Advertisement

Contribution of 30-day readmissions to the increasing costs of care for the diabetic foot

Open ArchivePublished:March 05, 2019DOI:https://doi.org/10.1016/j.jvs.2018.12.028

      Abstract

      Objective

      The inpatient cost of care for diabetic foot ulcers (DFUs) has been estimated to be $1.4 billion annually in the United States. We have previously demonstrated that the risk of 30-day unplanned readmission for patients with DFU is nearly 22%. Our aim was to quantify the cost of readmissions for patients admitted with DFU.

      Methods

      All patients presenting to our multidisciplinary diabetic limb preservation service from June 2012 to June 2016 were enrolled in a prospective database. Inpatient costs and net margins were calculated overall and for index admissions vs 30-day unplanned readmissions.

      Results

      A total of 249 admissions for 150 patients were included. Of these, 206 admissions were index admissions and 43 were 30-day readmissions. The most common reason for readmission was the foot wound (49%), followed by a bypass wound (14%), renal complications (9%), and other systemic complications. Surgical interventions during readmission were common (47%) and included both podiatric (37%) and vascular (23%). The wound healing outcomes were favorable, with 78% of all wounds achieving healing by 1 year. Limb salvage was 91% overall. The median hospital cost per admission was $20,111 (interquartile range, $12,589-$33,254) and did not differ between the index and readmissions ($22,165 vs $19,408; P = .46). However, the hospital net margins were lower after readmission ($3908 vs $1975; P = .02). The overall cost of care for patients requiring readmission was significantly greater than that for patients not readmitted ($79,315 vs $28,977; P < .001). During the study period, DFU care at our institution cost $7.9 million, of which $1.2 million (16%) was attributable to readmission costs.

      Conclusions

      Readmissions for patients with DFU are common and associated with a substantial cost burden. The cost of readmission for patients with DFU was as high as the cost of the index admission but with lower hospital net margins. When extrapolated to national data, the 15% readmission cost burden we have reported would be equivalent to $210 million hospital costs annually. Focused efforts at preventing readmissions in this high-risk patient population are essential to reducing the overall costs of care associated with DFUs.

      Graphical Abstract

      Keywords

      Article Highlights
      • Type of Research: Single-center retrospective analysis of prospectively collected institutional data
      • Key Findings: Thirty-day readmissions occurred in 17% of 249 diabetic foot ulcer (DFU) admissions and were associated with a similar cost, but lower net margin, compared with the index admissions. DFU care cost $7.9 million over 4 years, of which $1.2 million (16%) was attributable to readmission costs.
      • Take Home Message: Focused efforts at preventing readmissions in this high-risk patient population are essential to reducing the overall costs of care associated with DFU.
      The inpatient cost of care for diabetic foot ulcers (DFUs) has been estimated to be $1.4 billion annually in the United States.
      • Hicks C.W.
      • Selvarajah S.
      • Mathioudakis N.
      • Sherman R.E.
      • Hines K.F.
      • Black III, J.H.
      • et al.
      Burden of infected diabetic foot ulcers on hospital admissions and costs.
      From 2005 to 2010, the hospital costs associated with the inpatient care of a DFU increased by >30% owing to the increasing number of procedures aimed at limb preservation in a progressively sicker patient population.
      • Hicks C.W.
      • Selvarajah S.
      • Mathioudakis N.
      • Perler B.A.
      • Freischlag J.A.
      • Black III, J.H.
      • et al.
      Trends and determinants of costs associated with the inpatient care of diabetic foot ulcers.
      However, the inpatient length of the hospital stay has not changed over time,
      • Hicks C.W.
      • Selvarajah S.
      • Mathioudakis N.
      • Perler B.A.
      • Freischlag J.A.
      • Black III, J.H.
      • et al.
      Trends and determinants of costs associated with the inpatient care of diabetic foot ulcers.
      suggesting that multiple admissions and readmissions could be a contributing factor to the increasing burden of DFU care on the US healthcare system.
      Thirty-day readmissions after inpatient admissions for foot ulcer care vary depending on the population studied. Among a cohort of both diabetic and nondiabetic patients undergoing lower extremity bypass, the 30-day readmission rate was only 15% overall but increased to 19% for patients with chronic limb-threatening ischemia.
      • Jones C.E.
      • Richman J.S.
      • Chu D.I.
      • Gullick A.A.
      • Pearce B.J.
      • Morris M.S.
      Readmission rates after lower extremity bypass vary significantly by surgical indication.
      In contrast, the 30-day readmission rate among a group of patients with diabetes treated primarily with wound care and minor amputations has been estimated to be closer to 30%.
      • Remington A.C.
      • Hernandez-Boussard T.
      • Warstadt N.M.
      • Finnegan M.A.
      • Shaffer R.
      • Kwong J.Z.
      • et al.
      Analyzing treatment aggressiveness and identifying high-risk patients in diabetic foot ulcer return to care.
      At our institution, the incidence of 30-day readmissions for patients with DFU treated by a multidisciplinary team has been ∼22%.
      • Holscher C.M.
      • Hicks C.W.
      • Canner J.K.
      • Sherman R.L.
      • Malas M.B.
      • Black III, J.H.
      • et al.
      Unplanned 30-day readmission in patients with diabetic foot wounds treated in a multidisciplinary setting.
      Previous studies have reported on the cost of DFU per wound episode, including both inpatient and outpatient care.
      • Hicks C.W.
      • Selvarajah S.
      • Mathioudakis N.
      • Sherman R.E.
      • Hines K.F.
      • Black III, J.H.
      • et al.
      Burden of infected diabetic foot ulcers on hospital admissions and costs.
      • Hicks C.W.
      • Selvarajah S.
      • Mathioudakis N.
      • Perler B.A.
      • Freischlag J.A.
      • Black III, J.H.
      • et al.
      Trends and determinants of costs associated with the inpatient care of diabetic foot ulcers.
      • Hicks C.W.
      • Canner J.K.
      • Karagozlu H.
      • Mathioudakis N.
      • Sherman R.L.
      • Black III, J.H.
      • et al.
      The SVS WIfI classification system correlates with cost of care for diabetic foot ulcers treated in a multidisciplinary setting.
      • Hopkins R.B.
      • Burke N.
      • Harlock J.
      • Jegathisawaran J.
      • Goeree R.
      Economic burden of illness associated with diabetic foot ulcers in Canada.
      • Rice J.B.
      • Desai U.
      • Cummings A.K.
      • Birnbaum H.G.
      • Skornicki M.
      • Parsons N.B.
      Burden of diabetic foot ulcers for Medicare and private insurers.
      • Harrington C.
      • Zagari M.J.
      • Corea J.
      • Klitenic J.
      A cost analysis of diabetic lower-extremity ulcers.
      • Joret M.O.
      • Dean A.
      • Cao C.
      • Stewart J.
      • Bhamidipaty V.
      The financial burden of surgical and endovascular treatment of diabetic foot wounds.
      • Skrepnek G.H.
      • Mills Sr., J.L.
      • Armstrong D.G.
      A diabetic emergency one million feet long: Disparities and burdens of illness among diabetic foot ulcer cases within emergency departments in the United States, 2006-2010.
      From our institutional experience, we previously reported that the overall mean cost of care per DFU episode (from presentation to healing) was ∼$24,226.

