A collaborative approach in patient education for diabetes foot and wound care: A pragmatic randomised controlled trial

Abstract

Foot care education is an important strategy in reducing lower limb complications. There is evidence that contemporary communication approaches can improve patient education outcomes. To inform the potential of such methods in diabetic foot education, we trialled a collaborative approach in patient education counselling in a podiatry clinic. We conducted a single‐blind pragmatic randomised controlled trial on 52 diabetes patients who had an active foot ulcer. Participants were randomised to either collaborative education or traditional didactic education. Outcomes on knowledge and self‐care behaviours were collected via a pre and post study questionnaire (max score: 75). The study ended at 12 weeks or when the wound healed prior. 42 (80.7%) participants completed the study. The collaborative patient education group had a significant increase in score post‐study (38.8 ± 8.5) compared to pre‐study (32.8 ± 6.9; P < .001). The control group had no significant increase in score post study. The difference in scores between groups had a moderate effect size (d = 0.54). The use of a collaborative approach in patient education was able to produce significantly greater increase in knowledge retention and self‐care behaviours, without the need for additional consultation time in a podiatry clinic.

1. INTRODUCTION

Diabetes is a chronic lifestyle disease and the leading cause of death and amputations worldwide. In Singapore, an estimated 12.8% of the population ¹ has diabetes and approximately one‐sixth ² of these patients have foot complications related to the disease. Diabetic foot problems disproportionately accounts for an estimated 25% of total healthcare costs related to diabetes in Singapore. ² Poor socio‐economic status, including education and foot care knowledge, has been empirically accepted as a risk factor for diabetic foot complications. ³ , ⁴ Foot care education is one of the strategies employed throughout the world to reduce the incidences and complications of lower limb ulcerations and amputation. ⁵ , ⁶ , ⁷ , ⁸

Traditional didactic education is the default approach for clinicians who are not trained in counselling techniques for behavioural change. The traditional method which focuses largely on imparting knowledge and persuading patients to comply to a set behaviour is found to have little effect on self‐care habits. ⁹ , ¹⁰ Podiatrists are committed to their duties to help guide patients towards better self‐care, but do not necessarily have the skills to do so effectively. ⁹ It is reported that patients prefer to engage in a discussion and restate or rephrase their understanding ¹¹ to own the solutions, instead of a directive or didactic approach whereby the clinician provides “one‐way” information. In fact, directive persuasion is thought to lead to resistance to change and is counter‐effective. ⁹ , ¹² This is also known as “psychological reactance”, whereby a person adopts or strengthens a view or attitude that is contrary to what was intended. ⁹ , ¹²

Compared to traditional didactic education, contemporary approaches such as open‐ended communication, interview style, and a collaborative approach, were found to be advantageous in enacting behaviour change. These contemporary methods better engage patients in their own care and elicit intrinsic motivation to change health behaviours. Collaborative methods involving Motivational Interviewing (MI) and Solutions Focused (SF) counselling, have been found to be effective in various areas of behavioural change such as smoking, diet, physical activity, and addiction. ¹³ , ¹⁴ , ¹⁵ , ¹⁶ , ¹⁷ Such methods are increasingly being used in chronic disease patients: Healthcare workers in various countries have reported the use of patient engagement and collaborative counselling methods in diabetic foot care. ⁶ , ⁹ , ¹⁰ , ¹⁸ , ¹⁹ , ²⁰ , ²¹ However, a review paper by Binning and colleagues ²² on Motivational Interviewing for the prevention of diabetic foot complications was unable to draw strong conclusions due to insufficient and low level of evidence. This reflects a need for good quality studies on the effects of contemporary counselling methods in podiatric diabetic foot care.

While there is sporadic adoption of Motivational Interviewing and Solutions Focused counselling techniques in Asia, its efficacy in the Asian population have not been widely published. The confluence and common strengths of the Motivational Interviewing and Solutions Focused Counselling were described by Lewis & Osborn ¹⁶ : (a) adopting respectful posture by honouring patient stories, (b) tapping on patient's resources, know‐how & intrinsic preferences and incorporating them into tasks and (c) acknowledging that the patient‐clinician relationship is key to the change. The approach holds the belief that “non‐compliance” and “resistance” does not lie within the patient. The clinician plays a supportive role in helping the patients achieve their goals. ²³ Co‐created action plans become fruitful because it is the patients who hold critical beliefs on how change takes place relative to their goals. ¹⁶ , ²⁴ In our study, we applied the blended collaborative patient education ¹⁶ , ²⁵ approach in English language medium. The sample population consisted of a mix of Asian ethnicities, namely Chinese, Malay, and Indian, in a tertiary hospital setting.

