of foot ulcers in people with diabetes. Whilst the literature was searched up to March 2001, of the eight included studies (Barth 1991; Bloomgarden 1987; Kruger 1992; Rommemaa 1997 & Hamalainen 1998; Litzelman 1993; Malone 1989; Mazzuca 1986; Rettig 1986), all but two of these (Rommemaa 1997 & Hamalainen 1998, Mazzuca 1986) had been included in the first foot care guideline evidence. The publication dates of the other two fell within the time scope also of the first guideline. Following their systematic appraisal of the existing evidence, Valk et al concluded that the methodological quality of the evidence was poor and the results conflicting. The evidence suggests that patient education may have positive, but short lived effects on patient foot care knowledge and behaviour, and it may reduce foot ulceration and amputation particularly in high risk patients.

To determine if foot care education provided by a podiatrist is effective in the longer term, Hamalainen et al (1998) randomised 530 Finnish patients aged 10–80 years with diabetes and no obvious current need for podiatry into those (n=267) referred to a podiatrist for a 45 minute individual counselling session plus treatment as many times as judged appropriate over a one year period and 263 who received written instructions on foot care. Knowledge scores increased over time (measured at one and seven years of follow-up) in both groups and both sexes but differences between the intervention and control group were only significant at one year (p=0.025). For self care, scores improved significantly in time in both groups, although in men there were no differences between the intervention and control groups. However for women, scores were higher than those for the men at all time periods, they increased significantly more over time in the podiatrist group (p=0.011) and they were significantly higher in the podiatrist group compared with the control group at both one year (p<0.001) and at seven years (p<0.05). The biggest changes occurred over the first year of follow-up. Visits to a podiatrist were similar between intervention and control group at baseline but proportionately more patients in the intervention group had visited a podiatrist at the seven year follow-up (intervention 82.3% vs 49.7%) (p<0.001). In the six years preceding the seven-year follow-up, both men and women in the intervention group had visited a podiatrist more often (mean number of visits: men 5.2±SD 9.1, women 9.0±9.5) than those in the control group (men 2.6± 7.2, women 3.7± 5.8) (p<0.001), although the number of visits tailed off and was not significantly different between the two groups in the last two years (p=0.06). There were no differences in number of physician inspections of the feet, prevalence of foot or toe abnormalities or problems. The benefits of a counselling session plus one year of podiatrist support can be maintained for several years afterwards without further intervention.

Campbell and colleagues (1996) looked at a number of parameters associated with diabetes management and control over a 12 month period, including podiatrist consultations, in Type 2 patients randomised to provision of one of four education programmes. At early follow-ups (three and six months), the numbers consulting a podiatrist differed significantly between the four programmes: minimal; group education, individual education; and behavioural, being higher in the last group. However at 12 months, there were no statistical differences in numbers consulting between the four groups.

Pieber et al (1995) evaluated an education programme (general diabetes care and foot care) in general practice in 107 patients. Intervention patients (in groups of 4–8) received 4 weekly sessions (1½–2 hours each) covering a range of aspects of diabetes care. At 6 months, callus formation and other measures of skin condition were significantly improved in the education group.

Litzelman et al (1993) evaluated a foot care education programme conducted (in groups of 1–4) by nurse clinicians in a general practice setting with 396 patients. Intervention patients agreed to personalised behavioural contracts, reinforced by telephone and postcard reminders at 1 and 3 months. This study included an organisational intervention for health care providers who were given practice guidelines, informational flow sheets and prompts to perform visual examination and provide education. At 1 year, there was a significant reduction in serious lesions for patients receiving the educational intervention. There were 4 amputations in the control group and 1 in the intervention group (p=0.2). Considerably improved rates of visual examination and provision of education during clinician contact were reported in the intervention group. The agreement of individual management plans between health professionals and patients is a central tenet of consensus statements on care (Edmonds et al, 1996). The study by Litzelman and colleagues provides some evidence to support this view.

Kruger et al (1992) evaluated a participatory foot care education programme with 50 patients beginning hospital diabetes care. Both groups received a normal education package (a videotape and supplementary explanation from an instructor, instructions to examine feet and use a checklist daily). The intervention group additionally received participatory hands-on teaching and learning sessions (including actual foot washing, inspection, care of corns and calluses, cutting toe-nails, evaluating problems and suitable footwear) and an education kit with materials (buff pads and an inspection mirror). At 6 months no changes in knowledge or behaviour were observed.

Barth et al (1991) evaluated an intensive foot care education intervention, recruiting selectively 70 patients by radio and newspaper campaigns as well as from general practice and diabetes centres. Patients in both groups received a conventional education programme including 1 hour on foot care. The intervention group received an additional 9 hours of education over a 4-week period. These sessions used a motivational technique and were taken by a podiatrist and a psychologist. The intensive care group showed significantly greater compliance with advice to consult a podiatrist (p=0.008) and a greater reduction in foot problems (p<0.006) at 1 month but neither of these effects were significant at 6 months.

