Does in-shoe pressure analysis to assess and modify medical grade footwear improve patient adherence and understanding? A mixed methods study

Participants were patients referred from the Sydney Local Health District (SLHD) Podiatry and High Risk Foot Services to undergo in-shoe pressure analysis assessment, a relatively new service implemented at the beginning of 2021 as part of the Royal Prince Alfred Hospital (RPAH) Diabetes Centre High Risk Foot Service. Convenience sampling was used for this study, with patients who met the inclusion/exclusion criteria informed of the study during routine consultation and invited to participate. Patients interested in participating were provided with the Participant Information Sheet, which they were given a minimum of 24 h to read.

Inclusion criteria were Type 1 or Type 2 diabetes mellitus, and peripheral neuropathy diagnosed previously on routine neurovascular assessment (10 g monofilament and vibration perception using tuning fork, neurothesiometer or VibraTip™). All participants had a history of DFU on the plantar surface of the foot. Where participants had a history of multiple plantar ulcerations, all sites were included for assessment in this study. Participants were eligible if they owned MGF and custom orthoses or pre-fabricated accommodative orthoses issued within the last 12 months. MGF was defined according to the criteria described in the recent guidelines by Van Netten et al. [7] and included fully customised or pre-fabricated footwear with features that reduce the risk of DFU, such as extra depth or width, or a modified sole (e.g. rigid forefoot rocker). MGF had to contain removable orthoses to facilitate modifications where required. Footwear greater than 12 months of age was excluded as MGF of patients with a history of DFU is routinely replaced at 12 months within SLHD Podiatry due to material fatigue. A minimum period of two weeks wearing prescribed MGF prior to study commencement was included to allow for gradual and safe introduction of MGF and to allow for baseline adherence to be measured.

Patients with active DFU or Charcot neuroarthropathy were excluded from the study, as were patients considered to be at high risk of falls or unable to walk 10 m with or without assistive devices due to inability to safely complete the walking trials required for pressure measurement. Patients with a forefoot amputation or below- or above-knee amputation were excluded for technical reasons due to inability to collect data from one insole only, and to avoid damage to the Pedar insoles by inserting over an in-shoe filler. A large or raised filler used to accommodate a forefoot amputation would require bending of the insole in a way that may damage the equipment. Where patients had a small or shallow filler or a filler located in such a way that the Pedar insole could be safely positioned, they were included in the study.

Ethical approval for this study was sought and obtained from both Cardiff University School of Medicine Research Ethics Committee (SREC reference SMREC 20/98) and the RPAH Ethics Review Committee (Protocol number X20-0455) in December 2020. All assessments, including questionnaires, were conducted by the investigator (CM [a podiatrist employed by SLHD]). The questionnaires used in this study were original, as no previously validated questionnaires assessing the topics of patient adherence and understanding of MGF were identified. The questionnaires were developed in consultation with the SLHD Research Electronic Data Capture (REDCap) Data Management team to optimise functionality. The questionnaires were then piloted on the first three participants recruited for the study in order to assess acceptability and make necessary modifications for the final study. All questionnaires were read out to participants by the investigator (CM) from the REDCap database hosted by SLHD and developed by Harris et al. [14, 15]. REDCap is a software platform for data collection and management utilised for research studies in SLHD. Questionnaires were delivered in a standardised manner to minimise potential interviewer bias that Meadows [16] noted may occur if there is variability in delivery between participants.

Participant characteristics are provided in Additional file 1. Most participants had T2DM (n = 14) with a group mean HbA1c% of 7.5% (range 5.3%-11.5%). The majority of participants (n = 10) wore modified MGF, with a rigid forefoot rocker the most common modification (n = 9). None of the participants had undergone in-shoe pressure analysis previously. Thirty-two previous foot ulcers were reported with the most common location for both feet combined the forefoot (n = 15) followed by hallux (n = 12). The mean time since the most recent ulcer healed was 28.4 months (range 1–180 months). Most participants (n = 13) had not been diagnosed with peripheral arterial disease. All participants had one or more foot deformities affecting one or both feet, with the most common deformities consisting of claw/hammer toes (n = 9), hallux limitus/rigidus (n = 7), and prominent metatarsal heads (n = 7).

