Abstract
The aims of this study were to investigate the point prevalence, and associated independent factors, for foot disease (ulcers, infections and ischaemia) in a representative hospitalised population. We included 733 (83%) of 883 eligible adult inpatients across five representative Australian hospitals on one day. We collected an extensive range of self‐reported characteristics from participants. We examined all participants to clinically diagnose foot disease (ulcers, infections and ischaemia) and amputation procedures. Overall, 72 participants (9·8%) [95% confidence interval (CI):7·2–11·3%] had foot disease. Foot ulcers, in 49 participants (6·7%), were independently associated with peripheral neuropathy, peripheral arterial disease, previous foot ulcers, trauma and past surgeon treatment (P < 0·05). Foot infections, in 24 (3·3%), were independently associated with previous foot ulcers, trauma and past surgeon treatment (P < 0·01). Ischaemia, in 33 (4·5%), was independently associated with older age, smokers and past surgeon treatment (P < 0·01). Amputation procedures, in 14 (1·9%), were independently associated with foot infections (P < 0·01). We found that one in every ten inpatients had foot disease, and less than half of those had diabetes. After adjusting for diabetes, factors linked with foot disease were similar to those identified in diabetes‐related literature. The overall inpatient foot disease burden is similar in size to well‐known medical conditions and should receive similar attention.
Keywords: Factors, Foot disease, Foot wounds, Inpatients, Prevalence
Introduction
Foot‐related conditions have been frequently reported to be a leading cause of diabetes‐related hospitalisation 1, 2, 3. Our recent study went a step further and found that foot‐related conditions, particularly foot disease, were a leading cause of all hospitalisations 4. We reported that foot‐related conditions were the primary reason for hospitalisation in 7·4% of a representative Australian inpatient population 4. Two‐thirds of those were for the foot disease disorders of ulcers, infections and ischaemia; whilst the remainder were for foot trauma and amputation 4. While our study was the first to quantify the direct inpatient burden 4, the overall foot‐related inpatient burden made up of the direct (causing admission) and indirect burden (present during admission) remains to be quantified 5, 6.
Two recent reviews have interrogated the existing literature investigating the foot‐related conditions present in inpatients 5, 6. A narrative review confirmed foot disease, specifically foot ulcers, to be the leading foot‐related conditions present in both specific and representative inpatient populations 5. Additionally, a systematic review could only identify enough literature to calculate a pooled prevalence estimate for one sub‐group of collective foot‐related conditions in representative inpatient populations, that is, 4·7% for diabetes‐related foot disease in representative inpatient populations 6. However, this pooled prevalence estimate reported very high heterogeneity because of included studies reporting a variety of definitions for foot disease, with none including ischaemia in their definition 6. The contemporary definition of foot disease is now well‐recognised to include foot ulcers, infections and ischaemia 4, 7. Thus, neither review was able to identify any studies investigating the prevalence, or associated factors, for foot disease or all foot‐related conditions in a representative inpatient population 5, 6.
In order for policy makers, clinicians and researchers to fully appreciate the overall foot‐related inpatient burden, it appears necessary to investigate the common foot‐related conditions that are most likely to collectively make up this burden in representative inpatient populations. According to the literature, this burden is most likely to be made up predominantly of foot disease (ulcers, infection and ischaemia), with acute traumatic foot wounds and amputation procedures contributing the rest of this burden 4, 5, 6. Thus, the primary aims of this study were to investigate the point prevalence, and associated independent factors, for foot disease (ulcers, infection and ischaemia) present in a representative hospitalised population. The secondary aims were to investigate the point prevalence, and associated independent factors, for major foot‐related conditions (foot disease, acute wounds and amputation procedures) present in the same population.
Methods
Study design and settings
This study was part of the Foot Disease in Inpatients Study, a multi‐site observational point prevalence study carried out in five public hospitals in Queensland, Australia 4. The five hospitals have been described in detail elsewhere 4. In brief, each hospital was purposively selected to represent one of the five categories of peer‐group hospitals in Australia according to the National Health Performance Authority, including a major metropolitan general hospital (>500 beds in South Queensland), a major metropolitan specialist hospital (>500 beds in South Queensland), a major regional general hospital (>200 beds in Central Queensland), a large metropolitan general hospital (>200 beds in South Queensland) and a large regional general hospital (>50 beds in North Queensland) 4. Institutional ethics committees approved this study, and all participants gave written informed consent 4. This study's design, methodology and definitions have been described previously 4.
Participants
Eligible participants were all adult inpatients present on the designated day of data collection at each of the five hospitals. Excluded were those younger than 18 years; cognitively impaired; or in a paediatric, maternity or psychiatric ward 4. Sample size calculations determined that 750 participants were required to adequately power this study 4. Briefly, 1146 inpatients were present during data collection; 883 met the inclusion criteria. and 733 (83%) were included in this study 4.
Variables
The explanatory variables used for this study have been defined in detail elsewhere and included self‐reported and clinically diagnosed variables 4. In brief, the self‐reported explanatory variables were grouped into demographic (age and gender), social determinant (socioeconomic status, geographical remoteness, education levels achieved, country of birth and Australian indigenous status), medical condition history (diabetes, hypertension, dysiplidaemia, myocardial infarct, cerebrovascular accident, chronic kidney disease, cancer, arthritis, depression, acute foot trauma and smoking history), self‐care ability (mobility impairment, vision impairment, main footwear worn outside and main footwear worn inside house) and past foot treatment in the previous 12 months prior to hospitalisation (by podiatrist, general practitioner, specialist physician, surgeon, nurse, orthotists and other) variables 4. Main footwear worn variables were then collapsed into low‐risk footwear (walking shoe, runner, oxford shoe, boot or bespoke footwear), moderate‐risk footwear (moccasin, Ugg boot, slipper or backless slipper), high‐risk footwear (high heels, flip flops, court shoe, mule or sandal) and no footwear (socks only or barefoot) 4. The clinically diagnosed explanatory variables included previous foot ulcer (self‐reported with clinical examination to confirm), previous amputation (self‐reported with clinical examination to confirm), peripheral neuropathy (absence of sensation to a 10‐g monofilament on two or more of three plantar forefoot locations), peripheral arterial disease (absence of at least one foot pulse and a toe systolic pressure <70 mmHg) and foot deformity (three or more of small muscle wastage, bony prominence, prominent metatarsal heads, hammer/claw toes, limited joint mobility and Charcot deformity) on at least one foot 4. Peripheral arterial disease (PAD) was further categorised into mild PAD (toe systolic pressure of 51–70 mmHg), moderate PAD (toe systolic pressure 31–50 mmHg) and critical PAD/ischaemia (toe systolic pressure <30 mmHg) 4. The outcome variables for this study were clinically diagnosed foot‐related conditions, including foot disease disorders (ulcers, infections and ischaemia), acute foot wounds and new amputation procedures. Table 1 outlines the exact detailed criteria, definitions and supporting citations for each outcome variable used in this study 4, 7, 8, 9, 10, 11, 12, 13.
Table 1.
Definitions for each outcome variable
| Outcome variable | Definition |
|---|---|
| Foot‐related conditions 4 | Any foot ulcer, foot infection, ischaemia, acute foot wound or amputation procedure present |
| Foot disease 4, 8 | Any foot ulcer, foot infection or ischaemia present |
| Foot ulcer 8, 9 | An existing full‐thickness wound beneath the ankle of primarily neuropathic, ischaemic, pressure injury origin or post‐foot ulcer amputation site |
| Foot infection 8, 9, 10, 11 | At least two manifestations of inflammation (purulence, erythema, pain, tenderness, warmth or induration) |
| Mild foot infection | A foot infection with erythema extending <2 cm from the edge of the wound |
| Moderate foot infection | A foot infection with erythema extending >2 cm from the edge of the wound |
| Severe foot infection | Any foot infection with signs of systemic inflammatory response syndrome |
| Ischaemia 7, 8, 9 | At least one absent foot pulse and a toe systolic pressure <30 mmHg |
| Acute foot wound 8 | An existing full‐thickness wound beneath the ankle of primarily traumatic or post‐surgical origin (excluding amputation procedures because of foot ulcer) |
| Amputation procedure 12, 13 | A new lower‐extremity amputation procedure performed as part of the current inpatient admission, plus a clinical examination to verify a post‐surgical amputation wound site |
| Minor amputation procedure | An amputation procedure distal to the ankle |
| Major amputation procedure | An amputation procedure proximal to the ankle |
Data collection
The data collection procedure has been described elsewhere 4. In brief, the Queensland Foot Disease Form (QFDF) 4 was developed from a similar validated data collection instrument 8. The QFDF captured all variables through a survey of participants' self‐reported history and a physical examination for clinically diagnosed foot‐related conditions 4, 8. Data collectors were publicly employed podiatrists who received extensive training and scored at least 90% accuracy on assessment 4. Data were collected by teams of data collectors between 8 am and 5 pm on each hospital's single data collection day between June and December 2013 4. A 5% subsample of study data collected was tested against audited medical records, and high levels of agreement were reported 4.
