SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Aim

The aim of this study was to explore the cognitive representations of peripheral neuropathy and self-reported foot-care behaviour in an Australian sample of people with diabetes and peripheral neuropathy.

Methods

This cross-sectional study was undertaken with 121 participants with diabetes and peripheral neuropathy. Cognitive representations of peripheral neuropathy were measured by the Patients' Interpretation of Neuropathy questionnaire and two aspects of self-foot-care behaviour were measured using a self-report questionnaire. Hierarchical cluster analysis using the average linkage method was used to identify distinct illness schemata related to peripheral neuropathy.

Results

Three clusters of participants were identified who exhibited distinct illness schemata related to peripheral neuropathy. One cluster had more misperceptions about the nature of peripheral neuropathy, one cluster was generally realistic about the nature of their condition and the final cluster was uncertain about their condition. The cluster with high misperceptions of their condition undertook more potentially damaging foot-care behaviours than the other clusters (F = 4.98; P < 0.01).

Conclusions

People with diabetes and peripheral neuropathy have different illness schemata that may influence health-related behaviour. Education aimed at improving foot-care behaviour and foot-health outcomes should be tailored to specific illness schemata related to peripheral neuropathy.

What's new?
  • The problem of diabetes-related foot morbidity is significant and the success of interventions is variable.
  • There is currently a limited understanding of the role neuropathic-specific illness cognitions have in mediating foot-care behaviour in people with diabetes.
  • This study has identified three clusters of participants from an Australian high-risk population of people with diabetes who report distinct illness cognitive schemata associated with diabetes-related peripheral neuropathy.
  • This study has determined a relationship between cognitive representations of neuropathy and potentially damaging foot-care behaviour.

Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

The burden of foot morbidity secondary to diabetes is substantial and likely to grow with the increasing prevalence of diabetes. Although there has been rapid growth in our understanding of the aetiology of foot complications, success in decreasing the incidence rates of lower-limb amputation is limited to distinct populations [1]. The common-sense model of illness behaviour postulates that illness-specific belief systems determine people's behavioural responses to illness [2]. In diabetes, a number of studies have demonstrated support for this model, with interventions based on this model being shown to be efficacious [3, 4]. The model proposes that there are five domains of cognitive representations associated with the experience of a health condition—the diagnostic label (identity), the expected duration (timeline), the expected outcomes (consequences), the believed cause (cause) and perceived treatment/control of the condition [5].

Vileikyte and colleagues have developed a peripheral neuropathy-specific questionnaire to assess these beliefs [the Patients' Interpretation of Neuropathy (PIN) questionnaire [6]]. Although they showed relationships between beliefs and foot care, the dimensions were examined in isolation rather than as interactive multidimensional representations, as hypothesized by the common-sense model of illness [7]. One way to address this is to use cluster analysis, which can identify groups of individuals with similar multidimension beliefs systems, and may have the potential to improve the predictive value over looking at illness dimensions in isolation [7]. Using this approach, we sought to explore the relationship between neuropathic-specific illness representations and foot-care behaviour in adults with diabetes and neuropathy.

Patients and methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

After obtaining human ethics approval, a convenience sample was recruited from a multidisciplinary high-risk foot clinic and a podiatry teaching clinic in a regional city of Australia. Eligibility criteria were a prior diagnosis of diabetes and clinically determined peripheral neuropathy, defined as an inability to detect the 10-g Semmes Weinstein monofilament (Bailey, Salford, UK) on four or more sites on at least one foot, and/or a vibratory perception threshold of > 25 V on at least one foot (Biomedical Instrument Co., Newbury, OH, USA) [8].

Participants completed a questionnaire booklet containing the validated 39-item self-report instrument Patients' Interpretation of Neuropathy questionnaire, and a second 17-item questionnaire assessing foot-care behaviour [6]. The Patients' Interpretation of Neuropathy questionnaire is made up of 39 items and 11 factors (nine cognitive and two emotional), representing the cognitive and emotional domains of the common-sense model of illness for peripheral neuropathy (see Table 1 for details). The self-report behaviour questionnaire has strong face and content validity based on established international guidelines [9, 10], covering what are thought to be important self-care activities (e.g. foot examination, foot hygiene, skin and nail care and use of footwear), although it should be noted that there is a limited evidence base for these activities at present. At the time of study design and data collection, the behaviour questionnaire was the most robust available. Basic participant characteristics were recorded and included age, gender, diabetes type, self-reported duration of diabetes, previous education levels, living arrangements and a prior history of a diabetes-related foot pathology.

