Real-world treatment patterns and overall survival among men with Metastatic Castration-Resistant Prostate Cancer (mCRPC) in the US Medicare population

BACKGROUND

Real-world treatment patterns and survival in metastatic castration-resistant prostate cancer (mCRPC) have not been characterized for the full fee-for-service Medicare population.

METHODS

Men newly diagnosed with mCRPC were identified in Medicare fee-for-service claims during 1/1/2014-6/30/2019. Men had evidence of mCRPC and continuous insurance coverage ≥1 year before and ≥6 months after diagnosis unless patients died. Treatment patterns after diagnosis were described. Survival from mCRPC diagnosis and from start of first-line (1 L) therapy was modeled using Kaplan-Meier analysis.

RESULTS

Among 14,780 men with mCRPC, mean age was 76 and median follow-up after mCRPC was 17.0 months. 22% received no life-prolonging therapy after mCRPC, 78% received ≥1 line of therapy (LOT), 42% underwent ≥2 LOTs, and 20% had ≥3 LOTs. Median time from start of 1 L to next LOT or end of follow-up was 13.7 months, 10.9 months from 2 L start, and 8.9 months from 3 L start. The most common 1 L to 2 L treatment sequences among men with ≥2 lines were NHT followed by a different NHT (33%), chemotherapy followed by NHT (14%), and NHT followed by chemotherapy (13%). For those initiating 1 L treatment with NHTs, only 28% received subsequent treatment with a different class of therapy. Median survival was 25.6 months after mCRPC and 23.4 months following treatment initiation.

CONCLUSIONS

More than 1 in 5 Medicare patients with mCRPC did not receive any life-prolonging therapy, and less than half received 2 L therapy. NHTs were the most common 1 L and 2 L therapies, with patients treated with NHT as 1 L followed by a different NHT for 2 L as the most common treatment sequence. Median survival from diagnosis for all patients was 25.6 months. These data highlight the dramatic undertreatment that occurs for mCRPC patients, particularly for therapies beyond NHTs as well as the common use of sequential NHTs in real-world data.

Healthcare Costs in Men with Metastatic Castration-Resistant Prostate Cancer: An Analysis of US Medicare Fee-For-Service Claims

INTRODUCTION

To analyze healthcare resource utilization (HRU) and healthcare costs in men with metastatic castration-resistant prostate cancer (mCRPC) in the US Medicare population.

METHODS

A published claims-based algorithm was used to identify men with mCRPC in the fee-for-service Medicare population between January 1, 2014, and December 31, 2019. Unadjusted all-cause HRU (days) and healthcare costs paid by Medicare (medical and pharmacy) per patient per year (PPPY) are described for the periods before mCRPC diagnosis, after diagnosis, and from the start of first-line (1L), second-line (2L), and third-line (3L) therapy with mCRPC life-prolonging treatments to the start of subsequent therapy or end of follow-up/death.

RESULTS

A total of 14,780 men with mCRPC were identified. After mCRPC diagnosis, 11,528 men initiated 1L mCRPC therapy, 6275 initiated 2L, and 2945 initiated 3L. All-cause medical HRU (days PPPY) increased after mCRPC diagnosis and from 1L through 3L treatment, particularly for outpatient care (pre-diagnosis, 10.4; 1L, 16.2; 2L, 18.9; 3L, 22.0) and physician/other visits (pre-diagnosis, 30.1; 1L, 46.5; 2L, 50.2; 3L, 56.9). Similarly, mean all-cause healthcare costs PPPY were $27,468 in the year before mCRPC diagnosis and increased over four fold to $124,379 after mCRPC diagnosis and continued to rise from start of 1L ($148,325) to 2L ($160,118) to 3L ($165,186) therapy.

CONCLUSION

HRU and healthcare costs increased substantially following mCRPC diagnosis, and continued to increase even further through progression from 1L through 3L mCRPC therapy. These findings help to quantify the economic burden of mCRPC and to contextualize the economic value of treatments that delay disease progression.

