Estimating the Real-World Cost of Diabetes Mellitus in the United States During an 8-Year Period Using 2 Cost Methodologies.

BACKGROUND: Diabetes is associated with substantial clinical and economic burdens on patients and on the US healthcare system. Treatment options for patients with type 1 or type 2 diabetes have increased significantly, from only 3 drug classes in 1995 to more than 12 distinct classes today. Although several of the newer treatments are reported to have improved efficacy and safety profiles, they are often substantially more costly than older medications. Consequently, as drug options increase, the cost of diabetes management continues to grow. OBJECTIVES: To estimate the annual real-world costs of type 1 and 2 diabetes, as well as diabetes prevalence, treatment patterns, care quality, and resource utilization during 8 years. METHODS: In this cross-sectional study, we examined 8 annual cohorts of patients with type 1 or type 2 diabetes, on a biennial basis, using claims data from the HealthCore Integrated Research Database between 2006 and 2014. Patients were matched with controls by age, sex, residency, and health plan type. We assessed the prevalence of diabetes, treatment patterns, care quality measures, and all-cause and diabetes-related healthcare costs using 2 methods. Method 1 calculated the annual costs as the difference in all-cause costs between patients with diabetes and matched controls. Method 2 calculated the costs for healthcare encounters based on specific codes for a diabetes diagnosis or for antidiabetes medications. RESULTS: Between 346,486 and 410,234 patients with type 2 diabetes and between 21,176 and 26,228 patients with type 1 diabetes were included in each study year cohort. Between 2007 and 2014, the prevalence of type 2 diabetes increased from 4.9% to 6.3%. The costs associated with using Method 1 were almost double the cost estimates in Method 2 during most of the study period. For patients with type 1 diabetes, the associated costs were twice greater with Method 1 than with Method 2. Projections to the entire US population in 2014 indicated a total of 19.3 million individuals with diabetes and associated direct costs of $314.8 billion that year. CONCLUSION: Cost estimates can guide the prioritization of healthcare expenditures. The results of this study showed that costs attributable to diabetes differed by approximately 2-fold, depending on the estimation method. The management of the escalating expenses for diabetes management in the United States requires judicious selection of the methods for estimating costs.

Understanding Prediabetes in a Medicare Advantage Population Using Data Adaptive Techniques.

The objective was to identify individuals with undiagnosed prediabetes from administrative data using adaptive techniques. The data source was a national Medicare Advantage Prescription Drug (MAPD) plan administrative data set. A retrospective, cross-sectional study developed and evaluated data adaptive logistic regression, decision tree, neural network, and ensemble predictive models for metabolic syndrome and prediabetes using 3 mutually exclusive cohorts (N = 279,903). The misclassification rate (MCR), average squared error (ASE), c-statistics, sensitivity (SN), and false positive (FP) rates were compared to select the final predictive models. MAPD individuals with continuous enrollment from 2013 to 2014 were included. Metabolic syndrome and prediabetes were defined using clinical guidelines, diagnosis, and laboratory data. A total of 512 variables identified through subject matter expertise in addition to utilizing all data available were evaluated for the modeling. The ensemble model demonstrated better discrimination (c-statistics, MCR, and ASE of 0.83, 0.24, and 0.16, respectively), high SN, and low FP rate in predicting metabolic syndrome than the individual data adaptive modeling techniques. Logistic regression demonstrated better discrimination (c-statistics, MCR, and ASE of 0.67, 0.13, and 0.11 respectively), high SN, and low FP rate in predicting prediabetes than the other adaptive modeling techniques or ensemble methods. The scored data predicted prediabetes in 44% of the MAPD population, which is comparable to 2005-2006 National Health and Nutrition Examination Survey prediabetes rates of 41%. The logistic regression model demonstrated good performance in predicting undiagnosed prediabetes in MAPD individuals.

Positive predictive value between medical-chart body-mass-index category and obesity versus codes in a claims-data warehouse.