Statistical analysis
All data were analysed using SPSS 22·0 for Windows (SPSS Inc., Chicago, IL) or GraphPad Software. Prevalence proportions with 95% confidence intervals (95% CI) were calculated for all outcomes. Differences between groups were tested using χ 2 tests, Kruskal–Wallis tests, t‐tests or ANOVAs. Univariate logistic regression analyses tested for crude associations with outcomes (P < 0·2) 14, 15. Two multivariate models were used for each outcome: Model 1 used all variables, and Model 2 used all variables except past foot treatment variables. All variables crudely associated in univariate analyses (P < 0·2) were included in multivariate logistic regression analyses 14, 15. A backwards stepwise method was employed to remove non‐significant variables (P > 0·05) at each step until only variables reaching significance remained (P < 0·05) (unadjusted model) 14, 15. Omitted variables were re‐entered into the models and retained as confounders if they changed the Beta estimates of any independent explanatory variable by >20% (adjusted model) 14. Collinearity, goodness of fit, significance, parsimony and variance were assessed at each step 14, 15. Cases with missing data were excluded as missing data cases made up <5% in all models 14.
Results
Descriptive data for all 733 participants' outcome variables are displayed for each hospital in Table 2. Additionally, all explanatory variables and univariate analyses are summarised in Tables A1 and A2 in the Appendix. Table 2 shows that a foot‐related condition was present in 86 participants [11·8% (95% CI: 9·6–14·3%)], 34 of whom had diabetes [4·6% (3·3–6·4%)]. Foot disease was present in 72 participants [9·8% (7·2–11·3%)], 33 of whom had diabetes [4·5% (3·2–6·3%)]. No differences existed between hospitals for any foot‐related condition outcomes (P > 0·1).
Table 2.
Participant outcome variables for each hospital [number (%) unless otherwise stated]
| n | Total [95% CI] | Large regional (general) (n = 21) | Major regional (general) (n = 82) | Large metropolitan (general) (n = 90) | Major metropolitan (speciality) (n = 232) | Major metropolitan (general) (n = 308) | P value | |
|---|---|---|---|---|---|---|---|---|
| Foot‐related conditions | 732 | 86 (11·8%) [9·6–14·3] | 2 (9·5%) | 12 (14·6%) | 8 (8·9%) | 19 (8·2%) | 45 (14·7%) | 0·147 |
| Foot disease | 732 | 72 (9·8%) [7·9–12·2] | 1 (4·8%) | 9 (11·0%) | 7 (7·8%) | 19 (8·2%) | 36 (11·7%) | 0·546 |
| Foot ulcers | 732 | 49 (6·7%) [5·1–8·8] | 0 | 7 (8·5%) | 4 (4·4%) | 11 (4·8%) | 27 (8·8%) | 0·183 |
| Foot infections | 723 | 24 (3·3%) [2·2–4·9] | 0 | 5 (6·2%) | 3 (3·3%) | 5 (2·2%) | 11 (3·6%) | NA |
| Mild foot infection | 723 | 11 (1·5%) [0·8–2·7] | 0 | 2 (2·5%) | 2 (2·2%) | 1 (0·4%) | 6 (2·0%) | NA |
| Moderate foot infection | 723 | 7 (1·0%) [0·4–2·0] | 0 | 2 (2·5%) | 0 | 2 (0·9%) | 3 (1·0%) | NA |
| Severe foot infection | 723 | 6 (0·8%) [0·3–1·8] | 0 | 1 (1·3%) | 1 (1·1%) | 2 (0·9%) | 2 (0·7%) | NA |
| Ischaemia | 728 | 33 (4·5%) [3·2–6·3] | 1 (4·8%) | 2 (2·4%) | 3 (3·3%) | 8 (3·5%) | 19 (6·3%) | NA |
| Acute foot wounds | 732 | 17 (2·3%) [1·4–3·7] | 1 (4·8%) | 4 (4·9%) | 2 (2·2%) | 1 (0·4%) | 9 (2·9%) | NA |
| Amputation procedures | 732 | 14 (1·9%) [1·1–3·2] | 0 | 2 (2·4%) | 0 | 2 (0·9%) | 10 (3·2%) | NA |
| Minor amputation procedure | 732 | 8 (1·1%) [0·5–2·2] | 0 | 1 (1·2%) | 0 | 1 (0·4%) | 6 (1·9%) | NA |
| Major amputation procedure | 732 | 6 (0·8%) [0·3–1·8] | 0 | 1 (1·2%) | 0 | 1 (0·4%) | 4 (1·3%) | NA |
95% CI, 95% confidence interval; NA, not applicable to test as the assumption of χ 2 test is violated as two cells had an expected count < 5.
Foot disease
Foot ulcers were present in 49 participants [6·7% (5·1–8·8%)], 26 [3·6% (2·4–5·2%)] of whom had diabetes (Table 2). Foot ulcers made up 74% of the 66 participants with any type of foot wound [9·0% (7·2–11·3%)]. After univariate analysis, 15 explanatory variables were associated with foot ulcers (all, P < 0·05) (Table A1). Foot infection was excluded from foot ulcer models as foot infection was only present in wounds. Table 3 shows that after adjustment for socioeconomic status, foot ulcers were independently associated with foot trauma, previous foot ulcers, moderate‐critical PAD, peripheral neuropathy and past surgeon treatment in Model 1 (P < 0·05). All independent variables remained significant in Model 2 (P < 0·01).
Table 3.
Independent associated factors for foot ulcers using multivariate logistical regression (odds ratios [95% CI])
| Risk factor | Unadjusted | P value | Adjusted | P value |
|---|---|---|---|---|
| Model 1 | ||||
| Acute foot trauma | 25·35 [6·57–97·85] | <0·001 | 33·03 [7·29–149·67] | <0·001 |
| Previous foot ulcer | 18·84 [7·50–47·33] | <0·001 | 22·68 [8·27–62·15] | <0·001 |
| Peripheral neuropathy | 3·79 [1·59–9·00] | 0·003 | 5·12 [2·01–13·05] | 0·001 |
| PAD | <0·001 | <0·001 | ||
| Nil PAD | Referent | Referent | ||
| Mild PAD | 0·23 [0·03–1·62] | 0·141 | 0·27 [0·04–1·83] | 0·177 |
| Moderate PAD | 11·47 [3·95–33·30] | <0·001 | 15·91 [4·95–51·15] | <0·001 |
| Ischaemia (critical PAD) | 3·79 [1·03–13·93] | 0·045 | 5·02 [1·24–20·24] | 0·024 |
| Past surgeon treatment | 14·79 [4·52–48·33] | <0·001 | 12·01 [3·28–43·92] | <0·001 |
| Model 1 results | Pseudo R 2:0·600 omnibus: df = 7, P < 0·001 | Missing: 7 (1·0%); H&L: P = 0·495 | Pseudo R 2: 0·622 omnibus: df = 11, P < 0·001 | Missing: 29 (4·0%); H&L: P = 0·691 |
| Model 2 | ||||
| Acute foot trauma | 18·73 [4·93–71·20] | <0·001 | 23·29 [5·27–103·04] | <0·001 |
| Previous foot ulcer | 19·85 [8·51–46·29] | <0·001 | 24·39 [9·42–63·15] | <0·001 |
| Peripheral neuropathy | 4·19 [1·87–9·40] | 0·001 | 5·73 [2·37–13·85] | <0·001 |
| PAD | <0·001 | <0·001 | ||
| Nil PAD | Referent | Referent | ||
| Mild PAD | 0·35 [0·07–1·85] | 0·219 | 0·38 [0·07–2·08] | 0·262 |
| Moderate PAD | 10·71 [3·88–29·60] | <0·001 | 15·37 [5·03–46·91] | <0·001 |
| Ischaemia (critical PAD) | 8·36 [2·75–25·39] | <0·001 | 10·13 [3·03–33·84] | <0·001 |
| Model 2 results | Pseudo R 2:0·544 omnibus df = 6, P <0·001 | Missing: 7 (1·0%); H&L: P = 0·869 | Pseudo R 2: 0·579 omnibus: df = 10, P < 0·001 | Missing: 29 (4·0%); H&L: P = 0·890 |
CI, confidence interval; Pseudo R 2, Nagelkerke R 2; omnibus, omnibus tests of model coefficients; df, degrees of freedom; missing, excluded missing cases; H&L, Hosmer and Lemeshow test; PAD, peripheral arterial disease.
Foot infections were present in 24 participants [3·3% (2·2–4·9%)], 12 [1·7% (0·9–2·9%)] of whom had diabetes (Table 2). Of participants with foot infections, 11 (46%) were mild, 7 (29%) were moderate, and 6 (25%) were severe foot infections. After univariate analysis, 13 variables were associated with foot infection (all, P < 0·05) (Table A1). Foot wounds (foot ulcers and acute foot wounds) were excluded from infection models as infection was only present in wounds. Table 4 shows that after adjustment for socioeconomic status and PAD, foot infections were independently associated with previous foot ulcers, foot trauma and past surgeon treatment in Model 1 (P < 0·01). All independent variables remained significant in Model 2 (P < 0·01).
Table 4.
