Epidemiology and Health Care Cost of Diabetic Foot Problems
The diabetic lower extremity has long been a cause for both morbidity and mortality in patients afflicted with this multisystem disease. Unfortunately, the global prevalence of diabetes mellitus has been projected to nearly double from a baseline of 2.8% in 2000 to 4.4% by 2030, affecting over 350 million individuals (Wild et al. Diabetes Care. 2004;27(5):1047–53). In the decade beginning in 1997, the prevalence of diabetes in the USA has increased by 48% (http://apps.nccd.cdc.gov/DDTSTRS/default.aspx). Lower extremity morbidity contributes substantially to the toll diabetes takes on the individual and the health care system. This chapter focuses on the epidemiologic aspects of risk factors and complications in the diabetic lower extremity, particularly as they relate to the outcome of amputation. Included in the discussion is the influence of demographic factors, such as gender, age, race, and socioeconomic considerations, as well as the cost to the health care system of lower extremity disease in diabetes.
KeywordsDiabetic peripheral neuropathy Peripheral vascular disease Diabetic ulcer pathway Musculoskeletal deformities Ulcerations Amputations Lower extremity disease Diabetic limb
Epidemiology of Individual Risk Factors
The individual systems at risk that predispose an individual to ulceration are covered in greater detail throughout this textbook. In this chapter, a brief introduction to these risk factors is presented as they relate to the epidemiology of the at-risk foot.
Although many manifestations of neuropathy may go unrecognized by the patient, autonomic neuropathy is perhaps the most overlooked in the diabetic limb. In addition to contributing to impaired vasoregulation, it also may result in changes to the texture and turgor of the skin, such as dryness and fissuring. Dysregulation of local perspiration may contribute to increased moisture and increase the risk of fungal infections. With increased stiffness within the skin, areas of friction are less flexible and hyperkeratotic lesions may develop. Untreated, these lesions may progress with respect to thickness and induration, and exert increased pressure on deep tissues with resultant ulceration [18, 19].
Another form of neuropathy that influences the diabetic limb is reduced motor function. Frequently, this targets the intrinsic musculature of the foot resulting in joint instability. As innervation decreases, muscle wasting is observed. Over time, these imbalances lead to flexible deformities that become progressively more rigid. Rigid deformities are subject to greater pressure and predispose patients to ulcer formation .
Perhaps the most commonly recognized form of neuropathy among patients with diabetes is sensory neuropathy, resulting in the loss of sensation beginning in the most distal part of the extremity. This may manifest as an inability to detect temperature changes, vibration, proprioception, pressure, and, most seriously, pain. Some patients have a form of painful sensory neuropathy that includes symptoms, such as burning and tingling, known as paresthesias. This also contributes to the risk of ulcer formation as they may be unaware of pain associated with smaller injuries because of the persistent neuropathic pain . The prevalence of painful DPN is difficult to truly measure and define. NHANES estimated that 10.9% of adults with diabetes suffered from symptomatic DPN. Symptomatic DPN was defined as painful sensations, tingling, numbness, or loss of feeling. A population-based study through the Mayo clinic found that 20% of their diabetic cohort had painful DPN . In the UK, the prevalence of chronic painful DPN was found to be 16.2%  and the incidence, through a UK research database, was 15.3/100,000 patient-years (95% CI 14.9–15.7) . Although there is a lack of high-quality data available from a population health perspective, the prevalence of DPN is believed to increase with the duration of diabetes, poor glucose control, age, and smoking [12, 22, 23]. There is significant variability in the prevalence of DPN reported in the literature. This is most likely attributable to differences among each study’s population, geographic location, time period evaluated, definition of neuropathy, method of diagnosis, and source of data (i.e., patient self-report, billing codes, medical records, physician reports). It is important to note that peripheral neuropathy is likely the most important risk factor underlying the majority of diabetic lower extremity complications. Strong associations have been identified between peripheral neuropathy and DFU, diabetic foot infections, amputations, Charcot arthropathy, and surgical site infections over the last several decades [24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38] (REFS).