OBJECTIVE: To evaluate the positive predictive value of claims-based V85 codes for identifying individuals with varying degrees of BMI relative to their measured BMI obtained from medical record abstraction. METHODS: This was a retrospective validation study utilizing administrative claims and medical chart data from 1 January 2009 to 31 August 2015. Randomly selected samples of patients enrolled in a Medicare Advantage Prescription Drug (MAPD) or commercial health plan and with a V85 claim were identified. The claims-based BMI category (underweight, normal weight, overweight, obese class I-III) was determined via corresponding V85 codes and compared to the BMI category derived from chart abstracted height, weight and/or BMI. The positive predictive values (PPVs) of the claims-based BMI categories were calculated with the corresponding 95% confidence intervals (CIs). RESULTS: The overall PPVs (95% CIs) in the MAPD and commercial samples were 90.3% (86.3%-94.4%) and 91.1% (87.3%-94.9%), respectively. In each BMI category, the PPVs (95% CIs) for the MAPD and commercial samples, respectively, were: underweight, 71.0% (55.0%-87.0%) and 75.9% (60.3%-91.4%); normal, 93.8% (85.4%-100%) and 87.8% (77.8%-97.8%); overweight, 97.4% (92.5%-100%) and 93.5% (84.9%-100%); obese class I, 96.9 (90.9%-100%) and 97.2% (91.9%-100%); obese class II, 97.0% (91.1%-100%) and 93.0% (85.4%-100%); and obese class III, 85.0% (73.3%-96.1%) and 97.1% (91.4%-100%). CONCLUSIONS: BMI categories derived from administrative claims, when available, can be used successfully particularly in the context of obesity research.

A comparison of entropy balance and probability weighting methods to generalize observational cohorts to a population: a simulation and empirical example.

PURPOSE: We compared methods to control bias and confounding in observational studies including inverse probability weighting (IPW) and stabilized IPW (sIPW). These methods often require iteration and post-calibration to achieve covariate balance. In comparison, entropy balance (EB) optimizes covariate balance a priori by calibrating weights using the target's moments as constraints. METHODS: We measured covariate balance empirically and by simulation by using absolute standardized mean difference (ASMD), absolute bias (AB), and root mean square error (RMSE), investigating two scenarios: the size of the observed (exposed) cohort exceeds the target (unexposed) cohort and vice versa. The empirical application weighted a commercial health plan cohort to a nationally representative National Health and Nutrition Examination Survey target on the same covariates and compared average total health care cost estimates across methods. RESULTS: Entropy balance alone achieved balance (ASMD ≤ 0.10) on all covariates in simulation and empirically. In simulation scenario I, EB achieved the lowest AB and RMSE (13.64, 31.19) compared with IPW (263.05, 263.99) and sIPW (319.91, 320.71). In scenario II, EB outperformed IPW and sIPW with smaller AB and RMSE. In scenarios I and II, EB achieved the lowest mean estimate difference from the simulated population outcome ($490.05, $487.62) compared with IPW and sIPW, respectively. Empirically, only EB differed from the unweighted mean cost indicating IPW, and sIPW weighting was ineffective. CONCLUSION: Entropy balance demonstrated the bias-variance tradeoff achieving higher estimate accuracy, yet lower estimate precision, compared with IPW methods. EB weighting required no post-processing and effectively mitigated observed bias and confounding. Copyright © 2016 John Wiley & Sons, Ltd.

Characteristics Relating to Adherence and Persistence to Basal Insulin Regimens Among Elderly Insulin-Naïve Patients with Type 2 Diabetes: Pre-Filled Pens versus Vials/Syringes.

INTRODUCTION: Previous studies have found higher rates of adherence in patients with type 2 diabetes mellitus (T2DM) using insulin pens compared to vial and syringe administration; however, little evidence is available to support this observation in elderly patients. METHODS: This was a retrospective claims database analysis of a predominantly elderly Medicare Advantage with Prescription Drug (MAPD) insurance population consisting of 3172 insulin-naïve patients with T2DM who initiated basal insulin using pre-filled pens or vial and syringe ('vial'). The index date was defined by the first pharmacy claim for basal insulin. Adherence, measured as proportion of days covered (PDC) and medication possession ratio (MPR), and persistence were evaluated in a 12-month follow-up period using an adjusted days' supply. Multivariate regression analyses and a Cox proportional hazards model were used to identify characteristics associated with adherence and non-persistence, respectively, and compare findings between the pen and vial groups. RESULTS: The pen cohort was slightly younger than the vial cohort (69.4 vs. 70.1 years, respectively; P = 0.0338). Similar proportions of male patients (53.3% vs. 56.8%; P = 0.0529) occurred in both cohorts, and lower Deyo-Charlson Comorbidity Index (4.4 vs. 5.0; P < 0.0001) was found for the pen cohort. Adjusted mean PDC was significantly higher in the pen cohort than the vial cohort (0.67 vs. 0.50; P < 0.001), as was mean MPR (0.75 vs. 0.57; P < 0.0001). Adjusted odds for adherence (PDC ≥ 80%) showed a positive association with use of an insulin pen (odds ratio = 2.19, 95% CI: 1.86-2.59). The adjusted risk of non-persistence (discontinuation) was significantly lower (58%) in the pen cohort relative to the vial cohort (hazard ratio = 0.42, 95% CI: 0.38-0.45). Key limitations include assumptions related to accuracy and comprehensiveness of claims data, and specifically days' supply data used to measure insulin adherence. CONCLUSION: These findings suggest that pen devices improved insulin therapy adherence in a primarily elderly MAPD population with T2DM. FUNDING: Novo Nordisk Pharmaceuticals, Inc.