Independent associated factors for foot infections using multivariate logistical regression (odds ratios [95% CI])
| Risk factor | Unadjusted | P value | Adjusted | P value |
|---|---|---|---|---|
| Model 1 | ||||
| Acute foot trauma | 16·67 [3·73–74·43] | <0·001 | 19·16 [2·82–130·12] | 0·003 |
| Previous foot ulcer | 23·01 [7·97–66·44] | <0·001 | 39·81 [10·94–144·90] | <0·001 |
| Past surgeon treatment | 8·68 [2·71–27·83] | <0·001 | 19·88 [3·56–110·98] | 0·001 |
| Model 1 results | Pseudo R 2:0·452 omnibus: df = 3, P < 0·001 | Missing: 12 (1·6%); H&L: P = 1·00 | Pseudo R 2: 0·536 omnibus: df =10, P < 0·001 | Missing: 37 (5·0%); H&L: P = 0·904 |
| Model 2 | ||||
| Acute foot trauma | 13·65 [3·24–57·45] | <0·001 | No confounders identified | |
| Previous foot ulcer | 35·95 [13·11–98·56] | <0·001 | ||
| Model 2 results | Pseudo R 2:0·391 omnibus: df =2, P < 0·001 | Missing: 12 (1·6%); H&L: P = 1·00 | ||
Pseudo R 2, Nagelkerke R 2; omnibus, omnibus tests of model coefficients; df, degrees of freedom; missing, excluded missing cases; H&L, Hosmer and Lemeshow test.
Ischaemia was present in 33 participants [4·5% (3·2–6·3%)], 12 [1·7% (0·9–2·9%)] of whom had diabetes (Table 2). After univariate analysis, 16 variables were associated with ischaemia (all, P < 0·05) (Table A1). No confounders were identified. Table 5 shows ischaemia was independently associated with older age, current smoking and past surgeon treatment in Model 1 (P < 0·01); however, no independent variables remained significant in Model 2 (P > 0·05).
Table 5.
Independent associated factors for ischaemia using multivariate logistical regression (odds ratios [95% CI])
| Risk factor | Unadjusted | P value | Adjusted | P value |
|---|---|---|---|---|
| Model 1 | ||||
| Age (continuous year) | 1·06 [1·03–1·09] | <0·001 | No confounders identified | |
| Smoker | 4·92 [1·71–14·17] | 0·003 | ||
| Past surgeon treatment | 21·02 [8·51–51·91] | <0·001 | ||
| Model 1 results: | Pseudo R 2:0·235 omnibus: df = 3, P < 0·001 | Missing: 7 (1·0%); H&L: P = 0·141 | ||
| Model 2 | ||||
| Nil | All | >0·05 | All | >0·05 |
| Model 1 results | ||||
Pseudo R 2, Nagelkerke R2; omnibus, omnibus tests of model coefficients; df, degrees of freedom; missing, excluded missing cases; H&L, Hosmer and Lemeshow test.
Acute foot wounds
Acute foot wounds were present in 17 participants [2·3% (1·4–3·7%)], one [0·1% (0–0·9%)] of whom had diabetes (Table 2). Acute foot wounds made up 26% of the 66 participants with any type of foot wound. After univariate analysis, five variables were associated with acute foot wounds (all, P < 0·05) (Table A2). Foot trauma was excluded from acute foot wound models as all participants with acute foot wounds had acute foot trauma. Table 6 shows that acute foot wounds were not independently associated with any variables in Model 1 (P > 0·05). However, in Model 2 after adjustment for inside footwear worn and foot deformity, acute foot wounds were independently associated with younger age (<40 years) (P < 0·05).
Table 6.
Independent associated factors for acute foot wounds using multivariate logistical regression (odds ratios [95% CI])
| Risk factor | Unadjusted | P value | Adjusted | P value |
|---|---|---|---|---|
| Model 1 | ||||
| Nil | All | >0·05 | All | >0·05 |
| Model 1 results: | ||||
| Model 2 | ||||
| Age groups | 0·002 | 0·006 | ||
| 18–40 years | Referent | Referent | ||
| 41–60 years | 0·25 [0·07–0·82] | 0·023 | 0·24 [0·07–0·82] | 0·023 |
| 61–80 years | 0·11 [0·03–0·41] | 0·001 | 0·07 [0·01–0·38] | 0·002 |
| 81+ years | 0·10 [0·01–0·78] | 0·028 | 0·11 [0·01–1·04] | 0·054 |
| Model 2 results | Pseudo R 2:0·103 omnibus: df = 3, P = 0·002 | Missing: 4 (0·5%); H&L, P = 1·00 | Pseudo R 2: 0·166 omnibus: df = 7, P = 0·002 | Missing: 30 (4·1%); H&L: P = 0·642 |
Pseudo R 2, Nagelkerke R 2; omnibus, omnibus tests of model coefficients; df, degrees of freedom; missing, excluded missing cases; H&L, Hosmer and Lemeshow test.
Amputation procedures
Amputation procedures were present in 14 participants [1·9% (1·1–3·2%)], 10 [1·4% (0·7–2·5%)] of whom had diabetes (Table 2). Of those 14, eight (57%) were minor, and six (43%) were major amputations. Minor and major amputation procedures were combined for regression because of the limited number of amputation procedures. The reason for the amputation procedure was foot disease in 12 participants and foot trauma and multi‐organ failure in one participant each. After univariate analysis, 17 variables were associated with having an amputation procedure (all, P < 0·05) (Table A2). Foot ulcers and acute foot wounds were excluded as collinearity was identified with foot infection and foot trauma, respectively. Table 7 shows amputations were not independently associated with any variables in Model 1 (P > 0·05). However, in Model 2, after adjustment for previous foot ulcers, amputations were independently associated with foot infection in Model 2 (P < 0·01).
Table 7.
Independent associated factors for amputation procedures using multivariate logistical regression (odds ratios [95% CI])
| Risk factor | Unadjusted | P value | Adjusted | P value |
|---|---|---|---|---|
| Model 1 | ||||
| Nil | All | >0·05 | All | >0·05 |
| Model 1 results | ||||
| Model 2 | ||||
| Infection | 77·22 [17·88–333·43] | <0·001 | 18·14 [2·97–110·76] | 0·002 |
| Model 2 results | Pseudo R 2:0·336 omnibus: df =1, P < 0·001 | Missing: 11 (1·5%); H&L: P = 1·00 | Pseudo R 2: 0·390 omnibus: df = 2, P < 0·001 | Missing: 12 (1·6%); H&L: P = 1·00 |
Pseudo R 2, Nagelkerke R 2, omnibus, omnibus tests of model coefficients; df, degrees of freedom; missing, excluded missing cases; H&L, Hosmer and Lemeshow test.
Discussion
Our study is the first to investigate the overall inpatient burden of foot‐related conditions within a representative inpatient population. Our findings indicate that 11·8% of all inpatients had a major foot‐related condition present. Foot disease was present in 9·8% of participants (6·7% ulcers, 3·3% infections and 4·5% ischaemia); 2·3% had acute foot wounds and 1·9% new amputation procedures. Interestingly, 46% of participants with foot disease had diabetes, whereas 70% of those undergoing an amputation procedure had diabetes. Foot ulcers and infections were more likely in inpatients with a previous foot ulcer, trauma, PAD, peripheral neuropathy or those who had past foot treatment by a surgeon. Ischaemia was more likely to be found in inpatients of older age, smokers or those who had past surgeon treatment. Amputation procedures were more likely in those with a foot infection, whereas acute foot wounds were more likely in younger inpatients. These findings suggest that the overall foot‐related inpatient burden is considerably larger than historically appreciated and is mostly made up of foot disease.
Although this is the first study of its kind, our prevalence findings are generally consistent with the limited available previous reports on specific or subgroups of foot disease disorders from a recent systematic review 6. First, our 9·8% foot disease finding fell within the review's crude heterogeneous range (0·2–11·9%), as did our 6·7% foot ulcer (0·3–13·5%) and 3·3% foot infection finding (0·1–6·4%) 6. Our 4·5% ischaemia finding was much lower than the 7·2% from the only previous similar study 16. However, this may be explained by the previous study investigating only people over 40 years using ankle brachial indices and medical record audits, which have a higher false positive rate than the toe systolic pressures used by our study 16. Second, our 4·5% diabetes‐related foot disease finding was remarkably consistent with the 4·7% pooled prevalence estimate from the review 6. Also, our 1·4% diabetes‐related amputation procedure prevalence was very similar to the 1·5% reported from the review 6. While our 3·5% diabetes‐related foot ulcer prevalence was higher than the 2·4% pooled prevalence estimate 6, our 1·7% diabetes‐related foot infections was lower than the 3·4% pooled estimate 6. Lastly, consistent with diabetes‐related foot infection studies, all foot infections in our study were present within foot wounds, regardless of diabetes 6, 11. Additionally, our infection severity findings were similar to the largest prospective diabetes‐related foot infection study 11; 46% for mild (versus 47% in a previous study), 29% moderate (versus 34%) and 25% severe infections (versus 18%) 11.