Table 1. Participant characteristics
Variablen = 121
  1. Data are percentages or means ± standard deviations.

  2. a

    Vibratory perception threshold derived from the mean score of both feet. When a participant could not feel any vibratory perception, a score of 51 was given. This was the case for 45% of participants for their left foot and 49% of participants for their right foot.

  3. b

    The final eight scales that comprised the five cognitive domains of the Patients' Interpretation of Neuropathy (PIN) (illness identity, causes, timeline, cure/controllability and consequences) uses a five-point Likert scale response format (1 = strongly disagree, 2 = disagree, 3 = uncertain, 4 = agree, 5 = strongly agree). The data for the individual items of the cognitive PIN scales were pooled to provide an indicative mean score for each PIN scale. The seventeen-item questionnaire is split into two foot-care behavioural subscales: nine items pertaining to preventative behaviour and eight items to potentially damaging behaviour. Responses were rated on two different scales: a 6-point scale for ‘during the past week’ questions (twice a day, daily, every other day, twice a week, once a week, or never) and on a four-point scale for ‘in general’ questions (always, most of the time, occasionally, or never). Because of the difference in scaling, items were converted to a 0 to 1 scale before summating scores. Higher scores indicate both more preventative and potentially foot-damaging behaviours.

Male sex (%)80.2
Age (years)65.0 ± 10.39
Type 1 diabetes (%)9.1
Diabetes duration (years)15.64 ± 11.0
Vibratory perception thresholda44.58 ± 7.53
Completed secondary school (%)14
Lives alone (%)27.3
Previous diabetes-related foot pathology (%)67.8
Previous diabetes-related foot ulceration (%)57.9
Previous diabetes-related foot hospital admission (%)33.9
Previous Charcot neuro-oesteoarthropathy (%)14.9
Previous diabetes-related foot/lower limb amputation (%)16.5

PINb—ID1 (four items, α = 0.73): Good circulation = healthy feet

Example item: ‘Lost or reduced feeling means poor circulation in my feet’

3.65 ± 0.79

PIN—ID3 (three items, α = 0.77): Foot ulcers would be painful

Example item: ‘If I had a foot ulcer (an open sore) I would get pain in my feet’

2.87 ± 0.92

PIN—C1 (four items, α = 0.76): Physical causes of foot ulcers

Example item: ‘Changes in foot shape can cause foot ulcers (open sores)’

3.48 ± 0.63

PIN—C2 (four items, α = 0.66): Self/practitioner blame

Example item: ‘Foot ulcers (open sores) are caused by not taking care of oneself’

2.91 ± 0.78

PIN—TL (three items, α = 0.73): Accurate foot ulcer onset

Example item: ‘Foot ulcers (open sores) can develop very fast’

3.64 ± 0.66

PIN—CC1 (five items, α = 0.76): Efficacy of foot self-care

Example item: ‘Checking my feet every day can prevent foot ulcers (open sores) from occurring’

3.64 ± 0.58

PIN—CC2 (three items, α = 0.86): Practitioner foot ulcer control

Example item: ‘Foot care specialists such as podiatrists can prevent foot ulcers (open sores) from occurring’

2.96 ± 0.81

PIN—Cons (four items, α = 0.85): Anticipated consequences

Example item: ‘Lost or reduced feeling in my feet could lead to injuries to my feet’

4.11 ± 0.58

Preventative foot-care behaviour

Example item: ‘During the past week how often did you examine your feet?’

0.57 ± 0.16

Potentially damaging foot-care behaviour

Example item: ‘During the past week how often did you walk barefoot outdoors?’

0.17 ± 0.14

The scale structure of the cognitive factors of the Patients' Interpretation of Neuropathy questionnaire was checked using direct oblimin rotation of the correlation matrix, examining the scree plots for the point of inflection, with no item double loads > 0.4. Hierachical cluster analysis utilizing squared Euclidean distances of z-scores and the average linkage method was then undertaken. Number of clusters was determined by examination of the agglomeration schedule to determine the point of inflection. When comparing differences between clusters, analysis of variance was used for continuous measures, using Bonferroni correction, and χ2 analysis for categorical variables. All analysis was undertaken using SPSS version 16.0 (SPSS Inc., Chicago, IL, USA).

