Psychological interventions for treating foot ulcers, and preventing their recurrence, in people with diabetes

Abstract

Background

Diabetic foot ulceration (DFU) can be defined as a full‐thickness wound below the ankle and is a major complication of diabetes mellitus. Despite best practice, many wounds fail to heal, and when they do, the risk of recurrence of DFU remains high. Beliefs about personal control, or influence, on ulceration are associated with better engagement with self‐care in DFU. Psychological interventions aim to reduce levels of psychological distress and empower people to engage in self‐care, and there is some evidence to suggest that they can impact positively on the rate of wound healing.

Objectives

To evaluate the effects of psychological interventions on healing and recurrence of DFU.

Search methods

In September 2019, we searched the Cochrane Wounds Specialised Register; the Cochrane Central Register of Controlled Trials (CENTRAL), Ovid MEDLINE (including In‐Process & Other Non‐Indexed Citations), Ovid Embase, Ovid PsycINFO and EBSCO CINAHL Plus. We also searched clinical trials registries for ongoing and unpublished studies, and reviewed reference lists of relevant included studies as well as reviews, meta‐analyses and health technology reports to identify additional studies. There were no restrictions with respect to language, date of publication or study setting.

Selection criteria

We included randomised controlled trials (RCTs) and quasi‐RCTs that evaluated psychological interventions compared with standard care, education or another psychological intervention. Our primary outcomes were the proportion of wounds completely healed; time to complete wound healing; time to recurrence and number of recurrences.

Data collection and analysis

Four review authors independently screened titles and abstracts of the studies identified by the search strategy for eligibility. Three authors independently screened all potentially relevant studies using the inclusion criteria and carried out data extraction, assessment of risk of bias and GRADE assessment of the certainty of the evidence.

Main results

We identified seven trials that met the inclusion criteria with a total of 290 participants: six RCTs and one quasi‐RCT. The studies were conducted in Australia, the USA, the UK, Indonesia, Norway and South Africa. Three trials used a counselling‐style intervention and one assessed an intervention designed to enhance an understanding of well‐being. One RCT used a biofeedback relaxation training intervention and one used a psychosocial intervention based on cognitive behavioural therapy. A quasi‐RCT assessed motivation and tailored the intervention accordingly.

Due to the heterogeneity of the trials identified, pooling of data was judged inappropriate, and we therefore present a narrative synthesis. Comparisons were (1) psychological intervention compared with standard care and (2) psychological intervention compared with another psychological intervention.

We are uncertain whether there is a difference between psychological intervention and standard care for people with diabetic foot ulceration in the proportion of wounds completely healed (two trials, data not pooled, first trial RR 6.25, 95% CI 0.35 to 112.5; 16 participants, second trial RR 0.59, 95% CI 0.26 to 1.39; 60 participants), in foot ulcer recurrence after one year (two trials, data not pooled, first trial RR 0.67, 95% CI 0.32 to 1.41; 41 participants, second trial RR 0.63, 95% CI 0.05 to 7.90; 13 participants) or in health‐related quality of life (one trial, MD 5.52, 95% CI ‐5.80 to 16.84; 56 participants). This is based on very low‐certainty evidence which we downgraded for very serious study limitations, risk of bias and imprecision.

We are uncertain whether there is a difference in the proportion of wounds completely healed in people with diabetic foot ulceration depending on whether they receive a psychological intervention compared with another psychological intervention (one trial, RR 2.33, 95% CI 0.92 to 5.93; 16 participants). This is based on very low‐certainty evidence from one study which we downgraded for very serious study limitations, risk of bias and imprecision.

Time to complete wound healing was reported in two studies but not in a way that was suitable for inclusion in this review. One trial reported self‐efficacy and two trials reported quality of life, but only one reported quality of life in a manner that enabled us to extract data for this review. No studies explored the other primary outcome (time to recurrence) or secondary outcomes (amputations (major or distal) or cost).

Authors' conclusions

We are unable to determine whether psychological interventions are of any benefit to people with an active diabetic foot ulcer or a history of diabetic foot ulcers to achieve complete wound healing or prevent recurrence. This is because there are few trials of psychological interventions in this area. Of the trials we included, few measured all of our outcomes of interest and, where they did so, we judged the evidence, using GRADE criteria, to be of very low certainty.

Plain language summary

Do psychological therapies help diabetic foot ulcers to heal and prevent their recurrence?

Background

Diabetes is a condition that causes high levels of sugar in the blood. Blood sugar levels are controlled by insulin, a hormone made by the pancreas. Insulin instructs the liver, muscles and fat cells to remove sugar from the blood and store it. When the pancreas does not make enough insulin, or the body does not respond to insulin, too much sugar stays in the blood.

High blood sugar can damage the nerves in the body’s extremities (such as the hands or feet) and cause numbness. This means that if someone with diabetes cuts their foot by stepping on a sharp object, or develops blisters on their feet, they might not be aware if it. Blisters may develop into open wounds or sores, known as diabetic foot ulcers (DFU). These can be slow to heal, because diabetes damages blood vessels and this restricts blood supply – and the oxygen and nutrients blood carries, which are necessary for healing. If left untreated, ulcers can become infected. In severe cases, amputation of a toe, foot, or more, may be necessary.

People with DFU may feel distressed about their wounds and the impact these have on their life. This can reduce chances of ulcers healing, and make them more likely to reappear. Psychological therapies might improve ulcer healing and prevent reappearance, by helping people to feel that they can manage their diabetes and overcome DFU.

What did we want to find out?

We wanted to find out if psychological therapies improve DFU healing and prevent their reappearance. We also wanted to know if they affect the number of amputations, quality of life, cost of treatment and people's belief that they can manage the condition, in addition to comparing the effects of different psychological therapies.

Our methods

We searched for relevant randomised controlled trials, in which the treatment each person receives is chosen at random. These studies give the most reliable evidence about the effects of a treatment. We then compared the results, and summarised the evidence from all the studies. We assessed how certain the evidence was by considering factors such as the way studies were conducted, study sizes, and consistency of findings across studies. Based on our assessments, we categorised the evidence as being of very low, low, moderate or high certainty.

What we found

We found seven studies that involved 290 people with diabetes who were followed up for between six weeks and six months. The studies were conducted in Australia, the USA, Norway, Indonesia, South Africa and the UK. The psychological therapies investigated were:

‐ counselling (three studies);

‐ muscle relaxation (one study);

‐ individually‐tailored motivation (one study);

‐ a therapy that aims to develop a person’s understanding of well‐being (one study);

‐ group‐based cognitive behavioural therapy (one study).

Psychological therapies compared to usual care

We do not know if psychological therapies improve healing of DFU, or prevent ulcer reappearance, because the evidence is of very low certainty.

Different psychological therapies compared to each other

We do not know if some psychological therapies have more of an effect than others on healing of DFU or preventing ulcer reappearance. This is because either no studies investigated this, or the evidence is of very low certainty.

We do not know if psychological therapies have an effect on the time it takes for ulcers to reappear, amputation, quality of life or a person’s belief in their ability to manage their condition, because there were either no or too few studies investigating this. No studies reported information about the cost of psychological therapies.

What does this mean?

There is no robust evidence about the effects of psychological therapies on DFU healing and recurrence.

There is a need for high‐quality studies that include enough people to detect a potential effect of psychological therapies on ulcer healing or reappearance. It would be helpful to agree on a set of clear measures to include in all future studies, so that results from different studies could be compared and analysed together.

How‐up‐to date is this review?

The evidence in this Cochrane Review is current to September 2019.

Summary of findings

Summary of findings 1. Psychological intervention compared with standard care for treating and preventing recurrence of foot ulcers in people with diabetes.

Psychological intervention compared with standard care for treating and preventing recurrence of foot ulcers in people with diabetes
Patient or population: adults with diabetes with current or previous foot ulceration Setting: foot or podiatry clinics Intervention: psychological Intervention Comparison: standard care
Outcomes	Anticipated absolute effects* (95% CI)		Relative effect (95% CI)	№ of participants (studies)	Certainty of the evidence (GRADE)	Comments
	Risk with standard care	Risk with Psychological Intervention
Proportion of wounds completely healed at 3 months	Intervention: tailored occupational therapy treatment based on level of motivation)		Jansen 2009 (n = 16, RR 6.25, CI 0.35 to 112.52)	76 (2 RCTs)	⊕⊝⊝⊝ Very low 1 2	Two studies. Pooling was not appropriate within the comparison/outcome. Both studies found no difference between groups but the certainty of evidence was very low. Therefore, the evidence is very uncertain about the effect of psychological interventions on complete wound healing.
Proportion of wounds completely healed at 6 weeks	Intervention: 'Understanding Well‐being' treatment)		Imran 2018 (n = 60, RR 0.59, CI 0.26 to 1.39)
Time to complete wound healing	Not reported in a manner suitable for analysis
Time to recurrence	Not measured
Number of foot ulcer recurrences at 1 year	Intervention: theory‐based counselling		Skafjeld 2015 (n = 41, RR 0.67, CI 0.32 to 1.41)	54 (2 RCTs)	⊕⊝⊝⊝ Very low 2 3	Two studies. Pooling was not appropriate within the comparison/outcome. Both studies found no difference between groups but the certainty of evidence was very low. Therefore, the evidence is very uncertain about the effect of psychological intervention on number of foot ulcer recurrence.
Number of foot ulcer recurrences at 6 months	Intervention: group‐based psychosocial intervention		Vedhara 2012 (n = 13, RR 0.63, CI 0.05 to 7.90)
*The risk in the intervention group (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: Confidence interval; RR: Risk ratio
GRADE Working Group grades of evidence High certainty: we are very confident that the true effect lies close to that of the estimate of the effect Moderate certainty: we are moderately confident in the effect estimate: the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different Low certainty: our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect Very low certainty: we have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect

¹ Downgraded 2 levels due to high risk of selection (2 studies), performance (2 studies), detection (1 study) and attrition bias (2 studies) and unclear for risk of reporting bias (1 study)

² Downgraded 2 levels due to very serious imprecision (very small sample size and 95% CI estimate of effect included both little or no effect and appreciable benefit or harm) (2 studies)

³ Downgraded 2 levels due to high risk of performance (2 studies), detection (1 study), reporting (1 study) and other risk of biases (1 study) and unclear risk of selection (2 studies), detection, reporting and other biases.

Summary of findings 2. Psychological intervention compared with another psychological intervention for treating and preventing recurrence of foot ulcers in people with diabetes.

Psychological intervention compared with another psychological intervention for treating and preventing recurrence of foot ulcers in people with diabetes
Patient or population: adults with diabetes with current or previous foot ulceration Setting: foot or podiatry clinics Intervention: biofeedback‐assisted relaxation Comparison: relaxation
Outcomes	Anticipated absolute effects* (95% CI)		Relative effect (95% CI)	№ of participants (studies)	Certainty of the evidence (GRADE)	Comments
	Risk with relaxation	Risk with biofeedback‐assisted relaxation
Proportion of wounds completely healed at 3 months	Study population		RR 2.33, (CI 0.92 to 5.93)	16 (1 RCT)	⊕⊝⊝⊝ Very low 1 2	The evidence is very uncertain about the effect of biofeedback‐assisted relaxation on complete wound healing.
	375 per 1,000	874 per 1,000 (345 to 1,000)
Time to complete wound healing	This outcome was reported for the whole cohort in one study (Rice 2001), which included participants without diabetes and therefore did not contribute data to this outcome and comparison.
Time to foot ulcer recurrence	Not measured
Number of foot ulcer recurrences	Not measured
*The risk in the intervention group (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: Confidence interval; RR: Risk ratio
GRADE Working Group grades of evidence High certainty: we are very confident that the true effect lies close to that of the estimate of the effect Moderate certainty: we are moderately confident in the effect estimate: the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different Low certainty: our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect Very low certainty: we have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect

¹ Downgraded 2 levels due to high risk of selection (allocation concealment), performance and other biases and unclear risk of bias for selection (sequence generation) and reporting biases

² Downgraded 2 levels due to very serious imprecision (very small sample size and 95% CI estimate of effect included both little or no effect and appreciable benefit or harm)

Background

Description of the condition

Diabetes mellitus (DM) refers to a group of metabolic disorders resulting from defects in insulin secretion or insulin action, or both, giving rise to chronic hyperglycaemia (elevated blood glucose levels) (ADA 2009). There are two main types of DM: Type 1 (T1DM) usually develops in childhood, or early adulthood, and is characterised by insulin deficiency as a result of auto‐immune destruction of the insulin‐producing beta cells; Type 2 (T2DM) develops when there is a gradual beta cell destruction and reduced insulin secretion, and/or when the insulin produced does not work effectively (insulin resistance). Over the last decade, the global prevalence of DM has increased substantially in countries at all income levels, mirroring the worldwide obesity epidemic (WHO 2016). It is anticipated that almost 600 million people worldwide will have DM by the year 2035 (IDF 2017).

Diabetic foot disease, characterised by peripheral neuropathy (loss of sensation in the feet), ischaemia (reduced blood flow to the feet) and foot ulceration, is a major complication of diabetes. Approximately 50% of people with diabetes will develop peripheral neuropathy, while 15% to 25% will develop foot ulceration during the course of the disease (Bus 2016). Diabetic foot ulceration (DFU) can be defined as a full‐thickness wound (penetrates dermal tissues) below the ankle in a person with diabetes (Hoogeveen 2015) caused by peripheral neuropathy or ischaemia or both, and consequent trauma to the foot (Alexiadou 2012). Chronic ulceration can lead to adverse patient outcomes and complications including sepsis and increased foot morbidity, which may require distal (digital or transmetatarsal) or lower limb amputation (above or below knee). The progression and outcome from DFU is worsened by neuropathy and ischaemia and a myriad of other complications of diabetes such as impaired immune function, poor tissue oxygenation and defective healing (Falanga 2005). DFU is considered the most prevalent and costly of all diabetes complications and is associated with a five‐year mortality rate of 50% (Armstrong 2011).

The cost of treating DFU poses a significant economic burden. The EURODIALE study found the estimated costs associated with treatment of DFU to be EUR 10 billion per year in Europe (Schaper 2012). In the UK, DFU is estimated to account for 20% of the total cost of diabetes care (Wounds International Expert Working Group 2013). DFU also poses significant personal costs for individuals due to physical limitations and pain that impede work and social activities (Goodridge 2006).

There is a higher risk for ulceration and re‐ulceration in men (Iverson 2008). The underlying cause of this may be associated with different effects of ulceration on male behavioural, emotional, cognitive and social influences that affect healing and prevention of DFU recurrence (Vedhara 2012). Gender differences were seen in a meta‐analysis of the effects of psychological interventions in cardiac patients, suggesting that gender‐tailored programmes need to be developed and evaluated (Linden 2007).