      Hicks CW, Canner JK, Karagozlu H, Mathioudakis N, Sherman RL, Black JH 3rd, et al. Quantifying the costs and profitability of care for diabetic foot ulcers treated in a multidisciplinary setting [published online ahead of print January 1. 2019]. J Vasc Surg doi: 10.1046/j.jvs.2018.10.097.

      On average, we have found that patients required 2.6 inpatient admissions and 0.9 outpatient procedure to achieve complete healing, with a cost of ∼$41,420 and ∼$11,265 per event, respectively. However, the costs associated with 30-day hospital readmissions for patients with DFUs are unknown. The aim of the present study was to quantify the hospital costs associated with 30-day hospital readmissions for patients admitted with a DFU.

      Methods

       Study cohort

      All the patients presenting to our multidisciplinary diabetic limb preservation service from June 1, 2012 through June 31, 2016 were enrolled in a prospective database. The hospital institutional review board approved the study, and all the patients provided written informed consent to allow for the prospective collection of their demographic, socioeconomic, comorbidity, and wound-related information. All patients with a DFU were considered for inclusion, whether or not they had concomitant peripheral arterial disease (PAD) or chronic limb-threatening ischemia. For the purposes of the present study, PAD was defined as the presence of any symptoms consistent with ischemic lower extremity vascular disease (ie, claudication, rest pain, ischemic tissue loss) and noninvasive testing findings that confirmed impaired perfusion or angiographic evidence of arterial occlusive disease. Chronic limb-threatening ischemia was defined as PAD with rest pain or a wound. Patients without a DFU, those without hospital admissions, those with missing cost data, and those with Wound, Ischemia, and foot Infection stage 5 limbs (ie, unsalvageable limbs) were excluded from the present study.

       Multidisciplinary diabetic limb preservation service

      Our multidisciplinary diabetic limb preservation service primarily includes a vascular surgeon, surgical podiatrist, endocrinologist, wound nurse, and physician assistant. The patients see all providers at a single clinic location at each visit, and their care is coordinated. Infectious disease specialists, plastic surgeons, orthopedic foot and ankle surgeons, and orthotists or prosthetists are consulted on an as-needed basis. In brief, all patients presenting to our multidisciplinary diabetic limb preservation service undergo noninvasive vascular laboratory evaluations to test for PAD, radiography and/or magnetic resonance imaging to assess for osteomyelitis, and basic laboratory testing, including glycated hemoglobin, to assess glycemic control. Those patients with evidence of PAD will undergo angiography, and revascularization is performed whenever possible. Once perfusion has been achieved, the wound is debrided, with the goal of excising all infected tissue and bone, and aggressive local wound care is performed. After debridement, the wounds are offloaded with the assistance of an orthotist or prosthetist. Antibiotics are administered in accordance with the intraoperative culture data. In cases with forefoot or midfoot involvement, the input of our infectious diseases colleagues is obtained to assist with antibiotic decsisions. In general, all wounds involving the metatarsal head or more proximally will be treated with 6 weeks of intravenous antibiotics whether or not the bone margin finding was negative for osteomyelitis. Wounds limited to the soft tissue or the phalanx will be treated with a 2-week course of oral antibiotics.
      All the patients enrolled in our limb preservation service receive multidisciplinary care, regardless of their insurance status, including regularly scheduled outpatient appointments, noninvasive vascular laboratory testing, radiography, and wound care, as needed. A more thorough description of our team has been previously reported.
      • Mathioudakis N.
      • Hicks C.W.
      • Canner J.K.
      • Sherman R.L.
      • Hines K.F.
      • Lum Y.W.
      • et al.
      The Society for Vascular Surgery Wound, Ischemia, and foot Infection (WIfI) classification system predicts wound healing but not major amputation in patients with diabetic foot ulcers treated in a multidisciplinary setting.