Relationship of diabetes complications severity to healthcare utilization and costs among Medicare Advantage beneficiaries.

OBJECTIVES: This study evaluated the usefulness of the Diabetes Complications Severity Index (DCSI) in assessing healthcare resource utilization (HRU) and costs among Medicare Advantage plan members diagnosed with type 2 diabetes mellitus (T2DM). STUDY DESIGN: A retrospective cohort study of medical and pharmacy claims of 333,576 Medicare members aged 18 to 89 years with ≥1 medical claim with primary diagnosis or ≥2 medical claims with secondary diagnosis, of T2DM (International Classification of Diseases, Ninth Revision, Clinical Modification code 250.x0 or 250.x2) during the period of January 1, 2010, to December 31, 2011. METHODS: DCSI was assessed concurrently with HRU and healthcare costs (total, medical, and pharmacy). The cohort was subdivided into 6 DCSI groups: DCSI = 0 (no complications) through DCSI = 5+ (≥5). Associations of complication severity with HRU and costs of care were summarized using regression models. RESULTS: A 1-point increase in DCSI was associated with a $2744 increase in total costs (a $2480 increase in medical costs plus a $264 increase in pharmacy costs). Increasing DCSI was associated with greater use of inpatient and emergency department (ED) services. Among the higher complications subgroups, there were greater representations of older patients, men, and cases of depressive disorders and hypoglycemia. CONCLUSIONS: DCSI is useful for identifying Medicare plan members with T2DM who should be targeted for clinical programs. HRU and costs increased with DCSI severity. Increases in high-cost HRU, driven by inpatient and ED visits, suggest that preventing or delaying utilization of these services are essential to driving down costs in the T2DM population. Furthermore, high rates of depression and hypoglycemia warrant early screening and necessary treatment to improve patient outcomes.

Antidiabetic treatment patterns in a medicare advantage population in the United States.

BACKGROUND: Published guidelines for treatment of type 2 diabetes mellitus (T2DM) agree on initial pharmacotherapy. However, few specific recommendations on second-line agents are provided. OBJECTIVE: The objective of this study was to describe antidiabetic treatment patterns in Medicare Advantage patients with T2DM within 6 months of measurement of the glycosylated hemoglobin (HbA1c) level. RESEARCH DESIGN: This retrospective cross-sectional study utilized medical, pharmacy, and laboratory claims from a large Medicare Advantage with Prescription Drug (MAPD) coverage payer. MAPD members between 65 and 89 years old identified as having T2DM between 2009 and 2011 were eligible for inclusion. A 12-month baseline period before the first HbA1c value (index date) was evaluated for demographic and clinical differences. Antidiabetic therapy was evaluated for 6 months post-index. The study population was stratified into three cohorts based on index HbA1c value: controlled (<8%, 64 mmoL/mol), uncontrolled (≥ 8%, 64 mmoL/mol and <10%, 86 mmoL/mol), and severely uncontrolled (≥ 10%, 86 mmoL/mol). RESULTS: Despite elevated HbA1c values (≥ 8%, 64 mmoL/mol), 7-8% of patients did not receive antidiabetic therapy during the post-index period. Metformin and sulfonylureas were the oral antidiabetics (OADs) most frequently used as monotherapy. The majority of patients on combination therapy were on two or more OADs and higher injectable use was observed in the severely uncontrolled cohort. Metformin was included in >60% of the combination regimens with metformin + sulfonylurea being the most common. CONCLUSION: This study suggests suboptimal treatment of those not in glycemic control (HbA1c ≥ 8%, 64 mmoL/mol). Many patients classified as severely uncontrolled based on HbA1c received only monotherapy. Opportunities exist for treatment modification within this population to achieve tighter glycemic control.

The burden of diabetes mellitus for medicare beneficiaries.