Depression is common in people with DFU and the prevalence is double that of the non‐diabetic population (Ali 2006; Anderson 2001; Pearson 2014). The evidence to date suggests a bi‐directional relationship: depression is associated with increased risk of T2DM (Nouwen 2010), and the onset of diabetes is associated with subsequent depression (Mezuk 2008), poor glycaemic control (Lustman 2000) and diabetic complications (De Groot 2001). This may be attributed to the physiological stress of living with diabetes, the psychological burden of self‐managing diabetes and coping with the complications of the disease. Depression substantially increases the risk of DFU in a dose response manner: the greater the depressive symptoms, the greater the risk of DFU (Iverson 2015). We also know that depression is associated with a two‐fold increased risk of mortality in people with their first DFU at five years (Winkley 2012). Furthermore, health‐related quality of life among people with diabetes and foot ulcers is much lower than in the general population across a wide range of domains including restrictions in daily activities, physical restrictions and lower social functioning (Ribu 2007; Winkley 2009). Twenty per cent of people with severe mental illness have diabetes, usually T2DM, and these people, and people from other hard‐to‐reach groups, such as people with dementia or learning disabilities, are more likely to die prematurely from cardiovascular disease (Brown 2010; NHS Health Scotland 2004). They are also more likely to be at risk of suffering from the complications of diabetes such as diabetic foot disease, as they are less likely to receive adequate support with diabetes management (Mitchell 2009). However, as awareness is raised of these inequalities (Holt 2015), more complex interventions, such as psychological interventions, may be developed to support these individuals.

The development of a foot ulcer is a sign of progressive disease and, once present, DFU can prove challenging to heal (Wounds International Expert Working Group 2013). Strategies associated with the prevention of DFU and optimal management of active DFU are key to reducing the burden of diabetic foot disease. Specialist multidisciplinary teams have been shown to reduce the incidence of ulceration and amputation (Edmonds 1986); however, despite best practice, many wounds fail to heal and the risk of recurrence of DFU remains high with one study reporting a recurrence rate of 57% (Dubsky 2012). Holding certain beliefs about diabetes, such as an individual viewing themselves as having control or influence on ulceration, is associated with better engagement with self‐care in DFU (Vedhara 2014), and has also been shown to have a significant effect on survival from DFU (Vedhara 2016). Healthcare practitioners such as diabetes professionals, podiatrists, nurses, and people with DFU need the knowledge, skills and attitudes to facilitate empowerment and support self‐management.

Description of the intervention

Psychological interventions are by nature complex and difficult to define and precise definitions are frequently missing from intervention studies and reviews (Hodges 2011). New CONSORT guidelines have been developed to guide the reporting of psychological intervention studies (Montgomery 2018).

Psychological interventions are those that use 'the therapeutic alliance between the patient and therapist to bring about change in emotional, cognitive and behavioural functioning' (Ismail 2004). Psychological interventions are distinct from other types of intervention such as education or medication. Their aim is to improve the psychological and physical well‐being of patients using a form of communication, often talking therapy, to foster a supportive relationship in order to promote patient autonomy and empowerment in the self‐management of their chronic condition (Alam 2009; Kulzer 2007; McGloin 2015). A psychological intervention includes a psychotherapeutic (improved emotional, cognitive or behavioural functioning) or psychosomatic mechanism (addressing the stress of having a condition such as diabetes) (Goldbeck 2014) or both. Psychological interventions include, for example, supportive or counselling therapy, cognitive behavioural strategies (an umbrella term for problem‐solving, contract‐setting, goal‐setting, self‐monitoring of behaviours and enlisting social support), psychotherapy or psychological techniques such as motivational interviewing and newer techniques such as positive psychology and acceptance commitment therapy, both derived from cognitive behavioural therapy (CBT) (Goldbeck 2014; Ismail 2004). Despite the breadth of psychological interventions that might be used in the treatment of DFU in people with diabetes, there is no consensus on how such interventions should be classified. Psychological treatments may be categorised in a variety of ways including current behaviour change taxonomies or according to underlying theory.

How the intervention might work

Psychological interventions aim to reduce levels of psychological distress, including depression or perceived levels of stress of the individual, a factor that affects wound healing negatively (Ebrecht 2004; Walburn 2009). Chronic stress increases cortisol release, which has an anti‐inflammatory effect. This disrupts the normal functioning of immune cells required for the inflammatory phase of wound healing and therefore delays wound healing (Ebrecht 2004; Guo 2010; Kiecolt‐Glaser 1995; Vileikyte 2007). There is some evidence to suggest that psychological interventions can impact positively on the rate of wound healing (Weinman 2008).

The response to stress can also result in unhealthy behaviours (Guo 2010). Psychological interventions may improve peoples' levels of confidence in controlling and self‐managing their diabetes and reduce the potential for ulcer recurrence by enabling or supporting them to make lifestyle changes that promote wound healing, including adequate sleep, physical activity, limb elevation and offloading, wearing of a total contact cast and appropriate footwear, healthy nutrition, reduced alcohol intake, smoking cessation and improved social interaction (Armstrong 2005; Brown 2012; Guo 2010). Whilst psychological interventions for people with DFU may be beneficial, there may be obstacles to implementation, such as a lack of awareness of psychological issues on the part of health professionals, and this needs to be considered. For example, some health professionals may not be aware that psychological problems are more common in diabetes and that these are associated with adverse outcomes. Addressing these issues may involve additional training and ultimately provide access to psychological services.

Why it is important to do this review

Less healing and a higher rate of ulcer recurrence is associated with poor glycaemic control and this may reflect poor patient engagement and a lack of psychological support (Dubsky 2012). Illness beliefs about DFU have recently been shown to be linked to survival and predict foot self‐care practice in DFU (Vedhara 2014; Vedhara 2016). Psychological interventions may have the potential to improve self‐care and reduce the morbidity and costs associated with DFU. There is a need for a systematic review of the evidence on the effectiveness of psychological interventions in DFU. The aim of this is to allow people with DFUs, healthcare professionals caring for them and those trying to prevent recurrence to take informed decisions regarding their implementation.

A number of systematic reviews of psychological interventions in diabetes have been conducted or are planned (Alam 2009; Chew 2017; Ismail 2004; Steed 2003), but to our knowledge no systematic review of psychological interventions in promoting healing and preventing recurrence of DFU has been completed.

Objectives

To evaluate the effects of psychological interventions on healing and recurrence of DFU.

Methods

Criteria for considering studies for this review

Types of studies

Randomised controlled trials (RCTs) including quasi‐randomised, cross‐over and cluster‐RCTs. We would have only included cross‐over trials that reported outcome data at the end of the first treatment period and prior to cross‐over. We did not limit eligibility with respect to language of the report, year or place of publication.

Types of participants

People of 18 years or older, in any care setting, including their own home, with a DFU or a history of DFU.

Types of interventions

For the purposes of this review, a psychological intervention was considered for inclusion if it met the following criteria (adapted from Goldbeck 2014 and Ismail 2004):

• tailored to the individual and includes psychological interventions with a psychotherapeutic or psychosomatic mechanism of action. This included, but was not limited to cognitive behavioural therapy, cognitive therapy, psycho‐dynamic therapy, counselling (techniques of goal‐setting, problem‐solving, identifying strengths and barriers, motivational interviewing), family systems or systemic therapy, and others such as supportive, relaxation, and biofeedback;

• provided through structured interactions, face‐to‐face or via telephone in individual or group setting, between a participant and a facilitator;

• provided by facilitators who may or may not have had qualifications or training in psychology or psychotherapy;

• interactions between the participant and facilitator of any frequency or duration;

• the main target of the psychological intervention was promoting healing and/or preventing recurrence of DFU.

The following broad categories were used to classify the different types of interventions (Goldbeck 2014):

• cognitive behavioural therapy;

• cognitive therapy;

• psychodynamic therapy;

• counselling;

• family systems or systemic therapy;

• other interventions.

We excluded studies of the following interventions:

• interventions which rely only on a structured or unstructured intervention utilising teaching or instructional approaches providing information related to DFU;

• specific interventions focused on preventing the first incidence of foot ulceration.

Where data were available, we aimed to undertake the following comparisons: psychological interventions versus standard care or versus another psychological intervention or versus education on healing and/or recurrence of DFU.

Types of outcome measures

Primary outcomes

The primary outcome for this review was complete wound healing.

This may have been presented in either or both of the following formats:

• the proportion of wounds completely healed (frequency of complete healing by group);

• time to complete wound healing (survival data: study‐level data reported as a hazard ratio (HR) with standard error (SE)).

Recurrence was assessed as:

• time to recurrence (survival data: study‐level data reported as an HR with SE);

• number of recurrences.

We anticipated that trials would have different follow‐up times. We sought to group the timing of follow‐up data into:

• short‐term (up to 12 weeks following treatment);

• medium‐term (more than 12 weeks and up to six months after the end of treatment);

• long‐term follow‐up (longer than six months and less than 12 months after the end of treatment).

We sought to include non‐censored data, i.e. mean or median time to healing without survival analysis, if there was an indication that all participants experienced complete healing during the trial period but considered these as less robust assessments of these outcomes, and we did not plan to include these data in preparing the 'Summary of findings' tables.

Secondary outcomes

• Amputations (major or distal);

• Health‐related quality of life, as measured by tools such as SF‐36 or EQ‐5D and/or disease‐specific quality of life instruments designed for use with DFU patients;

• Self‐efficacy as measured by tools such as the Diabetes Empowerment Scale and/or the Footcare Confidence Scale;

• Cost.

Search methods for identification of studies

Electronic searches

We searched the following electronic databases to identify reports of relevant clinical trials:

• the Cochrane Wounds Specialised Register (searched 03 September 2019);

• the Cochrane Central Register of Controlled Trials (CENTRAL; 2019, Issue 8) in the Cochrane Library (03 September 2019);

• Ovid MEDLINE including In‐Process & Other Non‐Indexed Citations (1946 to 03 September 2019);

• Ovid Embase (1974 to 03 September 2019);

• Ovid PsycINFO (1806 to 03 September 2019);

• EBSCO CINAHL Plus (Cumulative Index to Nursing and Allied Health Literature; 1937 to 03 September 2019).

The search strategies for the Cochrane Wounds Specialised Register, CENTRAL, Ovid MEDLINE, Ovid Embase and EBSCO CINAHL Plus can be found in Appendix 1. We combined the Ovid MEDLINE search with the Cochrane Highly Sensitive Search Strategy for identifying randomised trials in MEDLINE: sensitivity‐ and precision‐maximising version (2008 revision) (Lefebvre 2011). We combined the Embase search with the Ovid Embase filter developed by the UK Cochrane Centre (Lefebvre 2011). We combined the CINAHL Plus searches with the trial filter developed by Glanville 2019. There were no restrictions with respect to language, date of publication or study setting.

We also searched the following clinical trials registries:

• ClinicalTrials.gov (www.clinicaltrials.gov) (searched 03 September 2019);

• World Health Organization (WHO) International Clinical Trials Registry Platform (www.who.int/trialsearch) (searched 03 September 2019).

Search strategies for clinical trial registries can be found in Appendix 1.

Searching other resources

Searching reference lists of included trials and relevant reviews

We aimed to identify other potentially eligible trials or ancillary publications by searching the reference lists of retrieved included trials, as well as relevant systematic reviews, meta‐analyses and health technology assessment reports.

Searching by contacting individuals or organisations

When necessary, we contacted authors of key papers and abstracts to request further information about their trials.

Adverse effects

We did not perform a separate search for adverse effects of interventions used; we considered adverse effects as described in included studies only.

Data collection and analysis

The methodology for data collection and analysis is based on the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011). We carried out data collection and analysis according to the methods outlined in the published protocol of this review (McGloin 2017).

Selection of studies

Four reviewers (HMG, CMI, KW and GG) independently assessed the title and abstract of all potentially relevant studies identified from the search strategy to identify those that might meet the inclusion criteria. We retrieved the full text of studies identified as potentially relevant by at least one author. HMG, CMI and GG independently screened full‐text articles for inclusion or exclusion. We resolved any disagreement by discussion or by consulting a fourth review author (KW). All relevant studies excluded at the full‐text stage were listed as excluded studies and reasons for their exclusion are presented in the Characteristics of excluded studies table. Our screening and selection processes, including findings at each stage, are presented in a PRISMA flowchart (Liberati 2009).

Data extraction and management

Three authors (HMG, CMI and GG) independently extracted data using a data extraction form designed specifically for this review. Any disagreements were resolved by discussion or we consulted with a third review author (DD). We extracted the following information from each included study:

• trial authors

• year of publication

• country where RCT performed

• setting of care

• unit of investigation – participant, leg or ulcer

• overall sample size

• participant selection criteria

• number of participants randomised to each treatment arm

• baseline characteristics of participants per treatment arm (gender, age, baseline ulcer area, ulcer duration and previous history of ulceration)

• details of the psychological intervention including: intervention content (components, techniques, treatment materials, tailoring to individual), proposed mechanism of action, method of delivery, number of sessions, length or time of session, background, qualifications and training of healthcare personnel delivering the intervention, target outcome

• duration of treatment

• duration of follow‐up

• statistical methods used for data analysis to inform decisions on whether or not baseline adjustments have been made and if data were censored

• primary and secondary outcomes measured

• withdrawals (per treatment arm with numbers and reasons)

• source of trial funding

We pilot tested the data extraction tool on two papers prior to the conduct of the full review and no amendments were necessary.

One reviewer (HMG) entered all data into Review Manager 5 (RevMan 5.3) software which was checked for accuracy against the data extraction sheets by a second reviewer (GG) (Review Manager 2014). Where additional information was needed, we contacted authors of the original reports to provide further information on study characteristics and data.

Assessment of risk of bias in included studies

Three reviewers (HMG, CMI and GG) independently assessed each study for risk of bias using the Cochrane 'Risk of bias' tool (Higgins 2011). This tool assesses risk of bias in sequence generation, allocation concealment, blinding, attrition, selective reporting and other topic‐ or design‐specific issues (e.g. extreme baseline imbalance). Any disagreements regarding the 'Risk of bias' assessment were resolved by discussion or by involving another author (DD). The risk of bias for each study across each domain was also summarised graphically. Criteria for judgements on each risk of bias domain are given in Appendix 2.

We planned to complete the risk of bias separately for self‐reported and objective outcome measurements, but only objective outcome measurements were included.

Where information on risk of bias was related to unpublished data or correspondence with trialists, we would have noted this in the 'Risk of bias' table.

Measures of treatment effect

Dichotomous data

For dichotomous data, we have presented results as summary relative risks (RR) with their corresponding 95% confidence intervals (95% CI) where available.

Continuous data

For continuous data, we used the mean difference where outcomes were measured in the same way between trials, and the standardised mean difference to combine outcomes from trials that measured the same outcome but had used different scales, with corresponding 95% CIs (Higgins 2011).

Time‐to‐event data

We had planned to measure the intervention effect for time‐to‐event data and express it as a hazard ratio but no study reported time to healing rates. We had aimed to use the methods used to analyse time‐to‐event outcomes described by Tierney 2007 and detailed in Chapter 7, section 7.7.6. of the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011).