       Study endpoints and definitions

      The study endpoints of interest included inpatient hospital charges, costs, professional fees, and net margins for a given admission or readmission. The hospital charges represent the stated price of care for a given service. The hospital costs represent the costs that the hospital incurred to provide care for that service. The hospital costs included fixed costs (ie, capital expenditures, employee salaries and benefits, building maintenance, utilities) and variable costs (ie, patient and provider supplies, diagnostic and therapeutic supplies, medications). They also included direct costs (ie, costs easily attributed to goods or service) and indirect costs (ie, costs not directly accountable to a cost object). The Johns Hopkins Hospital uses activity-based costing to attribute indirect costs to the patients. The hospital revenue represents the total amount of reimbursement the hospital received from the patients and insurance companies for a given admission or readmission. Net margins were calculated as the hospital revenue minus the hospital costs (ie, profit) for the admissions and readmissions. Professional fee charges were calculated as a cumulative charge for all members of the multidisciplinary team treating a particular patient during a given admission or readmission. The professional fees were not included in the net margin calculation, because they were only available as charges, rather than costs. All cost and charge data were collected from the Johns Hopkins financial office and represent hospital-based charges and costs reported in US dollars.
      The index admissions represent any admission that preceded a readmission. For patients for whom multiple readmissions occurred, the charges and costs were calculated for the first readmission only. All reported readmissions were unplanned readmissions that had occurred within 30 days of the index admission. The readmissions included those directly related to the foot wound as well as those for other causes (eg, acute kidney injury due to antibiotics, glycemic changes, cardiovascular complications). We did not include planned readmissions for the purposes of the present analysis. All index admissions and unplanned readmissions were reported and analyzed on a per-limb basis.

       Statistical analysis

      The aim of the present study was to compare the hospital costs and charges for index admissions vs 30-day readmissions. The baseline patient demographic, comorbidity, and wound data have been reported on a per-patient basis as the mean ± standard error or number of patients and percentages, as indicated. The hospital charges and costs have been reported as the median (interquartile range [IQR]) on an admission-related basis and captured for index admissions, index admissions with associated unplanned 30-day readmissions, and 30-day readmissions alone. The hospital costs and charges for the index admissions were compared with the 30-day unplanned readmissions using Wilcoxon signed-rank tests. We also compared the hospital charges and costs on a per-patient basis (ie, patients with readmissions vs those without) using Wilcoxon rank-sum tests.
      All hypothesis tests were 2-sided with α of 0.05. All analyses were performed using Stata, version 14 (StataCorp LP, College Station, Tex).

      Results

       Study cohort

      A total of 249 admissions for 150 patients were included in the present study. The median patient age was 57.7 years (IQR, 49.8-64.5 years). Of the 150 patients, 93 (62.0%) were men, and 96 (64.0%) were black; 77 (51.3%) had Medicare insurance, 41 (27.3%) had Medicaid insurance, and 32 (21.3%) had private or self-pay insurance. Also, 142 patients (94.7%) had type 2 diabetes, 141 (94.0%) had neuropathy with the loss of peripheral sensation, and 76 (50.7%) had PAD. Finally, 132 patients (88%) were independent at baseline. A complete summary of the baseline patient characteristics is provided in Table I.
      Table IBaseline demographic, comorbidity, and wound characteristics of our diabetic foot cohort evaluated for readmission costs (N = 150)
      CharacteristicValue
      Age, years
       Median57.7
       IQR49.8-64.5
      Male gender93 (62.0)
      Race
      Race was self-designated by the patient.
       White50 (33.3)
       Black96 (64.0)
       Other4 (2.7)
      Insurance
       Medicare77 (51.3)
       Medicaid41 (27.3)
       Private/self-pay32 (21.3)
      Comorbid conditions
       Type 1 diabetes8 (5.3)
       Type 2 diabetes142 (94.7)
       Hypertension126 (84.0)
       Dyslipidemia74 (49.3)
       Coronary artery disease51 (34.0)
       Congestive heart failure27 (18.0)
       Peripheral arterial disease76 (50.7)
       Cerebrovascular disease22 (14.7)
       Chronic kidney disease35 (23.3)
       Dialysis22 (14.7)
       Neuropathy with loss of peripheral sensation141 (94.0)
       History of any transplant14 (9.3)
      Tobacco history
       Current smoker45 (30.0)
       Former smoker47 (31.3)
       Never smoker58 (38.7)
      Functional status
       Independent132 (88.0)
       Partially or totally dependent17 (11.3)
      Limb characteristics
       Wound duration, days88.8 ± 12.2
       WIfI clinical stage
      14 (2.7)
      238 (25.3)
      357 (38.0)
      451 (34.0)
       Recurrent wound10 (6.8)
      No. of 30-day readmissions
       129 (19.3)
       24 (2.7)
       32 (1.3)
      WIfI, Wound, Ischemia, and foot Infection.
      Data presented as median and IQR, n (%), or mean ± standard error.
      a Race was self-designated by the patient.
      The mean wound duration at the initial presentation was 88.8 ± 12.2 days. The Wound, Ischemia, and foot Infection stage at presentation was stage 1 in 4 patients (2.7%), stage 2 in 38 (25.3%), stage 3 in 57 (38.0%), and stage 4 in 51 patients (34.0%). Surgical procedures were performed in 248 of the index admissions (99.6%). These included podiatric surgery (n = 205 [82.3%]), revascularization procedures (open, n = 33 [13.3%]; endovascular, n = 34 [13.7%]), and diagnostic angiography (n = 12 [4.8%]).
      Of the 150 patients, 35 had required readmission with 30 days (23.3%). Of these 35 patients, 29 (19.3%) had had 1 readmission, 4 (2.7%) had had 2, and 2 (1.3%) had had 3 readmissions. A statistically significant decrease was seen in the readmission rates over time (26.8% in 2013 vs 11.1% in 2016; P = .03; Fig). The wound healing outcomes were favorable, with 78.4% of all wounds achieving healing at a mean follow-up period of 158.1 ± 11.5 days. The limb salvage rate was 91.2% overall; however, of the 13 patients who required a major amputation, only four (30.8%) achieved independent ambulation postoperatively.
      Figure thumbnail gr1
      FigTrends in 30-day readmissions over time. A significant decrease was found in the readmission rates during the study period (P = .03).