One study (Malone et al, 1989) enrolled only patients with infection, ulceration or prior amputation and this is discussed in a following section (see: Education for patients with foot ulcers).

Bloomgarden et al (1987) evaluated a diabetes clinic programme consisting of 9 education sessions – covering general areas of diabetes care as well as foot care, and using various media: films, a card game, slides and role-playing – in 202 patients. No statistically significant differences in the occurrence of foot lesions were found between groups at 18 months. Of patients enrolled to the education intervention, 56% attended 7 or more sessions. These ‘graduates’ of the programme showed better knowledge and behaviour scores than those who didn’t graduate. Approximately 90% of patients in this study were black or Hispanic.

Rettig et al (1986) evaluated a diabetes home education programme (covering general areas of diabetes care as well as foot care) in 471 patients. Intervention patients were visited and assessed by a nurse, who tailored the education programme during the home visits to suit the individual needs of the patient. The findings at 6 months post enrolment, show no difference between groups in foot appearance scores (based upon a checklist of 16 abnormal conditions observed during foot inspection) or hospitalisation.

NICE Technology Appraisal 60 (2003)

NICE issued guidance about patient education models in diabetes in 2003. Although this was concerned with models for all of diabetes patient education, rather than foot care specifically, the findings and guidance are of relevance. The NICE guidance was based on eight trials (six randomised controlled trials [RCTs] and two controlled clinical trials [CCTs]) concerned with general self-management education for Type 2 diabetes. The outcomes considered were generally HbA1c, with body mass index (BMI) or weight also being reported in some studies. The guidance also reported trials considered with focused self-management, but foot care aspects of diabetes did not appear to have been addressed in the considered trials.

The following key messages are from the NICE technology appraisal.

“There is insufficient evidence currently available to recommend a specific type of education or provide guidance on the setting for, or frequency of, sessions.

However, to achieve maximum effectiveness some principles of good practice are clear:

educational interventions should reflect established principles of adult learning

education should be provided by an appropriately trained multidisciplinary team to groups of people with diabetes, unless group work is considered unsuitable for an individual

sessions should be accessible to the broadest range of people, taking into account culture, ethnicity, disability and geographical issues, and could be held either in the community or at a local diabetes centre

educational programmes should use a variety of techniques to promote active learning (engaging individuals in the process of learning and relating the content of programmes to personal experience), adapted wherever possible to meet the different needs, personal choices and learning styles of people with diabetes, and should be integrated into routine diabetes care over the longer term.

Multidisciplinary teams providing education should include, as a minimum, a diabetes specialist nurse (or a practice nurse with experience in diabetes) who has knowledge of the principles of patient education and a dietitian. Although not formally assessed in this appraisal, input from other disciplines, such as podiatry, has potential value. The composition of the team and the way that members interact may vary between programmes, but team functioning should be tailored to the needs of different groups of people with diabetes.”


The available studies provide inconsistent messages about the value of education and interpretation is weakened by small studies, different endpoints used by different investigators, and lack of a common approach. Only two of the larger studies reported serious lesions as endpoints. One indicated that where patients agree behavioural contracts and receive periodic reminders then foot ulcers could be prevented over 1 year (Litzelman et al 1993). Another study, spacing nine educational sessions over 18 months and using a range of teaching media, found no significant change, (Bloomgarden et al, 1987). A further factor complicating interpretation is that patients enrolled in studies may be at certain phases in disease processes making their clinical values atypical. Consequently both treatment and control groups may show considerable improvement over the period of follow-up (regression to the mean) making analyses of findings more complex.

Intensive, prompted education requiring action from both patients and clinicians may reduce foot complications over relatively short-term periods of time although the evidence is inconclusive concerning the best method. Available trials of education feature inadequate follow-up to assess the potential for primary care prevention of foot complications. The value of education interventions in the longer term is unknown and it is likely that important messages and habits will need reinforcing periodically in patients and health professionals. Better foot care awareness in patients and health professionals will be worthwhile if it increases the likelihood of early detection and appropriate response to factors raising the likelihood of ulceration.

The guideline development group also considered the issues of what key points/topics should be covered in patient education, particularly from the viewpoint of what would people with diabetes need to know about in relation to foot care. Whilst recognising that that it was not possible to give definitive answers in this area, they have produced a framework of key points that might provide a useful starting point. This was developed from a consensus of the guideline development group rather than from research literature, as this is not available This framework is presented in Appendix 26.



See Appendix 26 for issues and topics that might be covered in patient education.

Copyright © 2003, School of Health and Related Research (ScHARR), University of Sheffield.
Bookshelf ID: NBK51733


Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...