A Retrospective Review of Adverse Events Related to Diabetic Foot Ulcers

Login toDownload PDF version
Ostomy Wound Management 2017;63(6):30–33
Latricia Allen, DPM, MPH, FACFOAM; Gail Powell-Cope, PhD, ARNP, FAAN; Alfred Mbah, PhD; Tatjana Bulat, MD, CMD; and Eni Njoh, MPH


The risk of adverse events (AEs) such as infection and amputation related to diabetic foot ulcers (DFUs) has been studied, but less is known about rate of other AEs such as falls. As part of a quality improvement project, AEs in veterans with diabetes mellitus (DM) with and without a DFU were examined.

Demographic data including ICD-9 codes, AEs, and comorbidities for all patients with a diagnosis of DM and/or DFU treated between 2009 and 2014 at the James A. Haley Veterans’ Hospital, Tampa, FL, were reviewed retrospectively utilizing the Corporate Data Warehouse (CDW) database. Identifiable protected health information data using patient scrambled social security numbers were collected from the CDW to allow the longitudinal data to be linked at the unique patient level. Descriptive statistics (eg, mean, standard deviation) were determined for demographics, AE, and comorbidities. Adjusted models of AE association with DM and DFU were analyzed using regression modeling via a statistical analysis system. The models were adjusted for age, race, gender, marital status, and comorbidities. Data extracted included individuals diagnosed with diabetes with a DFU (n = 3238, average age 66.0 ± 10.6 years) and diabetes without DFU (n= 41 324, average age 64.4 ± 11.5 years). Participants were mostly Caucasian in both the DFU and non-DFU cohorts (2655 [82.0%] and 32 269 [78.1%], respectively) and male (3129 [96.6%] and 39 580 [95.8%], respectively). The most common comorbidities in this population were peripheral vascular disease (PVD, 39.5%) and peripheral neuropathy (PN, 23.2%). PVD was more common in veterans with (39.5%) than without a DFU (9.2%). Compared to individuals with DM, those with DM and DFU were significantly more likely to experience an infection (OR = 9.43; 95% CI 8.54-10.4), undergo an amputation (OR = 7.40; 95% CI 6.16-8.89), or experience a fracture (OR = 3.65; 95% CI 2.59-5.15) or fall (OR = 2.26; 95% CI 1.96-2.60) (P <.01 for all variables). Although the increased risk of infection and amputation among persons with DFUs has been documented, less is known about the rate of falls and fractures. The current  findings will serve as baseline data for future implementation trials to reduce DFU-associated AEs, and clinicians may want to consider expanding DFU patient education efforts to include fall risk.


Diabetes mellitus (DM) contributes to significant morbidity and mortality. According to the World Health Organization, DM prevalence worldwide is ~347 million.1 In 2014, the National Diabetes Statistics Report2 determined that 29.1 million Americans people were diagnosed with diabetes; in addition, DM and associated comorbidities are more common than some cancers in the United States and are the seventh leading cause of death. DFU programs have been developed with the goal of preventing ulcers and amputations.3 Comorbidities such as end-stage renal disease, cardiovascular disease, peripheral vascular disease (PVD), and peripheral neuropathy (PN) commonly contribute to adverse events (AEs), which may lead to decreased quality of life.4 Specifically, DFUs may put patients at risk for AEs such as falls, fractures, infection, amputation, and death.5 

According to the American Diabetes Association,2 PN affects approximately 60% to 70% of patients with diabetes.2 Patients with PN may have unsteady gait and/or impaired proprioception that may put them at risk for falls, fractures, contusions, and other types of AEs. PVD reduces blood flow to the extremity, which frequently leads to delayed wound healing and wound infection, putting the patient at risk for possible limb loss.6 A prospective study7 that examined patients with a history of a foot ulcer found that among 400 participants, 63% reported 923 falls over a 2-year follow-up period. Of persons needing medical attention post-fall, 17% had experienced at least 1 fall and >30% reported falling more than 1 time. This fall rate is higher than general population fall rates reported as 28.7% for same age group in 2014.8 

DFUs have been noted to increase the likelihood of lower extremity amputation.3 If treatment is delayed, it is not uncommon for a DFU to become infected.9 In 2010, approximately 73 000 nontraumatic amputations of a lower extremity occurred in patients with diabetes in the United States.2 Patients with DM-related amputation have been reported to have a higher mortality rate than persons with some cancers, with approximately 50% mortality within 5 years post-amputation.10 Patient safety improvement initiatives are needed to manage DFUs more effectively, improve quality of care, and reduce the likelihood of AEs in this vulnerable patient population.