Real-world treatment patterns among men with metastatic castration-resistant prostate cancer (mCRPC) in the U.S. Medicare population

406

Background: Multiple life-prolonging therapies have been approved for mCRPC, for example novel hormonal therapies (NHT; abiraterone [abi], enzalutamide [enza]), docetaxel (doc), cabazitaxel, sipuleucel-T, and radium-223. This study describes real-world treatment patterns with life-prolonging therapies and sequencing among men with mCRPC in the US Medicare population. Methods: Men newly diagnosed with mCRPC were identified in Medicare fee-for-service claims during 1/1/2014–6/30/2019. Adult men were required to have a diagnosis of prostate cancer, metastasis diagnosis, castration-resistance using a published claims-based algorithm, and continuous insurance coverage for ≥1 year before and ≥6 months after index mCRPC diagnosis unless patients died. Treatment patterns of life-prolonging therapies after mCRPC diagnosis and sequencing were described. Results: Among 14,780 men with mCRPC, median age was 75 years, 10% used NHT in the year prior to mCRPC, and 3% had prior taxane therapy. Median follow-up after mCRPC diagnosis was 17 months. 22% of men received no life-prolonging therapy after mCRPC diagnosis, 78% received ≥1 line of therapy with life-prolonging treatment after mCRPC diagnosis, 42% had ≥2, and 20% had ≥3. The most common first-line (1L) therapies were abi (36%), enza (28%), and doc (16%). The most common second-line (2L) therapies were enza (33%), abi (28%), and doc (15%). The most common third-line (3L) therapies were doc (24%), enza (19%), and abi (17%). Median time from start of 1L to next line of therapy or end of follow-up was 13.7 months, 10.9 months from the start of 2L, and 8.9 months from the start of 3L. The most common 1L to 2L treatment sequences among men with ≥2 lines were NHT followed by a different NHT (33%), chemotherapy followed by NHT (14%), and NHT followed by chemotherapy (13%). There were 5,630 men with ≥2 lines of therapy and ≥1 NHT, of whom 53% had ≥2 NHTs. Conclusions: Substantial proportions of men with mCRPC did not receive a life-prolonging therapy or had only 1L therapy after mCRPC diagnosis, with a 50% fall-off rate after each line of therapy. NHTs were the most common 1L and 2L therapies, and NHT followed by a different NHT was the most common treatment sequence. Further research is needed to understand how treatment patterns change as NHTs and doc are used earlier in the disease continuum and new therapies are introduced and ultimately to identify optimal treatment sequencing.

Health care costs among men with metastatic castration-resistant prostate cancer (mCRPC) in the U.S. Medicare population

10

Background: Studies have reported healthcare costs before and after mCRPC diagnosis in commercially insured populations, but no information is available about costs with progression through lines of therapy after mCRPC diagnosis. This study describes healthcare costs among men with mCRPC in the US Medicare population before mCRPC diagnosis, after diagnosis, and with progression through lines of therapy. Methods: Men newly diagnosed with mCRPC were identified in Medicare fee-for-service claims during 1/1/2014–6/30/2019. Adult men were required to have a diagnosis of prostate cancer, metastasis diagnosis, castration-resistance using a published claims-based algorithm, and continuous insurance coverage for ≥1 year before and ≥6 months after index mCRPC diagnosis unless patients died. Unadjusted all-cause healthcare costs (medical and pharmacy) per patient per year (PPPY) to Medicare inflated to 2019 dollars were described for the periods before mCRPC diagnosis, after diagnosis, and from the start of first-line (1L), second-line (2L), and third-line (3L) therapy with mCRPC life-prolonging treatments to the start of subsequent therapy or end of follow up. Results: Among 14,780 men with mCRPC, median age was 75 years, and the mean Quan-Charlson Comorbidity Index was 2.1. Median follow-up after mCRPC diagnosis was 17 months. During the follow up, 3,252 men had no life-prolonging treatment, 11,528 men initiated 1L mCRPC therapy, 6,275 initiated 2L, and 2,945 initiated 3L. Mean all-cause healthcare costs PPPY were $27,468 in the year before mCRPC diagnosis, $124,379 after mCRPC diagnosis, $102,380 among men without life-prolonging treatment after mCRPC diagnosis, $148,325 from the start of 1L to subsequent therapy or end of follow up, $160,118 from the start of 2L therapy, and $165,186 from the start of 3L therapy. Conclusions: Mean healthcare costs increased over 4-fold from before to after mCRPC diagnosis and increased steadily as patients progressed from first through third lines of mCRPC therapy. These findings help quantify the economic burden of mCRPC and contextualize the economic value of treatments that delay disease progression.[Table: see text]