A total of 24 sites where MPP > 200 kPa at baseline were targeted for pressure reduction. Pressure reductions were achieved for all but four sites (see Additional file 2 for the full list of pressure reductions and modifications performed). Following pressure analysis, 8 sites (33%) were successfully offloaded to < 200 kPa, with an average of 2 rounds of orthotic modification required. For two participants there were no sites > 200 kPa and no modifications were required.

The ranges for self-reported adherence inside the home prior to pressure analysis at Week 0 and at Week 4 are presented in Fig. 2. The range of self-reported adherence inside the home increased for seven (46.7%) participants, decreased for one (6.6%), and remained the same for seven (46.7%).

The data for adherence outside the home are presented in Fig. 3. The data show that self-reported adherence outside the home increased for five (33.3%) participants and stayed the same for 10 (66.7%), of which seven were already 100% at baseline.

The number of participants who reported ≥ 80% adherence at Week 0 was two (13.3%) for inside the home and eight (53.3%) for outside the home. This increased to three (20.0%) for inside the home and 12 (80.0%) for outside the home at Week 4.

Several themes surrounding non-adherence and adherence to MGF were identified in the responses to the Week 0 Pre-assessment Questionnaire A and Week 4 Questionnaire. The themes for non-adherence are presented in Table 2. The number of times each theme was identified in participant responses at Week 0 and Week 4 is presented. A significant proportion of themes for non-adherence to MGF related specifically to use inside the home.

Themes surrounding adherence to MGF are also presented in Table 2. Dominant themes at both Week 0 and 4 included ‘following healthcare advice’ and ‘comfort’. ‘Safety and protection of feet’ was reported more frequently at Week 4 compared to Week 0.

The mean score for patient understanding of MGF before and after pressure analysis at Week 0 is presented for each of the five statements related to MGF use in Table 3 (best possible score 5; worst possible score 1). The overall mean score represents the combined mean for all five statements. The data do not demonstrate large changes in patient understanding with the overall mean score increasing from 4.0 to 4.2 out of 5. With the exception of Statement 3 where the mean score increased from 3.9 pre-assessment to 4.5 post-assessment (p = 0.01), the differences between scores for the remaining statements were not statistically significant (p > 0.05).

There was no significant change in the proportion of participants who reported that they were or were not satisfied with medical grade footwear before and after the intervention. Prior to the intervention, 10 participants were satisfied, one was not, three were unsure, and one preferred not to answer. Following the intervention, the number who were satisfied increased to 11, while one remained not satisfied, two unsure and one participant still preferred not to answer. Several themes for satisfaction and dissatisfaction with MGF were identified in the responses to the Pre-assessment Questionnaire B and Post-assessment Questionnaire (see Additional file 3). Similar to the themes surrounding adherence to MGF, ‘comfort’ was a dominant theme for satisfaction with MGF and ‘designed to protect feet and prevent foot complications’ was more likely to be reported following the intervention.

Responses to the Week 0–1 Telephone Questionnaire are provided in Table 4, and the themes identified from the questionnaire in Additional file 4. The patient experience of in-shoe pressure analysis was largely positive, with all 15 participants reporting that results were communicated in a way they could understand and attending was worthwhile. When asked why pressure analysis was worthwhile, the primary reasons provided were that it demonstrated pressure points and improved offloading.

No adverse events occurred during the in-shoe pressure analysis session at Week 0. Four adverse events occurred over the course of the final study, of which three were attributed to use of medical grade footwear, consisting of a plantar forefoot fissure, dorsal skin trauma and dorsal neuropathic ulcer. The fourth adverse event consisted of haemorrhagic callus at the plantar hallux related to intrinsic foot deformity, not use of MGF. None of the participants were withdrawn from the study or follow-up questionnaires. Three participants were advised to cease use of MGF during the study due to risk of skin trauma and/or ulceration and returned to alternative footwear (n = 2) or offloading device (n = 1).