This general consistency with available findings from the literature reassures us of the reliability and validity of our overall findings. In our previous study, we found that 4·9% of all inpatients were in hospital for the primary reason of foot disease 4. The 9·8% findings from this study, in combination with the findings of the previous study 4, suggests that one in every 10 representative inpatients (9·8%) has foot disease present, and half (4·9%) of those are in hospital because of their foot disease 4. Interestingly, the equivalent diabetes‐related foot disease findings from these studies suggest that one in every 22 representative inpatients (4·5%) has diabetes‐related foot disease present, and nearly half (2·0%) of those are in hospital because of their foot disease 4. Alternatively, in the 23·5% of inpatients with diabetes, this suggests that one in every five diabetes inpatients (19·2%) has foot disease present, and nearly one in 10 diabetes inpatients (8·7%) are in hospital because of their foot disease 4.
When interpreted against other Australian inpatient literature, our previous findings indicated that foot disease was a top 10 direct cause of hospitalisation in Australia 4. The valid and reliable self‐reported medical history findings of this study now allows for a more direct comparison of the size of the overall burden of foot disease to other medical conditions in a representative sample of inpatients. This study identified that similar proportions of inpatients had a self‐reported foot ulcer (disease) history (10%) to those reporting a chronic kidney disease history (12%) or cerebrovascular disease history (12%). This suggests that the overall inpatient burden imposed by foot disease is comparable in size to those imposed by the more well‐known and resourced conditions of kidney and cerebrovascular disease. Furthermore, it is highly likely that our study under‐reported those with a foot disease history as some patients with previous amputations did not also have previous foot ulcers, and we did not capture those with previous foot infections or previous ischaemia. Considering that our previous study reported foot disease to be a leading cause of hospitalisation in Australia when compared to other inpatient disease in similar inpatient literature 4, 17, this paper reinforces the need for policy makers, clinicians and researchers to address the overall inpatient burden of foot disease in the same way they already do for other leading causes of the overall inpatient burden 4. This recommendation is further reinforced when considering that the population we investigated has been reported to have very similar demographic, diabetes and medical history characteristics to those reported in other large Australian and international inpatient studies 4.
This is one of the first studies to investigate associations with foot‐related conditions 6, 7, 18. Interestingly, although we adjusted for diabetes, our findings are consistent with studies reporting factors associated with diabetes‐related foot disease 19, 20. First, studies investigating diabetes‐related foot ulcers consistently identify the common risk factors of peripheral neuropathy, PAD, previous foot ulcers and trauma 19, 20. After controlling for diabetes, these factors were also independently associated with foot ulcers in our study. Second, consistently reported risk factors for diabetes‐related foot infection are foot ulcers, previous foot ulcers and trauma, and again after controlling for diabetes, these were the factors identified in our study 10, 11. Third, we identified older age and smoking to be independently associated with inpatient ischaemia, which was consistent with the only previous similar inpatient study of ischaemia 16. Fourth, while acute foot wounds in inpatients had yet to be studied, the independent factor of younger age identified in our study is consistent with trauma‐related amputation literature 12, 13. Lastly, nearly all amputation procedures were performed in patients with active foot disease, which is consistent with most amputation literature 7, 11, 12, 13. However, interestingly, 70% of these amputation procedures were on inpatients with diabetes‐related foot disease, yet people with diabetes only represented 46% of the inpatients with foot disease. This finding suggests that diabetes inpatients with foot disease have worse hospitalisation outcomes than non‐diabetes inpatients with foot disease. It could be hypothesised that this was because inpatients with diabetes present with an increased severity of foot disease, which was not reflected in the limited severity measures reported in our study (such as ischaemia or infection severity), or perhaps inpatients with foot disease are treated differently if they have diabetes compared to if they do not have diabetes while they are in hospital.
In addition to exploring the demographic, social determinant, comorbidity and foot disease history factors typically investigated in similar diabetes‐related foot disease studies, our study also explored associations with previous foot treatment, footwear and self‐care ability. Past foot treatment by a surgeon in the year prior to hospitalisation was the only factor identified from these modifiable variables to be independently associated with foot disease disorders. This is not surprising considering surgeons are a key member of recommended outpatient multidisciplinary foot disease teams 21, 22, 23. Perhaps more surprising was that no other footwear, self‐care or past foot treatment factor was also associated with inpatient foot disease disorders. In particular, there was no association found in our study with any of the other recommended outpatient multidisciplinary foot disease team members (medical, podiatry, nursing and orthotist 21, 22, 23, 24, 25, 26). It may be hypothesised that inpatients with foot disease were more likely to be in hospital because they had not been attended by such a multidisciplinary foot disease team prior to hospitalisation 4, 21, 22, 23, 24, 25, 26. This hypothesis is reinforced when considering that Australian regions implementing access to outpatient multidisciplinary foot disease teams have demonstrated significant reductions in hospitalisation and amputation rates in diabetes populations 22, 27. Regardless, as per previous recommendations, hospitalisation is seen as an ideal opportunity to triage inpatients with foot disease into the recommended care of multidisciplinary foot disease teams to reduce future inpatient burdens from re‐admission and amputation 4, 22, 24, 25, 26.
The findings of this study also have other potentially significant future policy, clinical and research implications. First, our findings suggest that an average hospital with 600 beds could expect to house at any given time 71 inpatients with a major foot‐related condition present, including 59 with foot disease and 11 recovering from an amputation procedure. Forecasting this across Australia's 49 153 available overnight public hospital beds 18 suggests that 4817 public hospital beds each night host a patient with foot disease, including 934 with a patient recovering from an amputation procedure. With the cost of an Australian hospital bed on average being AU$971 28, it can be extrapolated that each year foot disease contributes to an overall cost burden of AU$4·7 billion (directly and indirectly) on the Australian public hospital system. Second, with such a comparatively large inpatient burden of foot disease, policy makers should consider implementing similar inpatient continuum of care strategies that are commonly used for other large inpatient burdens, such as chronic kidney disease and cerebrovascular disease. Furthermore, strong consideration should be given to expanding coverage of multidisciplinary diabetic foot disease teams to all inpatients with foot disease regardless of diabetes. Third, with a self‐reported previous foot ulcer history being found to be independently associated with foot disease, simply questioning all inpatients on admission to hospital may be a very effective and efficient recommendation to identify the vast majority of inpatients with foot disease. Last, further research is recommended to investigate both the predictors and successful interventions for the very large inpatient population with foot disease, particularly those with non‐diabetic foot disease.
This study should be read cognisant of several strengths and limitations as previously reported 4. In brief, strengths included using existing literature to adequately power the study 4, 6; investigating an inpatient population that was highly representative of reported Australian inpatient populations 4, 17, 29 and comparable to international inpatient populations, particularly for diabetes prevalence 4, 29, 30, 31; employing trained and tested data collectors 4, 8 who used a valid and reliable data collection instrument 4, 8; and analysing data using recommended regression models, adjusting for a range of confounders 14, 15. Limitations included using a cross‐sectional study design that can only report associated factors; excluding older, cognitively impaired patients reported to have higher foot disease prevalence; only investigating for full‐thickness wounds, which under‐reports stage 1 pressure injuries 6; and using self‐reported past foot treatment variables. Additionally, amputation procedures were aggregated, and this may have affected our regression findings as minor and major amputation procedures are generally performed for different clinical reasons 7, 12, 13. Finally, it is acknowledged that performing a large number of statistical comparisons and using regression on outcomes with less than 20 cases does increase the likelihood of a type 1 statistical error 14, 15.
In conclusion, our methodologically robust point prevalence study is the first to report the overall foot‐related inpatient burden. Our study identified that foot‐related conditions, particularly foot disease, caused an overall inpatient burden that is comparable to other well‐known inpatient burdens, such as those caused by chronic kidney disease and cerebrovascular disease. Furthermore, although our study adjusted for diabetes, we still found similar independent factors associated with foot disease that had been previously found for diabetes‐related foot disease. It is recommended that policy makers, clinicians and researchers seriously consider adopting inpatient strategies that have been used with success in other large comparable well‐known diseases so as to reduce this large, yet seemingly silent, overall inpatient burden caused by foot disease.
Acknowledgements
This work was kindly supported by grant funding from Queensland Health (Queensland Government, Australia) and the Wound Management Innovation Cooperative Research Centre (Australia). The authors also warmly acknowledge the tireless work of the Queensland Health‐employed podiatrists and Queensland University of Technology podiatry students who undertook the training, testing and data collection for this project. Without their enthusiasm, this study would not have been possible. The authors declare they have no competing interests.
Table A1.