Peripheral Vascular Disease
Consequences of the compromised vascular system in diabetes can be among the most devastating complications. Both macro- and microvascular diseases are believed to contribute to the consequences of peripheral vascular disease, resulting in the inability of the dysvascular or ischemic limb to heal itself properly. Small injuries may progress to larger wounds because of reduced healing capacity. Delivery of systemic antibiotics can be compromised and leave infections uncontrolled. Among patients with diabetes, all blood vessels regardless of size and function are affected . The prevalence of peripheral arterial disease (PAD) is higher in people with diabetes compared to the general US population. NHANES found that the prevalence of PAD was 4.5% (95% CI 3.4–5.6) in the general population but increased to 9.5% (95% CI 5.5–13.4) in persons with diabetes . Figure 1.2 also illustrates that the largest disparity between genders was in 1996, and since that time the gap has reduced substantially with near equality of the rate per 1000 diabetic patients in 2003 . Studies have shown that peripheral vascular disease develops at a younger age among patients with diabetes as compared to the general population . In one large population-based study, over half of diabetic subjects were found to have absent pedal pulses, a common sign of impaired vascular function . Another study found that in patients with nonpalpable pulses, the relative risk of ulceration was 4.72 (95% CI 3.28, 6.78) as compared to a normal exam with all four pulses palpable . Ankle-brachial index (ABI), despite recognized limitations in the diabetic population, has also been used in diabetic screening. In patients with an ABI <0.90, the relative risk has been reported to be 1.25 (95% CI 1.05, 1.47) for developing an ulcer vs. diabetic patients with a normal ABI . In the widely published EURODIALE DFU study from Europe, patients with PAD had a 71% increased risk for failure to heal their ulcers and a 61% increased risk for infection compared to those foot ulcer patients without PAD (Prompers). The Society for Vascular Surgery published a clinical practice guideline in 2016 that reviews the association of PAD with diabetic foot complications and appropriate management strategies for the ischemic diabetic foot  (Hingorani).
Musculoskeletal deformities play an important role in the diabetic ulcer pathway. The presence or absence of a deformity, such as a hammertoe or bunion, predisposes the structures to increased pressure and friction. As noted above, motor neuropathy may contribute to such deformities, but other diabetes-associated complications such as glycation of collagen have also been indicted [45, 46, 47]. In a population-based study of a nationally representative sample, the prevalence of LED has been found to be significantly higher in those with diabetes (30.2% (95% CI 22.1–35.1)) compared to those without diabetes (18.7% (95% CI 15.9–21.4)) in the USA .
The prevalence of foot deformity in people with diabetes is not known, but the presence of foot deformity has been shown to increase the risk of developing a foot ulcer. One study found that 63% of patients who developed an ulcer had a fixed deformity beforehand . In one large population-based study of diabetes, the relative risk of ulcer occurrence was 2.56 (95% CI 2.04, 3.22) among patients with deformities as compared to individuals with no or few deformities . Boyko et al. identified the presence of an abnormally shaped foot as carrying a relative risk of 1.93 (95% CI 1.07, 3.48) for ulceration . A study by Mason and associates found that patients with diabetes had similar proportions of deformities to rheumatoid arthritis patients . Foot deformities, including limited joint mobility, lead to higher plantar foot pressures and these consequently often lead to an increased risk for DFU [38, 50, 51, 52]. Restriction of ankle joint dorsiflexion caused by a tight Achilles tendon (equinus deformity) has been found prevalent in neuropathic diabetic patients and is also associated with increased risks for forefoot ulcers [51, 53, 54, 55, 56].