Unit of analysis issues

Issues with multiple ulcers

Had the issue arisen, we would have assessed whether trials presented outcomes per participant or as multiple wounds per participant. For wound healing and amputation, we would have treated the participant as the unit of analysis. One study (McBride 2016) referred to more than one ulcer in their methods but no raw wound healing data were reported.

Cluster‐randomised trials

We did not include any cluster‐randomised trials as there were no studies of this type that met the inclusion criteria for this review. If included in future updates, we will adjust the sample sizes for all included arms using the methods described in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011) using an estimate of the intra‐cluster correlation co‐efficient (ICC) derived from the trial (where given), from a similar trial or from a trial of a similar population. If we use ICCs from other sources, we will report this and conduct sensitivity analyses to investigate the effect of variation in the ICC on intervention effect estimates. We will perform a sensitivity analysis to investigate the effects of the randomisation unit, i.e. to determine the sensitivity of the effect estimates to inclusion and exclusion of cluster trials.

Studies with multiple arms

We did not include studies with multiple treatment arms as there were no studies of this type that met the inclusion criteria for this review. If included in future updates, we will combine all relevant experimental intervention groups in the study (e.g. groups with psychological interventions of different duration) into a single group and all comparable relevant control intervention groups into a single control group. We will not combine control groups with different types of interventions (e.g. standard care and education on DFU healing) in a single meta‐analysis and will instead analyse these separately.

Repeated measures

We did not include any studies/outcomes with repeated outcome assessments as there were no studies of this type that met the inclusion criteria for this review. If included in future updates, we will group assessment time‐frames as 12 weeks or less (short‐term), more than 12 weeks up to six months (medium‐term) and more than 6 up to 12 months (long‐term).

Cross‐over trials

We did not include any cross‐over trials as there were no studies of this type that met the inclusion criteria for this review. If included in future updates, we will incorporate outcome data from the end of the first treatment period and prior to cross‐over only.

Dealing with missing data

We noted levels of attrition for all included trials. We contacted trial authors for missing data and if there was no response we considered the available data.

We carried out analyses, as far as possible, on an intention‐to‐treat basis, i.e. we attempted to include all participants in the group to which they were randomised, regardless of whether or not they received the allocated intervention. The denominator for each outcome in each trial was the number of people randomised minus any for whom outcomes were known to be missing. Where a randomised participant was not included in the analysis, we assumed that there was no ulcer healing i.e. the person contributed to the denominator only.

Assessment of heterogeneity

We judged meta‐analysis as inappropriate because included trials reported different comparators and outcomes.

We assessed clinical heterogeneity by exploring the variability of participants, interventions, comparators and outcomes. We had planned to explore the effect of gender and type of psychological approach through subgroup analyses (see Subgroup analysis and investigation of heterogeneity). Subgroup analyses, however, were not possible due to absence of data.

We had planned to assess statistical heterogeneity in each meta‐analysis using the I² and Chi² statistics. We would have regarded statistical heterogeneity as substantial/considerable if an I² was greater than 50% or if there was a low P value (less than 0.10) in the Chi² test for heterogeneity. Due to a lack of comparable outcome data in the trials identified, statistical heterogeneity was not assessed.

Assessment of reporting biases

We searched multiple databases, online trial registries and grey literature databases to locate all relevant publications. We included seven trials in our review and were, therefore, unable to test formally for reporting biases. If there were more than 10 studies included in a meta‐analysis, we had planned to use a funnel plot of all studies to visually assess reporting biases. If we detected funnel plot asymmetry visually, we would have performed exploratory analyses using the test proposed by Egger 1997 for continuous outcomes and that proposed by Harbord 2006 for dichotomous outcomes.

Data synthesis

As meta‐analysis was judged inappropriate, details of included studies have been combined in a narrative synthesis according to type of comparator and by primary outcomes.

In terms of meta‐analytical approach, we anticipate that future updates will contain clinical, methodological and statistical heterogeneity in included studies. Thus, where meta‐analysis is possible, we anticipate using a random‐effects approach. Conducting meta‐analysis with a fixed‐effect model in the presence of even minor heterogeneity may provide overly narrow confidence intervals. In future updates, we will only use a fixed‐effect approach when clinical and methodological heterogeneity is assessed to be minimal, and the assumption that a single underlying treatment effect is being estimated holds. We plan to use Chi² and I² to quantify heterogeneity but not to guide choice of model for meta‐analysis. We will exercise caution when meta‐analysed data are at risk of small study effects because a random‐effects model may be unsuitable. In this case, or where there are other reasons to question the selection of a fixed‐effect or random‐effects model, we will assess the impact of the approach using sensitivity analyses to compare results from alternate models. We will report any evidence that suggests that the use of a particular model might not be robust. We may meta‐analyse even when there is thought to be extensive heterogeneity. We will attempt to explore the causes behind this using meta‐regression, if possible (Thompson 1999).

Data have been presented, unpooled, using forest plots, where possible. For dichotomous outcomes, we presented the study estimate as a risk ratio (RR) with 95% CI. For continuous outcomes measured in the studies identified, we presented the mean difference (MD) with 95% CI; we would have pooled standardised mean difference (SMD) estimates where studies measured the same outcome using different methods. For time‐to‐event data, we planned to plot (and, if appropriate, pool) estimates of hazard ratios and 95% CIs as presented in the study reports using the generic inverse variance method in RevMan 5.3 but no study reported this type of data. Where time to healing was analysed as a continuous measure but it was not clear if all wounds were healed, we had planned to only document the outcome of the study and not summarise data or use them in any meta‐analysis, however no study reported this. Pooled estimates of treatment effect would have been obtained using Cochrane RevMan software (version 5.3) (Review Manager 2014), if required.

Subgroup analysis and investigation of heterogeneity

We expected the following to introduce clinical heterogeneity and planned to carry out the following subgroup analysis to explore the impact of treatment approach.

1. Male versus female participants

The rationale for the subgroup analysis between genders was because it has been suggested that there is a higher risk for ulceration and re‐ulceration in men due to different psychological and social responses to chronic illness (Iverson 2008). There may be gender differences in the effects of psychological interventions in DFU. Exploring these is an important first step to developing gender‐tailored programmes (Linden 2007).

2. Type of psychological approach (according to the classification by Goldbeck 2014)

(See Types of interventions in Methods section).

It is clinically important to understand which type of psychological intervention could be most successful.

We planned to limit subgroup analyses to primary outcomes and would have explored subgroup differences by interaction tests available within RevMan 5 including the Chi² statistic and P value, and the interaction test I² value (Review Manager 2014).

We did not perform any subgroup analyses as there were insufficient data.

Sensitivity analysis

A sensitivity analysis would have been conducted on use of a fixed‐ versus a random‐effects model and on trial quality, by excluding all studies at high risk of bias in sequence generation and allocation concealment. We planned to limit sensitivity analyses to primary outcomes, however, there were insufficient data for sensitivity analyses.

Summary of findings and assessment of the certainty of the evidence

For each comparison, we prepared a ‘Summary of findings’ table. Two reviewers (HMG and DD) graded the certainty of the evidence for each outcome independently using criteria devised by the GRADE Working Group 2004 and GRADEprofiler (GRADEpro) software (Guyatt 2008; Higgins 2011; Schünemann 2010). The four levels of evidence certainty are 'high', 'moderate', 'low' or 'very low'. Certainty may be downgraded due to study limitations (risk of bias), imprecision, inconsistency, indirectness or publication bias. When evaluating the 'Risk of bias' domain, we downgraded the GRADE assessment when we classified a study as being at high risk of bias for one or more domains, or when the 'Risk of bias' assessment for selection bias was unclear (this was classified as unclear for the generation of the randomisation sequence domain and the allocation concealment domain). For imprecision, we selected an informal optimal information size of 300 for binary outcomes, following the GRADE default value (Guyatt 2011). We also followed GRADE guidance and downgraded twice for imprecision when there were very few events and CIs around effects included both appreciable benefit and appreciable harm. We included only primary outcomes in the ‘Summary of findings’ tables, i.e. the proportion of wounds completely healed, time to complete wound healing, recurrence and time to recurrence.

Results

Description of studies

See Characteristics of included studies; Characteristics of excluded studies.

Results of the search

The searches yielded 652 citations. The search of the grey literature found 489 records. Thus, we screened 1141 records in all and after initial review excluded 1100 because they did not meet the inclusion criteria. We identified 41 studies for full‐text assessment. We excluded 29 studies, three studies are ongoing, two studies are awaiting classification and seven studies met the inclusion criteria (Barth 1991; Imran 2018; Jansen 2009; McBride 2016; Rice 2001; Skafjeld 2015; Vedhara 2012). See PRISMA flow chart (Figure 1)

1.

Study flow diagram.

Included studies

Seven studies were included (Barth 1991; Imran 2018; Jansen 2009; McBride 2016; Rice 2001; Skafjeld 2015; Vedhara 2012). These included two RCTs, of which one was conducted in Australia (Barth 1991), and one in the USA (Rice 2001); and four pilot RCTs conducted in the UK (McBride 2016; Vedhara 2012), Norway (Skafjeld 2015) and Indonesia (Imran 2018), respectively. The study conducted in South Africa employed a quasi‐randomised design (Jansen 2009).

Participants

Barth 1991 randomised 70 people, men accounted for 56% of 62 participants who were included in the final dataset. The two groups did not differ significantly in age, gender, time since diagnosis, oral hypoglycaemic or insulin therapy, glycosylated haemoglobin and number of foot problems requiring treatment. A range of foot problems was reported at baseline, of which, there was a prevalence of 3.3% (n = 1) of active foot ulceration in the control group and none in the intervention group. The intervention group had significantly higher cholesterol levels and there were significantly more participants with peripheral vascular disease than in the control group.

Jansen 2009, which randomised 16 people, reported the mean age of participants in the intervention group was 67 years and 57 years in the control group. Men accounted for 80% of the final study sample of 10 participants. Comorbidities or other demographics for the participants were not reported. All participants had an active foot ulcer on recruitment which ranged in size from 56 mm² to 652 mm².

In a pilot study in Indonesia, Imran 2018 randomised 60 people, of which 34 received the intervention, while 26 received standard care. According to the authors, the two groups did not differ at baseline on demographic characteristics except in terms of diabetes mellitus years duration (3.0 versus 8.0, P = 0.014) but details of age or gender were not provided, nor was there any information on the type of DFU, its size or duration. No significant difference was reported at baseline except in the distress characteristic (40.0 versus 37.4 P = 0.028).

McBride 2016 randomised 56 participants, of which seven were lost to follow‐up. All participants were white Britons and 73% were male. The groups did not differ significantly at baseline in age, employment status, education or marital status. The mean age of participants was 59 years in the control group and 62 years in the intervention group. All participants had a foot ulcer at baseline that was either new, current or a case of re‐ulceration.

Rice 2001 recruited 32 participants in total, of which 13 were male and 16 had diabetes. The age range for the study sample was 35 to 88 years with an average of 63 years. The average age differed significantly between the intervention and control group (57 versus 69 years). There were no differences in ulcer class, duration or area, skin condition, pedal pulses, vibratory sensation, pain, peroneal nerve fibre perception and levels of participant ambulation. The baseline ulcer perimeter was significantly higher in the intervention group (50.3 mm² versus 27.2 mm²).

Skafjeld 2015 randomised 41 participants with an average age of 59 years in the control group and 57 years in the intervention group. In the control and intervention groups, 75% and 86% were male, respectively. The groups did not differ significantly in education, BMI, waist circumference, smoking status, type 1 or type 2 diabetes, glycosylated haemoglobin and insulin therapy. There was a significantly higher level nephropathy and systemic vascular risk factors in the intervention group at baseline.

Vedhara 2012 randomised 15 participants with a mean age of 59 years in the control group (n = 5) and 69 years in the intervention group (n = 10). Men accounted for 77% (n = 6) of the final study sample of 13 participants. None of the intervention group participants were smokers compared with 2/5 (40%) of the control group. No analysis of baseline imbalances was undertaken but education level appeared higher in the intervention group and mean number of ulcers was higher in the control group.

Interventions

Three studies evaluated the effectiveness of a counselling‐style intervention (Barth 1991; McBride 2016; Skafjeld 2015).

Barth 1991 recruited adults with T2DM and suboptimal glycaemic control to an intensive foot care specific educational programme versus a conventional type 2 diabetes educational programme. The theoretical underpinning for the intervention group was based on cognitive motivation theory and the aim was to engage participants in performing daily foot care by increasing their motivation to perform it. Three sessions were conducted by a podiatrist and one session was conducted by a psychologist. An additional aim was to support them in the action phase once a decision had been reached about whether or not they wanted to follow the recommended foot care on a long‐term basis. Participants were given the necessary medical information they needed and practical support to perform foot care during the sessions.

Skafjeld 2015 conducted a pilot RCT to test the feasibility of monitoring foot temperature along with theory‐based counselling with adults with diabetic neuropathy and history of foot ulceration to reduce recurrent ulceration. Every three months, the nurse assessed their readiness to record skin temperature and then delivered stage‐appropriate counselling according to the transtheoretical model. It is not clear from the paper whether the nurse received training in the intervention from a psychologist but further information received from the author confirmed that the trainer did have continuing professional education in health counselling.

The study by McBride 2016 was a pilot RCT that included adults with a diabetic foot ulcer who were randomised to receive usual care or an intervention to increase shared decision‐making (Decision Navigation). The overall aim was to improve decision self‐efficacy and foot care at three months’ follow‐up. The intervention was delivered to individuals by a trained health psychologist via telephone.

One RCT tested a biofeedback relaxation training intervention to improve peripheral perfusion and healing (Rice 2001). The trialists used stratified randomisation to allocate adult participants with or without diabetes and with chronic non‐healing ulcers to the control or intervention group. It is not clear from the report who delivered the intervention or the training they received to deliver this.

The fifth study was a quasi‐randomised trial applying the Creative Ability Model (MCA) to adults with diabetic foot problems to improve healing at three‐month follow‐up (Jansen 2009). The treatment was delivered by an occupational therapist to the intervention group and was tailored to the level of motivation of the participant. It was not clear how many sessions were received by participants in either group. Nor was it clear whether this intervention could be classified as psychological based on the inclusion criteria for this systematic review. However, following contact with the author, we clarified that some psychological techniques were used which fall under the umbrella of CBT, namely problem‐solving, goal‐setting, activity‐scheduling and client‐centred practice, and therefore we decided to include this study.

The study by Vedhara 2012 was a feasibility RCT that implemented a group‐based cognitive behavioural intervention to adults with a history of foot ulceration who also exhibited evidence of psychological distress. The evaluation of the intervention was primarily qualitative as part of a feasibility study although, as the authors also reported ulcer recurrence rates, we included this study. The intervention was delivered by a diabetes specialist nurse and podiatrist who had received three days training in cognitive and behavioural competencies derived from CBT from an experienced CBT therapist.