The aim of the study was to evaluate the impact of collaborative patient education compared to that of traditional didactic education through a randomised controlled trial (RCT). Based on promising evidence on effective collaborative methods in other patient populations, we hypothesised that patients with diabetes in the collaborative patient education group will have better outcomes in terms of knowledge retention and behaviour modification. This could be a simple and effective way forward in achieving better “patient compliance” through self‐management of chronic diseases, thereby reducing the burden of chronic disease on the healthcare system.

2. METHODS

2.1. Study design

This was a two‐arm pragmatic randomised controlled trial to evaluate the effectiveness of a collaborative patient education in persons with diabetic foot wound. The two arms were (A) Collaborative Patient Education and (B) Traditional Didactic Patient Education as the control. Participants were recruited from a tertiary care clinic specialising in diabetes management in Singapore.

Ethics approval was received from Singhealth Centralised Institutional Review Board (Ref: 2018/2630). The study was also registered with Clinicaltrials.gov: NCT 04278742. All study participants provided consent to take part in the study.

2.2. Randomisation and blinding

Recruitment was done through recruitment posters and clinicians' referrals. Recruited participants were randomly allocated in into either intervention (Group A) collaborative patient education or (Group B) traditional didactic patient education. The randomisation of group allocation was done via an online randomiser. ²⁶ To ensure allocation blinding from participants, sealed envelopes containing a note of group allocation to each subject number were used. Envelopes were opened by a member of the research team after criteria was checked and written consent had been obtained. Participants were not informed whether they were undergoing the traditional didactic education or collaborative patient education. However, they were assured that clinicians attending to them had a minimum of 5 years of clinical experience.

2.3. Inclusion/exclusion criteria

We recruited patients with diabetes who had a foot wound. Patients were eligible for the study if they were at least 21 years of age, have type 2 diabetes, a foot ulcer, independent in activities of daily living (ADL) and able to communicate in English and understand written English. Exclusion criteria were cognitive impairment, blindness or deafness, and current mental health conditions like psychotic disorders or dementia. We also excluded patients whose pedal pulses were not palpable and did not have a minimum toe pressure of 30 mm Hg; This was so as not to confound wound healing which was the targeted study endpoint.

2.4. Attending clinicians

Podiatrists in both intervention and control groups were matched for their years of experience. This was to ensure that experience level of clinicians in both groups were comparable. The clinicians' experience in both groups ranged from 5 to 10 years.

2.5. Intervention group—Collaborative patient education and active listening

With training in essential counselling skills, Solutions Focused counselling and Motivational Interviewing skills enhancement courses, the study team members came together to develop an approach to patient education which involves (a) collaboration with the patient, (b) respecting that patients are the experts of their own lives and (c) drawing out patients' intrinsic self‐motivation and know‐hows to work towards co‐creating next steps in the treatment plan.

The study team developed a patient education communication approach that can be easily adapted into a clinical consultation (Figure 1).

FIGURE 1.

Communication workflow for a collaborative approach to patient education

Whilst the study team decided on the pragmatic approach of talking about the problem or concern of that day (no standardising the patient education content), the areas which were standardised were (a) the approach within each group and (b) essential wound care information is shared as part of the wound consultation. While clinicians are experts in their own fields, patients are experts of their own lives. Patients themselves hold the key to incorporate changes into their daily lives. Therefore, on the clinician's part, active listening, acknowledging patient's efforts, know‐how and constraints is crucial in a collaborative approach in patient education.

In the intervention group, the new method of collaborative communication is delivered alongside standard diabetes wound care treatment such as wound cleansing and dressing, scalpel‐debridement, offloading through paddings, orthotics or boot, as appropriate. No extra time was allocated for intervention counselling; The intervention collaborative communication was delivered within the typical wound consultation and treatment duration of 20 to 30 minutes.

2.6. Control group

The control group received standard diabetes wound care and education which was delivered in the traditional didactic style. Standard diabetes wound care treatment covered areas such as wound cleansing and dressing, scalpel‐debridement, offloading through paddings, orthotics or boot, as appropriate. The duration of a typical wound consultation and treatment for the clinic was 20 to 30 minutes.

