Costs of Lower-Extremity Ulcers Among Patients With Diabetes Free

This was a retrospective analysis of medical and pharmacy claims from a large MCO and Prescription Solutions, a pharmacy and medical management company. Longitudinal claims data were used from members residing in California, Oregon, Oklahoma, Texas, and Washington—approximately 2.7 million lives.

The study observation period was from 1 January 2000 through 31 December 2001. Patients aged 18 years or older were included if they had at least one medical claim with a diagnosis of diabetes (online appendix section 1 [available at http://care.diabetesjournals.org]) and at least one ulcer-related medical claim during the observation period. An ulcer-related medical claim was defined as a claim with a diagnostic or procedure code representing a lower-extremity ulcer (online appendix section 2); specific codes were selected from those described in previous publications (6,7).

Episodes of care were identified for each patient and consisted of three periods: pretherapy, therapy, and post-therapy. The therapy period began on the day of the first ulcer-related medical claim and continued until the end of service date of the last ulcer-related claim before the post-therapy period. The pretherapy period was defined as the 90-day period before the therapy period during which no ulcer-related medical claims occurred. The post-therapy period was the 90-day period following the therapy period during which no ulcer-related medical claims occurred. Episodes were excluded from the analysis if any of the following situations occurred: 1) the patient was not continuously enrolled for the entire episode (including the pretherapy, therapy, and post-therapy periods); 2) the 90-day pretherapy or post-therapy periods fell outside of the study observation period; 3) the ulcer was still in progress at the end of the study observation period; or 4) the duration of the therapy period was ≤7 days (because shorter episodes would likely represent a minor skin condition, not a lower-extremity ulcer). For patients with more than one episode, only the first episode was included in the analysis.

To stratify each episode according to severity, a severity grading system (based on the system used by Holzer et al. [6], which was adapted from the Wagner classification [9]) was used. Level 1 was defined as a superficial ulcer, level 2 as a deeper ulcer (i.e., cellulitis) without bone involvement, level 3 as a deep ulcer with bone involvement (i.e., osteomyelitis, periostitis), and level 4/5 as gangrene or amputation. Levels 4 and 5 were merged because they were difficult to distinguish on the basis of medical claims.

Episodes were grouped into mutually exclusive cohorts based on the presenting severity level, which was defined as the severity stratification code at the index date. If there was more than one severity code at index, the highest level of severity was chosen. In addition, episodes were stratified according to their outcome severity level, which was defined as the highest level of severity during the therapy period. Episodes that could not be classified according to presenting or outcome severity level were excluded.

The primary outcomes were ulcer-related pharmacy, medical, and total costs measured over each episode’s therapy period. Ulcer-related pharmacy costs were defined as the outpatient prescription costs for ulcer-related oral antibiotics (i.e., prescriptions filled within 7 days before or after an ulcer-related medical claim or a previous ulcer-related antibiotic prescription) plus the costs for topical recombinant human platelet-derived growth factor. Ulcer-related medical costs were defined as the medical charges with an ulcer-related diagnosis or procedure code listed as the primary diagnosis. Medical costs were further stratified into charges associated with acute inpatient hospitalization (i.e., at least 1 day length of stay), other inpatient hospitalization (i.e., <1 day length of stay), inpatient skilled nursing facility, emergency department, outpatient hospital, outpatient office, and home therapy. While skilled nursing and home therapy charges were analyzed, these data may not be complete. Total ulcer-related costs were calculated as the sum of the ulcer-related pharmacy and medical costs.

Other variables included Chronic Disease Score (10) (a measure of comorbidity calculated during the pretherapy period), vascular status (inadequate vs. adequate), presence of a diagnosis of a malignant neoplasm of the lymphatic and hematopoietic tissue (online appendix A, section 3) during the pretherapy or therapy period, and presence of a diagnosis of amputation (online appendix A, section 4) during the therapy period. Patients were considered to have inadequate vascular status if they had a diagnosis code listed in online appendix A, section 4, during the pretherapy or therapy period.

Statistical analyses were performed using SAS, version 8.2. Patient characteristics and costs were analyzed using descriptive statistics and χ², t, or F tests. Cost data were also analyzed using nonparametric tests. Because the results remained consistent, only the parametric test results are reported. All statistical tests were two sided with an α of 0.05.

A total of 2,332 patients were identified. After excluding 79 patients whose episode could not be classified according to severity level, the final study population consisted of 2,253 patients.

Patient and episode characteristics are displayed in Table 1. The mean age was 68.9 years, and 59% were male. Approximately 44% of patients had inadequate vascular status, and this proportion increased with severity level, ranging from 33% for level 1 to 70% for level 4/5.

The average episode duration was 87.3 ± 82.8 days (median 62 days; range 8–542 days). Amputations occurred in 682 (30.3%) patients. Approximately 24% experienced an ulcer-related acute inpatient hospitalization, with the lowest proportion among patients with severity level 1 (11%) and the highest among severity level 4/5 (49%) (P < 0.0001). Of 1,694 episodes presenting as level 1, 2, or 3, 498 patients (29.4%) progressed to a higher outcome severity level.