The objective was to estimate health care costs and utilization for Medicare beneficiaries with type 1 (T1DM) or type 2 (T2DM) diabetes and their respective matched control cohorts. A retrospective claims cohort analysis was used to assess direct health care cost and utilization of health services in 2009 for patients aged 65-89 who were enrolled in a Medicare Advantage Plus prescription drug plan. Patients were matched 1:1 with patients without diabetes. All-cause health care costs for 2009 were calculated as the sum of all medical and pharmacy claims. The analysis included 6562 patients with T1DM and an equal number of matched controls, and 194,775 patients with T2DM and an equal number of matched controls. There were no significant demographic differences between cohorts for matched variables. Patients with T2DM had significantly higher mean Deyo/Charlson Comorbidity Index scores compared with their controls (2.47 versus 0.77; P<0.001), although all groups reported a high rate of costly comorbidities such as hypertension and heart disease. Mean all-cause health care costs per patient per year were significantly higher for patients with T1DM and T2DM versus controls for inpatient hospitalizations; outpatient, office, and emergency room visits; pharmacy expenditures; and total health care costs for 2009 (T1DM group: $20,701±$30,201; T1DM-matched control group: $6,537±$10,441; T2DM group: $10,437±$18,518; T2DM-matched control group: $6,505±$11,140). Diabetes escalates health care costs for Medicare Advantage Plus patients compared with patients in the same plan without diabetes, regardless of comorbidities.

Retrospective real-world adherence in patients with type 2 diabetes initiating once-daily liraglutide 1.8 mg or twice-daily exenatide 10 µg.

BACKGROUND: The effectiveness of a drug is significantly influenced by a patient's adherence to the required regimen. OBJECTIVE: The goal of this retrospective database analysis was to determine the factors affecting adherence over a 12-month follow-up period in adults with type 2 diabetes mellitus (DM) initiating once-daily liraglutide (1.8 mg) or twice-daily exenatide (10 µg). METHODS: A patient-centric claims database was used, covering the period January 2009 to December 2011. Patients were included if they had ≥1 claim of once-daily liraglutide 1.8 mg or twice-daily exenatide 10 µg from January to December 2010 (index date [ID]), ≥2 diagnoses of type 2 DM before ID, continuous enrollment for 12 months before and after ID, and age ≥18 years at ID. Patients were required to be glucagon-like peptide-1 receptor agonist treatment-naive in the 12 months preceding ID and have a second prescription for once-daily liraglutide 1.8 mg or twice-daily exenatide 10 µg during the 12 months after ID. The medication possession ratio (MPR) was used as a continuous variable and to categorize patients as high-adherent (MPR ≥80%) or low-adherent (MPR <80%). Regression analyses were conducted to determine the predictors for nonadherence in the type 2 DM population, with bivariate testing of the MPR categories conducted initially to determine the predictors to be included in the final regression model. RESULTS: A total of 3623 patients (once-daily liraglutide 1.8 mg, n = 2036; twice-daily exenatide 10 µg, n = 1587) were identified. Variables found to reduce adherence were younger age, female sex, Southern geographic region, twice-daily exenatide treatment, and higher percentage of copayment from the claimant. After adjusting for confounding factors, patients receiving once-daily liraglutide 1.8 mg were ∼11% more adherent than patients receiving twice-daily exenatide 10 µg (95% CI, 7-14; P < 0.0001). The odds ratio for "poor" adherence (MPR <80%) with twice-daily exenatide 10 µg therapy compared with liraglutide 1.8 mg once-daily was 1.33 (95% CI, 1.16-1.53; P < 0.0001). CONCLUSIONS: This study found that adherence to once-daily liraglutide 1.8 mg treatment was superior to twice-daily exenatide 10 µg over a 12-month follow-up period. Nonadherence has important implications to the health care system, both in terms of clinical effectiveness and economic burden (eg, hospitalization, productivity losses). Using strategies to increase adherence is vital to reduce the future clinical and economic burden of diabetes.

Impact of pen utilization on insulin cost reduction in long-term care facilities.