Overall survival by race and ethnicity among men with metastatic castration-resistant prostate cancer (mCRPC) in the U.S. Medicare population

144

Background: Previous studies reported mixed findings about racial and ethnic disparities in overall survival (OS) in mCRPC. This study describes OS by race and ethnicity among men with mCRPC in the US Medicare population. Methods: Men newly diagnosed with mCRPC were identified in Medicare fee-for-service claims during 1/1/2014–6/30/2019. Adult men were required to have a diagnosis of prostate cancer, metastasis diagnosis, castration-resistance using a published claims-based algorithm, and continuous insurance coverage for ≥1 year before and ≥6 months after index mCRPC diagnosis unless patients died. OS from mCRPC diagnosis and from start of first-line (1L) therapy for mCRPC for White (W), Black (B), Hispanic (H), and Asian (A) men were estimated using Kaplan-Meier analysis and Cox proportional hazards models adjusting for patient characteristics and 1L mCRPC therapy type or no treatment. Results: Among 14,780 men with mCRPC in this study, 75% were W, 14% were B, 6% were H, 3% were A, and 3% were of other or unknown race. Mean age at mCRPC diagnosis was 76 years among W men; B men had similar age while H and A men were slightly older than W men (Table). B, H, and A men had higher Quan-Charlson Comorbidity Index (CCI) than W men. Median follow-up after mCRPC diagnosis was 17 months. Similar proportions of W, H and A men (78%, 78%, and 79%, respectively) and lower proportion of B men (75%) initiated 1L life-prolonging therapy after mCRPC diagnosis. Among treated men, higher proportions of B, H, and A men (71%, 74%, and 73%, respectively) initiated 1L therapy with novel hormonal therapy than W men (64%). Median OS after mCRPC diagnosis was 26.0, 22.3, 22.9, and 24.2 months among W, B, H, and A men, respectively. Median OS after initiation of 1L mCRPC therapy was 23.8, 21.1, 19.9, and 24.1 months among W, B, H, and A men, respectively. After adjusting for patient characteristics and 1L treatment, OS was not different for B and H men relative to W men, while A men had lower risk of death. (Table). Conclusions: This study found no statistically significant differences in overall survival in mCRPC for B and H men and lower risk of death for A men relative to W men after adjusting for patient characteristics and treatment in the US Medicare population.[Table: see text]

The cost-effectiveness of dacomitinib in first-line treatment of advanced/metastatic epidermal growth factor receptor mutation-positive non-small-cell lung cancer (EGFRm NSCLC) in Sweden

AIMS

Although the benefit of first-line epidermal growth factor receptor (EGFR) tyrosine kinase inhibitors (TKIs) over chemotherapy in EGFR mutation-positive (EGFRm) non-small-cell lung cancer (NSCLC) has been demonstrated in clinical trials, the optimal treatment sequence remains unclear. The objective of our study was to evaluate the cost-effectiveness of dacomitinib in Sweden vs afatinib and osimertinib in first-line treatment of EGFRm NSCLC.

MATERIALS AND METHODS

A partitioned survival model was developed with three health states: progression-free, post-progression, and death. Progression-free and overall survival curves were used to inform movements between states. Clinical data were taken from randomized trials, compared via a network meta-analysis (NMA). Utility data were taken from published studies and costs from national Swedish sources. The model used a 15-year time horizon and a Swedish healthcare payer perspective. Sensitivity and scenario analyses were performed.

RESULTS

The base-case analysis showed that dacomitinib accrued a total of 2.10 quality-adjusted life-years (QALYs) at a total cost of Swedish krona (SEK) 874,615. The incremental cost-effectiveness ratio (ICER) for dacomitinib vs afatinib was SEK 461,556 per QALY gained. The ICER of osimertinib vs dacomitinib, where the small QALY gains of the former came at a high additional cost, was SEK 11,444,709. Deterministic and probabilistic sensitivity analyses confirmed the robustness of these results; changes to drug and medical resource use costs and overall survival had the greatest impact on ICER estimates.