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Participation rate for potential participants conveniently assessed was 93%, with a total of 121 people enrolled in the study. Participant characteristics and mean Patients' Interpretation of Neuropathy scale and behaviour scores are reported in Table 1. The sample consisted of a high proportion of men with Type 2 diabetes and only 14% reported completing secondary school. There was a high mean vibratory perception threshold (44.58 ± 7.53) and a majority of the sample had a history of diabetes-related foot ulceration.

Factor analysis of the Patients' Interpretation of Neuropathy provided partial support for the scale structure reported by Vileiktye et al. [6]. The second of the three identity-related scales hypothesized to be present (ID2) [6] was excluded from further analysis, as the factor analysis suggested that the items did not load consistently onto a single factor and the hypothesized items demonstrated poor internal consistency (α = 0.39). The remaining six cognitive scales were demonstrated to be robust with an acceptable internal consistency of α > 0.65 (see Table 1 for details).

Initial analysis of the foot-care scales indicated they had low internal consistency, with one item from the preventative scale and two from the damaging scale demonstrating low inter-item correlations. The preventative behaviour item removed (‘How often do you cut your toenails straight across?’) and one of the potentially damaging behaviour items (‘When your feet feel cold at night, how often do you use hot water bottles/heating pads to warm them?’) had a high response rate indicating ‘not applicable’. The second potentially damaging item removed asked about a behaviour that is undertaken infrequently (‘How often do you rely on feeling the fit of the shoes when buying a new pair?’). Removal of these items improved the internal consistency of the scales to an acceptable level (preventative α = 0.60, potentially damaging α = 0.58).

Cluster analysis indicated that a three cluster represented the data well, based on the point of inflection on the agglomeration schedule. The three cluster solution included 46 participants grouped into cluster 1, 51 into cluster 2 and 24 into cluster 3, and are characterized by their respective mean cognitive Patients' Interpretation of Neuropathy scale scores in Fig. 1. Cluster 1 reported higher levels of misperception about their condition; particularly, that good circulation equals healthy feet. The participants in cluster 1 also believed the strongest that ulcers are caused by poor self or medical care (C2) and have the strongest beliefs both in the efficacy of self foot-care (CC1) and that health professionals can prevent foot ulcers (CC2). The participants in cluster 2 appeared to be more realistic about their foot health, with scores particularly low on three scales, indicating they are less likely to have the misperception that foot ulcers are painful (ID3), are less likely to agree that foot ulcers are caused by poor care (C2) and are less likely to agree that health professionals can prevent foot ulcers (CC2). In comparison, those in cluster 3 had mean Patients' Interpretation of Neuropathy scale scores that were all clustered around the midpoint of the range, indicating a substantial degree of uncertainty.

image

Figure 1. Mean scores and standard deviations for Patients' Interpretation of Neuropathy factors by cluster. (●) ID1: misperception (good circulation = healthy feet); (▼) ID3: misperception (foot ulcers are painful); (image) C1: physical cause of ulcers; (■) C2: self/practitioner blame; (▲) TL: acute ulcer onset; (image) CC1: efficacy of self-foot-care; (♦) CC2: practitioner ulcer control; (image) Cons: anticipated consequences.

Download figure to PowerPoint

Cluster membership was not associated with gender, age, type of diabetes, living arrangements or level of education. Cluster 2 had a larger proportion of participants with a history of Charcot neuro-osteoarthropathy (χ2 = 11.08, P < 0.01). Cluster 2 had a larger proportion of participants who had a history of diabetes-related ulceration, although this did not reach statistical significance. Cluster membership showed some differences in reported foot-care behaviour. There was little difference between the clusters with respect to preventative behaviour (F = 1.21, P = 0.3). However, there was a large difference across the clusters for reported potentially damaging behaviour (F = 4.98, P < 0.01), with cluster 1 reporting significantly more potentially damaging behaviours (0.21 ± 0.16, P < 0.01) than cluster 2 (0.13 ± 0.11).

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Previous research has identified specific cognitive representations that a person with diabetes has with respect to their peripheral neuropathy [6]. The structure of these beliefs was largely replicated in this study of participants from regional Australia, who reported low levels of education and who were at a high risk of diabetes-related foot complications.