OBJECTIVE: To determine the impact on insulin acquisition cost of a pharmacy program to convert insulin utilization from multidose vials to pen-delivery systems for long-term care residents covered by Medicare Part A, and managed care plans. DESIGN: Retrospective cost comparison. SETTING: Long-term care facilities. PATIENTS: Residents covered by Medicare Part A and managed care plans. INTERVENTIONS: Policy to replace insulin vials with pen devices, effective July 2009. MAIN OUTCOME MEASURES: Mean insulin cost-per-patient day (total insulin purchases divided by patient admission days) and pen utilization (pen purchases as a percent of total insulin purchases). RESULTS: Insulin purchase data covered 2,405 admissions in 75 facilities over the 12-month period ending June 2010. Pen device purchases increased from less than 1% to almost 35% of total insulin purchases over the study period during which insulin cost per patient-day declined from $10.29 to $4.08. For Medicare Part A patients with admissions of 30 days or fewer, the most frequent visit type, mean cost per patient-day decreased from $13.73 to $9.19 as pen purchases increased from less than 1% to about 32%. For these same patients, mean cost per patient-day for admissions using only pen devices was $7.04, compared with $11.79 for admissions using only vials (P < 0.001). Significant differences in mean cost per patient-day were also found for residents covered by managed care and for longer admissions. CONCLUSION: Total insulin costs can be reduced through higher utilization of pen devices by patients in long-term care facilities.

Economic benefits of improved insulin stability in insulin pumps.

PURPOSE: Insulin pump users discard unused medication and infusion sets according to labeling and manufacturer's instructions. The stability labeling for insulin aspart (rDNA origin] (Novolog) was increased from two days to six. The associated savings was modeled from the perspective of a hypothetical one-million member health plan and the total United States population. DESIGN: The discarded insulin volume and the number of infusion sets used under a two-day stability scenario versus six were modeled. METHODS: A mix of insulin pumps of various reservoir capacities with a range of daily insulin dosages was used. Average daily insulin dose was 65 units ranging from 10 to 150 units. Costs of discarded insulin aspart [rDNA origin] were calculated using WAC (Average Wholesale Price minus 16.67%). The cost of pump supplies was computed for the two-day scenario assuming a complete infusion set change, including reservoirs, every two days. Under the six-day scenario complete infusion sets were discarded every six days while cannulas at the insertion site were changed midway between complete changes. AWP of least expensive supplies was used to compute their costs. PRINCIPAL FINDINGS: For the hypothetical health plan (1,182 pump users) the annual reduction in discarded insulin volume between scenarios was 19.8 million units. The corresponding cost reduction for the plan due to drug and supply savings was $3.4 million. From the U.S. population perspective, savings of over $1 billion were estimated. CONCLUSIONS: Using insulin that is stable for six days in pump reservoirs can yield substantial savings to health plans and other payers, including patients.

Does type of bolus insulin matter in the hospital? Retrospective cohort analysis of outcomes between patients receiving analogue versus human insulin.

BACKGROUND: Poor glycemic control in hospitalized patients has been associated with increased morbidity and mortality. Research suggests that analogue bolus insulin may be more effective in achieving blood glucose (BG) control compared with human bolus insulin. OBJECTIVE: This study compares mortality, length of stay (LOS), costs, and BG control in hospitalized patients receiving either analogue or human bolus insulin. METHODS: This retrospective cohort analysis used data from January 1, 2004, to December 31, 2007, within the Health Facts database (Cerner Corporation, Kansas City, Missouri). Nonsurgical adult patients who received exclusively analogue or human bolus insulin during hospitalization were included in the study. Propensity score matching and multivariate regression analyses were used to compare patients treated with analogue versus human bolus insulin. The study outcomes were in-hospital mortality, hospital LOS among survivors (to avoid potentially short hospitalizations among nonsurvivors distorting results), and hospitalized BG control (present vs absent), defined as having a mean BG of 70 to <200 mg/dL during hospitalization. RESULTS: In total, 35,049 participants met the inclusion criteria and 5568 of 7754 patients in the analogue group were matched by their propensity scores to patients in the human bolus group (mean age, 67.1 years; 53% women; 77% white). On propensity score analysis, analogue bolus insulin was associated with lower mortality (relative risk [RR] = 0.52; 95% CI, 0.45-0.61) and shorter LOS (0.668-day reduction; 95% CI, 0.44-0.89) compared with human bolus insulin. However, analogue insulin was associated with only a modest benefit for BG control (RR = 0.88; 95% CI, 0.81-0.95). The multivariate regression analysis produced similar findings. CONCLUSIONS: In this cohort of hospitalized patients, analogue bolus insulin was associated with lower mortality, shorter LOS, and modestly better BG control compared with patients treated with human bolus insulin. These results highlight the need for a randomized controlled clinical trial comparing outcomes by bolus insulin type in the hospital setting to determine a true mortality benefit.

Cost sharing, adherence, and health outcomes in patients with diabetes.