LIMITATIONS

This model is subject to uncertainty associated with extrapolating long-term treatment effects from shorter trial follow-up periods, although this would also be a limitation when using direct comparison or time-dependent hazard ratios. The NMA was limited by the use of indirect comparison, although sensitivity analyses supported the robustness of our findings.

CONCLUSIONS

Our model demonstrated that dacomitinib is cost-effective for first-line EGFRm NSCLC treatment in Sweden vs afatinib and osimertinib.

Network meta analysis of first-line therapy for advanced EGFR mutation positive non-small-cell lung cancer: updated overall survival

Aim: To update overall survival (OS) results from a previous network meta analysis comparing the relative clinical efficacy of epidermal growth factor receptor-targeted tyrosine kinase inhibitors ( EGFR TKIs) for EGFR mutation positive ( EGFR+) advanced non-small-cell lung cancer (NSCLC). Materials & methods: A Bayesian network meta analysis was conducted using updated/mature randomized controlled trial OS results in response to first-line EGFR TKI therapies. Results: Dacomitinib showed a numerical improvement of OS relative to other EGFR TKIs: afatinib (hazard ratio [HR]: 0.87; 95% credible interval [CrI]: 0.61–1.24), erlotinib (HR: 0.79; 95% CrI: 0.44–1.42), gefitinib (HR: 0.75; 95% CrI: 0.59–0.95) and osimertinib (HR: 0.94; 95% CrI: 0.68–1.29). Conclusion: Dacomitinib should be considered as a first-line treatment option for patients diagnosed with advanced EGFR+ NSCLC.

Indirect analysis of first-line therapy for advanced non-small-cell lung cancer with activating mutations in a Japanese population

Background: Five EGFR-tyrosine kinase inhibitors (EGFR TKIs) are currently available in the first-line setting for non-small-cell lung cancer (NSCLC) in Japan. The aim here was to compare the relative efficacy of EGFR TKIs in the Japanese population. Materials & methods: A systematic review identified randomized controlled trials examining the efficacy of first-line EGFR TKIs. A Bayesian network meta-analysis was used to assess these EGFR TKI comparisons for progression-free survival (PFS). Results: A total of seven randomized controlled trials were identified and considered for network meta-analysis. Dacomitinib showed a trend toward improved PFS versus all comparators. Conclusion: Dacomitinib demonstrated a trend toward improved PFS and therefore, should be considered one of the standard first-line therapies for Japanese patients diagnosed with EGFR+ non-small-cell lung cancer.

Burden of illness among patients with severe aplastic anemia who have had insufficient response to immunosuppressive therapy: a multicenter retrospective chart review study

This study assessed treatment patterns and healthcare resource utilization (HRU) of patients with severe aplastic anemia (SAA) with insufficient response to immunosuppressive therapy (IST). A retrospective chart review was conducted at Dana-Farber Cancer Institute (DFCI), United States, and Hôpital Saint-Louis (HSL), France. Eligible patients were ≥ 18 years old, diagnosed with acquired SAA between January 1, 2006, and July 31, 2016, had insufficient response to IST, and had ≥ 12 months of follow-up post-diagnosis. Overall survival (OS) was estimated using the Kaplan-Meier method. Among the 40 patients, mean age at diagnosis was 44 years and 53% were women. Median follow-up time after SAA diagnosis was 48.3 months. Ninety-five percent of patients received antithymocyte globulin (ATG) as primary therapy prior to hematopoietic stem cell transplant (HSCT). Most common secondary SAA therapies prior to HSCT were eltrombopag (28%) and androgens (15%). Seventy-five percent of patients received HSCT. Prior to HSCT, patients received an average of 2.7 red blood cell (RBC) and 3.3 platelet transfusions per month; patients had 0.9 hospitalizations, 0.4 emergency room visits, and 12.8 office visits per year. Five-year OS was 75%, with infection as the primary cause of death. Additionally, this study provides information on the subgroup of patients receiving eltrombopag which was the most common secondary therapy. This study quantified transfusion and HRU burden associated with SAA and demonstrated high 5-year survival in a recently treated cohort.

Basal Insulin Initiation and Maintenance in Adults with Type 2 Diabetes Mellitus in the United States

OBJECTIVE

A survey of US adults with type 2 diabetes mellitus was conducted to better understand patients' insulin initiation experiences and treatment persistence behaviors.