       Readmission characteristics

      Of the 249 admissions analyzed, 206 were index admissions without readmission and 43 were 30-day readmissions. The most common reason for readmission was the foot wound (n = 21 [48.8%]), followed by bypass wound (n = 6 [14.0%]), renal complications (n = 4 [9.3%]), and other systemic complications (Table II). Surgical interventions performed during readmission were common (n = 20 [46.5%]) and included both podiatric (n = 16 [37.2%]) and vascular (n = 10 [23.3%]) interventions. Vascular interventions consisted of secondary 30-day reinterventions in four patients and primary revascularization procedures for poor wound healing in the setting of borderline findings for noninvasive studies in six patients. No patient required a major amputation during a readmission. The hospital length of stay was not significantly different between the index admission and readmission (7.6 ± 0.8 vs 9.0 ± 1.3 days; P = .80). A full description of our experience with DFU readmissions, including details of the etiology and risk factors, has been previously reported.
      • Holscher C.M.
      • Hicks C.W.
      • Canner J.K.
      • Sherman R.L.
      • Malas M.B.
      • Black III, J.H.
      • et al.
      Unplanned 30-day readmission in patients with diabetic foot wounds treated in a multidisciplinary setting.
      Table IIReasons for 30-day readmissions in patients with diabetic foot ulcers (N = 43)
      Reason for readmissionNo. (%)
      Wound, foot21 (48.8)
      Wound, bypass6 (14.0)
      Renal (acute kidney injury)4 (9.3)
      Gastrointestinal4 (9.3)
      Cardiac2 (4.7)
      Neurologic1 (2.3)
      Pulmonary1 (2.3)
      Bypass graft thrombosis1 (2.3)
      Open revascularization after failed endovascular intervention1 (2.3)
      Other2 (4.7)

       Hospital costs

       Index admissions

      The median hospital charges and costs for all index admissions were $25,771 (IQR, $14,919-$39,719) and $20,111 (IQR, $12,589-$33,254), respectively (Table III). The median professional fee charges for all index admissions was $2102 (IQR, $1228-$4088). The median hospital revenue was $23,200 (IQR, $12,777-$37,090), and the median hospital net margin was $2284 (IQR, $−24 to $6058).
      Table IIIComparison of hospital costs associated with initial admission vs 30-day readmission
      Charges/costsAll admissions and readmissions (N = 249)Index admission with readmission (n = 43)Readmission (n = 43)P value
      Total charges$25,771 ($14,919-$39,719)$29,164 ($15,554-$43,641)$23,054 ($13,674-$36,576).25
      Total professional fee charges$2102 ($1228-$4088)$2102 ($1379-$4933)$1209 ($666-$4990).65
      Total cost$20,111 ($12,589-$33,254)$22,165 ($13,935-$34,622)$19,408 ($11,094-$31,478).46
      Total revenue$23,200 ($12,777-$37,090)$27,169 ($14,139-$40,615)$19,425 ($11,361-$34,609).18
      Net margin$2284 ($24-$6058)$3908 ($638-$8886)$1975 ($21-$5621).02
      Data presented as median (interquartile range).

       Index admissions with readmissions

      The median hospital charges and cost per index admission associated with a readmission were $29,164 (IQR, $15,554-$43,641) and $22,165 (IQR, $13,935-$34,622), respectively. These did not significantly differ from the mean hospital charges or cost for readmission (mean, $23,054 [IQR, $13,674-$36,576]; and mean, $19,408 [IQR $11,094-$31,478], respectively; P ≥ .25; Table III). The median hospital revenue was ∼$7700 greater for admissions compared with that for readmissions ($27,169 [IQR, $14,139-$40,615] vs $19,425 [IQR, $11,361-$34,609]); however, this difference was not statistically significant (P = .18). However, the hospital net margins were lower after readmission ($3908 [IQR, $638-$8886] vs $1975 [IQR, $−21 to $5621]; P = .02).