In a pilot RCT in Indonesia (Imran 2018), those in the intervention group received their first treatment in the clinic during which they were interviewed, with the interviews including five trigger questions that would enhance the participants' understanding of well‐being based on the International Consensus (International Consensus 2012). The qualifications of the interventionist and the training they received was not described. In the second treatment, the participant wrote a diary at home to answer six structured questions. The interviews and diary writing were performed three times, once every two weeks, for each participant.

Outcomes

There are several ways to measure and report on wound healing including time to complete healing, change in ulcer size and/or the proportion of ulcers healed in the study period. We included studies that reported on all or some of the following outcomes: ulcer healing rates, change in the size of the ulcer(s) and the frequency of complete healing (Imran 2018, Jansen 2009, Rice 2001). We included studies that reported on foot ulcer occurrence and incidence in the follow‐up period (Skafjeld 2015; Vedhara 2012). Barth 1991 reported ulceration at baseline but not at follow‐up and McBride 2016 reported healing rate but no raw data were available for inclusion in analysis.

We aimed to include studies that reported on the following secondary outcomes: amputations (major or distal), health‐related quality of life, self‐efficacy and cost. Two studies reported health‐related quality of life at baseline and follow‐up (Imran 2018; McBride 2016) and one reported changes in self‐efficacy (Imran 2018).

Excluded studies

We excluded 29 studies for the following reasons: did not include people with DFU or a history of DFU (13 studies); did not include interventions with a psychological component (16 studies) (see Characteristics of excluded studies).

Studies awaiting classification

Two studies (McMurray 2002; Simson 2008) were deemed to have a psychological component. McMurray 2002 included people with diabetes and end‐stage renal failure but it was unclear if the study population had a current or previous history of DFU. In Simson 2008, a supportive psychotherapy intervention was implemented for people with DFU and comorbid depression, however, it was unclear if any outcome relevant to this review was measured (see Characteristics of studies awaiting classification).

Ongoing Studies

Three potentially relevant ongoing trials were identified in our search. One trial (IRCT20100628004256N8) aims to explore the impact of nurse‐led care incorporating counselling on quality of care and healing of diabetic foot ulcers. The second trial (NCT03909802) is investigating the effect of self‐ and family management of diabetic foot ulcer programmes on wound healing. A further trial (NL7406) was identified as potentially relevant as it aims to test a psychological intervention on the quality of life and wound healing in patients with DFU (see Characteristics of ongoing studies).

Risk of bias in included studies

We have presented details of risk of bias in separate tables for each study (see Characteristics of included studies) and in a summary table and graph (see Figure 2; Figure 3).

2.

Risk of bias summary: review authors' judgements about each risk of bias item for each included study.

3.

Risk of bias graph: review authors' judgements about each risk of bias item presented as percentages across all included studies.

Allocation

Three studies (Imran 2018; McBride 2016; Skafjeld 2015) described methods of random allocation. The study by Imran 2018, reported that the mechanism of random allocation sequence was implemented by computer and thus was deemed to be at low risk of bias. In McBride 2016, the random sequence was determined by an external researcher through the use of a computer‐generated random number table. Skafjeld 2015 used block randomisation to assign each four subjects to blocks with two in each group. Randomisation was stratified for participants with a history of Charcot foot, who have an extra high risk of recurrence of DFU. All study‐related procedures, including randomisation, were performed at the Diabetes Clinic, Oslo University Hospital. Trialists did not specify how the random sequence was generated but we assumed it was computer‐generated from the description given.

Jansen 2009 allocated participants to the experimental or control group alternately as they were referred for occupational therapy and therefore was classified at high risk of bias. In Barth 1991, Rice 2001 and Vedhara 2012, the random sequence generation procedure was not described, therefore the risk of bias was deemed unclear.

Allocation concealment was not described in four studies (Barth 1991; Rice 2001; Skafjeld 2015; Vedhara 2012). We assumed that allocation was not concealed in Jansen 2009 as participants were alternately assigned to the experimental or the control group. McBride 2016 ensured allocation concealment from the research team and participants by using opaque, sealed envelopes containing a note of group allocation. Envelopes were opened by a member of the research team after baseline measures had been taken. Similar methods were used by Imran 2018, in that sequentially numbered envelopes were placed in a container and opened in the clinic once eligible participants had arrived, however, as these envelopes could potentially be viewed by those allocating participants, this study was deemed to have a high risk of bias.

Blinding

Due to the interactive nature of the interventions for all studies, neither the intervention staff or the participants were blinded to group allocation. However, it should be possible to ensure blinding of outcome assessors. Barth 1991, Jansen 2009 and Rice 2001 all employed independent assessors who were blinded to group allocation to assess foot and wound outcomes. McBride 2016 and Vedhara 2012 did not state if the podiatrist assessing wound outcomes was blinded to whether the participant was in the experimental or control group. Skafjeld 2015 disclosed that the study nurse performed all assessments, therefore this study was deemed at high risk of bias. The study by Imran 2018 stated it was "open blinding" and therefore all personnel involved in assessment would have been aware of group allocation.

Incomplete outcome data

Barth 1991, Imran 2018 and Jansen 2009 were deemed at high risk of attrition bias. In Barth 1991, although 62 out of 70 completed the programme and follow‐up, they did not carry out an intention‐to‐treat analysis. In Jansen 2009, six were lost to follow‐up out of 16 recruited and an intention‐to‐treat analysis was not specified, therefore, we judged the attrition higher than acceptable. Out of the 60 participants randomised in the study by Imran 2018, only 29 were included in the final analysis and an intention‐to‐treat analysis was not performed. We deemed McBride 2016 at unclear risk although they carried out an intention‐to‐treat analysis and 49 out 56 participants completed the follow‐up. On closer inspection, wound healing rate was analysed on 45 participants but the data on four patients was missing and reasons for missing outcome data were unclear. We contacted the study author but the data were unavailable. Rice 2001, Skafjeld 2015 and Vedhara 2012 were all judged to be at low risk of attrition bias.

Selective reporting

We judged Barth 1991 to be at an unclear risk of selective reporting bias. The study reported the reduction in foot problems at one month as statistically significant but did not provide the numerical data that supported the findings. In the measurements section and in Table 2 of baseline measures, the number of foot ulcers were listed but they were not specifically reported in the follow‐up data. McBride 2016 was deemed to be at high risk of selective reporting bias as the authors stated in the methods section that ulcer size was measured at baseline and follow‐up and that the analysis revealed no significant differences for wound healing rate. However, the wound healing data section was empty in Table 4. Vedhara 2012 was deemed to be at high risk as re‐ulceration was reported but this was not planned in the Methods section.

Imran 2018, Jansen 2009, Rice 2001, and Skafjeld 2015 were all judged to be at unclear risk as no protocol for the study was available but all outcomes measured were reported as expected.

Other potential sources of bias

No other potential sources of bias were apparent in the studies.

Effects of interventions

See: Table 1; Table 2

See Table 1 for the main comparison, psychological intervention compared with standard care, and Table 2 for the comparison, psychological intervention compared with another psychological intervention.

Comparison 1: Psychological intervention compared with standard care (six trials)

Six studies (258 participants) were included in this comparison, of which, one trial (Barth 1991) did not report any outcomes relevant to this review. Five trials (Imran 2018; Jansen 2009; McBride 2016; Skafjeld 2015; Vedhara 2012) reported the following outcomes for this comparison:

Primary outcomes

Proportion of wounds completely healed

Two studies (Imran 2018; Jansen 2009) reported the primary outcome, proportion of wounds completely healed. However, as Jansen 2009 used the Creative Ability Model (MCA) approach and Imran 2018 used interviews using five trigger questions that would enhance the participants' understanding of well‐being, we decided not to pool the data due to the heterogeneity of the interventions. Jansen 2009 evaluated the effect on wound healing by measuring ulcer size at baseline and three months (intervention end). We extracted data on the proportion of ulcers healed. Ulcers reported as 0 mm² were considered fully healed. In the psychological intervention group, 28.6% (2/7) of participants achieved complete wound healing, whereas, in the standard care group, no participants achieved this (0/9) ( n = 16, RR 6.25, 95% CI 0.35 to 112.5, very low‐certainty evidence). Imran 2018 reported on the number of ulcers healed at two, four and six weeks. In the intervention group, 20.6% (7/34) healed after six weeks. This compared with a total of 34.6% (9/26) healed at week six in the control group (n = 60, RR 0.59, 95% CI 0.26 to 1.39, very low‐certainty evidence). It is unclear whether a psychological intervention compared with standard care has any impact on complete wound healing as we assessed the level of evidence as very low (downgraded by two levels due to high risk of selection, performance, detection and attrition bias and unclear for risk of reporting bias and downgraded by two levels due to very serious imprecision (very small sample size and 95% CI estimate of effect included both little or no effect and appreciable benefit or harm)).

See Table 1.

Time to complete wound healing

One trial (Imran 2018) reported the numbers healed after two, four and six weeks but this was not reported in a manner suitable for analysis within this review.

Length of time to recurrence

Not measured.

Number of foot ulcer recurrences

Two trials (Skafjeld 2015, Vedhara 2012) reported the primary outcome, foot ulcer recurrence. Foot ulcer recurrence was defined as an end point (after one year) in one study (Skafjeld 2015) and was classified according to the Wagner foot classification system. In Vedhara 2012, foot ulcer recurrence was assessed in the six‐month period following a ten‐week intervention. However, as Skafjeld used a counselling approach and Vedhara used CBT, we decided not to pool the data due to the heterogeneity of the interventions.

Skafjeld 2015 reported that in the psychological intervention group, 33% (7/21) of participants developed a foot ulcer compared with 50% (10/20) in the control group (n = 41, RR 0.67, 95% CI 0.32 to 1.41), very low‐certainty evidence). Vedhara 2012 reported a foot ulcer recurrence of 12.5% (1/8) in the psychological intervention group and 20% (1/5) in the control group (n = 13, RR 0.63, 95% CI 0.05 to 7.90), very low‐certainty evidence). It is unclear whether a psychological intervention compared with standard care has any impact on foot ulcer recurrence as we assessed the level of evidence as very low (downgraded by two levels due to high risk of performance (2 studies), detection (1 study), reporting (1 study) and other risk of biases (1 study) and unclear risk of selection (2 studies), detection, reporting and other biases and downgraded by two levels due to very serious imprecision (very small sample size and 95% CI estimate of effect included both little or no effect and appreciable benefit or harm)). See Table 1.

Secondary outcomes

Amputations (major or distal)

Not reported.

Health‐related quality of life

Two trials reported quality of life (Imran 2018; McBride 2016).

McBride 2016 reported health‐related quality of life (HRQOL) at 12 weeks post‐intervention using the European Quality of Life Scale (EQ‐5D). The standard deviation (SD) of mean change for each group was not reported, therefore we have presented the comparison of group mean quality of life score at 12 weeks. The mean HRQOL score for the psychological intervention group was 69.17 (SD 20.05) and for the control group was 63.65 (SD 22.78) (one trial, n = 56, MD 5.52, 95% CI ‐5.80 to 16.84, very low‐certainty evidence). It is unclear whether a psychological intervention compared with standard care has any impact on HRQOL as we assessed the level of evidence as very low (downgraded by two levels due to high risk of performance, detection and reporting bias and unclear risk of attrition bias and downgraded by one level due to serious imprecision (very small study and information is likely to be insufficient i.e. total number of participants is less than 400)).

A second study (Imran 2018) reported changes in the quality of life over six weeks using the Quality of Life Short Form questionnaire consisting of 17 items with a score range of 17‐85 (a high score indicating good quality of life). According to the authors, there was no statistically significant difference at two, four or six weeks between groups. The results were reported at each time point but those whose wounds had healed or dropped out of the study were excluded in the final analysis leaving only 16 in the intervention group and 13 in the control group. The direction of change was unclear and there were no data on actual baseline or end of study period quality of life scores and the results were therefore not presented in a manner suitable for inclusion in this review.

Self‐efficacy

One trial (Imran 2018) reported changes in self‐efficacy as assessed "by adapting an instrument, consisting of seven items with a score range of 7‐28 ‐ a high score indicated good quality of life", but it should be noted that quality of life was reported elsewhere using another tool. For this outcome, from baseline to end of week six, the mean change in self‐efficacy was ‐1.4 (+/‐ 4.0) in the intervention group versus ‐1.5 (+/‐ 3.5) in the control group, P = 0.987. Data were analysed on a per protocol basis as only 16 remained in the intervention and 13 in the control groups. However, as baseline scores or end of study period scores were not provided in a format suitable for data extraction in this review it is unclear if self‐efficacy improved healing outcomes in DFU.

Cost

Not reported.

Comparison 2: Psychological intervention compared with another psychological intervention (one trial).

One trial (16 participants) contributed to this comparison (Rice 2001).

Primary outcomes

Proportion of wounds completely healed

One study (Rice 2001), which compared biofeedback‐assisted relaxation training with relaxation training alone, reported the primary outcome, proportion of wounds completely healed. The number of healed ulcers were measured by digital planimetry (computerised area measurement) at three months. In the intervention group, 87.5% (7/8) of participants achieved complete wound healing whereas in the control group 37.5% (3/8) participants achieved this (one trial, n = 16, RR 2.33, 95% CI 0.92 to 5.93, very low‐certainty evidence). It is unclear whether one psychological intervention versus another psychological intervention has any impact on wound healing as we assessed the level of evidence as very low (downgraded by two levels due to high risk of selection (allocation concealment), performance and other biases and unclear risk of bias for selection (sequence generation) and reporting biases and downgraded by two levels due to very serious imprecision (very small sample size and 95% CI estimate of effect included both little or no effect and appreciable benefit or harm)). See Table 2.

Time to complete wound healing

This outcome was reported for the whole cohort in one study (Rice 2001), which included participants without diabetes and therefore did not contribute data to this outcome and comparison.

Length of time to recurrence

Not measured.

Number of foot ulcer recurrences

Not measured.

Secondary outcomes

Amputations (major or distal)

Not reported.

Health‐related quality of life

Not reported.

Self‐efficacy

Not reported.

Cost

Not reported.

Discussion

Summary of main results

Overall, the findings of the review show the dearth and the low certainty of existing evidence on the effects of psychological interventions for wound healing and preventing ulcer recurrence in people with DFU.

We identified seven studies that met the inclusion criteria for this review (Barth 1991; Imran 2018; Jansen 2009; McBride 2016; Rice 2001; Skafjeld 2015 and Vedhara 2012). Meta‐analysis was not possible due to heterogeneity across comparators and outcomes. Details of included studies have been combined in a narrative synthesis by type of comparator and by primary outcomes. We explored two comparisons (1) psychological Intervention compared with standard care and (2) psychological intervention compared with another psychological intervention.

Very low certainty evidence means we are uncertain whether there is a difference between a psychological intervention or standard care for people with diabetic foot ulceration in complete wound healing, foot ulcer recurrence after one year or health‐related quality of life.

We are also uncertain whether there is a difference in complete wound healing in people with diabetic foot ulceration depending on which type of psychological intervention they receive.