The Veterans Health Administration (VHA) has a growing number of patients affected by diabetes, and the number of veterans with a diagnosis of DM in 2001 was 2 times greater than the general population, which was approaching 20% at that time.11 A retrospective study examining all VHA hospitals determined that more than half of VHA hospitalized patients had a diagnosis of DM.12 This study also determined that more than one half and two thirds of VHA hospitalizations were due to lower extremity ulcers and amputations, respectively.

Given the prevalence of diabetes-related ulcers in the VHA, the purpose of this retrospective study was to evaluate the rate of AEs (diabetes-related amputations, falls, fractures, and infection) in a population of veterans with diabetes to guide plans for future patient care improvement efforts. 


Data retrieval and analysis involved in this quality improvement project took place over a 9-month period at the James A. Haley Veterans’ Hospital, VISN 8 Patient Safety Center of Inquiry in Tampa, FL. The project was approved by the local Research and Development Committee and was classified as an operations/quality improvement effort. Data for veterans with diabetes with and without DFU were retrieved from the Corporate Data Warehouse (CDW) using administrative ICD-9 code data for all patients with a diagnosis of diabetes and DFUs for the years 2009 through 2014. Data were collected to describe AEs in patients with diagnosis of a DFU and without DFU. All data were collected via an Excel (Microsoft Corporation, Redmond, WA) spreadsheet by a doctorate level trained statistician and stored in a secure location, maintaining patient anonymity. Only study personnel had access to the data.

DFU, non-DFU, lower extremity diabetes-related amputation, falls, fracture, and infection ICD-9 codes for the above-mentioned study time period were extracted from the CDW using structured query language then entered into SAS (SAS Institute, Inc, Cary, NC) for analysis. Adjusted models of AE association with DM and DFU were abstracted and analyzed. Descriptive statistics were used to examine the following variables: DFU rate, demographics, falls, fractures, infection, hospitalization, amputation, and comorbidity disease (acute limb ischemia, cardiovascular disease, end-stage renal disease, PN, PVD, and visual impairment) to determine whether an association could be found between DM and DFU. 


The total sample size for individuals diagnosed with diabetes with DFU and diabetes without DFU were 3586 and 40 938, respectively. Participants were mostly Caucasian in both the diabetic ulcer and non-DFU cohorts (2931 [81.7%] and 31 938 [78.1%], respectively) and male (3476 [96.9%] and 39 197 [95.8%], respectively) (see Table 1). A statistically significant difference was found between the distribution of men and women in the DFU and non-DFU cohorts (P = .02); however, the Cramer’s effect size value (v = 0.0001) indicated no effect. Participants with a DFU were less likely to be married (2002, 55.8%) than those without a DFU (26 007, 63.5%). The distribution of marital status for patients with and without a DFU was statistically significant, (P <.00001); however, the Cramer’s effect size value (v = 0.0001) suggested a low practical difference between cohorts. Caucasians (81.7%) with diabetes were more likely to have a DFU than African Americans (10.9%) in this population. Comorbidities were more prevalent among individuals with than without a DFU. A diagnosis of PVD was the most prevalent (1415 [39.5%]), followed by PN (833 [23.2%]). Among participants without DFU, the most prevalent comorbidity was PVD (3784 [9.2%]), followed by PN (2489 [6.1%]). A statistically significant difference was noted between the DFU and non-DFU cohort with regard to PVD (P <.0000001), and the Cramer’s effect size value (v = 0.0001) was significantly below the conventional level of low effect size. A statistically significant difference also was found between the DFU and non-DFU cohort with regard to PN (P <.0000001) and the Cramer’s effect size value (v = 0.0001), suggesting low meaningful differences between cohorts. owm_0617_allen_table1

Infection was the most frequent AE for participants with a DFU (1280 [39.5%]) compared to falls for participants without a DFU (1467 [3.5%]) (see Table 2). Subsequently, persons with diabetes and a foot ulcer have increased risk for the AEs considered in the study (amputation, falls, fracture, and infection) compared to persons without a DFU (see Table 3). Further, compared to individuals with diabetes, those with diabetes and a foot ulcer were more than 9 times as likely to experience an infection (OR = 9.43; 95% CI 8.54-10.4), more than 7 times as likely to undergo an amputation (OR = 7.40; 95% CI 6.16-8.89), more than 3 times as likely to experience a fracture (OR = 3.65; 95% CI 2.59-5.15), and more than 2 times as likely to fall (OR = 2.26; 95% CI 1.96-2.60). These results were highly significant, with P <.001 (see Table 3).