OBJECTIVES: To assess the relationship between cost sharing and adherence to antidiabetic medications in patients with type 2 diabetes and to examine the relationship between medication adherence and outcomes, including complication rates, medical service utilization, and workplace productivity measures. STUDY DESIGN: A retrospective, cross-sectional study analyzing the healthcare experience of patients with type 2 diabetes on oral antidiabetic medication (OAD) with or without insulin (n = 96,734) and patients on OAD only (n = 55,356) with employer-sponsored insurance in the 2003-2006 MarketScan Database. METHODS: Using a 2-stage residual inclusion model, the first stage estimated the effects of cost sharing on adherence to antidiabetic medications in an 18-month time frame (January 2003 through June 2004). Adherence was determined from the percentage of days covered. The second stage estimated the effects of adherence on complication rates (eg, retinopathy, neuropathy, peripheral vascular disease), medical service utilization rates, and measures of productivity (absence days and short-term disability days) in the subsequent 2 years (July 2004 through June 2006). RESULTS: A $10 increase in the patient cost-sharing index resulted in a 5.4% reduction in adherence to antidiabetic medications for patients on OAD only and a 6.2% reduction in adherence for patients on OAD with or without insulin. Adherence was associated with lower rates of complications (eg, amputation/ulcers, retinopathy) and also was associated with fewer emergency department visits and short-term disability days. CONCLUSIONS: Medical plans, employers, and policy makers should consider implementing interventions targeted to improve antidiabetic medication adherence, which may translate to better outcomes.

Cost of diabetes: comparison of disease-attributable and matched cohort cost estimation methods.

OBJECTIVE: To estimate and compare the annual direct healthcare cost among Type 1 (T1DM) and Type 2 (T2DM) diabetes patients using two cost estimation methods: (1) DM-attributable cost and (2) all cause case-control cost. RESEARCH DESIGN AND METHODS: An administrative claims cohort study using the HealthCore Integrated Research Database (HIRD(R)) identified T1DM and T2DM patients age >or=18 and <65 years between 1/1/2006 - 12/31/2006. DM patients (cases) were matched 1:1 with non-DM patients (controls) by age, gender, state, and commercial plan type (HMO, PPO, POS). All patients had continuous eligibility for calendar years 2006-07. DM-attributable cost was assessed by summing medical claims for DM (ICD-9-CM codes 250.xx) and pharmacy claims for anti-hyperglycemic agents, and all cause health care cost was assessed for cases and controls, for the calendar year 2007. RESULTS: A total of 12,096 T1DM and 256,245 T2DM cases and matched controls were identified. T1DM and T2DM cases had significantly higher average baseline comorbidities and Deyo-Charleson Comorbidity scores than controls (2.17 vs. 0.23 and 1.62 vs. 0.39, respectively, p < 0.0001 for both).While DM attributable cost estimation resulted in a mean annual cost of $6247 for T1DM and $3002 for T2DM in 2007, the mean annual (per patient) all-cause total cost estimation using the case-control method resulted in a difference of $10,837 ($14,060 for cases, vs. $3223 for controls) for T1DM; and $4217 ($8070 for cases, vs. $3853 for controls) for T2DM. CONCLUSIONS: The DM-attributable cost method underestimated costs by 42% for T1DM and 29% for T2DM compared to the case-control method. The difference was smaller but still significant (33% for T1DM and 14% for T2DM) when multivariate technique was used. Patients with DM may use a substantial amount of medical and pharmacy services not directly attributable to DM, and attributable cost method may underestimate the total cost of DM. This study has limitations inherent to the retrospective claims data analysis and generalizability of results is limited to those from similar population.

Direct health care costs of patients with type 2 diabetes within a privately insured employed population, 2000 and 2005.

OBJECTIVE: To assess the incremental economic burden of privately insured patients with type 2 diabetes (T2DM) in 2000 and 2005 in the United States. METHODS: Adults with T2DM and 24 months of continuous health plan enrollment were identified in the MarketScan databases (2000 and 2005). Control groups of persons without diabetes were selected for comparison using propensity score matching. Total adjusted health care costs were estimated using generalized linear modeling. RESULTS: Adjusted health care costs of patients with T2DM in 2005 were 136% higher than those of the matched controls ($12,733 vs $5406, P < 0.001). Similarly, costs of patients with T2DM in 2000 were 146% higher than those of the matched controls ($12,423 vs $5058, P < 0.001). Expenditures were similar for individuals with T2DM in 2000 and 2005. CONCLUSIONS: T2DM continues to impose a substantial economic burden to self-insured employers.