Univariate analysis for participants with a foot ulcer, foot infection or ischaemia
| Foot ulcer | Foot infection | Ischaemia | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Variables | All | n (%) | Odds ratio [95% CI] | P value | n (%) | Odds ratio [95% CI] | P value | n (%) | Odds ratio [95% CI] | P value |
| Participants* | 733 | 49 (6·7%) | [5·1–8·7%] | 24 (3·3%) | [2·2–4·9%] | 33 (4·5%) | [3·2–6·3%] | |||
| Medical ward | 459 (62·6%) | 26 (53·1%) | 1·00 | 13 (54·2%) | 1·00 | 16 (48·5%) | 1·00 | |||
| Surgical ward | 274 (37·4%) | 23 (46·9%) | 1·53 [0·85–2·74] | 0·153*, † | 11 (45·8%) | 1·46 [0·65–3·32] | 0·361† | 17 (51·5%) | 1·83 [0·91–3·69] | 0·090*, † |
| Demographics | ||||||||||
| Age (SD) years | 62·0 (18·6) | 66·0 (15·7) | 1·01 [0·99–1·03] | 0·128* | 61·1 (18·8) | 1·00 [0·98–1·02] | 0·799 | 71·1 (13·4) | 1·03 [1·01–1·06] | 0·005** |
| Age groups | 0·124* | 0·450 | 0·069* | |||||||
| 18–40 years | 110 (15·0%) | 4 (8·2%) | 1·00 | 3 (12·5%) | 1·00 | 1 (3·0%) | 1·00 | |||
| 41–60 years | 188 (25·7%) | 11 (22·4%) | 1·67 [0·52–5·36] | 0·392 | 7 (29·2%) | 1·36 [0·34–5·36] | 0·664 | 5 (15·2%) | 3·01 [0·35–26·11] | |
| 61–80 years | 316 (43·2%) | 29 (59·2%) | 2·68 [0·92–7·80] | 0·071 | 13 (54·2%) | 1·51 [0·42–5·41] | 0·535 | 18 (54·5%) | 6·65 [0·88–50·42] | 0·067 |
| 81+ years | 117 (16·0%) | 5 (10·2%) | 1·18 [0·31–4·52] | 0·806 | 1 (4·2%) | 0·30 [0·03–2·92] | 0·299 | 9 (27·3%) | 9·08 [1·13–72·93] | 0·038 |
| Male gender | 408 (55·8) | 28 (57·1%) | 1·06 [0·59–1·90] | 0·848 | 16 (66·7%) | 1·61 [0·68–3·82] | 0·278 | 20 (60·6%) | 1·22 [0·60–2·50] | 0·580 |
| Social determinants | ||||||||||
| Socioeconomic status | 711 | 0·328 | 0·586 | 0·181* | ||||||
| Most disadvantaged | 102 (14·3%) | 10 (22·2%) | 1·00 | 3 (14·3%) | 1·00 | 8 (25·8%) | 1·00 | |||
| Second most disadvantaged | 159 (22·4%) | 12 (26·7%) | 0·74 [0·31–1·79] | 0·507 | 7 (33·3%) | 1·52 [0·38–6·02] | 0·552 | 6 (19·4%) | 0·46 [0·15–1·36] | 0·159 |
| Middle | 98 (13·8%) | 4 (8·9%) | 0·39 [0·12–1·28] | 0·120 | 1 (4·8%) | 0·34 [0·03–3·30] | 0·350 | 3 (9·7%) | 0·36 [0·09–1·41] | 0·144 |
| Second least disadvantaged | 240 (33·8%) | 15 (33·3%) | 0·61 [0·26–1·41] | 0·246 | 8 (38·1%) | 1·13 [0·29–4·33] | 0·864 | 13 (41·9%) | 0·66 [0·27–1·65] | 0·375 |
| Least disadvantaged | 112 (15·8%) | 4 (8·9%) | 0·34 [0·10–1·11] | 0·074 | 2 (9·5%) | 0·60 [0·10–3·66] | 0·579 | 1 (3·2%) | 0·10 [0·01–0·84] | 0·034 |
| Geographic remoteness | 711 | 0·907 | 0·807 | 0·345 | ||||||
| Major city | 435 (61·2%) | 26 (57·8%) | 1·00 | 13 (61·9%) | 1·00 | 17 (54·8%) | 1·00 | |||
| Inner regional area | 153 (21·5%) | 9 (20·0%) | 0·99 [0·62–3·97] | 0·975 | 4 (19·0%) | 0·90 [0·29–2·81] | 0·858 | 5 (16·1%) | 0·84 [0·30–2·31] | 0·729 |
| Outer regional area | 66 (9·3%) | 6 (13·3%) | 1·57 [0·62–3·97] | 0·341 | 2 (9·5%) | 1·02 [0·27–4·64] | 0·976 | 6 (19·4%) | 2·44 [0·93–6·44] | 0·071 |
| Remote area | 30 (4·2%) | 2 (4·4%) | 1·12 [0·25–4·97] | 0·881 | 0 | 0 | NA | 1 (3·2%) | 0·84 [0·11–6·55] | 0·869 |
| Very remote area | 27 (3·8%) | 2 (4·4%) | 1·26 [0·28–5·59] | 0·765 | 2 (9·5%) | 2·58 [0·55–12·06] | 0·229 | 2 (6·5%) | 1·95 [0·43–8·93] | 0·388 |
| <Year 10 education level | 395 (54·0%) | 32 (65·3%) | 1·65 [0·91–3·04] | 0·105* | 14 (58·3%) | 1·21 [0·53–2·76] | 0·652 | 25 (75·8%) | 2·79 [1·24–6·28] | 0·013** |
| Indigenous | 34 (4·6%) | 3 (6·1%) | 1·37 [0·40–4·64] | 0·616 | 2 (8·3%) | 1·89 [0·43–8·40] | 0·402 | 2 (6·1%) | 1·34 [0·31–5·82] | 0·701 |
| Born overseas | 161 (22·0%) | 5 (10·2%) | 0·38 [0·15–0·98] | 0·045** | 2 (8·3%) | 0·31 [0·07–1·32] | 0·113* | 4 (12·1%) | 0·47 [0·16–1·36] | 0·164* |
| Medical condition history | ||||||||||
| Diabetes | 172 (23·5%) | 26 (53·1%) | 4·15 [2·30–7·49] | <0·001** | 12 (50·0%) | 3·45 [1·52–7·84] | 0·003** | 12 (36·4%) | 1·93 [0·93–4·00] | 0·079* |
| Hypertension | 359 (49·0%) | 29 (59·2%) | 1·56 [0·86–2·80] | 0·141* | 12 (50·0%) | 1·04 [0·46–2·34] | 0·929 | 22 (66·7%) | 2·15 [1·03–4·50] | 0·042** |
| Dyslipidaemia | 234 (31·9%) | 17 (34·7%) | 1·15 [0·62–2·11] | 0·661 | 6 (25·0%) | 0·71 [0·28–1·81] | 0·468 | 17 (51·5%) | 2·37 [1·18–4·78] | 0·016** |
| Myocardial infarction | 146 (19·9%) | 9 (18·4%) | 0·90 [0·42–1·89] | 0·771 | 4 (16·7%) | 0·79 [0·27–2·35] | 0·674 | 12 (36·4%) | 2·42 [1·16–5·03] | 0·019* |
| Cerebrovascular accident | 85 (11·6%) | 6 (12·2%) | 10·7 [0·44–2·58] | 0·889 | 4 (16·7%) | 1·55 [0·52–4·64] | 0·436 | 7 (21·2%) | 2·13 [0·90–5·07] | 0·088* |
| Chronic kidney disease | 89 (12·1%) | 10 (20·4%) | 1·96 [0·94–4·07] | 0·073* | 3 (12·5%) | 1·03 [0·31–3·53] | 0·960 | 7 (21·2%) | 2·04 [0·86–4·85] | 0·106* |
| Cancer | 174 (23·7%) | 15 (30·6%) | 1·45 [0·77–2·73] | 0·249 | 7 (29·2%) | 1·31 [0·54–3·22] | 0·553 | 6 (18·2%) | 0·71 [0·29–1·75] | 0·453 |
| Arthritis | 274 (37·4%) | 23 (46·9%) | 1·55 [0·86–2·75] | 0·147* | 12 (50·0%) | 1·68 [0·74–3·79] | 0·213 | 17 (51·5%) | 1·85 [0·92–3·72] | 0·086* |
| Depression | 191 (26·1%) | 11 (22·4%) | 0·81 [0·40–1·61] | 0·545 | 6 (25·0%) | 0·94 [0·37–2·40] | 0·897 | 8 (24·2%) | 0·91 [0·40–2·05] | 0·818 |