Metabolic and Systemic Risk Factors
In addition to specific risk factors noted above, the prevalence of LED is also increased among patients with several modifiable systemic risk factors. Cross-sectional and cohort studies have established that better glycemic control is associated with reduced risk of lower extremity amputation (LEA), but this has been difficult to demonstrate in randomized trials [57, 58]. Nonetheless, one systematic review investigating the associations between glycated hemoglobin and amputation found an overall relative risk for LEA of 1.26 (95% CI 1.16–1.36) for each percentage increase in HgbA1c . The American Diabetes Association recommends that many complications, including LED, may be reduced by maintaining HbA1c <7.0%, blood pressure <130/80, HDL cholesterol >50 mg/dL, normal weight (BMI 18.5–25 kg/m ), and not smoking. Using data from 1999 to 2004 NHANES, Dorsey et al. reported that diabetic patients with LED were less likely to have met HbA1c (39.5% vs. 53.5%) and HDL cholesterol targets (29.7% vs. 41.1%) than patients without LED. Among non-Hispanic (NH) Blacks with LED, it was also noted that systolic and diastolic blood pressure was significantly less likely to be controlled than among non-Hispanic Whites . A recent review on diabetic foot ulcers also indicates the important role played by elevated glycated hemoglobin levels on their recurrence and the importance of maintaining optimal glucose control in this regard .
The Perfect Storm
Thus far, the presence of individual risk factors leading to ulceration has been described, but this fails to capture the interaction of these risk factors in the clinical setting. Reiber and colleagues have proposed a widely accepted causal pathway, which incorporates the relationships between risk factors and ulcerations . They advocate that the singular presence of individual risk factors represents a component but not a sufficient cause for acute ulceration. Rather, they found that the presence of two or more risk factors increased the risk of ulceration between 35 and 78% depending on the component risk factors. Furthermore, they noted that a “clinical triad” comprising neuropathy, minor foot trauma, and foot deformity was present in more than 63% of cohort patients who developed an ulcer. Peripheral neuropathy as represented by loss of protective threshold was evident in 78% of ulcer pathways, while peripheral vascular disease was a component cause in 35% of the pathways. Foot deformities were identified as a component cause in 63% of ulcer pathways .
As evidenced in Table 1.1 and Fig. 1.4, the development of a foot ulcer is a major risk factor for LEAs . In fact, it has been proposed that foot ulcers precede 84% of diabetes-related amputations and are a common diabetes-related cause of hospitalization [75, 76]. Moreover, patients with neuropathic diabetic foot ulceration have a 7% risk of amputation in the next 10 years .
Racial and Ethnic Disparities
Differences in the incidence of diabetic amputation vary substantially among racial and ethnic groups, although the overall incidence rate has decreased over time. The racial and ethnic divisions that follow are broadly defined in the manner most frequently employed by the CDC, AHRQ, and other monitoring agencies. Because these sources are updated on an annual basis, they provide readers with a consistent reference for these figures. Incidence is discussed in terms of the rate per 1000 persons with diabetes and the rate per 100,000 total population. Although the former calculation is the more informative from an epidemiologic perspective, it is also less accurate because of estimates made about the prevalence of diabetes. In most population-level studies, white non-Hispanic individuals frequently serve as the reference group in the USA. With this common reference, the risk of a White/NH diabetic patient would be equal to 1.0.
Gender and racial/ethnic differences have been presented above, but beyond the scope of clinical characteristics are regional and socioeconomic determinants, which have also been reported as a source of disparate outcomes . Socioeconomic status is a term that attempts to capture an individual’s capacity to function within society. This is often measured using their level of education, annual income, or community of residence. Several studies support the proposal that lower socioeconomic status carries a higher likelihood of amputation [97, 102]. This impacts the overall health of an individual in many ways. Lower education can reduce an individual’s health literacy, the understanding of one’s health, and behaviors that promote a healthy lifestyle. It may also impair early recognition of pathology before it becomes limb threatening. Annual income may impact the means to seek or obtain care or purchase supplies/medications to carry out treatments prescribed by the medical team. Lower income may also reflect an occupation that does not permit the absence from work in order to seek care.
The wealth of a community also can contribute to limitations in access to care and resources that can be directed to remove obstacles. An individual with an ulceration living in a wealthy community, where a specialized wound center was present and easily accessible to provide treatment, would be more likely to obtain care than an individual living in a resource-poor community, where the treatment options may be more limited and less effective in reducing the likelihood of progression. A frequently used proxy for community resources is the median income of a given zip code. Again, comparing data from 2004, the incidence of amputations was 33% higher in communities where the median income was less than $25,000 as compared to the incidence where the median income was $25,000–$34,999. This difference becomes even more substantial when compared to communities with a median income of $45,000 or more, where the incidence is 2.4 times greater in the under-$25,000 categories. Since the median income for the USA was $44,389 in 2004, these data suggests that the age- and gender-adjusted relative risk of amputation is between 25 and 240% higher for communities where the median income is below the national median than for communities where the median income is above the national median.