       Overall costs of care

      The overall cost of care for patients requiring readmission was significantly greater than that for patients who were not readmitted ($79,315 [IQR, $41,842-$127,642] vs $28,977 [IQR, $14,576-$56,018]; P < .001). The hospital charges, revenue, and net margins were also significantly greater for patients requiring readmission (P < .001 for all; Table IV), although the professional fee charges did not differ between the two groups (P = .19). During the study period, DFU care at our institution cost $7.9 million, of which $1.2 million (15.8%) was attributable to the readmission costs (Table V).
      Table IVComparison of hospital costs associated with care for diabetic foot ulcers stratified by 30-day readmission
      Charges/costsAll patients (N = 150)No readmission (n = 115)Readmission (n = 35)P value
      Total charges$43,364 ($23,016-$101,191)$34,695 ($19,159-$70,762)$101,346 ($50,952-$150,134)<.001
      Total professional fee charges$2652 ($1228-$5800)$2418 ($971-$4088)$4934 ($1379-$6411).19
      Total cost$34,893 ($18,716-$76,530)$28,977 ($14,576-$56,018)$79,315 ($41,842-$127,642)<.001
      Total revenue$38,349 ($20,097-$88,942)$31,951 ($16,001-$61,881)$93,866 ($47,436-$139,161)<.001
      Net margin$3568 ($192-$9980)$2965 ($233-$8201)$12,447 ($2679-$24,691)<.001
      Data presented as median (interquartile range).
      Table VSum total costs of care for index admissions and readmissions in patients with diabetic foot ulcers
      Charges/costsAll admissions and readmissions (N = 249)Readmissions alone (n = 43)Readmissions costs, %
      Total charges$9,739,268 ± $516,718$1,519,947 ± $257,84315.6
      Total professional fee charges$453,086 ± $59,562$6865 ± $51301.5
      Total cost$7,857,370 ± $417,027$1,238,057 ± $210,81215.8
      Total revenue$8,800,247 ± $480,958$1,354,137 ± $234,13215.4
      Net margin$942,878 ± $121,552$116,080 ± $55,42212.3
      Data presented as sum total ± standard error.