| Acute foot trauma | 26 (3·5%) | 10 (20·4%) | 10·67 [4·55–25·06] | <0·001** | 6 (25·0%) | 16·31 [5·62–47·30] | <0·001** | 1 (3·0%) | 0·84 [0·11–6·38] | 0·864 |
| Smoker | 104 (14·2%) | 7 (14·3%) | 1·02 [0·44–2·33] | 0·968 | 7 (29·2%) | 2·65 [1·07–6·56] | 0·035** | 8 (24·2%) | 2·02 [0·89–4·61] | 0·095* |
| Ex‐smoker | 304 (41·5%) | 19 (38·8%) | 0·89 [0·49–1·61] | 0·694 | 10 (41·7%) | 1·00 [0·44–2·29] | 0·997 | 12 (36·4%) | 0·800 [0·39–1·65] | 0·542 |
| Self‐care ability | ||||||||||
| Mobility impairment | 242 (33·2%) | 29 (59·2%) | 3·17 [1·75–5·74] | <0·001** | 11 (45·8%) | 1·77 [0·78–4·00] | 0·173* | 17 (51·5%) | 2·25 [1·12–4·54] | 0·023** |
| Vision impairment | 110 (15·1%) | 10 (20·4%) | 1·49 [0·72–3·08] | 0·284 | 6 (25·0%) | 1·92 [0·74–4·95] | 0·177* | 12 (36·4%) | 3·51 [1·67–7·37] | 0·001** |
| Footwear worn: inside | 0·655 | 0·202 | 0·470 | |||||||
| Low‐risk footwear | 81 (11·1%) | 6 (12·5%) | 1·00 | 4 (16·7%) | 1·00 | 2 (6·1%) | 1·00 | |||
| Moderate‐risk footwear | 263 (36·1%) | 20 (41·7%) | 1·03 [0·40–2·66] | 0·953 | 6 (25·0%) | 0·45 [0·12–1·63] | 0·224 | 16 (48·5%) | 2·56 [0·58–11·37] | 0·217 |
| High‐risk footwear | 139 (19·1%) | 6 (12·5%) | 0·57 [0·18–1·82] | 0·342 | 2 (8·3%) | 0·28 [0·05–1·57] | 0·149 | 5 (15·2%) | 1·50 [0·28–7·90] | 0·635 |
| No footwear worn | 245 (33·7%) | 16 (33·3%) | 0·87 [0·33–2·31] | 0·785 | 12 (50·0%) | 0·99 [0·31–3·17] | 0·988 | 10 (30·3%) | 1·68 [0·36–7·84] | 0·509 |
| Footwear worn: outside | 0·885 | 0·888 | 0·224 | |||||||
| Low‐risk footwear | 386 (53·2%) | 23 (47·9%) | 1·00 | 12 (50·0%) | 1·00 | 15 (45·5%) | 1·00 | |||
| Moderate‐risk footwear | 75 (10·3%) | 6 (12·5%) | 1·37 [0·54–3·49] | 0·507 | 3 (12·5%) | 1·27 [0·35–4·63] | 0·713 | 7 (21·2%) | 2·55 [1·00–6·48] | 0·050 |
| High‐risk footwear | 250 (34·4%) | 18 (37·5%) | 1·23 [0·65–2·33] | 0·525 | 8 (33·3%) | 1·02 [0·41–2·53] | 0·967 | 10 (30·3%) | 10·4 [0·46–2·35] | 0·926 |
| No footwear worn | 15 (2·1%) | 1 (2·1%) | 1·13 [0·14–8·95] | 0·910 | 1 (4·2%) | 2·19 [0·27–17·99] | 0·468 | 1 (3·0%) | 1·77 [0·22–14·33] | 0·594 |
| Past foot treatment | ||||||||||
| Yes | 256 (34·9%) | 39 (79·6%) | 8·36 [4·10–17·05] | <0·001** | 19 (79·2%) | 7·65 [2·82–20·74] | <0·001** | 24 (72·7%) | 5·39 [2·47–11·79] | <0·001** |
| Podiatry | 180 (24·6%) | 25 (51·0%) | 3·54 [1·97–6·38] | <0·001** | 11 (45·8%) | 2·72 [1·20–6·18] | 0·017** | 17 (51·5%) | 3·50 [1·73–7·08] | 0·001** |
| GP | 93 (12·7%) | 26 (53·1%) | 10·38 [5·61–19·20] | <0·001** | 16 (66·7%) | 16·64 [6·89–40·19] | <0·001** | 9 (27·3%) | 2·77 [1·24–6·15] | 0·013** |
| Surgeon | 36 (4·9%) | 19 (38·8%) | 24·78 [11·71–52·43] | <0·001** | 10 (41·7%) | 20·09 [8·10–49·80] | <0·001** | 12 (36·4%) | 16·70 [7·34–37·99] | <0·001** |
| Physician | 21 (2·9%) | 6 (12·2%) | 6·21 [2·29–16·79] | <0·001** | 2 (8·3%) | 3·44 [0·75–15·75] | 0·112* | 2 (6·1%) | 2·30 [0·51–10·30] | 0·278 |
| Nurse | 20 (2·7%) | 12 (24·5%) | 27·32 [10·53–70·93] | <0·001** | 6 (25·0%) | 17·59 [6·01–51·52] | <0·001** | 5 (15·2%) | 8·10 [2·75–23·85] | <0·001** |
| Orthotist | 4 (0·5%) | 2 (4·1%) | 14·47 [1·99–105·01] | 0·008 | 2 (8·3%) | 31·68 [4·27–235·36] | 0·001 | 1 (3·0%) | 7·21 [0·73–71·24] | 0·091* |
| Other | 9 (1·2%) | 0 | 0 | NA | 0 | 0 | NA | 0 | 0 | NA |
| Foot disease history | ||||||||||
| Previous foot ulcer | 72 (9·8%) | 33 (67·3%) | 33·95 [17·22–66·95] | <0·001** | 18 (75·0%) | 38·00 [14·45–99·93] | <0·001** | 16 (48·5%) | 10·94 [5·24–22·83] | <0·001**, † |
| Previous amputation | 30 (4·1%) | 14 (29·2%) | 17·14 [7·73–37·98] | <0·001** | 7 (29·2%) | 13·25 [4·97–35·37] | <0·001** | 9 (27·3%) | 12·64 [5·21–30·65] | <0·001**, † |
| Foot risk factors | ||||||||||
| Peripheral neuropathy | 160 (22·0%) | 30 (62·5%) | 7·05 [3·81–13·04] | <0·001** | 15 (62·5%) | 6·55 [2·81–15·27] | <0·001** | 14 (42·4%) | 2·79 [1·37–5·70] | 0·005** |
| PAD | <0·001** | <0·001** | ||||||||
| Nil PAD | 575 (79·0%) | 18 (37·5%) | 1·00 | 12 (50·0%) | 1·00 | – | – | – | ||
| Mild PAD | 69 (9·5%) | 2 (4·2%) | 0·92 [0·21–4·07] | 0·916 | 0 | 0 | NA | – | – | – |
| Moderate PAD | 51 (7·0%) | 15 (31·3%) | 12·89 [6·00–27·67] | <0·001** | 8 (33·3%) | 8·84 [3·43–22·82] | <0·001** | – | – | – |
| Ischaemic (Critical PAD) | 33 (4·5%) | 13 (27·1%) | 20·11 [8·67–46·65] | <0·001** | 4 (16·7%) | 6·88 [2·08–22·73] | 0·002** | – | – | – |
| Foot deformity | 158 (22·4%) | 19 (41·3%) | 2·64 [1·43–4·88] | 0·002** | 8 (34·8%) | 1·92 [0·80–4·61] | 0·146* | 11 (36·7%) | 2·10 [0·98–4·51] | 0·058*, † |
| Foot disease disorders | ||||||||||
| Foot ulcer | 49 (6·7%) | – | – | – | 21 (87·5%) | 196·29 [54·8–703·5] | <0·001** | 13 (39·4%) | 12·26 [5·64–26·65] | <0·001**, † |
| Acute foot wound | 17 (2·3%) | – | – | – | 3 (12·5%) | 12·30 [3·05–49·70] | <0·001** | 0 | 0 | NA |
| Foot infection | 24 (3·3%) | 21 (46·7%) | 196·29 [54·8–703·5] | <0·001**, † | – | – | – | 4 (12·9%) | 4·95 [1·58–15·48] | 0·006**, † |
GP, general practitioner; PAD, peripheral arterial disease; SD, standard deviation.
95% CI are for prevalence figures.
Explanatory variable excluded from multivariate model as considered not on causal pathway for outcome.
P < 0·2;
P < 0·05.
Table A2.