Amputation outcomes most often vary based on the level and location of procedure along with the corresponding postamputation complications. Generally, minor amputations are considered as limb salvage procedures and are associated with longer survival than major amputations. Each level carries with it different consequences ranging from recurrent foot ulcerations to death. The two most pressing consequences are those of subsequent amputation and death. One study found that the overall reamputation rates were 26.7, 48.3, and 60.7% after 1, 3, and 5 years following the index amputation, respectively. In general, the more proximal the amputation, the higher the likelihood of a more severe complication . During the first 12 months following a toe amputation, the risk of another amputation is 22.8% on the ipsilateral side and 3.5% on the contralateral side. Over a 5-year period, the risk increases to 52.3 and 29.5%, respectively. For midfoot amputations, 18.8% of patients required another amputation on the same side during the first year, and 9.4% required an amputation on the opposing limb during that same time. After 5 years, the incidence of amputation increases to 42.9% on the same limb and 33.3% on the contralateral limb. Individuals with either a transtibial or more proximal amputation had a reamputation proportion of 4.7 and 13.3% of the same extremity after 1 and 5 years, respectively. Surprisingly, a subsequent amputation of the contralateral limb occurred in 11.6% after 1 year and 53.3% after 5 years. It would be expected that a higher occurrence of additional amputations would be seen after distal procedures given the presence of more at-risk structures. A more recent study of 116 Veterans who underwent a forefoot amputation found that 49% underwent ipsilateral reamputation within 3 years after the initial procedure, with 79% having the reamputation within the first 6 months . These findings support an approach using frequent surveillance, careful monitoring, and postamputation education to reduce the risk of subsequent amputations [86, 105].
A direct causal relationship between amputation and short-term mortality has not been proven, but a strong association between these variables has been shown in several studies [72, 107, 108]. One proposed mechanism is that the postamputation exertion of gait stresses the cardiovascular system and increases the risk of a fatal cardiac event.
Amputation is not a benign outcome for either diabetic or nondiabetic patients. One study noted that the 1-, 5-, and 10-year mortality rates for nondiabetic individuals were 27.3%, 57.2%, and 77.1%, respectively. The study also noted that diabetic mortality was reported to be 32.8% after 1 year, 68.1% after 5 years, and 91.6% after 10 years. In the observed populations, the gap between the respective groups increased mortality rate from 5.5 to 14.5%. The authors concluded that diabetic patients had a 55% greater risk of death following amputation than nondiabetics, and that median survival was 27.2 months and 46.7 months, respectively . More recently, a very large study of diabetic patients in the UK was reported in 2015 that specifically investigated the association between LEA and risk for death. In this study of more than 416,000 persons with diabetes, 6566 (1.6%) had an LEA and 77,215 persons died during the 10-year study period. After adjusting for all known covariates that might also predict death in this population, there was a greater than twofold independent risk for death in patients who had undergone LEA (HR 2.37 (95% CI 2.27–2.48) . Hence it seems that diabetes-related LEA portends a significant risk of death even after controlling for major cardiovascular risk factors when compared to those diabetic patients not suffering lower extremity amputations.
As noted with the risk of reamputation, the risk of mortality is also influenced by the level of the index amputation. Within 1 year of the index amputation, mortality rates for diabetic patients were 6.6% after digital amputations, 4.4% after ray amputations, 10.5% after midfoot amputations, and 18.2% after a major amputation. Extending this to 5 years from the initial amputation, toe and ray amputations had mortality rates of 26.2% and 15.8%, respectively. Five-year mortality after a major amputation was found to be 36%, while midfoot amputations carried a risk of 21% .