Table 2 displays the ulcer-related pharmacy, medical, and total costs. Total ulcer-related costs averaged $13,179 per episode and increased according to outcome severity, ranging from $1,892 (level 1) to $27,721 (level 4/5). Inpatient acute hospitalization costs represented the majority of costs, averaging $10,188 per episode.

Total ulcer-related costs were higher among patients with inadequate vascular status than among patients with adequate vascular status ($23,372 vs. $5,218, respectively, P < 0.0001). Total ulcer-related costs were also higher for patients <65 years of age than for patients 65 years and older ($16,390 vs. $11,925, P = 0.02) (Table 3).

Total ulcer-related costs were analyzed for patients who presented with severity level 1, 2, or 3 and then progressed to a higher outcome severity level, as compared with patients who did not progress (Table 4). Patients who progressed to a higher severity level had higher ulcer-related costs compared with patients who did not progress ($20,136 vs. $3,063, P < 0.0001).

In this population of patients with diabetes and lower-extremity ulcers, total ulcer-related costs averaged $13,179 per episode. Consistent with other studies (6,7), acute inpatient hospital charges accounted for 77% of the overall charges, indicating that hospitalization was a major cost driver.

Episode costs increased with severity level, and these increases appear to be ulcer-driven since comorbidity scores were similar across ulcer severity levels. Higher costs were observed among younger patients, patients with inadequate vascular status, and patients whose ulcer progressed to a higher severity level. Costs averaged $4,465 higher for patients <65 years compared with older patients. On average, costs were $18,154 higher for patients with inadequate vascular status compared with patients with adequate vascular status. This cost difference is consistent with other studies (6) and is not surprising, given that ischemia and compromised circulation owing to inadequate vascular status can prolong ulcer healing and increase the risk of amputation (1,11). Patients presenting with severity level 1, 2, or 3 and whose ulcer progressed to a higher severity experienced more than $17,000 higher costs on average than those whose ulcers did not progress.

Of the identified episodes, 22% progressed to a higher severity level, 30% required amputation, and 24% resulted in an inpatient hospitalization. These numbers are high, considering that patients in this study had medical insurance and therefore access to health care. These results emphasize the importance of early detection and aggressive treatment of lower-extremity ulcers, and they suggest the need for effective intervention programs designed to prevent hospitalization and ulcer progression.

Episode cost ($13,179) in this study was higher than in previously published studies. In an analysis of 1991 and 1992 data from patients under the age of 65 years with diabetes and lower-extremity ulcers, average payment per episode was $4,595 (6). Another analysis of 1995 data estimated that ulcer-related Medicare spending averaged $3,609 per year among patients with diabetic lower-extremity ulcers (7). There are several explanations for the differences in these data. Because costs in the present analysis were from 2000 and 2001 data, they would be expected to be higher than costs from earlier years. Costs in this analysis were based on charges, which are higher than actual costs. In contrast, studies using Medicare data may underestimate the costs of treatment because they only include costs for services that qualify for reimbursement. Definition of episode could contribute to differences in episode duration and costs among the different studies. In the present study, episode duration ranged from 8 to 542 days. To evaluate whether costs were inflated by the longer episodes, total ulcer-related costs were recalculated after excluding 25 patients (1% of the study population) with episodes longer than 365 days, which resulted in costs similar to those of the study population ($13,017 ± $34,513 vs. $13,179 ± $34,511).

The medical costs in this analysis were those associated with a primary (i.e., first-listed) ulcer-related diagnosis, which may underestimate the costs of a lower-extremity ulcer episode. In contrast, an analysis using costs associated with any ulcer-related diagnosis may have greater limitations because it likely overestimates the costs by including costs associated with other diagnoses. For this analysis, the costs associated with a primary diagnosis of lower-extremity ulcer were reported because they were less likely to inappropriately include costs associated with other high-cost complications of diabetes, such as cardiovascular and renal diseases.

Because this study was conducted among patients in the western United States who have employer-based insurance or Medicare + Choice, the results may not be representative of all patients with diabetic lower-extremity ulcers. As with other database analyses, these data are subject to errors in diagnosis coding, incomplete claims, and unobservable factors that may have influenced outcomes. Costs associated with outpatient intravenous antibiotics may not have been captured since the database does not always contain claims for outpatient intravenous medications. Classification of episode severity level was attempted using a coding stratification based on the Wagner system (6,9), but this method may not sufficiently characterize patients’ level of disease. In addition, this analysis did not assess patients’ degree of glucose control, a factor that has been associated with several diabetes complications, including neuropathy (12).

This study provides insight into the current health care costs associated with treating diabetic lower-extremity ulcers. Treatment of diabetic lower-extremity ulcers is costly, especially among younger patients, patients with inadequate vascular status, and patients whose ulcer progressed to a higher severity. These results emphasize the value of intensive outpatient interventions designed to prevent hospitalization and ulcer progression.

This study was supported by a grant from Bristol-Myers Squibb.