The primary aim of this study was to determine if patient adherence to MGF improved following in-shoe pressure analysis and modification of MGF. The results of this study indicated that self-reported adherence inside the home increased for 46.7% of participants, decreased for 6.6%, and stayed the same for 46.7%. However, the number of participants who reported ≥ 80% adherence to MGF inside the home was low, increasing from 13.3% at Week 0 to 20.0% at Week 4. Adherence outside the home increased for 33.3% of participants and stayed the same for 66.7%. However, baseline adherence outside the home was high compared to previous research, with 53.3% (n = 8) participants reporting ≥ 80% adherence, compared to 26% (n = 13) reported by Macfarlane and Jensen [11].

When examining the number of participants who reported ≥ 80% adherence to MGF outside the home, this increased from eight (53.3%) at Week 0 to 12 (80.0%) at Week 4, which indicates a clinically significant improvement in adherence (reaching ≥ 80%) for four participants. In contrast, while adherence inside the home improved for seven participants (46.7%), adherence only increased to a clinically significant level (≥ 80%) for one participant. The finding of low adherence to MGF inside the home compared to outside the home is consistent with the results of previous studies. Macfarlane and Jensen [11] reported that 12% (n = 6) of participants wore prescribed MGF > 80% of the time at home compared to 26% (n = 13) outside the home, while Waaijman et al. [10] reported mean adherence of 61% at home compared to 87% adherence away from home for 107 participants with previous DFU, and this difference was statistically significant (p < 0.01). The challenge of comparing the findings of the current study with previous research is that adherence is not frequently reported separately for indoor and outdoor use, however, this is important given the significant difference in reported adherence between these two conditions [23].

We are not aware of any previous research investigating in-shoe pressure analysis as a tool to improve patient adherence to MGF. Bus et al. [8] and Abbott et al. [24] reported patient adherence following in-shoe pressure analysis, however, the study aims and application of in-shoe pressure analysis were different from the current study. Bus and colleagues sought to blind participants to the intervention by evaluating the results of pressure analysis and modifying footwear out of view, which prevented bias from differences in how groups perceived treatment. The findings of the current study indicate that communicating the results of in-shoe pressure analysis and demonstrating the offloading capacity of MGF may improve adherence. However, a larger study incorporating an objective and accurate measure of adherence is needed to explore this further. For example, Bus et al. [8] utilised an in-shoe temperature monitor combined with a step activity monitor around the ankle. Alternatively, dual accelerometers worn on the body and attached to the shoe can provide an objective measure of adherence [23].

Abbott et al. [24] studied patient adherence to the use of an intelligent insole inserted into MGF, which provided an alert via smartwatch when excessive plantar pressures were detected and prompted participants to modify weight-bearing behaviour. Abbott and colleagues reported that 69% of participants were ‘good compliers’ (defined as a mean of ≥ 4.5 h wear per day). When analysing adherence between the intervention group and control group (which received no alerts), there was no difference in the percentage of ‘good compliers’ between groups (both 69%). However, Bus [25] highlighted the high drop-out rate of 35% during the wearing-in period and 50% in the intervention group during follow-up, which Abbot and colleagues attributed to difficulty adapting to the smartwatch technology, as opposed to the higher frequency of alerts. This notion is supported by the findings of Najafi et al. [26], who conducted a study assessing adherence to the use of the same intelligent insole system and found that adherence increased for those participants who received a high number of alerts per hour compared to those who received a low number of alerts. However, while the current study assessed the impact of a single session of in-shoe pressure analysis measuring mean peak pressure during walking, the insole system assessed by Abbott and colleagues and Najafi et al. [26] consisted of a continuous monitoring system that measured sustained low pressures (around 35 mmHg) during daily activities including sitting, standing and walking. This therefore limits the capacity to compare the findings of these studies to the present study.