Time to complete wound healing was reported in two studies but not in a way that was suitable for inclusion in this review. One trial reported self‐efficacy but data were not reported in a way that enabled us to include them in this review. No studies explored the other primary outcome (time to recurrence) or secondary outcomes (amputations (major or distal) or cost).

Overall completeness and applicability of evidence

There is no high quality evidence available to assess the benefits of psychological interventions in the treatment and prevention of recurrence in diabetic foot ulceration. We were unable to undertake any meta‐analysis due to heterogeneity in outcomes and comparators. All seven studies were community‐based. Six (Barth 1991; Imran 2018; Jansen 2009; McBride 2016 ; Skafjeld 2015; Vedhara 2012) used psychological interventions that aimed to enhance the participants’ motivation to take action to achieve the goals required for optimal foot self‐care. Rice 2001 used biofeedback‐assisted relaxation therapy to increase blood flow and thus skin temperature. All seven studies used different approaches and content, although they met the criteria set for inclusion. Only three studies reported wound healing and each used different approaches, the effects of which could not be pooled (Imran 2018; Jansen 2009; Rice 2001). McBride 2016 measured wound healing but the data were not reported in sufficient detail to allow any further interpretation or analysis. Barth 1991 reported foot problems, of which some were ulcers but these data were not reported in a way that allowed data extraction; however, we felt it was important to include it as it met the inclusion criteria for this review. Two studies reported foot ulcer recurrence (Skafjeld 2015; Vedhara 2012). Two studies (Imran 2018; McBride 2016) reported health‐related quality of life. All interventions were delivered by various healthcare practitioners including podiatrists (Barth 1991; Rice 2001; Vedhara 2012), psychologists (Barth 1991; McBride 2016), occupational therapists (Jansen 2009) and nurses (Skafjeld 2015; Vedhara 2012). In one trial (Imran 2018), it was unclear who delivered the intervention.

Quality of the evidence

We assessed the quality of the evidence using the GRADE methodology and results for primary outcomes are presented in the 'Summary of findings' tables (Table 1; Table 2). Across the included studies, evidence for the effects of psychological interventions versus standard care is of very low certainty. In one trial that assessed biofeedback‐assisted relaxation versus relaxation, the evidence is also of very low certainty. The overall certainty of the evidence is very low for the outcomes complete wound healing and foot ulcer recurrence. Due to the interactive nature of the intervention in all of the studies, it is very difficult to blind the intervention staff and the participants to group allocation. However, it should be possible to avoid detection bias by blinding the outcome assessors and three trials achieved this using independent assessors to assess foot and wound outcomes. We judged the other trials as unclear or at high risk for detection bias. We downgraded for very serious imprecision for all the primary outcomes because of the very small sample size and because the confidence intervals included both little or no effect and appreciable benefit or harm.

Potential biases in the review process

We are confident that the search procedure that was followed identified all relevant literature and it is unlikely that we have missed any trials. However, despite comprehensive searching, it is possible that we missed some trials that met our criteria.

Agreements and disagreements with other studies or reviews

Since the publication of the protocol for this review, one other recently published systematic review has been identified which examines psychosocial interventions in the treatment and prevention of recurrence of DFU Norman 2020.  In contrast to this review, interventions for people without a history of DFU were included, hence thirty‐one RCTs were identified of which 24 were prevention studies. Most of the studies used educational interventions. The evidence was found to be of low or very low quality and the authors mirror our recommendation for more high quality research of interventions which are underpinned by theoretical approaches for supporting people with DFU.

Authors' conclusions

Implications for practice.

Self‐care is considered to be an important part of foot care in DFU (IWGDF Working Group 2015) and it is recommended that new interventions are developed and evaluated that address the psychological and behavioural aspects of treating and preventing recurrence of DFU, as the evidence for these is limited (NICE 2016). Findings from this review show uncertainty as to whether psychological interventions over standard care enhance healing and prevent recurrence of diabetic foot ulceration. Practitioners should therefore weigh up the potential benefits of employing psychological interventions when treating individuals with DFU against the current level of evidence. Our conclusions are limited by the very small number of studies and by the lack of reporting of important outcomes. The small number of trials identified are also hampered by very low‐certainty evidence.

Implications for research.

There is a need for high‐quality research to be conducted in this area before conclusions can be made about the usefulness of psychological interventions in improving healing and preventing recurrence of diabetic foot ulceration. Future trials need to be adequately powered with clearly reported baseline wound size, wound healing and recurrence variables. Synthesis of evidence has been hampered by heterogeneity in outcomes measured and reported. Development of a core set of outcomes that should be measured in all trials evaluating the effects of psychological interventions in the care of people DFU would allow trial results to be compared and combined.

History

Protocol first published: Issue 10, 2017
Review first published: Issue 2, 2021

Appendices

Appendix 1. Search Strategies

Cochrane Wounds Specialised Register

1 MESH DESCRIPTOR Foot Ulcer EXPLODE ALL AND INREGISTER
2 MESH DESCRIPTOR Diabetic Foot EXPLODE ALL AND INREGISTER
3 (diabet* near3 ulcer*) AND INREGISTER
4 (diabet* near3 (foot or feet)) AND INREGISTER
5 (diabet* near3 wound*) AND INREGISTER
6 #1 OR #2 OR #3 OR #4 OR #5 AND INREGISTER
7 MESH DESCRIPTOR Behavior and Behavior Mechanisms EXPLODE ALL AND INREGISTER
8 MESH DESCRIPTOR Behavioral Disciplines and Activities EXPLODE ALL AND INREGISTER
9 MESH DESCRIPTOR Psychological Phenomena EXPLODE ALL AND INREGISTER
10 MESH DESCRIPTOR Psychology EXPLODE ALL AND INREGISTER
11 ((cognitive* or interpersonal or systems or famil* or acceptance or commitment or supportive or behavi* or psycho*) near2 (therap* or intervention*)) AND INREGISTER
12 (CBT) AND INREGISTER
13 ((Contract* or goal*) near2 set*) AND INREGISTER
14 (Support* near3 (social or famil* or friend* or profession*)) AND INREGISTER
15 (counsel*) AND INREGISTER
16 (Motivational near3 interview*) AND INREGISTER
17 ((chang* or improv* or increas* or decreas* or reduc* or modif* or activat* or encourag* or control* or manag*) near2 (emotion* or cogniti* or behavi* or style* or self* or relaxation or stress or anxiety or depress*)) AND INREGISTER
18 (biofeedback*) AND INREGISTER
19 (problem* near2 (solv* or sort* or control* or manag*)) AND INREGISTER
20 #7 OR #8 OR #9 OR #10 OR #11 OR #12 OR #13 OR #14 OR #15 OR #16 OR #17 OR #18 OR #19 AND INREGISTER
21 #6 AND #20 AND INREGISTER

The Cochrane Central Register of Controlled Clinical Trials (CENTRAL)

#1 MeSH descriptor: [Foot Ulcer] explode all trees

#2 MeSH descriptor: [Diabetic Foot] explode all trees

#3 (diabet* near/3 ulcer*):ti,ab,kw (Word variations have been searched)

#4 (diabet* near/3 (foot or feet)):ti,ab,kw (Word variations have been searched)

#5 (diabet* near/3 wound*):ti,ab,kw (Word variations have been searched)

#6 {or #1‐#5}

#7 MeSH descriptor: [Behavior and Behavior Mechanisms] explode all trees

#8 MeSH descriptor: [Behavioral Disciplines and Activities] explode all trees

#9 MeSH descriptor: [Psychological Phenomena] explode all trees

#10 MeSH descriptor: [Psychology] explode all trees

#11 ((cognitive* or interpersonal or systems or famil* or acceptance or commitment or supportive or behavi* or psycho*) near/2 (therap* or intervention*)):ti,ab,kw

#12 (CBT):ti,ab,kw

#13 ((Contract* or goal*) near/3 set*):ti,ab,kw

#14 (Support* near/3 (social or famil* or friend* or profession*)):ti,ab,kw

#15 (counsel*):ti,ab,kw

#16 (Motivational near/3 interview*):ti,ab,kw

#17 ((chang* or improv* or increas* or decreas* or reduc* or modif* or activat* or encourag* or control* or manag*) near/2 (emotion* or cogniti* or behavi* or style* or self* or relaxation or stress or anxiety or depress*)):ti,ab,kw

#18 (biofeedback*):ti,ab,kw

#19 (problem* near/2 (solv* or sort* or control* or manag*)):ti,ab,kw

#20 {or #7‐#19}

#21 #6 and #20

Ovid MEDLINE

1 exp Diabetic Foot/

2 exp Foot Ulcer/

3 (diabet$ adj3 ulcer$).ti,ab.

4 (diabet$ adj3 (foot or feet)).ti,ab.

5 (diabet* adj3 wound*).ti,ab.

6 or/1‐5

7 exp "Behavior and Behavior Mechanisms"/

8 exp "Behavioral Disciplines and Activities"/

9 exp psychological phenomena/

10 exp Psychology/

11 px.fs.

12 ((cognitive* or interpersonal or systems or famil* or acceptance or commitment or supportive or behavi* or psycho*) adj3 (therap* or intervention*)).ti,ab.

13 CBT.ti,ab.

14 ((Contract* or goal*) adj3 set*).ti,ab.

15 (support* adj3 (social or famil* or friend* or profession*)).ti,ab.

16 counsel*.ti,ab.

17 (Motivational adj3 interview*).ti,ab.

18 ((chang* or improv* or increas* or decreas* or reduc* or modif* or activat* or encourag* or control* or manag*) adj2 (emotion* or cogniti* or behaviour* or behavior* or style or self* or relaxation or stress or anxiety or depress*)).ti,ab.

19 biofeedback.ab,ti.

20 (problem* adj2 (solv* or sort* or control* or manag*)).ti,ab.

21 7 or 8 or 9 or 10 or 11 or 12 or 13 or 14 or 15 or 16 or 17 or 18 or 19 or 20

22 6 and 21

23 randomized controlled trial.pt.

24 controlled clinical trial.pt.

25 randomi?ed.ab.

26 placebo.ab.

27 clinical trials as topic.sh.

28 randomly.ab.

29 trial.ti.

30 or/23‐29

31 exp animals/ not humans.sh.

32 30 not 31

33 22 and 32

Ovid Embase

1 exp diabetic foot/

2 exp foot ulcer/

3 (diabet* adj3 ulcer*).ti,ab.

4 (diabet$ adj3 (foot or feet)).ti,ab.

5 (diabet* adj3 wound*).ti,ab.

6 or/1‐5

7 exp behavioral science/

8 exp behavior/

9 ((cognitive* or interpersonal or systems or famil* or acceptance or commitment or supportive or behavi* or psycho*) adj2 (therap* or intervention*)).ti,ab.

10 CBT.ti,ab.

11 ((Contract* or goal*) adj3 set*).ti,ab.

12 (Support* adj3 (social or famil* or friend* or profession*)).ti,ab.

13 counsel*.ti,ab.

14 (Motivational adj3 interview*).ti,ab.

15 ((chang* or improv* or increas* or decreas* or reduc* or modif* or activat* or encourag* or control* or manag*) adj2 (emotion* or cogniti* or behavi* or style or self* or relaxation or stress or anxiety or depress*)).ti,ab.

16 biofeedback*.ti,ab.

17 (problem* adj2 (solv* or sort* or control* or manag*)).ti,ab.

18 or/7‐17

19 6 and 18

20 Randomized controlled trials/

21 Single‐Blind Method/

22 Double‐Blind Method/

23 Crossover Procedure/

24 (random* or factorial* or crossover* or cross over* or cross‐over* or placebo* or assign* or allocat* or volunteer*).ti,ab.

25 (doubl* adj blind*).ti,ab.

26 (singl* adj blind*).ti,ab.

27 or/20‐26

28 exp animals/ or exp invertebrate/ or animal experiment/ or animal model/ or animal tissue/ or animal cell/ or nonhuman/

29 human/ or human cell/

30 and/28‐29

31 28 not 30

32 27 not 31

33 19 and 32

Ovid PsycINFO

1 exp DIABETES/

2 exp wounds/

3 1 and 2

4 (diabet* adj3 ulcer*).ti,ab.

5 (diabet* adj3 (foot or feet)).ti,ab.

6 (diabet* adj3 wound*).ti,ab.

7 3 or 4 or 5 or 6

8 exp behavioral sciences/

9 exp behavior/

10 exp treatment/

11 ((cognitive* or interpersonal or systemic or systems or famil* or acceptance or commitment or supportive or behavi* or psycho*) adj2 (therap* or intervention*)).ab,ti.

12 CBT.ti,ab.

13 ((Contract* or goal*) adj3 set*).ti,ab.

14 (support* adj3 (social or famil* or friend* or profession*)).ti,ab.

15 counsel*.ti,ab.

16 (Motivational adj3 interview*).ti,ab.

17 ((chang* or improv* or increas* or decreas* or reduc* or modif* or activat* or encourag* or control* or manag*) adj2 (emotion* or cogniti* or behaviour* or behavior* or style or self* or relaxation or stress or anxiety or depress*)).ti,ab.

18 biofeedback*.ti,ab.

19 (problem* adj2 (solv* or sort* or control* or manag*)).ti,ab.

20 or/8‐19

21 7 and 20

22 clinical trials/

23 (clin$ adj25 trial$).ti,ab.

24 random$.ti,ab.

25 ((randomised adj controlled adj trial$) or (randomized adj controlled adj trial$)).mp.

26 (controlled adj clinical adj trial$).mp.

27 (random adj allocat$).mp.

28 ((singl$ or doubl$ or trebl$ or tripl$) adj25 (blind$ or mask$)).ti,ab.

29 (control$ adj4 trial$).mp.

30 or/22‐29

31 (ANIMALS not HUMANS).sh.