RESEARCH DESIGN AND METHODS

Participants were recruited from consumer panels and grouped by basal insulin treatment pattern: continuers (no gap of ≥7 days within 6 months of initiation); interrupters (gap ≥7 days, resumed treatment); discontinuers (stopped for ≥7 days, not resumed). A quota of approximately 50 respondents per persistence category was set.

RESULTS

A total of 154 respondents (52 continuers, 52 interrupters, 50 discontinuers) completed the survey. Mean age was 51.4 years; 51.9% male. Continuers were more likely to report their views being considered during initiation, and less likely to report a sense of failure. Concerns included insulin dependence (64.3% agree/strongly agree), frequent blood glucose monitoring (55.2%), costs/ability to pay (53.9%), fears of or mistakes during self-injection (52.6%), and weight gain (52.6%). Continuers were motivated by benefits of insulin therapy; experienced or potential side effects were notable factors for interruption/discontinuation. Healthcare provider instruction was indicated as a reason for continuing, stopping, and restarting therapy.

CONCLUSION

Benefits of basal insulin therapy motivated continuers while side effects impacted interruption/discontinuation. Persistence on basal insulin is often influenced by provider actions. Earlier provider intervention upon signs of treatment discontinuation may promote persistence.

Basal insulin persistence in Brazilian participants with T2DM

OBJECTIVE

Real-world effectiveness of basal insulin therapy is affected by poor treatment persistence, often occurring soon after initiation. This analysis is part of an international cross-sectional study conducted in T2DM patients and is intended to describe the reasons behind non-persistence to insulin therapy in Brasil.

METHODS

Responders to an online survey in seven countries were classified as continuers (no gap of ≥7 days), interrupters (interrupted therapy for ≥7 days within first 6 months, then restarted), and discontinuers (terminated therapy for ≥7 days within first 6 months, and did not start it again before the survey). We present the results from the Brazilian cohort.

RESULTS

Of 942 global respondents, 156 were from Brasil, with a mean age of 34 years and a mean of 5.8 years since T2DM diagnosis. Reasons contributing to insulin continuation (n=50) were improved glycemic control (82%) and improved physical feeling (50%). Common reasons for interruption (n=51) or discontinuation (n=55) were, respectively, weight gain (47.1%, 43.6%), hypoglycemia (45.1%, 38.2%), and pain from injections (39.2%, 49.1%). However, not all patients who reported weight gain and hypoglycemia as a reason for interruption or discontinuation experienced these: 16/24 (66.7%) and 22/24 (91.7%) participants had weight gain, and 13/23 (56.5%) and 15/21 (71.4%) had hypoglycemia, respectively. The most important reason for possible re-initiation for interrupters and discontinuers, respectively, was persuasion by the physician/HCP (80.4%, 72.7%).

CONCLUSION

The benefits of basal insulin therapy motivated continuers to persist with the treatment; experienced or anticipated side effects contributed to interruption and discontinuation. Physician and patient training is key in the treatment of diabetes.

Real-world treatment patterns among patients with advanced gastric cancer in Russia: a chart review study

Objective: Little evidence is available on the management of patients with metastatic and/or unresectable gastric cancer (mGC) after the failure of first-line treatment. This study presents real-world data on characteristics and treatment patterns of patients with mGC in Russia.

Methods: Eligible patients were ≥18 years old, diagnosed with mGC ≥ January 1, 2012, received first-line chemotherapy followed by second-line chemotherapy or best supportive care (BSC), had ≥3 months of follow-up after the start of second-line chemotherapy or BSC (except in cases of death), and had not participated in a clinical trial. Data were summarized using descriptive statistics.

Results: A total of 88 physicians provided data from 202 charts. Mean age at mGC diagnosis was 53.7 (standard deviation: 11.2) years; 70.8% of patients were male. Reasons for first-line treatment discontinuation included disease progression (50.5%) and adverse events/toxicity (39.1%). There were 52 unique treatment regimens prescribed in second-line; capecitabine (14.5%), paclitaxel (9.3%), and capecitabine + oxaliplatin (8.7%) were the most frequent. Reasons for second-line treatment discontinuation included disease progression (39.8%) and patient refusal to continue (37.5%). During 2nd-line treatment, the most common treatment-related symptoms were nausea/vomiting (75.0%), while pain (73.8%) was the most common disease-related symptom. Antiemetics (63.4%), chemotherapy (61.6%), non-narcotic analgesics (48.3%), endoscopy (45.9%), and nutritional support (35.5%) were most frequently used as supportive care.