While this study demonstrated that adherence outside the home improved to ≥ 80% for four participants at Week 4, it is not known whether the improved adherence would have been sustained at 12 weeks or beyond. Keukenkamp et al. [27] studied the impact of motivational interviewing on footwear adherence in 13 patients with a history of DFU. Participants were randomised to receive either standard education or two 45-min motivational interview sessions. Keukenkamp and colleagues found that adherence outside the home was high at baseline (91% for the intervention group) and remained high at three months (92% intervention group, 93% standard education group). However, adherence inside the home for the intervention group increased from 49 to 84% at week 1, before returning to 40% at three months. This suggests that interventions to modify patient adherence to MGF may need to be repeated to have a sustained impact. This could be explored in a future study where in-shoe pressure analysis and visual demonstration of findings is repeated at monthly intervals, and where adherence is measured at three and six months to determine if repeated sessions provide a sustained impact on adherence.

The potential for in-shoe pressure analysis to modify patient understanding of the role of high pressure areas and improve adherence was suggested by Najafi et al. [28]. However, Najafi and colleagues were referring specifically to an intelligent insole system providing continuous feedback. The current study sought to investigate the impact of a single session of in-shoe pressure analysis on patient understanding. For the Pre-assessment Questionnaire B and Post-assessment Questionnaire, the mean score for Statement 3 increased from 3.9 to 4.5 and the difference in scores was statistically significant (p < 0.05). As Statement 3 referred specifically to using MGF > 80% of the time to avoid ulceration, this indicates that the intervention may have increased participants’ understanding of the need to wear MGF most of the time. However, the scores for the remainder of the statements did not demonstrate a statistically significant difference. Patient understanding at baseline was already at a high level (4 out of 5), which may explain the lack of statistically significant improvement in scores. As participants had already been fitted with MGF and educated on use prior to enrolment in the study, their level of understanding at study entry was already high. Furthermore, it is possible that the questionnaire was not sensitive to detecting change in patient understanding. Questionnaires with more simplified language, and which participants read and complete themselves may prove more sensitive to detecting change. Further studies are required to develop a reliable and valid tool to measure patient understanding of MGF.

The area in which participants had the poorest understanding was use of slippers or other house shoes at home, which is consistent with previous studies [29, 30]. There was minimal change in responses for this statement before and after a single session of pressure analysis, which is reflected in the minimal change in adherence inside the home and the persistence of themes related to non-adherence at home over the four weeks. The themes identified in this study are consistent with the findings of Paton and colleagues [29] who reported that participants viewed the home as a safe and familiar place where there was a low risk of injury. Participants for the current study also expressed a feeling of safety at home where they reported ‘sitting or keeping their feet up’. Further studies exploring the impact of continuous monitoring insole systems—such as the smartwatch employed by Abbott et al. [24]—on patient understanding of the high plantar pressures experienced with inappropriate footwear use at home may be beneficial. However, this would rely upon patient compliance with transferring of insoles between shoes where alternative footwear is worn at home. A barefoot pressure platform, as utilised by Gurney and colleagues [31], could be employed in future studies to illustrate the pressures to which bare feet are subjected and may be useful to compare with in-shoe pressure analysis to demonstrate the attenuation of pressures with MGF. While this may assist in modifying patient behaviour towards MGF use inside the home, the additional cost of a barefoot pressure platform is a limitation.

The response rate for the current study was high, with all 26 patients who were informed of the study expressing interest in learning more, of which 22 agreed to participate. The positive patient engagement with in-shoe pressure analysis was reflected in participant responses to the Week 0–1 Telephone Questionnaire, which demonstrated that 100% of participants felt the intervention was worthwhile. The dominant reasons provided by participants for why they found attending worthwhile was the ‘demonstration of pressure points’ (n = 4) and ‘improved offloading of pressure points’ (n = 3). This suggests that participants understood the purpose of in-shoe pressure analysis, were interested to learn more about the offloading properties of their footwear and valued having their footwear optimised.

While the target of 200kPa used in this study is commonly used to guide footwear and orthotic modification in patients with a history of DFU, Jones et al. [32] noted the limited evidence to support the use of this threshold; consisting of two cohort studies [9, 33] and one RCT [8]. It is also recognised that the findings of this study may not be generalisable to those using pressure measurement systems other than the Pedar, as different systems have different numbers of sensors, measurement range and sampling rate, which limits comparison [32].