      Discussion

      Readmissions for patients with DFU are common, occurring in 15% to 30% of patients.
      • Jones C.E.
      • Richman J.S.
      • Chu D.I.
      • Gullick A.A.
      • Pearce B.J.
      • Morris M.S.
      Readmission rates after lower extremity bypass vary significantly by surgical indication.
      • Remington A.C.
      • Hernandez-Boussard T.
      • Warstadt N.M.
      • Finnegan M.A.
      • Shaffer R.
      • Kwong J.Z.
      • et al.
      Analyzing treatment aggressiveness and identifying high-risk patients in diabetic foot ulcer return to care.
      • Holscher C.M.
      • Hicks C.W.
      • Canner J.K.
      • Sherman R.L.
      • Malas M.B.
      • Black III, J.H.
      • et al.
      Unplanned 30-day readmission in patients with diabetic foot wounds treated in a multidisciplinary setting.
      • McPhee J.T.
      • Nguyen L.L.
      • Ho K.J.
      • Ozaki C.K.
      • Conte M.S.
      • Belkin M.
      Risk prediction of 30-day readmission after infrainguinal bypass for critical limb ischemia.
      In our study, we found that the 30-day hospital readmissions were also associated with a substantial cost burden. The median hospital cost per DFU admission was >$20,000 and did not significantly differ between the index admissions and readmissions. The total cost of care for patients requiring readmissions was >2.5 times greater than that for those without readmissions, and the hospital net margins were lower after readmissions compared with the index admissions. More than 15% of our hospital's DFU costs were attributable to readmissions during the study period, equaling >$1 million in additional hospital costs. To the best of our knowledge, the present study is the first to quantify the substantial burden of DFU care specifically as it relates to hospital readmissions.
      We, and other groups, have previously reported the overall costs of care associated with DFUs in the United States.
      • Hicks C.W.
      • Selvarajah S.
      • Mathioudakis N.
      • Sherman R.E.
      • Hines K.F.
      • Black III, J.H.
      • et al.
      Burden of infected diabetic foot ulcers on hospital admissions and costs.
      • Hicks C.W.
      • Selvarajah S.
      • Mathioudakis N.
      • Perler B.A.
      • Freischlag J.A.
      • Black III, J.H.
      • et al.
      Trends and determinants of costs associated with the inpatient care of diabetic foot ulcers.
      • Hicks C.W.
      • Canner J.K.
      • Karagozlu H.
      • Mathioudakis N.
      • Sherman R.L.
      • Black III, J.H.
      • et al.
      The SVS WIfI classification system correlates with cost of care for diabetic foot ulcers treated in a multidisciplinary setting.
      • Hopkins R.B.
      • Burke N.
      • Harlock J.
      • Jegathisawaran J.
      • Goeree R.
      Economic burden of illness associated with diabetic foot ulcers in Canada.
      • Rice J.B.
      • Desai U.
      • Cummings A.K.
      • Birnbaum H.G.
      • Skornicki M.
      • Parsons N.B.
      Burden of diabetic foot ulcers for Medicare and private insurers.
      • Harrington C.
      • Zagari M.J.
      • Corea J.
      • Klitenic J.
      A cost analysis of diabetic lower-extremity ulcers.
      • Joret M.O.
      • Dean A.
      • Cao C.
      • Stewart J.
      • Bhamidipaty V.
      The financial burden of surgical and endovascular treatment of diabetic foot wounds.
      • Skrepnek G.H.
      • Mills Sr., J.L.
      • Armstrong D.G.
      A diabetic emergency one million feet long: Disparities and burdens of illness among diabetic foot ulcer cases within emergency departments in the United States, 2006-2010.
      Most of these studies focused on the burden of inpatient hospitalization costs using diagnosis codes and insurance claims, which provide a global overview of the costs of DFU care. Although no previous study has quantified the costs or margins associated with DFU readmissions, a couple of studies have documented the costliness of emergency evaluations of DFUs. Skrepnek et al
      • Skrepnek G.H.
      • Mills Sr., J.L.
      • Armstrong D.G.
      A diabetic emergency one million feet long: Disparities and burdens of illness among diabetic foot ulcer cases within emergency departments in the United States, 2006-2010.
      quantified the costs of care associated with patients with DFUs presenting to the emergency room using the Healthcare Cost and Utilization Project data. They found that ∼$1.9 billion annually was spent on DFU evaluation and care in the emergency room compared with $8.8 billion spent on inpatient DFU care annually. Hopkins et al
      • Hopkins R.B.
      • Burke N.
      • Harlock J.
      • Jegathisawaran J.
      • Goeree R.
      Economic burden of illness associated with diabetic foot ulcers in Canada.
      reported that, in Canada, the average number of emergency room visits per patient with DFU was 2.8 and accounted for 5% of all DFU resource usage in Canada. In our study, we found that 23% of patients with DFU required an unplanned readmission within 30 days and that the readmission costs were ∼$19,000 per incident. Taken together, these data suggest that DFU care is unpredictable, variable, and expensive. Furthermore, readmissions account for a substantial portion of DFU costs and should be considered a significant burden on the healthcare system.
      We found that the cost of readmissions for patients with DFUs was just as great as the cost of the index admissions. This is concerning because it suggests that the intensity of care required for a readmitted patient is similar to that required during an index admission. We have previously shown that most readmissions were for wound care issues that developed after discharge
      • Holscher C.M.
      • Hicks C.W.
      • Canner J.K.
      • Sherman R.L.
      • Malas M.B.
      • Black III, J.H.
      • et al.
      Unplanned 30-day readmission in patients with diabetic foot wounds treated in a multidisciplinary setting.
      ; thus, we assumed that the readmission costs would be reflective of medical management and aggressive local wound care. However, although 63% of readmissions were for wound-related issues in the present study, 23% of the readmitted patients required vascular surgical intervention. In addition, more than one third of the readmitted patients required subsequent podiatric surgery, which could be reflective of worsening infection, inadequate initial excision of osteomyelitis, and/or poor perfusion. The procedural costs for DFUs were greater than the medical costs (eg, admission for intravenous antibiotics),
      • Hicks C.W.
      • Selvarajah S.
      • Mathioudakis N.
      • Perler B.A.
      • Freischlag J.A.
      • Black III, J.H.
      • et al.
      Trends and determinants of costs associated with the inpatient care of diabetic foot ulcers.
      which likely increased the readmission costs in our study to a level similar to those for the index admission. Although the cost of readmission was similar to that of the index admission, the overall decrease in readmissions over time might represent an overall healthcare cost savings. However, the contribution of rehabilitation facility costs to the overall healthcare costs were not captured in the present study; thus, we could not confirm this hypothesis.
      All the patients evaluated by our multidisciplinary limb preservation team were treated via an algorithmic approach; thus, their vascular evaluation, vascular and podiatric interventions, and wound care were standardized.
      • Mathioudakis N.
      • Hicks C.W.
      • Canner J.K.
      • Sherman R.L.
      • Hines K.F.
      • Lum Y.W.
      • et al.
      The Society for Vascular Surgery Wound, Ischemia, and foot Infection (WIfI) classification system predicts wound healing but not major amputation in patients with diabetic foot ulcers treated in a multidisciplinary setting.