Perioperative mortality has been reported to be quite high following amputation. Mortality rates have ranged between 5.8 and 23% during the first 30 days following amputation [97, 110, 111, 112, 113, 114]. Patients requiring a guillotine amputation secondary to sepsis have a particularly high perioperative mortality rate of 14.3% . The most frequently cited 30-day mortality causes have been cardiac events and sepsis . Short-term mortality following amputation is primarily related to cardiac events, with rates ranging from 28.5 to 52.2% [104, 110]. Sepsis is the second most frequent cause of death, with rates ranging from 14.2 to 26.1% [104, 110, 115]. The level of amputation again has an influence on this outcome. Two distinct studies demonstrate similar 30-day mortality rates following above-the-knee (AKA) or below-the-knee amputations. Subramaniam et al. reported 17.5 and 4.2% mortality, while Stone et al. reported 17.6% and 3.6%, respectively . The results by Stone et al. were more comprehensive and demonstrated a trend of increasing perioperative mortality as amputations became more proximal starting at the metatarsals and ending at the hip .
Cost of Lower Extremity Disease in Diabetes
Cost to the Health Care System
Thus far, this chapter has covered the epidemiologic aspects of the at-risk foot. The remaining portion focuses on the costs attributable to these conditions. Boulton et al. commented on the substantial economic burden that the diabetic foot places on the afflicted patient and the health care system, although they recognized that most estimates fail to account for preventive care, lost productivity, and rehabilitation. They further proposed that if these aspects were also added to the current estimates as much as 20% of diabetes costs could be associated with diabetic foot ulcers . The excess costs are primarily attributable to more frequent hospitalization, use of antibiotics, and need for amputations and other surgical procedures .
Harrington and colleagues examined excess costs attributable to patients with diabetic foot ulcers vs. those with diabetes alone. Among the Medicare population sampled, they found that the direct costs per patient per year were $15,300 among patients with ulcers vs. $5200 for patients without an ulcer .
Similar findings were noted in a health maintenance organization (HMO) population, where diabetic patients without ulcers had a cost per patient per year of $5080 while it remained substantially higher for patients with an ulcer at $26,490 per patient per year [63, 120]. Costs also vary considerably based on ulcer grade. In a large insurance claims database, Stockl et al. observed that the cost of an ulcer episode ranged from $1892 for a level 1 ulcer to $27,721 for level 4/5 ulcers . Overall, inpatient hospital charges comprised 77% of total costs.
Costs of various diabetic foot complications adjusted to the US currency in 1998 and 2010
Diabetes without ulcer
Diabetes with ulcer
DM ulcer with primary healing
DM ulcer with amputation
DM major amputation
DM minor amputation
Charges to hospitals for patients with diabetes by amputation level, 2008 
Diabetes with complications
Length of stay
Average charge ($)
Length of stay
Average charge ($)
Amputation through foot
Length of stay
Average charge ($)
Length of stay
Average charge ($)
Length of stay
Average charge ($)
Skin ulcers in diabetes with complications
Skin ulcers in diabetes without complications
Chronic ulcer in diabetes with complications
Chronic ulcer in diabetes without complications
Lower extremity amputation in diabetes with complications
Cost-Effectiveness of Prevention
Most physicians and patients agree that prevention of lower extremity ulceration, infection, and amputation is the most desirable clinical strategy, and several studies have shown that this approach is either highly cost-effective or cost saving. In the UK, a 2-year prospective cohort study of 2000 patients comparing a diabetic foot protection and screening program with conventional diabetes care demonstrated that only 24 patients in the protection program developed ulcers vs. 35 patients receiving conventional care. More importantly, only 7 of the patients with ulcers in the specialized program progressed to amputation, whereas 23 progressed in the conventional care group (p < 0.01). The total cost of the screening program was only £100 per patient per year while producing a savings of 11 amputations in 1000 patients at a cost of £12,084/amputation . A retrospective cohort study from Austria using a Markov model to estimate long-term costs and outcomes in a dedicated screening program compared with conventional care similarly concluded that the screening program would reduce costs by 29.8% for mild (grade A) ulcers and by 49.7% for severe (grade D) ulcers, primarily due to lower amputation rates . In a systematic review from the CDC on the cost-effectiveness of interventions to prevent diabetes and its complications, the use of comprehensive foot care to prevent ulcers was one of the few interventions found to be cost saving .