2.7. Data collection and management

Demographics such as age, gender, ethnicity, HbA1c, wound size were obtained from patients' electronic medical records. Subject's educational level was obtained during the consent taking and recruitment phase. All data were anonymised. To ensure the accuracy of outcome measures, a study team member was in attendance when the survey was attempted by the patient. The patient had opportunity to clarify the meaning of questions.

2.8. Questionnaire

To our knowledge, there was no available validated questionnaire to examine patient's diabetes knowledge and wound care behaviours. The study team which had representatives from endocrinology, podiatry, social work, and psychology constructed a brief self‐administered questionnaire that includes knowledge, self‐care behaviour and self‐efficacy (personal judgement of confidence in dealing with situations) areas.

2.9. Outcome measure

The questionnaire consisted of 35 questions, with a maximum possible score of 75. A higher score indicated better overall knowledge, behaviour, and efficacy. For behaviour and efficacy domain, the score ranged from 0 (no uptake of behaviour) to 3 (most ideal behaviour). A tiered scoring allowed progression in behaviour change to be reflected: e.g. “sometimes” (score 1) to “most of the time” (score 2) and “all the time” (score 3). In the knowledge domain, each correct answer scored 1 and an incorrect answer scored 0. Each participant completed the questionnaire at the start of the study and at the end of the study, which was week‐12 ⁶ , ¹⁰ , ¹⁸ or when the wound healed, whichever occurred first.

2.10. Statistical analysis

Data were analysed on SPSS Statistics for Windows, version 24.0 (IBM Corp., Armonk, N.Y., USA). The primary outcome of interest was the knowledge and behaviour test scores, which were compared using student's t‐test between groups. Paired t‐test was also conducted to analyse the difference between post‐study and pre‐study within each subject. The internal consistency of the questionnaire was analysed using Cronbach's Alpha. Cronbach's Alpha coefficient of >0.7 reflects satisfactory internal validity. ²⁷ All evaluations were made assuming a two‐sided type 1 error rate set at 0.05.

3. RESULTS

Between December 2018 to February 2020, 60 subjects were recruited. The participants were randomised into two groups: Group A (intervention): 36 subjects and Group B (control): 24 subjects. Of the 60 subjects, 4 withdrew, 2 were lost to follow up, 1 participant's condition worsened and did not fit the criteria and 1 participant's case was escalated for further management by another department. Results were based on 52 subjects who participated in the study, of which, 42 subjects completed the study (Figure 2).

FIGURE 2.

Consort flowchart

3.1. Demographics

Participants had a mean age of 56.9 years and average of 15.2 years from first diagnosis of diabetes with mean HbA1c of 9.1% (Table 1).

TABLE 1.

Group demographics

	Group A intervention	Group B control
N	33	19
Age	55.2 ± 10.7	60.1 ± 10.6
Diabetes (y)	14.7 ± 9.5	16.0 ± 12.5
HbA1c (%)	8.8 ± 1.9	9.5 ± 2.4
Presenting wound size (mm)	15.6 ± 15.4	8.5 ± 14.6
Gender
Female	14 (42.4%)	2 (10.5%)
Male	19 (57.6%)	17 (89.5%)
Educational level
Primary	3 (9.1%)	4 (21.1%)
Secondary	17 (51.5%)	10 (52.6%)
Tertiary	13 (39.4%)	5 (26.3%)
Ethnicity
Chinese	12 (36.4%)	8 (42.1%)
Malay	10 (30.3%)	3 (15.8%)
Indian	9 (27.3%)	7 (36.8%)
Pakistani	1 (3.0%)	1 (5.3%)
Sikh	1 (3.0%)	0 (0.0%)
Note: Data expressed in mean (SD) or count (%)

3.2. Internal consistency of questionnaire tool

An analysis of the responses on the completed surveys indicated that the questionnaire achieved a Cronbach's Alpha value of .861, which reflected good internal consistency.

3.3. Knowledge and self‐care behaviour outcome measure

Table 2 shows the questionnaire scores at baseline. There was no significant difference in baseline scores between groups (P = .21; Table 2). Table 3 compares the results at baseline and post study. Participants in the collaborative education group experienced a significant improvement in post‐study score (P < .001; Table 3). While participants in the control group also experienced some improvement in scores after going through traditional didactic education, the improvement had a small effect size (d = 0.28) ²⁸ and did not reach statistical significance (P = .32; Table 2). The difference between mean scores of the two groups had a moderate effect size of 0.54. ²⁸

TABLE 2.