Univariate analysis for participants with an acute foot wound or new amputation procedure
| Acute foot wound | Amputation procedure | ||||||
|---|---|---|---|---|---|---|---|
| Variables | All | n (%) | Odds ratio [95% CI] | P value | n (%) | Odds ratio [95% CI] | P value |
| Participants* | 733 | 17 (2·3%) | [1·4–3·7%] | 14 (1·9%) | [1·1–3·2%] | ||
| Medical ward | 459 (62·6%) | 6 (35·9%) | 1·00 | 5 (35·7%) | 1·00 | ||
| Surgical ward | 274 (37·4%) | 11 (64·7%) | 3·61 [1·16–8·65] | 0·025**, † | 9 (64·3%) | 3·08 [1·02–9·28] | 0·046*, † |
| Demographics | |||||||
| Age (SD) years | 62·0(18·6) | 42·4(19·4) | 0·95 [0·93–0·97] | <0·001** | 65·5(12·7) | 1·01 [0·98–1·04] | 0·481 |
| Age groups | 0·002** | 0·702 | |||||
| 18–40 years | 110 (15·0%) | 9 (52·9%) | 1·00 | 0 | 1·00 | ||
| 41–60 years | 188 (25·7%) | 4 (23·5%) | 0·24 [0·07–0·82] | 0·023 | 4 (28·6%) | 0 | NA |
| 61–80 years | 316 (43·2%) | 3 (17·6%) | 0·11 [0·03–0·41] | 0·001 | 9 (64·3%) | 0 | NA |
| 81+ years | 117 (16·0%) | 1 (5·9%) | 0·10 [0·01–0·78] | 0·028 | 1 (7·1%) | 0 | NA |
| Male sex | 408 (55·8) | 13 (76·5%) | 2·62 [0·85–8·12] | 0·094* | 9 (64·3%) | 1·44 [0·48–4·33] | 0·519 |
| Social determinants | |||||||
| Socioeconomic Status | 711 | 0·657 | 0·983 | ||||
| Most disadvantaged | 102 (14·3%) | 1 (5·9%) | 1·00 | 3 (21·4%) | 1·00 | ||
| Second most disadvantaged | 159 (22·4%) | 5 (29·4%) | 3·25 [0·37–28·20] | 0·286 | 4 (28·6%) | 0·85 [0·19–3·87] | 0·836 |
| Middle | 98 (13·8%) | 3(17·6%) | 3·16 [0·32–30·89] | 0·323 | 0 | 0 | NA |
| Second least disadvantaged | 240 (33·8%) | 4 (1·7%) | 1·70 [0·19–15·42] | 0·636 | 5 (35·7%) | 0·71 [0·17–3·01] | 0·637 |
| Least disadvantaged | 112 (15·8%) | 4 (1·7%) | 3·70 [0·41–33·70] | 0·245 | 2 (14·3%) | 0·60 [0·10–3·67] | 0·580 |
| Geographic remoteness | 711 | 0·741 | 0·942 | ||||
| Major city | 435 (61·2%) | 9 (52·9%) | 1·00 | 8 (57·1%) | 1·00 | ||
| Inner regional area | 153 (21·5%) | 6 (35·3%) | 1·94 [0·68–5·55] | 0·216 | 4 (28·6%) | 1·43 [0·42–4·82] | 0·564 |
| Outer regional area | 66 (9·3%) | 1 (5·9%) | 0·73 [0·09–5·83] | 0·764 | 1 (7·1%) | 0·82 [0·10–6·66] | 0·852 |
| Remote area | 30 (4·2%) | 0 | 0 | NA | 0 | 0 | NA |
| Very remote area | 27 (3·8%) | 1 (5·9%) | 1·82 [0·22–14·89] | 0·578 | 1 (7·1%) | 2·05 [0·25–17·00] | 0·507 |
| <Year 10 education level | 395 (54·0%) | 8 (47·1%) | 0·75 [0·29–1·97] | 0·560 | 11 (78·6%) | 3·17 [0·88–11·46] | 0·078* |
| Indigenous | 34 (4·6%) | 1 (5·9%) | 1·29 [0·17–10·00] | 0·809 | 1 (7·1%) | 1·59 [0·20–12·56] | 0·658 |
| Born overseas | 161 (22·0%) | 2 (11·8%) | [0·11–2·05] | 0·311 | 2 (14·3%) | 0·58 [0·13–2·64] | 0·484 |
| Medical condition history | |||||||
| Diabetes | 172 (23·5%) | 1 (5·9%) | 0·20 [0·03–1·51] | 0·118* | 10 (71·4%) | 8·58 [2·66–27·72] | <0·001** |
| Hypertension | 359 (49·0%) | 4 (23·5%) | 0·31 [0·10–0·97] | 0·044** | 8 (57·1%) | 1·40 [0·48–4·08] | 0·535 |
| Dyslipidaemia | 234 (31·9%) | 2 (11·8%) | 0·28 [0·06–1·23] | 0·092* | 7 (50·5%) | 2·16 [0·75–6·24] | 0·153* |
| Myocardial Infarct | 146 (19·9%) | 0 | 0 | NA | 4 (28·6%) | 1·62 [0·50–5·25] | 0·419 |
| Cerebrovascular accident | 85 (11·6%) | 1 (5·9%) | 0·47 [0·06–3·58] | 0·465 | 3 (21·4%) | 2·12 [0·58–7·74] | 0·258 |
| Chronic kidney disease | 89 (12·1%) | 0 | 0 | NA | 4 (28·6%) | 2·98 [0·91–9·71] | 0·070* |
| Cancer | 174 (23·7%) | 0 | 0 | NA | 4 (28·6%) | 1·29 [0·40–4·14] | 0·671 |
| Arthritis | 274 (37·4%) | 3 (17·6%) | 0·35 [0·10–1·25] | 0·106* | 4 (28·6%) | 0·67 [0·21–2·15] | 0·498 |
| Depression | 191 (26·1%) | 8 (47·1%) | 2·58 [0·98–6·78] | 0·055* | 2 (14·3%) | 0·47 [0·10–2·10] | 0·321 |
| Acute foot trauma | 26 (3·5%) | 14 (82·4%) | 273·0 [69·28–1075·75] | <0·001 | 4 (28·6%) | 12·66 [3·68–43·50] | <0·001** |
| Smoker | 104 (14·2%) | 4 (23·5%) | 1·91 [0·61–5·98] | 0·266 | 3 (21·4%) | 1·67 [0·46–6·08] | 0·439 |
| Ex‐smoker | 304 (41·5%) | 7 (41·2%) | 0·99 [0·37–2·63] | 0·982 | 6 (42·9%) | 1·06 [0·37–3·10] | 0·911 |
| Self‐care ability | |||||||
| Mobility impairment | 242 (33·2%) | 4 (23·5%) | 0·61 [0·20–1·90] | 0·394 | 10 (71·4%) | 5·21 [1·62–16·77] | 0·006** |
| Vision impairment | 110 (15·1%) | 0 | 0 | NA | 3 (21·4%) | 1·55 [0·43–5·66] | 0·505 |
| Footwear worn: inside | 0·152* | 0·589 | |||||
| Low‐risk footwear | 81 (11·1%) | 2 (12·5%) | 1·00 | 2 (16·7%) | 1·00 | ||
| Moderate‐risk footwear | 263 (36·1%) | 1 (6·3%) | 0·15 [0·01–1·69] | 0·124 | 6 (50·0%) | 0·92 [0·18–4·66] | 0·922 |
| High‐risk footwear | 139 (19·1%) | 6 (37·5%) | 1·18 [0·35–9·11] | 0·480 | 2 (16·7%) | 0·58 [0·08–4·17] | 0·586 |
| No footwear worn | 245 (33·7%) | 7 (43·8%) | 1·16 [0·24–5·71] | 0·854 | 2(16·7%) | 0·33 [0·05–2·35] | 0·265 |
| Footwear worn: outside | 0·957 | 0·613 | |||||
| Low‐risk footwear | 386 (53·2%) | 9 (56·3%) | 1·00 | 8 (66·7%) | 1·00 | ||
| Moderate‐risk footwear | 75 (10·3%) | 1 (6·3%) | 0·57 [0·07–4·54] | 0·592 | 2 (16·7%) | 1·30 [0·27–6·22] | 0·747 |
| High‐risk footwear | 250 (34·4%) | 6 (37·5%) | 1·03 [0·36–2·94] | 0·950 | 2 (16·7%) | 0·38 [0·08–1·81] | 0·225 |
| No footwear worn | 15 (2·1%) | 0 | 0 | NA | 0 | 0 | NA |
| Past foot treatment | |||||||
| Yes | 256 (34·9%) | 3 (17·6%) | 0·39 [0·11–1·37] | 0·142* | 13 (92·9%) | 25·41 [3·31–195·39] | 0·002** |
| Podiatry | 180 (24·6%) | 1 (5·9%) | 0·19 [0·03–1·42] | 0·105* | 9 (64·3%) | 5·76 [1·90–17·41] | 0·002** |
| GP | 93 (12·7%) | 1 (5·9%) | 0·42 [0·06–3·22] | 0·406 | 11 (78·6%) | 28·44 [7·77–104·05] | <0·001** |
| Surgeon | 36 (4·9%) | 1 (5·9%) | 1·21 [0·16–9·41] | 0·854 | 9 (64·3%) | 46·07 [14·46–146·78] | <0·001** |
| Physician | 21 (2·9%) | 0 | 0 | NA | 2 (14·3%) | 6·13 [1·28–29·32] | 0·023** |
| Nurse | 20 (2·7%) | 1 (5·9%) | 2·29 [0·29–18·14] | 0·434 | 4 (28·6%) | 17·55 [4·97–61·93] | <0·001** |
| Orthotist | 4 (0·5%) | 1 (5·9%) | 14·81 [1·46–150·24] | 0·023 | 1 (7·1%) | 18·33 [1·77–188·18] | 0·014 |
| Other | 9 (1·2%) | 0 | 0 | NA | 0 | 0 | NA |
| Foot disease history | |||||||
| Previous foot ulcer | 72 (9·8%) | 1 (5·9%) | 0·57 [0·07–4·33] | 0·583 | 10 (71·4%) | 26·41 [8·05–86·68] | <0·001** |
| Previous amputation | 30 (4·1%) | 0 | 0 | NA | 7 (50·0%) | 30·17 [9·78–93·13] | <0·001** |
| Foot risk factors· | |||||||
| Peripheral neuropathy | 160 (22·0%) | 1 (6·3%) | 0·23 [0·03–1·77] | 0·159* | 10 (83·3%) | 18·87 [4·09–87·03] | <0·001** |
| PAD | NA | <0·001** | |||||
| Nil PAD | 575 (79·0%) | 16 (100%) | 1·00 | 4 (33·3%) | 1·00 | ||
| Mild PAD | 69 (9·5%) | 0 | 0 | NA | 0 | 0 | NA |
| Moderate PAD | 51 (7·0%) | 0 | 0 | NA | 4 (33·3%) | 12·15 [2·94–50·13] | 0·001** |
| Ischaemic (critical PAD) | 33 (4·5%) | 0 | 0 | NA | 4 (33·3%) | 19·69 [4·69–82·71] | <0·001** |
| Foot deformity | 158 (22·4%) | 3 (18·8%) | 0·80 [0·22–2·83] | 0·725 | 7 (63·6%) | 6·31 [1·82–21·82] | 0·004** |
| Foot disease disorders | |||||||
| Foot ulcer | 49 (6·7%) | – | – | – | 12 (85·7%) | 110·27 [23·80–510·82] | <0·001** |
| Acute foot wound | 17 (2·3%) | – | – | – | 2 (14·3%) | 7·80 [1·60–37·94] | 0·011** |
| Foot infection | 24 (3·3%) | 3 (27·3%) | 12·30 [3·05–49·70] | <0·001**, † | 6 (66·7%) | 77·22 [17·88–333·44] | <0·001** |
GP, general practitioner; PAD, peripheral arterial disease; SD, standard deviation.