Evaluation of changes in quality of life, as reflected in cost–utility analysis, has shown similar results. Ortegon et al. used a Markov model to estimate lifetime risk of developing foot disease among newly diagnosed patients with type 2 diabetes receiving optimal foot care guidelines, intensive glycemic control, or standard care . In all simulations using a wide range of assumptions in the sensitivity analysis, use of guidelines for foot care resulted in longer life expectancy, improved quality of life, lower incidence of foot ulcers, and fewer LEAs when compared with standard care. Most simulations demonstrated that the costs were less than $25,000 per QALY gained compared to standard care. The best results were obtained when foot care guidelines were combined with intensive glycemic control, with a cost of $7860 per QALY gained .
The diabetic limb is vulnerable to a variety of risk factors which have the potential to culminate in the onset of ulceration. Among patients with diabetes, the lifetime incidence of developing an ulcer is 15–25%. Wound healing may be a protracted process, and recurrent wounds are common during the first 2 months after closure .
Amputation is a devastating consequence of diabetic complications. Because of the intrinsic morbidity and mortality associated with amputations, diverse organizations have worked toward implementing plans to reduce amputation rates. In the USA, one such program includes the Healthy People 2010 objective to reduce the annual incidence of diabetic LEAs by 55%. By 2005, participating researchers had projected that a 29% reduction had been achieved despite an increase in diabetes prevalence by 35% during that same period . The incidence for amputations consistently appears to be approximately twice as high for males as females . Along racial and ethnic divisions, gender- and age-adjusted figures identify black non-Hispanics as the highest risk group and Asian/Pacific Islanders as the lowest . Another high-risk group includes diabetic patients living in poor areas, where the median income is less than $25,000 annually. Although a large gap between the wealthiest and poorest quartiles persists, the largest magnitude of reduction has occurred in the poorest group .
The consequences of LEA can be severe, particularly in diabetic patients, where their 10-year mortality rate is nearly 20% higher than that in similar nondiabetic populations . Even perioperative mortality is high, with rates between 5 and 23% reported in the first 30 days [97, 110, 111, 112, 113, 114]. This proportion can change depending on the level of the amputation performed. Digital and other “minor” amputations have a substantially lower mortality rate associated as compared to major amputations which may have a 5-year mortality rate of 36–69%. Subsequent amputations are also problematic, and as many as 68% of amputees will require further amputation within 5 years. This may be influenced by the level of the initial amputation, where digital amputations have a greater risk of reamputation than major amputations .
Health care costs associated with diabetic ulcers and amputations contribute significantly to the financial burden of diabetes. According to the US national inpatient sample, as of 2008, the total number of discharges attributed to diabetes-related amputations was projected to be 45,000. The average length of stay was 10.1 days with an inhospital mortality proportion of 1.29%. The most frequent discharge statuses were to a rehabilitation facility (37.9%), routine discharge (31.5%), or home health care (26.9%). The mean charges were $56,216 while the aggregate charges for the year 2008 had a total of $2,548,319,965. However, it is worth noting that charges and actual cost frequently are separated by a wide margin. Length of stay in the hospital was 47% longer after a major amputation than a toe amputation. During that same time, charges following a major amputation were 53% higher than those after a digital amputation . Importantly, measures aimed at preventing LED, including simple interventions such as following recommended guidelines, have been shown to be highly cost-effective in preventing ulcers and subsequent amputations.
Zimmet may have been correct to call diabetes a worldwide epidemic, as prevalence has climbed higher over the years [2, 4]. Even though the “at-risk” population has increased, the rates of limb-threatening complications have trended downward. The progressive deployment of the “team approach” to limb preservation (as Joslin had first employed) has been touted as a contributing factor, but patient education and vigilance should not be discounted for this success.
- 4.Centers for Disease Control and Prevention (CDC). Diabetes data and trends. http://apps.nccd.cdc.gov/DDTSTRS/default.aspx. Updated 2010. Accessed 30 Aug 2010.