Pre‐study baseline score for intervention and control groups

	Intervention group	Control group	Difference of mean	P‐value	95% CI for mean difference	Cohen's d effect size (95% CI)
Baseline score	32.8 ± 6.9	35.5 ± 7.6	−2.65	.21	−6.8 to 1.6	0.37 (−0.9 to 0.2)

Note: Results expressed in mean ± SD. t‐test was used to compare group means.

TABLE 3.

Pre‐ and post‐study results for full questionnaire of 35 questions, with maximum score of 75

	Post‐study score	Pre‐study score	Mean difference	P‐value	95% CI for mean difference	Cohen's d effect size (95% CI)
Collaborative education	38.8 ± 8.5	32.8 ± 6.9	5.45	<.001	2.85 to 8.04	0.81 (0.38 to 1.24)
Traditional didactic education (control)	39.1 ± 6.6	35.5 ± 7.6	1.86	.32	−2.03 to 5.75	0.28 (−0.26 to 0.81)
		Difference of means by groups	3.59 d = 0.54 (−0.1 to 1.2)

Note: Results expressed in mean ± SD and analysed using paired t‐test for post‐study and pre‐study differences. t‐test was used to compare group means.

3.4. Confidence perception in wound care

Table 4 gives information on the wound care efficacy perceptions of participants from both groups. The groups were further sorted according to the education level of the subjects. The greatest increase in confidence score (+0.25, P = .72) was experienced by participants in the control group with primary level of education, although not statistically significant. Participants with primary level of education in the intervention group experienced no difference in confidence score. Participants with secondary and above education in the intervention group participants experienced an increase in confidence scores, although not statistically significant. Interestingly, participants with secondary and tertiary level of education in the control group reported a decrease in confidence score (−0.14, P = .36 and  ‐0.33, P = .42 respectively) with respect to self‐management of small wounds. The decrease was also not statistically significant.

TABLE 4.

Results on the “confidence in taking care of a small cut (a small wound)”

Group	Education level	Post‐study score	Pre‐study score	Mean difference	P‐value	95% CI for mean difference
Collaborative education (intervention)	Primary	2.67 ± 0.58	2.67 ± 0.58	0.00 ± 1.00	1.00	(−2.48, 2.48)
	Secondary	2.27 ± 0.80	2.20 ± 0.56	0.07 ± 0.70	.72	(−0.32, 0.46)
	Tertiary	2.40 ± 0.70	2.30 ± 0.68	0.10 ± 0.88	.73	(−0.53, 0.73)
Traditional didactic education (control)	Primary	2.75 ± 0.50	2.50 ± 1.00	0.25 ± 1.26	.72	(−1.75, 2.25)
	Secondary	2.71 ± 0.49	2.86 ± 0.38	‐ 0.14 ± 0.38	.36	(−0.50, 0.21)
	Tertiary	1.67 ± 1.53	2.00 ± 1.00	‐ 0.33 ± 0.58	.42	(−1.77, 1.10)

Note: Results expressed in mean ± SD and analysed using repeated measures ANOVA.

4. DISCUSSION

We evaluated the difference of knowledge and self‐care behaviours in participants who had received collaborative communication style of education versus those who received traditional (one‐way) education from podiatrists.

A larger increase in knowledge retention and self‐care behaviours was observed in participants in intervention group using the collaborative approach compared to the control group. Intervention group has significant increase in knowledge retention and self‐care behaviours score (+5.45) post‐study, compared to baseline (P < .001). The increase (+1.9, P = .32) was not significant for participants in control group. Effect size of the difference in score between group was moderate (d = 0.54). On average, this implies that participants who underwent collaborative education had an improvement of score in about five areas of knowledge or behaviour, whereas traditional education produced improvement in only one to two areas in knowledge or behaviour.

In the survey, participants were asked how confident they were on caring for a small wound (Table 4). Interestingly, it was observed that participants with primary education had the largest increase (+0.25) in behavioural efficacy on caring for small wounds when exposed to traditional didactic style of information transfer. Comparatively, participants with primary education who had undergone collaborative style of education had no mean change in scores, pre and post study. This may suggest that participants with lower level of education may benefit from a didactic style of education for a stronger foundational knowledge base. It is also interesting to observe that participant with secondary and tertiary level of education in the control group experienced a decrease in confidence in caring for small wound (−0.14 and −0.33) after didactic persuasive style of communication. This is consistent with cautionary advice that direct persuasion may bring negative value to patients. ⁹ , ¹²

In the past two decades, there have been several case–control studies on the effects of education and counselling techniques in patients with diabetes foot conditions, involving occupations such as nurses, social workers, psychologists and rehabilitation physicians. ⁶ , ¹⁰ , ¹⁸ , ¹⁹ , ²⁰ To our best knowledge, this is the first study reporting the effects of such counselling techniques delivered by podiatrists.