      In addition, we were aggressive in our treatment of PAD. Thus, all patients with a toe pressure ≤60 mm Hg underwent angiography to ensure adequate angiosomal perfusion before undergoing podiatric debridement. With the surprising readmission costs we found in the present study and the data suggesting that operative interventions will be required for nearly one half of patients at readmission, we began been placing more emphasis on the postoperative care paradigm, including greater integration of home healthcare services or rehabilitation placement, whenever necessary, to ensure adequate wound care. We have also changed our paradigm to be more aggressive with intravenous antibiotics after foot wound debridement and minor amputation. Initially, we followed the Infectious Diseases Society of America 2012 guidelines for the treatment of osteomyelitis.
      • Lipsky B.A.
      • Berendt A.R.
      • Cornia P.B.
      • Pile J.C.
      • Peters E.J.
      • Armstrong D.G.
      • et al.
      2012 Infectious Diseases Society of America clinical practice guideline for the diagnosis and treatment of diabetic foot infections.
      However, after we found that most unplanned readmissions were attributable to deterioration of the foot wound, our infectious diseases colleagues recommended more aggressive treatment with prolonged intravenous antibiotics. Thus, we have changed our practice such that we now treat forefoot and midfoot wounds with 6 weeks of intravenous antibiotics, regardless of whether the final bone margin finding was negative. Patients with toe amputations receive a total of 2 weeks of antibiotics. Also, the decision to stop antibiotic therapy is now a group determination based on wound healing, inflammatory markers, and risk. Antibiotic therapy is almost never continued for >8 weeks. We have attributed our decrease in readmission rates over time to these changes in our treatment paradigm.
      Despite the similar hospital costs for index admissions compared with readmissions, the hospital net margins were lower for readmissions. This finding suggests that readmissions are costly for hospitals from a profitability standpoint, and the burden of this issue is likely to only progress in the coming years. Starting in 2012, the Centers for Medicare and Medicaid implemented the Hospital Readmissions Reduction Program, which reduces payments to hospitals with excess readmissions for acute myocardial infarction, heart failure, pneumonia, chronic obstructive pulmonary disease, total hip or knee replacement, and coronary artery bypass graft surgery.
      Centers for Medicare and Medicaid Services (CMS)
      Readmissions reduction program.
      Although DFU admissions are not currently included in the Hospital Readmissions Reduction Program, it is probable that readmission rates will be assessed and penalized across all diagnoses. The findings from our present study showed that 30-day readmissions are associated with a 50% reduction in net margins. In 2019, the payment reduction for hospitals with excess readmissions was capped at 3%. If applied to DFU care, this would lower our hospital's net margins to 47%. Given the increasing number of DFU admissions occurring annually in the United States,
      • Hicks C.W.
      • Selvarajah S.
      • Mathioudakis N.
      • Perler B.A.
      • Freischlag J.A.
      • Black III, J.H.
      • et al.
      Trends and determinants of costs associated with the inpatient care of diabetic foot ulcers.
      this penalty could potentially be devastating to the financial sustainability of treating institutions.
      The inpatient cost of care for DFU has been previously estimated to be $1.4 billion annually in the United States.
      • Hicks C.W.
      • Selvarajah S.
      • Mathioudakis N.
      • Sherman R.E.
      • Hines K.F.
      • Black III, J.H.
      • et al.
      Burden of infected diabetic foot ulcers on hospital admissions and costs.
      When extrapolated to the national data, the 15% readmission cost burden we have reported would be equivalent to an estimated $210 million in hospital costs annually. Given the clinical and financial effects of DFU readmissions on our healthcare system, focused efforts at reducing readmissions is essential. Rumenapf et al
      • Rumenapf G.
      • Geiger S.
      • Schneider B.
      • Amendt K.
      • Wilhelm N.
      • Morbach S.
      • et al.
      Readmissions of patients with diabetes mellitus and foot ulcers after infra-popliteal bypass surgery—attacking the problem by an integrated case management model.
      previously demonstrated that the implementation of an integrated case management system responsible for coordinating outpatient care can reduce hospital readmissions and lower costs for patients with DFUs undergoing bypass surgery. Driver et al
      • Driver V.R.
      • Fabbi M.
      • Lavery L.A.
      • Gibbons G.
      The costs of diabetic foot: the economic case for the limb salvage team.
      have similarly reported that the establishment of multidisciplinary diabetic foot teams can improve patients' access to care and reduce complications in the long term, justifying the upfront costs associated with the model. We sought to establish a similar treatment paradigm with our multidisciplinary limb preservation team. Although the upfront costs of a multidisciplinary approach to DFU management might be greater in the short term,
      • Apelqvist J.
      • Larsson J.
      What is the most effective way to reduce incidence of amputation in the diabetic foot?.
      • Rinkel W.D.
      • Luiten J.
      • van Dongen J.
      • Kuppens B.
      • Van Neck J.W.
      • Polinder S.
      • et al.
      In-hospital costs of diabetic foot disease treated by a multidisciplinary foot team.
      the long-term wound healing and limb salvage rates we have achieved have, ultimately, been cost-effective.
      • Ortegon M.M.
      • Redekop W.K.
      • Niessen L.W.
      Cost-effectiveness of prevention and treatment of the diabetic foot: a Markov analysis.
      The limb salvage rate was 91% in the present study, supporting our use of aggressive limb preservation strategies even in patients with advanced wounds. We are hesitant to resort to major amputations unless absolutely necessary, because the long-term functional morbidity of these procedures has been extremely high. In our experience, only 31% of patients will achieve independent ambulation after a major amputation. However, no clinically relevant independent predictors of readmission have been established that can be used to target high-risk patients.
      • Holscher C.M.
      • Hicks C.W.
      • Canner J.K.
      • Sherman R.L.
      • Malas M.B.
      • Black III, J.H.
      • et al.
      Unplanned 30-day readmission in patients with diabetic foot wounds treated in a multidisciplinary setting.
      Thus, we would suggest that frequent evaluations by the dedicated limb preservation team be used to identify patients whose wound healing is not progressing appropriately to allow for early reintervention, as necessary.
      The limitations of our study include that is was a single-institution study, the lack of a control group, and cost data were limited to 2012 through 2016. We initiated our multidisciplinary limb preservation service shortly after 2012, and all patients with DFU in our hospital system were treated by our team. Thus, the data we have presented represent a specific care model at a single hospital. Cost data for patients treated before the implementation of our team were not available; thus, we could not comment on whether the readmission costs we have reported are representative of the DFU readmission costs nationally. However, to the best of our knowledge, no previous studies have reported on the cost and net margins associated with 30-day readmissions among patients with DFU patients. Thus, our study is novel and provides a baseline from which future reports can be compared. Furthermore, our use of an institutional prospective database allowed us to accurately capture all 30-day readmissions to our hospital. Thus, we did not have to rely on generic diagnosis codes to identify the affected patients and their associated admissions.