- 5.Centers for Disease Control and Prevention, National Diabetes Statistics Report: Estimates of diabetes and its burden in the United States, 2014. 2014, U.S. Department of Health and Human Services: Atlanta, GA.Google Scholar
- 9.LeQuesne P, Parkshouse N, Faris I. Neuropathy. In: Faris I, editor. The management of the diabetic foot. 2nd ed. Edinburgh: Churchill Livingstone; 1991. p. 41.Google Scholar
- 17.Centers for Disease Control and Prevention (CDC). Hospitalizations for lower extremity conditions. http://www.cdc.gov/diabetes/statistics/hospitalization_national.htm. Updated 2007. Accessed 30 May 2010.
- 18.Cutting K. Glossary. In: Miller M, Glover G, editors. Wound management theory and practice. London: Johnson & Johnson Medical; 1999. p. 170.Google Scholar
- 19.Edmonds M, Foster A (eds). The high-risk foot. In: Managing the diabetic foot. London: Blackwell Science; 2000. p. 35–44.Google Scholar
- 32.Reiber GE. Epidemiology of foot ulcers and amputations in the diabetic foot. In: Bowker JH, Pfeifer MA, editors. The diabetic foot. St. Louis: Mosby; 2001. p. 13–32.Google Scholar
- 39.Faris I. Vascular disease. In: Faris I, editor. The management of the diabetic foot. 2nd ed. Edinburgh: Churchill Livingstone; 1991. p. 9.Google Scholar
- 54.Lin SS, Lee TH, Wapner KL. Plantar forefoot ulceration with equinus deformity of the ankle in diabetic patients: the effect of tendo-achilles lengthening and total contact casting. Orthopaedics. 1996;19(5):465–75.Google Scholar
- 74.Apelqvist J, Bakker K, van Houtum WH, Nabuurs-Franssen MH, Schaper NC. International consensus and practical guidelines on the management and the prevention of the diabetic foot. International working group on the diabetic foot. Diabetes Metab Res Rev. 2000;16(Suppl 1):S84–92.PubMedCrossRefPubMedCentralGoogle Scholar
- 90.Agency for Healthcare Research and Quality. 2009 State snapshots. Updated 2009. Accessed 30 Aug 2010.Google Scholar
- 91.Margolis DJ, et al. Incidence of diabetic foot ulcer and lower extremity amputation among Medicare beneficiaries, 2006 to 2008: Data Points #2, in Data Points Publication Series. 2011: Rockville (MD).Google Scholar
- 92.Centers for Disease Control and Prevention (CDC). Healthy people 2010 midcourse review. http://www.healthypeople.gov/Data/midcourse/pdf/fa05.pdf. Updated 2007. Accessed 20 June 2010.
- 95.Sumpio BE, Armstrong DG, Lavery LA, Andros G, SVS/APMA Writing Group. The role of interdisciplinary team approach in the management of the diabetic foot: a joint statement from the Society for Vascular Surgery and the American podiatric medical association. J Vasc Surg. 2010;51(6):1504–6.PubMedCrossRefGoogle Scholar
- 98.Centers for Disease Control and Prevention (CDC). CDC WONDER Data. http://wonder.cdc.gov/data2010/. Updated 2010. Accessed 10 July 2010.
- 99.Agency for Healthcare Research and Quality. 2008 National Healthcare Quality & Disparities Reports. http://www.ahrq.gov/qual/qrdr08.htm#toc. Updated 2008. Accessed 10 Aug 2010.
- 100.U.S. Census Bureau. American Community Survey Web site. http://www.census.gov/acs/www/. Updated 2004. Accessed 30 Aug 2010.
- 103.Agency for Healthcare Research and Quality. HCUP Quality Indicators Archive. AHRQ Quality Indicators, Rockville, MD. http://www.qualityindicators.ahrq.gov/hcup_archive.htm. Updated 2004. Accessed 10 May 2010.
- 127.Agency for Healthcare Research and Quality. Measuring healthcare quality. http://www.ahrq.gov/qual/measurix.htm. Updated 2009. Accessed 30 Aug 2010.