32 30 not 31

33 21 and 32

EBSCO CINAHL Plus

S45 S21 AND S44

S44 S43 NOT S42

S43 S22 OR S23 OR S24 OR S25 OR S26 OR S27 OR S28 OR S29 OR S30 OR S31 OR S32 OR S33 OR S34 OR S35 OR S36

S42 S40 NOT S41

S41 MH (human)

S40 S37 OR S38 OR S39

S39 TI (animal model*)

S38 MH (animal studies)

S37 MH animals+

S36 AB (cluster W3 RCT)

S35 MH (crossover design) OR MH (comparative studies)

S34 AB (control W5 group)

S33 PT (randomized controlled trial)

S32 MH (placebos)

S31 MH (sample size) AND AB (assigned OR allocated OR control)

S30 TI (trial)

S29 AB (random*)

S28 TI (randomised OR randomized)

S27 MH cluster sample

S26 MH pretest‐posttest design

S25 MH random assignment

S24 MH single‐blind studies

S23 MH double‐blind studies

S22 MH randomized controlled trials

S21 S6 AND S20

S20 S7 OR S8 OR S9 OR S10 OR S11 OR S12 OR S13 OR S14 OR S15 OR S16 OR S17 OR S18 OR S19

S19 TI ( (problem* N2 (solv* or sort* or control* or manag*)) ) OR AB ( (problem* N2 (solv* or sort* or control* or manag*)) )

S18 TI biofeedback* OR AB biofeedback*

S17 TI ( ((chang* or improv* or increas* or decreas* or reduc* or modif* or activat* or encourag* or control* or manag*) N2 (emotion* or cogniti* or behaviour* or behavior* or style or self* or relaxation or stress or anxiety or depress*)) ) OR AB ( ((chang* or improv* or increas* or decreas* or reduc* or modif* or activat* or encourag* or control* or manag*) N2 (emotion* or cogniti* or behaviour* or behavior* or style or self* or relaxation or stress or anxiety or depress*)) )

S16 TI (Motivational N3 interview*) OR AB (Motivational N3 interview*)

S15 TI ( (counsel*) ) OR AB ( (counsel*) )

S14 TI ( (support* N3 (social or famil* or friend* or profession*)) ) OR AB ( (support* N3 (social or famil* or friend* or profession*)) )

S13 TI ( (Contract* or goal*) N3 set*) ) OR AB ( (Contract* or goal*) N3 set*) )

S12 TI CBT OR AB CBT

S11 TI ( ((cognitive* or interpersonal or systems or famil* or acceptance or commitment or supportive or behavi*) N2 (therap* or intervention*) ) OR AB ( ((cognitive* or interpersonal or systems or famil* or acceptance or commitment or supportive or behavi*) N2 (therap* or intervention) )

S10 (MH "Psychology+")

S9 (MH "Psychological Processes and Principles+")

S8 (MH "Disciplines, Tests, Therapy, Services+")

S7 (MH "Behavior and Behavior Mechanisms+")

S6 S1 OR S2 OR S3 OR S4 OR S5

S5 TI (diabet* N3 wound*) OR AB (diabet* N3 wound*)

S4 TI ( (diabet* N3 (foot or feet)) ) OR AB ( (diabet* N3 (foot or feet)) )

S3 TI (diabet* N3 ulcer*) OR AB (diabet* N3 ulcer*)

S2 (MH "Foot Ulcer+")

S1 (MH "Diabetic Foot")

US National Institutes of Health Ongoing Trials Register (ClinicalTrials.gov)

psychological OR psychotherapy OR psychotherapies OR psychosocial OR cognitive OR family therapy OR supportive OR behavioural OR behavioral OR counselling OR counseling OR biofeedback OR motivational interviewing OR CBT | Diabetic Foot

psychological OR psychotherapy OR psychotherapies OR psychosocial OR cognitive OR family therapy OR supportive OR behavioural OR behavioral OR counselling OR counseling OR biofeedback OR motivational interviewing OR CBT | Diabetic Foot Ulcer

psychological OR psychotherapy OR psychotherapies OR psychosocial OR cognitive OR family therapy OR supportive OR behavioural OR behavioral OR counselling OR counseling OR biofeedback OR motivational interviewing OR CBT | Foot Ulcer

psychological OR psychotherapy OR psychotherapies OR psychosocial OR cognitive OR family therapy OR supportive OR behavioural OR behavioral OR counselling OR counseling OR biofeedback OR motivational interviewing OR CBT | Foot Ulcer, Diabetic

psychological OR psychotherapy OR psychotherapies OR psychosocial OR cognitive OR family therapy OR supportive OR behavioural OR behavioral OR counselling OR counseling OR biofeedback OR motivational interviewing OR CBT | Ulcer Foot

psychological OR psychotherapy OR psychotherapies OR psychosocial OR cognitive OR family therapy OR supportive OR behavioural OR behavioral OR counselling OR counseling OR biofeedback OR motivational interviewing OR CBT | Diabetic Foot Ulcer Mixed

psychological OR psychotherapy OR psychotherapies OR psychosocial OR cognitive OR family therapy OR supportive OR behavioural OR behavioral OR counselling OR counseling OR biofeedback OR motivational interviewing OR CBT | Diabetic Foot Ulcer Neuropathic

World Health Organization International Clinical Trials Registry Platform

foot ulcer [Title] AND psychological OR psychotherapy OR psychotherapies OR psychosocial OR cognitive OR family therapy OR supportive OR behavioural OR behavioral OR counselling OR counseling OR biofeedback OR motivational interviewing OR CBT [intervention]

foot ulcer [Condition] AND psychological OR psychotherapy OR psychotherapies OR psychosocial OR cognitive OR family therapy OR supportive OR behavioural OR behavioral OR counselling OR counseling OR biofeedback OR motivational interviewing OR CBT [intervention]

foot ulceration [Title] AND psychological OR psychotherapy OR psychotherapies OR psychosocial OR cognitive OR family therapy OR supportive OR behavioural OR behavioral OR counselling OR counseling OR biofeedback OR motivational interviewing OR CBT [intervention]

foot ulceration [Condition] AND psychological OR psychotherapy OR psychotherapies OR psychosocial OR cognitive OR family therapy OR supportive OR behavioural OR behavioral OR counselling OR counseling OR biofeedback OR motivational interviewing OR CBT [intervention]

diabetic foot [Title] AND psychological OR psychotherapy OR psychotherapies OR psychosocial OR cognitive OR family therapy OR supportive OR behavioural OR behavioral OR counselling OR counseling OR biofeedback OR motivational interviewing OR CBT [intervention]

diabetic foot [Condition] AND psychological OR psychotherapy OR psychotherapies OR psychosocial OR cognitive OR family therapy OR supportive OR behavioural OR behavioral OR counselling OR counseling OR biofeedback OR motivational interviewing OR CBT [intervention]

diabetes AND ulcer [Title] AND psychological OR psychotherapy OR psychotherapies OR psychosocial OR cognitive OR family therapy OR supportive OR behavioural OR behavioral OR counselling OR counseling OR biofeedback OR motivational interviewing OR CBT [intervention]

diabetes AND ulcer [Condition] AND psychological OR psychotherapy OR psychotherapies OR psychosocial OR cognitive OR family therapy OR supportive OR behavioural OR behavioral OR counselling OR counseling OR biofeedback OR motivational interviewing OR CBT [intervention]

Appendix 2. The Cochrane tool for assessing risk of bias

1. Was the allocation sequence randomly generated?

Low risk of bias

The investigators describe a random component in the sequence generation process such as: referring to a random number table; using a computer random number generator; coin tossing; shuffling cards or envelopes; throwing dice; drawing of lots.

High risk of bias

The investigators describe a nonrandom component in the sequence generation process. Usually, the description would involve some systematic, nonrandom approach, for example: sequence generated by odd or even date of birth; sequence generated by some rule based on date (or day) of admission; sequence generated by some rule based on hospital or clinic record number.

Unclear

Insufficient information about the sequence generation process provided to permit a judgement of low or high risk of bias.

2. Was the treatment allocation adequately concealed?

Low risk of bias

Participants and investigators enrolling participants could not foresee assignment because one of the following, or an equivalent method, was used to conceal allocation: central allocation (including telephone, web‐based and pharmacy‐controlled randomisation); sequentially‐numbered drug containers of identical appearance; sequentially‐numbered, opaque, sealed envelopes.

High risk of bias

Participants or investigators enrolling participants could possibly foresee assignments and thus introduce selection bias, such as allocation based on: use of an open random allocation schedule (e.g. a list of random numbers); assignment envelopes without appropriate safeguards (e.g. envelopes were unsealed, non‐opaque, or not sequentially numbered); alternation or rotation; date of birth; case record number; any other explicitly unconcealed procedure.

Unclear

Insufficient information provided to permit a judgement of low or high risk of bias. This is usually the case if the method of concealment is not described, or not described in sufficient detail to allow a definite judgement, for example, if the use of assignment envelopes is described, but it remains unclear whether envelopes were sequentially numbered, opaque and sealed.

3. Blinding ‐ was knowledge of the allocated interventions adequately prevented during the study?

Low risk of bias

Any one of the following.

• No blinding, but the review authors judge that the outcome and the outcome measurement are not likely to be influenced by lack of blinding.

• Blinding of participants and key study personnel ensured, and unlikely that the blinding could have been broken.

• Either participants or some key study personnel were not blinded, but outcome assessment was blinded and the non‐blinding of others unlikely to introduce bias.

High risk of bias

Any one of the following.

• No blinding or incomplete blinding, and the outcome or outcome measurement is likely to be influenced by lack of blinding.

• Blinding of key study participants and personnel attempted, but likely that the blinding could have been broken.

• Either participants or some key study personnel were not blinded, and the non‐blinding of others likely to introduce bias.

Unclear

Either of the following.

• Insufficient information to permit judgement of low or high risk of bias.

• The study did not address this outcome.

4. Were incomplete outcome data adequately addressed?

Low risk of bias

Any one of the following.

• No missing outcome data.

• Reasons for missing outcome data are unlikely to be related to true outcome (for survival data, censoring unlikely to be introducing bias).

• Missing outcome data are balanced in numbers across intervention groups, with similar reasons for missing data across groups.

• For dichotomous outcome data, the proportion of missing outcomes compared with the observed event risk is not enough to have a clinically relevant impact on the intervention effect estimate.

• For continuous outcome data, a plausible effect size (difference in means or standardised difference in means) among missing outcomes is not enough to have a clinically relevant impact on the observed effect size.

• Missing data have been imputed using appropriate methods.

High risk of bias

Any one of the following.

• Reason for missing outcome data are likely to be related to the true outcome, with either an imbalance in numbers or reasons for missing data across intervention groups.

• For dichotomous outcome data, the proportion of missing outcomes compared with the observed event risk is enough to induce clinically relevant bias in the intervention effect estimate.

• For continuous outcome data, a plausible effect size (difference in means or standardised difference in means) among missing outcomes is enough to induce a clinically relevant bias in the observed effect size.

• 'As‐treated' analysis done with a substantial departure of the intervention received from that assigned at randomisation.

• Potentially inappropriate application of simple imputation.

Unclear

Either of the following.

• Insufficient reporting of attrition/exclusions to permit a judgement of low or high risk of bias (e.g. number randomised not stated, no reasons for missing data provided).

• The study did not address this outcome.

5. Are reports of the study free of suggestion of selective outcome reporting?

Low risk of bias

Either of the following.

• The study protocol is available and all of the study’s prespecified (primary and secondary) outcomes that are of interest in the review have been reported in the prespecified way.

• The study protocol is not available but it is clear that the published reports include all expected outcomes, including those that were prespecified (convincing text of this nature may be uncommon)

High risk of bias

Any one of the following.

• Not all of the study’s prespecified primary outcomes have been reported.

• One or more primary outcomes is/are reported using measurements, analysis methods, or subsets of the data (e.g. subscales) that were not prespecified.

• One or more reported primary outcomes was/were not prespecified (unless clear justification for their reporting is provided, such as an unexpected adverse effect).

• One or more outcomes of interest in the review is/are reported incompletely so that they cannot be entered in a meta‐analysis.

• The study report fails to include results for a key outcome that would be expected to have been reported for such a study.

Unclear

Insufficient information provided to permit a judgement of low or high risk of bias. It is likely that the majority of studies will fall into this category.

6. Other sources of potential bias

Low risk of bias

The study appears to be free of other sources of bias.

High risk of bias

There is at least one important risk of bias. For example, the study:

• had a potential source of bias related to the specific study design used; or

• has been claimed to have been fraudulent; or

• had some other problem.

Unclear

There may be a risk of bias, but there is either:

• insufficient information to assess whether an important risk of bias exists; or

• insufficient rationale or evidence that an identified problem will introduce bias.

Data and analyses

Comparison 1. Psychological intervention versus standard care.

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1.1 Complete wound healing	2		Risk Ratio (M‐H, Fixed, 95% CI)	Subtotals only
1.2 Number of foot ulcer recurrence	2		Risk Ratio (M‐H, Fixed, 95% CI)	Subtotals only
1.3 Health‐related quality of life	1	56	Mean Difference (IV, Fixed, 95% CI)	5.52 [‐5.80, 16.84]

1.1. Analysis.

Comparison 1: Psychological intervention versus standard care, Outcome 1: Complete wound healing

1.2. Analysis.

Comparison 1: Psychological intervention versus standard care, Outcome 2: Number of foot ulcer recurrence

1.3. Analysis.

Comparison 1: Psychological intervention versus standard care, Outcome 3: Health‐related quality of life

Comparison 2. Psychological intervention versus another psychological intervention.

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
2.1 Complete wound healing	1		Risk Ratio (M‐H, Fixed, 95% CI)	Subtotals only

2.1. Analysis.

Comparison 2: Psychological intervention versus another psychological intervention, Outcome 1: Complete wound healing

Characteristics of studies

Characteristics of included studies [ordered by study ID]

Barth 1991.

Study characteristics
Methods	Design: parallel RCT to determine whether a comprehensive and intensive foot care education programme would have a greater impact on knowledge, compliance, and the number of foot problems than a conventional programme Setting: community, hospital diabetes centres and General Practitioners in Sydney, Australia
Participants	Sample: 70 randomised. 62 completed (Intervention (I)‐33; Control (C)‐29) Inclusion criteria: Type 2 diabetic patients (on any type of treatment); age of onset > 30 years; duration of diabetes > 3 months; duration of current type of treatment > 1 month; suboptimal blood glucose control (glycosylated haemoglobin (HbA1c > 9.5 %, normal reference range 6.0‐9.0%); overweight (body mass index (BMI) > 25 kg/m2); total energy intake as fat > 35 % (the last four criteria were necessary for the dietary component of the programme); no attendance at a diabetes education programme in the previous 6 months; competence in the English language and no major physical or mental disabilities preventing full participation in the programme Exclusion criteria: not reported
Interventions	Intervention group: received four weekly sessions of 1.5‐2.5 h duration. 3 sessions were conducted by a podiatrist and one session was conducted by a psychologist. The motivational techniques used in the programme were based on a cognitive motivation theory of Heckhausen and Kuhl. This theory delineates the mental processes which take place between the very beginning of a motivational tendency (the desire to achieve a particular goal) and the onset of actions directed towards achieving that goal. Control group: designed to reflect current standard educational practice in Australia for people with type 2 diabetes mellitus. The programme totaled 14 hours and was held on three consecutive days with 8‐10 participants per group. A wide range of topics typically covered in standard programmes was included (what is diabetes, complications, diet, exercise, etc.). A 1‐hour session was given by a podiatrist. Main areas covered were: washing, drying, and inspecting feet; cutting toe nails; treating minor foot problems; suitable footwear; and dealing with temperature extremes.
Outcomes	Primary outcomes: foot ulceration but not reported in a way that could be used Secondary outcomes: none measured Timepoints: baseline, 1 month, 3 months and 6 months after completion of the programmes
Notes
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Randomisation procedure not described
Allocation concealment (selection bias)	Unclear risk	Allocation concealment not described
Blinding of participants and personnel (performance bias) All outcomes	High risk	Not described but due to the interactive nature of the intervention, it is reasonable to assume it was not possible for group allocation to be blinded
Blinding of outcome assessment (detection bias) All outcomes	Low risk	Outcome personnel blinded "A documentation sheet was developed to record the results of participants’ foot assessments (the number of foot problems present and the number of foot problems requiring treatment) which were performed by an independent podiatrist, who was not aware of the patients’ experimental conditions" (p.114).
Incomplete outcome data (attrition bias) All outcomes	High risk	62 out of 70 randomised completed. Reasons for missing data described and unlikely to be related to outcome. No ITT analysis performed
Selective reporting (reporting bias)	Unclear risk	No study protocol available. Expected outcomes were reported but they were done in such a way that they could not be included in the review analyses.The study reported the reduction in foot problems at one month as statistically significant but did not provide the numerical data that could be included in the analysis of that outcome (p.115).
Other bias	High risk	"Significantly more participants of the intensive treatment condition had peripheral vascular disease and significantly higher baseline cholesterol levels" (p.112).