95% CI are for prevalence figures.
Explanatory variable excluded from multivariate model as considered not on causal pathway for outcome.
P < 0·2;
P < 0·05.
References
- 1. Boulton AJM, Vileikyte L, Ragnarson‐Tennvall G, Apelqvist J. The global burden of diabetic foot disease. Lancet 2005;366:1719–24. [DOI] [PubMed] [Google Scholar]
- 2. Currie CJ, Morgan CL, Peters JR. The epidemiology and cost of inpatient care for peripheral vascular disease, infection, neuropathy, and ulceration in diabetes. Diabetes Care 1998;21:42–8. [DOI] [PubMed] [Google Scholar]
- 3. Lazzarini PA, Gurr JM, Rogers JR, Schox A, Bergin SM. Diabetes foot disease: the Cinderella of Australian diabetes management? J Foot Ankle Res 2012;5:24. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Lazzarini PA, Hurn SE, Kuys SS, Kamp MC, Ng V, Thomas C, et al. Direct inpatient burden caused by foot‐related conditions: a multisite point‐prevalence study. BMJ Open 2016;6:e010811. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Lazzarini PA, Hurn SE, Kuys SS, Kamp MC, Reed L. Foot‐related conditions in hospitalised populations: a literature review. Wound 2016;24:16–35. [Google Scholar]
- 6. Lazzarini PA, Hurn SE, Fernando M, Jen S, Kuys SS, Kamp MC, et al. Prevalence of foot disease and risk factors in general inpatient populations: a systematic review and meta‐analysis. BMJ Open 2015;5:e008544. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Mills JL Sr, Conte MS, Armstrong DG, Pomposelli FB, Schanzer A, Sidawy AN, et al. The Society for Vascular Surgery Lower Extremity Threatened Limb Classification System: risk stratification based on wound, ischemia, and foot infection (WIfI). J Vasc Surg 2014;59:220–34. e1–2. [DOI] [PubMed] [Google Scholar]
- 8. Lazzarini PA, Ng V, Kinnear EM, Kamp MC, Kuys SS, Hurst C, et al. The Queensland high risk foot form (QHRFF) ‐ is it a reliable and valid clinical research tool for foot disease? J Foot Ankle Res 2014;7:7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Schaper NC. Diabetic foot ulcer classification system for research purposes: a progress report on criteria for including patients in research studies. Diabetes Metab Res Rev 2004;20 Suppl 1:S90–5. [DOI] [PubMed] [Google Scholar]
- 10. Lipsky BA, Berendt AR, Cornia PB, Pile JC, Peters EJG, Armstrong DG, et al. Executive summary: 2012 Infectious Diseases Society of America clinical practice guideline for the diagnosis and treatment of diabetic foot infections. Clin Infect Dis 2012;54:1679–84. [DOI] [PubMed] [Google Scholar]
- 11. Lavery LA, Armstrong DG, Murdoch DP, Peters EJG, Lipsky BA. Validation of the Infectious Diseases Society of America's diabetic foot infection classification system. Clin Infect Dis 2007;44:562–5. [DOI] [PubMed] [Google Scholar]
- 12. Moxey PW, Gogalniceanu P, Hinchliffe RJ, Loftus IM, Jones KJ, Thompson MM, et al. Lower extremity amputations‐‐a review of global variability in incidence. Diabet Med 2011;28:1144–53. [DOI] [PubMed] [Google Scholar]
- 13. Lazzarini PA, O'Rourke SR, Russell AW, Clark D, Kuys SS. What are the key conditions associated with lower limb amputations in a major Australian teaching hospital? J Foot Ankle Res 2012;5:12. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Tabachnick BG, Fidell LS. Using Multivariate Statistics, 5th edn. Boston: Allyn and Bacon, 2007. [Google Scholar]
- 15. Hosmer D, Lemeshow S. Applied Logistic Regression, 2nd edn. New York: John Wiley & Sons, 2000. [Google Scholar]
- 16. Lacroix P, Aboyans V, Voronin D, Le Guyader A, Cautrès M, Laskar M. High prevalence of undiagnosed patients with peripheral arterial disease in patients hospitalised for non‐vascular disorders. Int J Clin Pract 2008;62:59–64. [DOI] [PubMed] [Google Scholar]
- 17. Australian Institute of Health & Welfare (AIHW) . Australian Hospital Statistics 2011–12. Canberra: AIHW, 2013. URL http://www.aihw.gov.au/publication‐detail/?id=60129543133 [accessed on 22 July 2016]. [Google Scholar]
- 18. Australian Institute of Health & Welfare (AIHW) . Hospital Resources 2013–2014: Australian Hospital Statistics. Canberra: AIHW, 2015. URL http://www.aihw.gov.au/publication‐detail/?id=60129551442 [accessed on 22 July 2016]. [Google Scholar]
- 19. Crawford F, Cezard G, Chappell FM, Murray GD, Price JF, Sheikh A, et al. A systematic review and individual patient data meta‐analysis of prognostic factors for foot ulceration in people with diabetes: the international research collaboration for the prediction of diabetic foot ulcerations (PODUS). Health Technol Assess 2015;19:1–210. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Monteiro‐Soares M, Boyko E, Ribeiro J, Ribeiro I, Dinis‐Ribeiro M. Risk stratification systems for diabetic foot ulcers: a systematic review. Diabetologia 2011;54:1190–9. [DOI] [PubMed] [Google Scholar]
- 21. National Health & Medical Research Council (NHMRC) . National Evidence‐based Guideline on Prevention, Identification and Management of Foot Complications in Diabetes (Part of the Guidelines on Management of Type 2 Diabetes). Melbourne: Baker IDI Heart & Diabetes Institute, 2011. URL https://www.nhmrc.gov.au/guidelines‐publications/di21 [accessed on 22 July 2016]. [Google Scholar]
- 22. Lazzarini PA, O'Rourke SR, Russell AW, Derhy PH, Kamp MC. Reduced incidence of foot‐related hospitalisation and amputation amongst persons with diabetes in Queensland, Australia. PLoS One 2015;10:e0130609. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Prompers L. Diabetic foot disease in European perspective: results from the Eurodiale study [PhD Thesis]. Maastricht: Maastricht University, 2008.
- 24. Wraight P. Improving clinical outcomes for patients with diabetes related foot complications [PhD Thesis]. Melbourne: The University of Melbourne, 2005.
- 25. Wraight PR, Lawrence SM, Campbell DA, Colman PG. Creation of a multidisciplinary, evidence based, clinical guideline for the assessment, investigation and management of acute diabetes related foot complications. Diabet Med 2005;22:127–36. [DOI] [PubMed] [Google Scholar]
- 26. Wukich DK, Armstrong DG, Attinger CE, Boulton AJM, Burns PR, Frykberg RG, et al. Inpatient management of diabetic foot disorders: a clinical guide. Diabetes Care 2013;36:2862–71. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27. Baba M, Davis WA, Norman PE, et al. Temporal changes in the prevalence and associates of diabetes‐related lower extremity amputations in patients with type 2 diabetes: the Fremantle diabetes study. Cardiovasc Diabetol 2015;14:152. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28. Graves N, Halton K, Doidge S, Clements A, Lairson D, Whitby M. Who bears the cost of healthcare‐acquired surgical site infection? J Hosp Infect 2008;69:274–82. [DOI] [PubMed] [Google Scholar]
- 29. Bach LA, Ekinci EI, Engler D, Gilfillan C, Hamblin PS, MacIsaac RJ, et al. The high burden of inpatient diabetes mellitus: the Melbourne public hospitals diabetes inpatient audit. Med J Aust 2014;201:334–8. [DOI] [PubMed] [Google Scholar]
- 30. American Diabetes Association . Economic costs of diabetes in the U.S. in 2012. Diabetes Care 2013;36:1033–46. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31. National Health Service . National Diabetes Inpatient Audit 2013. Leeds: Health & Social Care Information Centre, 2014. URL http://www.hscic.gov.uk/searchcatalogue?productid=14305 [accessed on 22 July 2016]. [Google Scholar]