      Conclusions

      Readmissions for patients with DFU are common and are associated with a substantial cost burden. Our study showed that the cost of readmissions for patients with DFU was just as great as the cost of the index admissions but with lower hospital net margins. When extrapolated to the national data, the 15% readmission cost burden we have reported would be equivalent to $210 million in hospital costs annually. Focused efforts at preventing readmissions for this high-risk patient population, including implementation of a multidisciplinary care team, coordination of outpatient care, and early reevaluation after discharge, are essential to reducing the overall costs of care associated with DFUs.

      Author contributions

      Conception and design: CH
      Analysis and interpretation: CH, JK, HK, NM, RS, JB, CA
      Data collection: CH, JK, HK, NM, RS, JB, CA
      Writing the article: CH, CA
      Critical revision of the article: CH, JK, HK, NM, RS, JB, CA
      Final approval of the article: CH, JK, HK, NM, RS, JB, CA
      Statistical analysis: JK
      Obtained funding: Not applicable
      Overall responsibility: CH

      References

        • Hicks C.W.
        • Selvarajah S.
        • Mathioudakis N.
        • Sherman R.E.
        • Hines K.F.
        • Black III, J.H.
        • et al.
        Burden of infected diabetic foot ulcers on hospital admissions and costs.
        Ann Vasc Surg. 2016; 33: 149-158
        • Hicks C.W.
        • Selvarajah S.
        • Mathioudakis N.
        • Perler B.A.
        • Freischlag J.A.
        • Black III, J.H.
        • et al.
        Trends and determinants of costs associated with the inpatient care of diabetic foot ulcers.
        J Vasc Surg. 2014; 60: 1247-1254.e2
        • Jones C.E.
        • Richman J.S.
        • Chu D.I.
        • Gullick A.A.
        • Pearce B.J.
        • Morris M.S.
        Readmission rates after lower extremity bypass vary significantly by surgical indication.
        J Vasc Surg. 2016; 64: 458-464
        • Remington A.C.
        • Hernandez-Boussard T.
        • Warstadt N.M.
        • Finnegan M.A.
        • Shaffer R.
        • Kwong J.Z.
        • et al.
        Analyzing treatment aggressiveness and identifying high-risk patients in diabetic foot ulcer return to care.
        Wound Repair Regen. 2016; 24: 731-736
        • Holscher C.M.
        • Hicks C.W.
        • Canner J.K.
        • Sherman R.L.
        • Malas M.B.
        • Black III, J.H.
        • et al.
        Unplanned 30-day readmission in patients with diabetic foot wounds treated in a multidisciplinary setting.
        J Vasc Surg. 2018; 67: 876-886
        • Hicks C.W.
        • Canner J.K.
        • Karagozlu H.
        • Mathioudakis N.
        • Sherman R.L.
        • Black III, J.H.
        • et al.
        The SVS WIfI classification system correlates with cost of care for diabetic foot ulcers treated in a multidisciplinary setting.
        J Vasc Surg. 2018; 67: 1455-1462
        • Hopkins R.B.
        • Burke N.
        • Harlock J.
        • Jegathisawaran J.
        • Goeree R.
        Economic burden of illness associated with diabetic foot ulcers in Canada.
        BMC Health Serv Res. 2015; 15: 13
        • Rice J.B.
        • Desai U.
        • Cummings A.K.
        • Birnbaum H.G.
        • Skornicki M.
        • Parsons N.B.
        Burden of diabetic foot ulcers for Medicare and private insurers.
        Diabetes Care. 2014; 37: 651-658
        • Harrington C.
        • Zagari M.J.
        • Corea J.
        • Klitenic J.
        A cost analysis of diabetic lower-extremity ulcers.
        Diabetes Care. 2000; 23: 1333-1338
        • Joret M.O.
        • Dean A.
        • Cao C.
        • Stewart J.
        • Bhamidipaty V.
        The financial burden of surgical and endovascular treatment of diabetic foot wounds.
        J Vasc Surg. 2016; 64: 648-655
        • Skrepnek G.H.
        • Mills Sr., J.L.
        • Armstrong D.G.
        A diabetic emergency one million feet long: Disparities and burdens of illness among diabetic foot ulcer cases within emergency departments in the United States, 2006-2010.
        PLoS One. 2015; 10: e0134914
      1. Hicks CW, Canner JK, Karagozlu H, Mathioudakis N, Sherman RL, Black JH 3rd, et al. Quantifying the costs and profitability of care for diabetic foot ulcers treated in a multidisciplinary setting [published online ahead of print January 1. 2019]. J Vasc Surg doi: 10.1046/j.jvs.2018.10.097.

        • Mathioudakis N.
        • Hicks C.W.
        • Canner J.K.
        • Sherman R.L.
        • Hines K.F.
        • Lum Y.W.
        • et al.
        The Society for Vascular Surgery Wound, Ischemia, and foot Infection (WIfI) classification system predicts wound healing but not major amputation in patients with diabetic foot ulcers treated in a multidisciplinary setting.
        J Vasc Surg. 2017; 65: 1698-1705.e1
        • McPhee J.T.
        • Nguyen L.L.
        • Ho K.J.
        • Ozaki C.K.
        • Conte M.S.
        • Belkin M.
        Risk prediction of 30-day readmission after infrainguinal bypass for critical limb ischemia.
        J Vasc Surg. 2013; 57: 1481-1488
        • Lipsky B.A.
        • Berendt A.R.
        • Cornia P.B.
        • Pile J.C.
        • Peters E.J.
        • Armstrong D.G.
        • et al.
        2012 Infectious Diseases Society of America clinical practice guideline for the diagnosis and treatment of diabetic foot infections.
        Clin Infect Dis. 2012; 54: e132-e173
        • Centers for Medicare and Medicaid Services (CMS)
        Readmissions reduction program.
        (Updated 2016. Available at:)
        • Rumenapf G.
        • Geiger S.
        • Schneider B.
        • Amendt K.
        • Wilhelm N.
        • Morbach S.
        • et al.
        Readmissions of patients with diabetes mellitus and foot ulcers after infra-popliteal bypass surgery—attacking the problem by an integrated case management model.
        Vasa. 2013; 42: 56-67
        • Driver V.R.
        • Fabbi M.
        • Lavery L.A.
        • Gibbons G.
        The costs of diabetic foot: the economic case for the limb salvage team.
        J Vasc Surg. 2010; 52: 17S-22S
        • Apelqvist J.
        • Larsson J.
        What is the most effective way to reduce incidence of amputation in the diabetic foot?.
        Diabetes Metab Res Rev. 2000; 16: S75-S83
        • Rinkel W.D.
        • Luiten J.
        • van Dongen J.
        • Kuppens B.
        • Van Neck J.W.
        • Polinder S.
        • et al.
        In-hospital costs of diabetic foot disease treated by a multidisciplinary foot team.
        Diabetes Res Clin Pract. 2017; 132: 68-78
        • Ortegon M.M.
        • Redekop W.K.
        • Niessen L.W.
        Cost-effectiveness of prevention and treatment of the diabetic foot: a Markov analysis.
        Diabetes Care. 2004; 27: 901-907