The most common comorbidities among the diabetes with DFU group in this study were PVD and PN. When examining AEs among veterans with a history of DFUs, the most common were infection (39.5%), amputations (16%), and falls (14.6%). Infection and amputation were expected AEs in the diabetes with DFU patient population due to the nature of disease. However, this population also demonstrated an increased risk of falls that is not well documented in the literature. The rate of PN in this patient population was also high in the DFU population and can have a negative effect on gait performance based on the literature.13 The increased risk of falls in persons with diabetes and DFU has not been well noted in the literature and needs to be examined further. Data from this project will be used as baseline information for quality implementation efforts to improve fall risk screening, assessment, and interventions to decrease fall and injury risk in this vulnerable patient population. 


A limitation to this quality improvement project is inherent in the retrospective design of the study and further examination should take place for a larger robust prospective trial. 


The results of this retrospective examination showed that, compared to veterans with DM, those with diabetes and a DFU were more likely to have a higher risk of infection, amputation, falls, and fractures. Participants were mostly male Caucasians in both the DFU and non-DFU cohorts. The most common comorbidities in this population were PVD and PN, with PVD being more common in veterans with DFU. Additional studies are needed to further examine these outcomes and the potential effect of patient care improvements to reduce these AEs. In the meantime, clinicians might consider including AE education efforts in their plan of care for patients with DFUs, particularly as it relates to preventing falls and fractures. n


1. World Health Organization. 10 Facts about Diabetes. Available at: www.who.int/features/factfiles/diabetes/facts/en/. Accessed November 4, 2015.

2. Data from the National Diabetes Statistics Report, 2014. Available at: www.diabetes.org/diabetes-basics/statistics. Accessed June 10, 2014.

3. Driver VR, Fabbi M, Lavery L, Gibbons G. The costs of diabetic foot: the economic cases for the limb salvage team. J Vasc Surg. 2010;52(3  suppl):17S–22S.

4. Corriere M, Rooparinesingh N, Kalyani RR. Epidemiology of diabetes and diabetes complications in the elderly: an emerging public health burden. Curr Diab Rep. 2013;13(6):805–813.

5. Kalyani RR, Saudek CD, Brancati FL, Selvin E. Association of diabetes, comorbidities, and A1c with functional disability in older adults: results from the National Health and Nutrition Examination Survey (NHANES), 1999-2006. Diabetes Care. 2010;33(5):1055–1060. 

6. Armstrong DG, Cohen K, Courric S, Bharara M, Marston W. Diabetic foot ulcers and vascular insufficiency: our population has changed, but our methods have not. J Diabetes Sci Technol 2011;5(6):1591-1595.

7. Wallace C, Reiber GE, LeMaster J, et al. Incidence of falls, risk factors for falls, and fall-related fractures in individuals with diabetes and a prior foot ulcer. Diabetes Care. 2002;25(11):1983–1986.

8. Bergen G, Stevens MR, Burns ER. Falls and fall injuries among adults aged ≥65 years — United States, 2014. MMWR Morb Mortal Wkly Rep. 2016;65(37):993–998. 

9. Allen LL, Kalmar G, Driver VR. Treatment of a high-risk diabetic patient with peripheral vascular disease and osteomyelitis. Tech Vasc Interven Radiol. 2016;19(2):96–100.

10. Ziegler-Graham K, MacKenzie EJ, Ephraim PL, Travison TG, Brookmeyer R. Estimating the prevalence of limb loss in the United States: 2005 to 2050. Arch Phys Med Rehabil. 2008;89(3):422-429.

11. Reiber GE, Smith DG, Carter J, et al. A comparison of diabetic foot ulcer patients managed in VHA and non-VHA settings. J Rehabil Res Develop. 2001;38(3):309–317.

12. Mayfield JA, Reiber GE, Maynard C, Czermiecki J, Sangeotzan B. The epidemiology of lower-extremity disease in veterans with diabetes. Diabetes Care. 2004;27(2 suppl):b39–b44.

13. Mustapa A, Justine M, Musafah N, Jamil N, Manaf H. Postural control and gait performance in diabetic peripheral neuropathy: a systematic review.BioMed Res Int. 2016;(2016). 


Potential Conflicts of Interest: none disclosed


Dr. Allen is a podiatrist and HSR&D postdoctoral fellow; Dr. Powell-Cope is the Tampa Site Co-Director, Center of Innovation on Disability and Rehabilitation Research (CINDRR) and the Site Director for Interprofessional Patient Safety; Dr. Bulat is the Director, Tampa Patient Safety Center of Inquiry and Co-Director, Interprofessional Fellowship in Patient Safety; and Drs. Mbah and Njoh are statistitians at CINDRR, James A. Haley Veterans’ Hospital, Tampa, FL. Please address correspondence to: Latricia Allen, DPM, MPH, FACFOAM, James A. Haley Veterans’ Hospital, HSR&D Center of Innovation on Disability and Rehabilitation Research (CINDRR), 8900 Grand Oaks Circle Tampa, FL 33637; email: Latricia.allen@va.gov.