Imran 2018.

Study characteristics
Methods	Design: RCT
Participants	Sample: 60 randomised (I‐34, C‐26) Inclusion criteria: participants with type 2 DFU who met the following criteria: age ≥ 26 years, level of haemoglobin A1c (HbA1c) ≥ 6.5%, wound recurrence, and a wound grade of 1–5 according to the Wagner classification
Interventions	Intervention group: participants in the intervention group received local wound management along with Understanding Wellbeing treatment. The first treatment was performed in the clinic; during this treatment, the intervention group was interviewed, with the interviews including five trigger questions that would enhance the participants’ understanding of well‐being. In the second treatment, the participant wrote a diary at home to answer six structured questions. Control group: the control group received local wound management. Participants received wound treatment according to best practice at the wound care protocol Kitamura clinic: wound assessment, cleansing, debridement and wound dressing. Participants also received treatment advice, infection control and medicine according to the signs and symptoms they exhibited.
Outcomes	Primary outcomes: 1) self‐efficacy assessed by adapting an instrument (SE), consisting of seven items with a score range of 7–28 — a high score indicated good quality of life; 2) psychological distress assessed by adapting the psychological distress in diabetes (DDS), consisting of 12 items with a score range of 12–48 — a low score indicated distress; 3) self‐care assessed by the summary of diabetes self‐care activities (SDSCA), consisting of 11 items with a score range of 0–77; the items in this tool measure are diet, physical activity, blood sugar testing, foot care and smoking — a high score indicating good self‐care; 4) quality of life assessed by the wound QOL Short Form (QOL) , consisting of 17 items with a score range of 17–85 — a high score indicated good quality of life. Secondary outcomes: Wound closure or changes in DFU status, assessed by the diabetic foot ulcer assessment scale (DFUAS), consisting of 11 items with a score range of 0–98; subscales of this tool are depth of wound, wound size, size score, inflammation/infection, proportion of granulation tissue, type of necrotic tissue, proportion of necrotic tissue, proportion of slough, maceration, type of wound age, and tunnelling; low scores indicating a healed wound and high scores indicating a severe wound Time points: 2, 4 & 6 weeks
Notes
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	The mechanism of random allocation sequence was implemented by computer.
Allocation concealment (selection bias)	High risk	The computer generated a list of random numbers, which was arranged sequentially and then placed in an envelope and sealed. When eligible participants arrived at the clinic, the envelope was taken from the container and opened. There is the potential for envelopes to be accessed and opened and allocation revealed.
Blinding of participants and personnel (performance bias) All outcomes	High risk	Open blinding (both the researcher and participant knew the treatment that was administered). Those providing the treatment and the patients knew the allocation.
Blinding of outcome assessment (detection bias) All outcomes	High risk	Those evaluating the wound did not appear to be blinded to treatment allocation; as stated previously, this was an open trial.
Incomplete outcome data (attrition bias) All outcomes	High risk	14 participants dropped out of the intervention and seven participants dropped out of the control group. Therefore, the self‐efficacy and quality of life scores were not assessed at the end of the study period for all participants, i.e. week 6.
Selective reporting (reporting bias)	Unclear risk	All outcomes reported however, no study protocol available. Expected outcomes were reported but they were done in such a way that they could not be included in this review's analyses.
Other bias	Unclear risk	At baseline, diabetes duration was longer in the control group: 3.0 vs 8.0 years, P = 0.014.

Jansen 2009.

Study characteristics
Methods	Design: quasi‐randomised control trial (an experimental pre‐test‐post‐test design with an experimental and control group) to determine if an occupational therapy programme tailored by the Model of Creative Ability (MCA) achieved better treatment outcomes as compared to a traditional occupational therapy treatment (treatment‐as‐usual) regimen Setting: patients attending the Pretoria Academic Hospital diabetic foot clinic with diabetic foot ulcerations
Participants	Sample: total randomised 16 (I‐7 C‐9) Six lost to follow‐up (I‐1 C‐5). Four defaulted on treatment (I‐1 C‐3) and 2 from the control group underwent amputation. Inclusion criteria: adults with diabetic foot ulceration attending the PAH Diabetic Foot Clinic with diabetic foot ulcerations; new or previously untreated diabetic foot ulcers Exclusion criteria: healed ulcers or healed amputations; ulcers not related to the complications of diabetes
Interventions	Intervention: tailored occupational therapy treatment presented to the participant based on their level of motivation. Underpinned by Du Toit’s Model of Creative Ability (MCA) which provided a means to measure the level of motivation and treatment guidelines based on the level of motivation and action of the participant. The Creative Participation Assessment (CPA) was used to assess the participant's level of motivation. The intervention was implemented based on the level of motivation using the principles outlined in the MCA and included 6 sessions covering four main topics: a ‘diabetic foot’ and how the ulcer developed; knowledge and execution of correct foot hygiene and nail care; correct shoe wear and compensation for loss of sensation; adaptation to daily routine at home/work and discussion about mobility difficulties. Patients were also engaged in activities that were relevant to their standard care; these included mobility exercises, activities such as basic foot and nail care, basic activities of daily living and mobility. The delivery of topics was tailored to the individual depending on the result of the CPA. Participants classified as either passive or imitative; participation and treatment programme implemented accordingly Control: educating the participant about foot care (as part of standard hygiene and personal management) using a pamphlet; use of assistive devices; reinforcing treatment prescribed by the doctor and other team members; a written exercise programme to improve mobility and blood circulation and to maintain range of motion in preparation to do personal management tasks and move around the community
Outcomes	Primary outcomes: complete wound healing Secondary outcomes: none measured Time points: baseline and end intervention (3 months)
Notes	We extracted the proportion of ulcers healed from Table III, page 31. Ulcers reported as 0 mm2 were considered fully healed.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	High risk	"subjects were alternately assigned to the experimental or the control group as they were referred to occupational therapy".
Allocation concealment (selection bias)	High risk	"subjects were alternately assigned to the experimental or the control group as they were referred to occupational therapy".
Blinding of participants and personnel (performance bias) All outcomes	High risk	Not described but due to the interactive nature of the intervention, it is reasonable to assume it was not possible for group allocation to be blinded.
Blinding of outcome assessment (detection bias) All outcomes	Low risk	Doctor assessing wound healing blinded
Incomplete outcome data (attrition bias) All outcomes	High risk	1/7 participants were lost to follow‐up from the experimental group; 3/9 were lost from the control group. Unclear if intention‐to‐treat analysis used
Selective reporting (reporting bias)	Unclear risk	No protocol for the study available
Other bias	Low risk	No other sources of bias noted

McBride 2016.

Study characteristics
Methods	Design: parallel RCT to test whether the use of an intervention to facilitate shared decision‐making (Decision Navigation) increased decision self‐efficacy and adherence to foot treatment versus standard care in patients with a diabetic foot ulcer Setting: participants were recruited from a single‐site diabetes foot clinic in the Edinburgh Royal Infirmary, UK.
Participants	Sample: 56 patients with diabetes (I‐30, C‐26) Inclusion Criteria: newly diagnosed DFU; and/or considering a new treatment (as identified by a consultant clinician); did not display the normal anticipated healing rate (≤ 30% improvement in ulcer healing rate 2 weeks post‐initial contact with the diabetes foot clinic); any type of diabetes Exclusion Criteria: unable to give informed consent; severe ischaemic foot ulcer; identifiable severe psychiatric morbidity; < 16 years old
Interventions	Control and intervention group: received standard care which comprised: formal assessment of ulcer; treatment plan formed incorporating the use of debridement, off‐loading, infection control and/or a vascular intervention; treatment advice; and attendance at foot clinic at regular intervals for check‐ups as advised by the clinician Intervention: Decision Navigation, which is a multi‐component intervention developed to promote informed treatment decision‐making with the following four components: information booklet (treatment decision aid); personalised patient goals; consultation; audio‐recording and written summary of the consultation. Decision Navigation was delivered to participants (n = 30) by a trainee health psychologist and four assistant psychologists (navigators), who were formally trained in the methods by the founder of the intervention who also conducted fidelity tests. The navigators collaborated by telephone with the participants to produce a consultation plan which was then forwarded to their diabetic foot consultant physician. The navigators accompanied the participant to the appointment with the consultant foot care clinician and the subsequent consultation was audio‐taped. Following the consultation, the participants received an audio copy of the consultation and a written summary of the consultation from the navigator within 4 days. The navigators received one day of training and weekly case supervision from the founder of the intervention for a 2‐month period.
Outcomes	Primary outcomes: wound healing rate (at baseline (immediately after recruitment) and at 14 weeks defined as difference in ulcer size over time measured by means of planimetry in cm² but this data was not reported in the findings, see Table 4) Secondary outcomes: health‐related quality of life Time points: baseline (immediately after recruitment), 2 weeks after recruitment) and 14 weeks post recruitment
Notes	Contacted authors for wound healing data. Data not available
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Random sequence was determined by an external researcher through the use of a computer‐generated random number table
Allocation concealment (selection bias)	Low risk	To ensure allocation concealment from the research team and participants, opaque, sealed envelopes were employed which contained a note of group allocation. Envelopes were opened by a member of the research team after baseline measures had been taken.
Blinding of participants and personnel (performance bias) All outcomes	High risk	Not described but due to the interactive nature of the intervention, it is reasonable to assume it was not possible for group allocation to be blinded.
Blinding of outcome assessment (detection bias) All outcomes	Unclear risk	Blinding of podiatrist assessing wound healing not stated
Incomplete outcome data (attrition bias) All outcomes	Unclear risk	Stated they carried out an intention‐to‐treat analysis on 49 out 56 participants who completed follow‐up. On closer inspection, wound healing rate was analysed on 45 participants and it was unclear what the reasons for the missing data were (see Table 4).
Selective reporting (reporting bias)	High risk	No study protocol available. Stated in the methods section that ulcer size was measured at baseline and follow‐up and that the analysis revealed no significant differences for wound healing rate. However, the raw wound healing data section was blank in Table 4.
Other bias	Low risk	No other sources of bias noted

Rice 2001.

Study characteristics
Methods	Design: parallel randomised controlled trial to determine the effect of biofeedback‐assisted relaxation training on foot ulcer healing Setting: departments of podiatric medicine at the following institutions in Wisconsin: 1) University of Wisconsin Hospital and Clinics, Madison; 2) Kalker Podiatric Medicine Clinic, Middleton; 3) Gundersen Clinic, La Crosse; and 4) Skemp Clinic, La Crosse
Participants	Sample: 32 patients with or without diabetes and chronic non‐healing ulcers on the lower extremities. Randomisation was stratified according to diabetes hence the decision to include in this review (with diabetes I‐8 C‐8). Inclusion criteria: ulcer duration of more than 8 weeks; continuous care by a podiatric physician for more than 2 months prior to entering the study; ulcer after debridement categorised as class 2 through 6 under the Seattle Wound Classification System (class 2 refers to an ulcer or abscess due to acute soft‐tissue infection, and class 6 refers to a full‐thickness ulcer covered by eschar) Exclusion criteria: bone involvement and osteomyelitis; patients who later needed reconstructive or vascular surgery or who did not meet the criteria for compliance with experimental management
Interventions	Intervention: in addition to usual care, all patients in the experimental group received one session of training in biofeedback‐assisted relaxation. Homework for the experimental group used a 16‐minute audiocassette tape recording of the relaxation technique for a minimum of 5 days each week. Patients monitored their success at relaxing in the home environment by attaching an alcohol thermometer to the fleshy part of a great toe. Great toe skin temperature was measured both before and after relaxation (assisted thermal biofeedback) and recorded. The temperature logs were collected biweekly to verify compliance with the intervention. The temperature values were used for patient participation and encouragement. Patients were aware that successful relaxation would increase blood flow and thus skin temperature. They were encouraged to remember the sensations of comfort and warmth that occurred with the relaxation and to recall and strengthen those feelings in future sessions. Control: patients randomised to the control group were instructed to relax 15 to 20 minutes daily using a self‐selected method of relaxation (e.g. listening to music, watching television, or daydreaming) while off their feet.
Outcomes	Primary outcomes: complete wound healing Secondary outcomes: none measured Time points: baseline and end of intervention (3 months)
Notes
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Random sequence generation procedure not described. "Patients with non‐healing lower limb ulcerations were randomly assigned to the experimental or control group as they were seen in the clinics and followed for 3 months".
Allocation concealment (selection bias)	Unclear risk	Did not specify how allocation was concealed from those allocating participants
Blinding of participants and personnel (performance bias) All outcomes	High risk	Not described but due to the interactive nature of the intervention, it is reasonable to assume it was not possible for group allocation to be blinded.
Blinding of outcome assessment (detection bias) All outcomes	Low risk	The podiatric physicians were blinded to whether or not the patients were in the experimental intervention or control group.
Incomplete outcome data (attrition bias) All outcomes	Low risk	No reported withdrawals; no missing data
Selective reporting (reporting bias)	Unclear risk	No protocol for the study available
Other bias	High risk	Significant baseline imbalance in ulcer perimeter (see table 2, p. 136 mean ulcer perimeter at baseline: Intervention group (mean 50.3 mm) vs control (27.2 mm) P = 0.01)

Skafjeld 2015.