Bus et al. [5] successfully offloaded all 35 ROIs targeted for pressure reduction in their study, with successful offloading defined as 25% pressure reduction or MPP < 200 kPa. However, 19 of 35 sites (54%) were successfully offloaded according to the criterion of MPP < 200 kPa. In contrast, Waaijman et al. [6] reported 51–59% of sites were successfully offloaded according to the criterion of 25% reduction or < 200 kPa, and attribute the lower percentage compared to Bus et al. [5] to the larger number of regions targeted for modification. It is acknowledged that the pressure reduction results for the current study were lower with 8 of 24 ROIs (33%) offloaded to < 200 kPa. Possibly contributing to this was the absence of a pedorthist to assist with footwear modification, time restraints preventing further orthotic modifications, and the newly established nature of the clinic. However, the aims of the study were to investigate adherence and understanding of medical grade footwear in response to the intervention as it is used in routine clinical practice, therefore, while the aim was to reduce pressure < 200Kpa this was not deemed essential.

A further limitation of the current study was the subjective measure of self-reported adherence. While Bus et al. [8] and Abbott et al. [24] utilised objective methods of measuring adherence, adherence for the current study was measured subjectively and relied upon participants’ recollection of adherence over the previous four weeks. Therefore, the findings for patient adherence should be interpreted with caution given the risk of bias associated with self-reported adherence to MGF [23, 26]. A further limitation of this study was that plantar pressures were measured in a controlled clinical setting that may not reflect the cumulative forces a patient is subject to over a day [28, 34]. Importantly, other forces such as pressure from the dorsum of the shoe and shear pressure were not able to be captured with the in-shoe pressure technology utilised, however, they may contribute to ulcer formation [35].

Limitations of sample size and sampling method also weaken the findings of this study. The sample size was small and was not calculated to determine statistical significance, therefore, the statistical significance of the results related to patient adherence is not known. The sample size for the pilot and final study combined (n = 18) was smaller than the anticipated n = 20. Contributing to this was the delay in supply of MGF to patients within SLHD related to the COVID-19 pandemic. Furthermore, all 15 participants included in the final study were male, therefore, the findings of this study reflect an entirely male sample and may not be generalisable to female patients.

The questionnaires used for this study were created for the purpose of this study, therefore, the reliability and validity of these questionnaires is unknown. A pilot study was undertaken to assess participant comprehension and acceptability of the questionnaires. However, as these questionnaires were new, there were no previous studies to directly compare the findings to, limiting the conclusions that could be drawn, particularly regarding patient understanding of MGF. Quantifying changes in patient understanding through analysing differences in Likert score responses may not reflect a true change in understanding, and the clinical significance of the change in scores for patient understanding is unknown. The Week 0 questionnaires were delivered verbally by the investigator who conducted the pressure analysis and orthotic modifications. Despite questionnaires being delivered in a standardised format for all participants, there is the possibility of interviewer bias in how participant responses to short answer questions were noted down and at times summarised.

With regards to the adverse events that occurred during the study, three were attributed to use of MGF, two of which consisted of a dorsal lesion. It is possible that increased use of MGF as a result of participation in the study may have contributed to these events. Changes in physical activity could also have contributed, with participants possibly perceiving that they were more protected in MGF, however, this was not measured in the study and warrants investigation in future studies. A limitation of plantar pressure analysis is that dorsal pressures against the shoe are not able to be measured, which highlights the importance of regular visual inspection of the feet while transitioning into MGF.

This study confirmed previous research findings that have demonstrated poor adherence to MGF inside the home [10, 11, 27, 29, 30]. In light of the preference of patients to use slippers or simple house shoes at home, the provision of a medical grade slipper in addition to outdoor MGF may increase adherence inside the home, which was also recommended by Paton et al. [29]. However, the inevitable challenge with such footwear is the need to offload areas at risk of DFU, which may necessitate footwear features not compatible with a slipper design [36]. Ultimately, an informed discussion between footwear supplier and patient in which patient goals are addressed and any potential compromise reached may result in the best outcomes for the patient. Clinicians may find it useful to employ in-shoe pressure analysis to demonstrate the difference between a patient’s standard indoor footwear and the recommended MGF, as this may further reinforce the benefit of prescribed footwear.