This study was conducted in a tertiary acute hospital setting, where patients generally have a longer duration of diabetes and established diabetes‐related complications, compared to those managed in the community. It is encouraging to see positive results in this study, whereby participants who are more chronically ill and entrenched in their lifestyles were able to respond positively to collaborative discussion with their clinicians. This differs from the results of the study by McBride and colleagues, ¹⁸ whereby no significant difference was reported in the decision confidence and behaviours between the intervention (decision navigation) versus the control groups. Contrary to our experience, McBride and colleagues suggested that such personalised care may not improve health outcomes at later stages of disease timeline.

As methods such as Motivational Interviewing and Solutions Focused counselling have been proven effective in persons suffering from chronic schizophrenia, obesity, alcoholism, and drug abuse ¹⁴ , ¹⁵ , ¹⁶ , ¹⁷ , ²⁹ , ³⁰ , ³¹ , ³² , ³³ , ³⁴ , ³⁵ we are optimistic that such methods can be efficacious in chronic disease patient education such as diabetes. ²⁵ This is also supported by the results of our study. Patients with diabetes typically visit a podiatrist 3 to 4 times a year. For patients with an active foot wound, they may have podiatry visits as often as once weekly while the wound persists. Podiatrists are on the frontline of care, regularly seeing patients with diabetes for screening, treatment, and education. These clinical encounters provide an excellent opportunity to influence self‐care. From the mixed‐ethnicity Asian sample in this study, it can be inferred that such a collaborative communication style is also efficacious in the Asian population.

In our study, no additional time was given to podiatrists in the intervention group for collaborative style of communication. In other words, the podiatrists merely changed the way they approached their communication style and delivered the counselling within the usual treatment duration. Active listening and acknowledging that patients are experts of their own lives were keys to the change in communication style. In other previously reported studies, especially those involving the use of full Motivational Interviewing techniques, separate sessions of up to 60 minutes were required. ⁶ , ¹⁰ , ²⁹ Our technique presents a method that does not cost the clinician any extra time with the patient.

4.1. Limitations

There are several limitations to this study. First, being a pragmatic randomised controlled trial, the contents for education cannot be standardised. It takes the shape and pace of the interaction between the clinician and patient. In spite of the known limitations and variability, the study team went ahead with the pragmatic approach as we would like the results to be more directly applicable to actual clinical situations. Secondly, although the participants were randomised, the baseline mean score of the collaborative patient education group was moderately lower than the control group (d = 0.37). However, this difference between groups was not statistically significant. Thirdly, volunteers who step forward and consent to participating in such a study may already be more intrinsically motivated than those who opted not to volunteer for the study. We did however approach all suitable cases to minimise this effect. Finally, this study did not follow up on subjects beyond the study period of 12 weeks. Many of the study participants also completed the study before 12 weeks as their wound had healed earlier. Hence, we are unable to comment on longer term or retention effects of such an approach.

5. CONCLUSION

We found that the use of the collaborative approach in patient education produced better knowledge retention and self‐care behaviours, without the need for additional consultation time. We recommend that healthcare institutions consider including such communication skills into the training of clinicians involved in diabetes foot diseases to effectively educate patients on self‐management information and practice.

CONFLICT OF INTEREST

The authors declare no conflicts of interest.

AUTHOR CONTRIBUTIONS

Marabelle Heng: securing funding; study design; counselling method training; data collection; data processing; results interpretation; drafting and final approval of manuscript. Yu Heng Kwan: study design; data processing; results interpretation; drafting and final approval of manuscript. Nik Ilya: study design; data collection; data processing; drafting and final approval of manuscript. Izza Atiqa: securing funding; study design; counselling method training; results interpretation; drafting and final approval of manuscript. Patricia Jin: securing funding; study design; counselling method training; results interpretation; critical evaluation and final approval of manuscript. Debbie Hogan: study design; counselling method coaching; results interpretation; critical evaluation and final approval of manuscript. David Carmody: overseeing the project; securing funding; study design; data processing; results interpretation; drafting and final approval of manuscript.

Heng ML, Kwan YH, Ilya N, et al. A collaborative approach in patient education for diabetes foot and wound care: A pragmatic randomised controlled trial. Int Wound J. 2020;17:1678–1686. 10.1111/iwj.13450