Study characteristics
Methods	Design: parallel randomised controlled trial to test the feasibility of monitoring foot temperature in combination with theory‐based counselling with the ultimate aim of preventing recurrent diabetic foot ulcers Setting: participants recruited from six diabetes specialty outpatient clinics and one chiropodist in Oslo, Norway
Participants	Sample: 41 participants with type 1 and type 2 diabetes with previous history of foot ulcer Inclusion criteria: Type 1 or type 2 diabetes; age 18–80 years; belonging to group 3 of the Diabetes Foot Risk classification system (previous history of foot ulcer and peripheral neuropathy); capable of providing informed consent and completing a written questionnaire; willing to measure foot skin temperature if assigned to the intervention group Exclusion criteria: patients with open wounds, active Charcot disease, active osteomyelitis, or ischaemia (not palpable pulses or ankle/arm index < 0.7)
Interventions	Control and intervention group received the following standard care: instructed to inspect their feet under, below and between the toes, and record their observations in a log book daily; contact the study nurse if changes in their feet including a new ulcer were observed; advised to always wear their customised footwear; for general medical and diabetes care, they consulted their usual general practitioners. Intervention group also received the following: counselling tailored to the stage of change underpinned by the Transtheoretical Model (TTM) training to use a digital infrared thermometer to monitor foot temperature and how to conduct daily self‐monitoring of skin temperatures. If subjects observed a difference in skin temperature of > 2.0° C (2.2° C = 4° F) on the same spot in the sole of the foot compared to the same spot on the opposite foot on two consecutive days, they were advised to contact the study nurse and to reduce physical activity by one‐half until the temperature difference normalised (to < 2.0° C). record daily physical activity using a step‐counter during the first week of the study at three‐month visits, the study nurse assessed each subject’s readiness to record skin temperatures according to TTM stages, followed by tailored stage‐based counselling.
Outcomes	Primary outcomes: foot ulcer recurrence was defined as an end point in the study and was classified according to Wagner foot classification system. Time points: baseline and at study end Secondary outcomes: none measured
Notes
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Block randomisation was used to assign each four subjects to blocks with two in each group. Randomisation was stratified for patients with a history of Charcot foot, who have an extra high risk of recurrence. All study‐related procedures including randomisation were performed at the Diabetes Clinic, Oslo University Hospital. There was no specification as to how the random sequence was generated, but we assumed it was computer‐generated
Allocation concealment (selection bias)	Unclear risk	Did not specify how allocation was concealed from those allocating participants
Blinding of participants and personnel (performance bias) All outcomes	High risk	Not described but due to the interactive nature of the intervention, it is reasonable to assume it was not possible for group allocation to be blinded.
Blinding of outcome assessment (detection bias) All outcomes	High risk	The study nurse performed all assessments plus patients completed questionnaires. Neither the patients or the nurse were blinded to treatment group.
Incomplete outcome data (attrition bias) All outcomes	Low risk	Foot ulcer recurrence data were complete.
Selective reporting (reporting bias)	Unclear risk	No protocol for the study available
Other bias	High risk	Reported a significantly higher level of nephropathy (P < 0.01) and a higher level of systemic vascular risk factors in the intervention group at baseline (see table 1). Both these baseline imbalances may have happened by chance but this could suggest an issue with randomisation procedures.

Vedhara 2012.

Study characteristics
Methods	Design: parallel randomised controlled trial to test the feasibility of a psychosocial intervention for modifying psychosocial risk factors associated with foot re‐ulceration in diabetes Setting: participants recruited from specialist podiatry services in secondary care
Participants	Sample: 15 individuals who had been discharged with a healed ulcer in the previous 12 months Inclusion Criteria: diagnosed with type 1 or type 2 diabetes; exhibited evidence of psychological distress (score 6 on the Hospital Anxiety Depression Scale); had a history of foot ulceration; without a current ulcer, and had at least one foot Exclusion Criteria: diagnosed with Charcot foot; suffering from severe mental illness e.g. psychosis; had insufficient communication skills in the English language to participate in the intervention, or any other severe communication difficulties
Interventions	Intervention: individuals allocated to the intervention group participated in a group‐based psychosocial intervention. The intervention was conducted in two phases. The first phase sought to initiate changes in the psychosocial risk factors identified in the therapeutic model which influence diabetic foot ulceration outcomes and included behavioural, emotional, cognitive and social factors. The second aimed to maintain any gains made in the first phase of treatment. The first phase took place once a week over a ten‐week period. The second phase commenced two months after the end of the first phase and involved three maintenance sessions held at two‐month intervals. All sessions lasted between 60 and 90 min. Control: participants received ‘usual care’ during the period of the intervention which involved being discharged back into primary care for the management of their feet.
Outcomes	Primary outcomes: re‐ulceration Secondary outcomes: none measured Time points: within the 6‐month follow‐up period after intervention ended
Notes
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Random sequence generation procedure not described; "eligible individuals were randomised 2:1 to the intervention or control groups".
Allocation concealment (selection bias)	Unclear risk	Allocation concealment not described
Blinding of participants and personnel (performance bias) All outcomes	High risk	Not described but due to the interactive nature of the intervention it is reasonable to assume it was not possible for group allocation to be blinded.
Blinding of outcome assessment (detection bias) All outcomes	Unclear risk	Not described
Incomplete outcome data (attrition bias) All outcomes	Low risk	No missing data
Selective reporting (reporting bias)	High risk	No protocol for the study was available. Re‐ulceration reported but this was not planned in the methods
Other bias	Unclear risk	Education level appeared higher in the intervention group and there were a higher number of ulcers in the control group but no analysis of baseline imbalance reported (p.327)

^{CPA: Creative Participation Assessment
DFUAS:Diabetic Foot Ulcer Assessment Scale 
DDS: Diabetes Distress Scale
DFU: Diabetic Foot Ulceration
HbA1c: Glycosylated Haemoglobin
ITT: Intention to Treat
QOL: Quality of Life
SDSCA:Summary of diabetes self‐care activities
SE: Self‐efficacy
TTM:Transtheoretical Model of Behaviour change}

Characteristics of excluded studies [ordered by study ID]

Study	Reason for exclusion
Bahador 2017	No psychological component in the intervention
Bian 2012	No psychological component in the intervention
Borges 2004	Did not include participants with DFU or a history of DFU
Borges 2008	Did not include participants with DFU or a history of DFU
Cisneros 2010	No psychological component in the intervention
Corbett 2003	Did not include participants with DFU and no psychological component in the intervention
Fardazar 2018	Did not include participants with DFU or a history of DFU
Frank 2003	Did not include participants with DFU or a history of DFU
Fresinus 2009	Education only. Did not have a psychological intervention
Gershater 2011	Education only. Did not have a psychological intervention
Huang 2015	Did not include participants with DFU or a history of DFU
IRCT20180726040606N1	Education only. Did not have a psychological intervention
Keukenkamp 2018	Footwear adherence intervention
Kim 2016	Education only. Did not have a psychological intervention
Kirupa 2015	No psychological component in the intervention
Kruse 2010	Physical activity intervention to increase physical activity and reduce risk of falls
LeMaster 2003	No psychological component in the intervention
LeMaster 2008	Lifestyle physical activity intervention to increase physical activity ‐ ulcer incidence was measured to ascertain if weight‐bearing activity increased the risk of foot ulceration in people with diabetes and insensate feet
Liang 2012	Did not include participants with DFU or a history of DFU
Lincoln 2007	No psychological component in the intervention
Litzelman 1993	Education only. Did not have a psychological intervention
Martinez Marcos 1999	Did not include participants with DFU or a history of DFU
Monami 2015	No psychological component in the intervention
Qin 2015	Did not include participants with DFU or a history of DFU
Rosario 2017	Abstract only. Did not include participants with DFU or a history of DFU
Seyyedrasooli 2015	Did not include participants with DFU or a history of DFU
Wei 2016	Did not include participants with DFU or a history of DFU
Xiao 2017	Did not include participants with DFU or a history of DFU
Zhenghua 2011	Abstract only. No psychological component in the intervention

DFU: Diabetic Foot Ulceration

Characteristics of studies awaiting classification [ordered by study ID]

McMurray 2002.

Methods	Quasi‐RCT
Participants	Type 1 or Type 2 diabetes with end‐stage renal disease requiring dialysis, n = 83
Interventions	Diabetes care management including education, monitoring and motivational coaching, n = 45 versus standard diabetes care, n = 38
Outcomes	Self‐management knowledge, quality of life, self‐management behaviours, glycaemic control, foot risk assessment, amputations
Notes	Contacted authors to clarify if population met the review selection criteria ‐ no response received

Simson 2008.

Methods	RCT
Participants	Diabetic foot syndrome and comorbid depression, n = 30
Interventions	Supportive psychotherapy n = 15 versus standard treatment n = 15
Outcomes	Hospital Anxiety and Depression Scale, Problem Areas in Diabetes Scale, Diabetic Foot Syndrome (Wagners classification)
Notes	Authors contacted to check if they measured any outcomes relevant to this review ‐ no response received

Characteristics of ongoing studies [ordered by study ID]

IRCT20100628004256N8.

Study name	Implementation of nurse‐led care on quality of care and healing of diabetic foot ulcer
Methods	Design: RCT
Participants	Patients with type 1 & 2 diabetes including men and women 18 years of age and older with foot ulcer
Interventions	Intervention group: nurse‐led interventions will be done with integrated, interdisciplinary and comprehensive steps. In the integrated phase, care delivery will be step by step and in accordance with the guideline re care for diabetic foot ulcer. In the interdisciplinary phase, the counselling required by the treatment team will be carried out. In the comprehensive phase, supportive and educative care services will be provided to the participant and their caregiver depending on the patient's need. Control group: treated and provided with routine care which includes control of blood sugar, appropriate antibiotic treatment if it is necessary, off‐loading foot ulcer, removing the foot ulcer and dressing with normal saline.
Outcomes	Primary outcome: quality of care. Time point: before intervention, 6 weeks after intervention and follow‐up 12 weeks after the intervention. Method of measurement: QUALPAC (Quality Patient Care Scale) Secondary outcome: healing of diabetic foot ulcer. Time point: before intervention/6 weeks after intervention and follow‐up 12 weeks after the intervention. Method of measurement: Wound Assessment Check list
Starting date	January 2019
Contact information	na.samadi@arums.ac.ir
Notes

NCT03909802.

Study name	Effect of self‐ and family management of diabetic foot ulcers programs on health outcomes
Methods	Design: RCT
Participants	Individuals diagnosed with type 2 diabetes and DFU by certified wound care nurses An individual who either regularly or irregularly visits the selected wound care clinics Individual not in the fasting period either in Ramadan or another occasion An individual who is able to read and communicate in the Indonesian language
Interventions	Intervention group: the intervention combines self‐management and family management focussing on behavioural changes in patients with diabetic foot ulcers. Self‐management focusses on intensive health education of physical activities, diet, medication, foot care, and blood glucose control. Family management focusses on problem‐solving, establishing roles, and effective involvement in providing care amid individuals suffering DFU.
Outcomes	Primary outcome: the purpose of this study is to investigate the effect of a self‐ and family management of DFU programmes on health outcomes as follows: behaviours (adherence to physical activities, medications, diet, foot care, and blood glucose control), family supports, biomarkers (HbA1c, wound size) in patients suffering DFU in Indonesia.
Starting date	April 2019
Contact information	Sumarno Adi Subrata, PhD (Candidate), Universitas Muhammadiyah
Notes

NL7406.

Study name	The effect of psychological intervention on the quality of life and wound healing in patients with diabetic foot ulcer
Methods	RCT
Participants	Patients visiting the multidisciplinary outpatient diabetic foot clinic with ≥ 1 foot ulcer on or below the malleolus, who are diagnosed with diabetes mellitus and aged between 18 and 65 years are eligible for inclusion.
Interventions	One group of patients will receive therapy for 45 minutes once a week for 8 weeks. The therapy includes cognitive behavioral therapy, lifestyle and stress management, coping behavior therapy and treatment for depression. The other group will not receive psychological therapy.
Outcomes	The primary outcome is the score on the Short form 36 (SF‐36) 6 months after finishing psychological intervention. Secondary outcomes are the result of Diabetic Foot Ulcer scale short form (DFS‐SF) after finishing therapy and after 6 months follow‐up, the occurrence of wound healing and the occurrence of minor and major amputation.
Starting date	January 2019
Contact information	Not available
Notes

HbA1c: Glycosylated Haemoglobin
QUALPAC: Quality Patient Care Scale

SF‐36: Short form 36

Differences between protocol and review

Planned meta‐analyses, subgroup and sensitivity analyses could not be performed, as we included only seven trials which reported different comparators and outcomes. Instead, details of included studies have been combined in a narrative synthesis according to type of comparator and by primary outcomes. We updated our search terms by removing unhelpful search terms that identified many irrelevant studies regarding educational interventions in the original search, and adding terms for psychological therapies.

Contributions of authors

Helen McGloin: conceived the review; designed the review; coordinated and developed the review; screened studies for inclusion; extracted data; checked quality of data extraction; analysed or interpreted data; undertook quality assessment; checked quality assessment; performed statistical analysis; checked quality of statistical analysis; produced the first draft of the review; contributed to writing or editing the review; advised on the review; wrote to study authors/experts/companies; approved final review prior to publication; is guarantor of the review.

Declan Devane: designed the review; contributed to the co‐ordination of the review process; provided methodological support; extracted data; checked quality of data extraction; analysed or interpreted data; undertook quality assessment; checked quality assessment; performed statistical analysis; checked quality of statistical analysis; produced the first draft of the review; contributed to writing or editing the review; advised on the review; approved final review prior to publication.

Caroline McIntosh: designed the review; screened studies for inclusion; extracted data; checked quality of data extraction; analysed or interpreted data; produced the first draft of the review; contributed to writing or editing the review; advised on the review; approved final review prior to publication.

Kirsty Winkley: designed the review; screened studies for inclusion; acted as third author for agreeing study inclusion; analysed or interpreted data; produced the first draft of the review; contributed to writing or editing the review; advised on the review; wrote to study authors/experts/companies; approved final review prior to publication.

Georgina Gethin: conceived the review; designed the review; contributed to the co‐ordination of the review process; screened studies for inclusion; extracted data; checked quality of data extraction and entry; analysed or interpreted data; produced the first draft of the review; contributed to writing or editing the review; advised on the review; approved final review prior to publication.

Contributions of the editorial base

Anne‐Marie Glenny (Editor): edited the protocol; advised on methodology and protocol content; approved the final protocol prior to submission.

Jo Dumville (Joint Co‐ordinating Editor): edited the review; advised on methodology and review content; approved the final review prior to submission.

Gill Rizzello (Managing Editor): co‐ordinated the editorial process; advised on content; edited the protocol and the review.

Naomi Shaw and Sophie Bishop (Information Specialists): designed the search strategy, ran the searches and edited the search methods section.

Ursula Gonthier and Tom Patterson (Editorial Assistants): edited the reference sections.

Sources of support

Internal sources

• Division of Nursing, Midwifery and Social Work, School of Health Sciences, Faculty of Biology, Medicine and Health, University of Manchester, UK

External sources

• National Institute for Health Research (NIHR), UK

  This project was supported by the National Institute for Health Research, via Cochrane Infrastructure funding to Cochrane Wounds. The views and opinions expressed are those of the authors and not necessarily those of the NIHR, NHS or the Department of Health and Social Care.

Declarations of interest

Helen McGloin: none known.

Declan Devane: none known.

Caroline McIntosh: none known.

Kirsty Winkley: I receive consulting fees/honoraria from ValoTec. I am co‐applicant on a NIHR Programme Development grant to reduce the impact of diabetic foot ulceration.

Georgina Gethin: I received a grant from Curam/Fleming Medical and honoraria for presenting at industry‐sponsored education meetings from Healthcare 21, Lohmann & Rauscher and KCI Acelity in topics unrelated to this review.

New