The analysis identified three distinct clusters of participants, based on their beliefs about peripheral neuropathy, which is consistent with the findings in the illness perceptions literature [7]. Those in cluster 1 reported stronger misperceptions about peripheral neuropathy, tended to associate ulcers with poor health care in the past and believed strongly in the efficacy of self foot-care activities and health professionals in the prevention of ulcers. This group also reported significantly higher levels of potentially damaging foot-care behaviours. This suggests these participants may be motivated, but their beliefs about the nature of their neuropathy leads to potentially damaging foot care. A more intensive foot-care self-management programme might be warranted for this group, to help illicit and inform accurate beliefs about neuropathy.

The participants in cluster 2 were more likely to have had a history of a serious diabetes-related foot complication, were more realistic about the nature of peripheral neuropathy and reported the least potentially damaging behaviours. The adverse consequences of previous serious foot problems (and the associated regular contact with a high-risk foot clinic that this entails) may have had a significant impact on this group. Thus, the participants might be more likely to undertake more appropriate foot-care behaviour in the future.

The participants in cluster 3 were generally uncertain about all aspects of peripheral neuropathy as measured by the Patients' Interpretation of Neuropathy, including much more uncertainty about the potential consequences of peripheral neuropathy compared with the other two clusters. There may be significant scope for benefit by offering participants in this cluster focused, simple, educational tools on the nature of peripheral neuropathy that may be beneficial to enhance appropriate foot-care behaviour [11].

To the authors' knowledge, this the first study to identify distinct cognitive illness schemata of peripheral neuropathy among people with diabetes as measured by a validated instrument. In addition, although the tool used to measure foot-care behaviour has not undergone as an extensive validation process, this study has also identified different levels of reported foot-care behaviour across each of the three distinct illness schemata, with significant differences found between the clusters with respect to levels of potentially damaging foot-care behaviour reported. Any educational approach aimed at improving patient foot-care behaviours and foot-health outcomes needs to take into consideration different illness schemata people with diabetes and peripheral neuropathy may have toward their condition.

Funding sources

This project was supported by the Australasian Podiatry Education and Research Fund.

Competing interests

None declared.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

The authors acknowledge Dr Loretta Vileikyte and Dr Jeffrey Gonzalez for their advice regarding the use of the Patients' Interpretation of Neuropathy (PIN) and self-report foot-care behaviour questionnaires.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  • 1
    Boulton AJM, Vileikyte L, Ragnarson-Tennvall G, Apelqvist J. The global burden of diabetic foot disease. Lancet 2005; 366: 17191724.
  • 2
    Leventhal H, Meyer D, Nerenz D. The common-sense representation of illness danger. In: Rachman S, ed. Contributions to Medical Psychology. New York: Pergamon Press, 1980: 730.
  • 3
    Harvey JN, Lawson VL. The importance of health belief models in determining self-care behaviour in diabetes. Diabet Med 2009; 26: 513.
  • 4
    Petrie KJ, Broadbent E, Meechan G. Self-regulatory interventions for improving the management of chronic illness. In: Cameron LD, Leventhal H, eds. The Self-Regulation of Health and Illness Behaviour. London: Routledge, 2004: 257277.
  • 5
    Leventhal H, Brissette I, Leventhal E. The common-sense model of self-regulation of health and illness. In: Cameron LD, Leventhal H, eds. The Self-Regulation of Health and Illness Behaviour. London: Routledge, 2003: 4965.
  • 6
    Vileikyte L, Gonzalez J, Leventhal H, Peyrot M, Rubin R, Garrow A et al. Patient interpretation of neuropathy (PIN) questionnaire: an instrument for assessment of cognitive and emotional factors associated with foot self-care. Diabetes Care 2006; 29: 26172624.
  • 7
    Clatworthy J, Hankins M, Buik D, Weinman J, Horne R. Cluster analysis in illness perception research: a Monte Carlo study to identify the most appropriate method. Psychol Health 2007; 22: 123142.
  • 8
    Armstrong DG, Lavery LA, Vela SA, Quebedeaux TL, Fleischli JG. Choosing a practical screening instrument to identify patients at risk for diabetic foot ulceration. Arch Intern Med 1998; 158: 289292.
  • 9
    American Diabetes Association. Preventive foot care in diabetes. Diabetes Care 2004; 27: S6364.
  • 10
    IWGDF. International Consensus on the Diabetic Foot. Amsterdam: International Diabetes Federation Consultative Section on the Diabetic Foot/International Working Group on the Diabetic Foot, 1999.
  • 11
    Valk G, Kriergsman D, Assendelf W. Patient education for preventing diabetic foot ulceration: a systematic review. Endocrinol Metab Clin North Am 2002; 31: 633658.