Conclusions: Second-line treatment patterns for patients with mGC in Russia are heterogeneous. Results of this study indicate the need for more intensive implementation of the most active regimens in second-line treatment of mGC according to international and national guidelines.

Clinical burden of asynchrony in patients with asthma when using metered-dose inhalers for control

Background: Asynchrony, or lack of coordination between inhalation and actuation when using a pressurized metered-dose inhaler (MDI), could theoretically impact the delivery of inhaled medications and treatment efficacy. Objective: To assess the real-world association between asynchrony and clinical outcomes among patients with asthma who receive controller therapy delivered by MDIs. Methods: A cohort of patients was assembled via electronic health records. The patients were aged ≥12 years, with one or more documentations of an asthma diagnosis, no diagnosis of chronic obstructive pulmonary disease, and two or more prescriptions for an inhalation aerosol corticosteroid alone or with long-acting beta-2-agonist delivered via MDI. Their inhaler technique, demonstrated by using a placebo MDI, was evaluated at a clinic visit by study nurses who used a standardized 10-step checklist. Asynchrony was defined as any gap in timing between inhalation and actuation. Clinical outcomes were assessed via electronic health records during the 6 months before the clinic visit and were compared between patients with and patients without asynchrony by using multivariable regression analyses adjusted for age, gender, asthma severity proxy, and baseline comorbidities. Results: Of the total 254 eligible patients, mean age of 49.3 years, 90 males (35.4%), 32 (12.6%) had asynchrony. Patients with asynchrony had higher odds of an asthma exacerbation (adjusted odds ratio, 2.99; p = 0.009), and lower odds of risk domain asthma control (adjusted odds ratio, 0.41; p = 0.04) compared with patients without asynchrony. Conclusion: This study provided real-world evidence that asynchrony in MDI use among patients with asthma who were treated with controller MDIs was associated with clinical burden in terms of asthma exacerbations and control.

Key factors for overcoming psychological insulin resistance: an examination of patient perspectives through content analysis

OBJECTIVE

To understand participant perceptions about insulin and identify key behaviors of healthcare professionals (HCPs) that motivated initially reluctant adults from seven countries (n=40) who had type 2 diabetes (T2D) to start insulin treatment.

RESEARCH DESIGN AND METHODS

Telephone interviews were conducted with a subset of participants from an international investigation of adults with T2D who were reluctant to start insulin (EMOTION). Questions related to: (a) participants' thoughts about insulin before and after initiation; (b) reasons behind responses on the survey that were either 'not helpful at all' or 'helped a lot'; (c) actions their HCP may have taken to help start insulin treatment; and (d) advice they would give to others in a similar situation of starting insulin. Responses were coded by two independent reviewers (kappa 0.992).

RESULTS

Starting insulin treatment was perceived as a negative experience that would be painful and would lead down a 'slippery slope' to complications. HCPs engaged in four primary behaviors that helped with insulin acceptance: (1) showed the insulin pen/needle and demonstrated the injection process; (2) explained how insulin could help with diabetes control and reduce risk of complications; (3) used collaborative communication style; and (4) offered support and willingness to answer questions so that participants would not be 'on their own'. Following initiation, most participants noted that insulin was not 'as bad as they thought' and recommended insulin to other adults with T2D.

CONCLUSIONS

Based on these themes, two actionable strategies are suggested for HCPs to help people with psychological insulin resistance: (1) demonstrate the injection process and discuss negative perceptions of insulin as well as potential benefits; (2) offer autonomy in a person-centred collaborative approach, but provide support and accessibility to address concerns. These findings help HCPs to better understand ways in which they can engage reluctant people with T2D with specific strategies.

Insulin non-persistence among people with type 2 diabetes: how to get your patients to stay on insulin therapy

Continuing use of medication is key to effective treatment and positive health outcomes, particularly in chronic conditions such as diabetes. However, in primary care, non-persistence (i.e. discontinuing or interrupting treatment) with insulin therapy is a common problem among patients with type 2 diabetes. To help primary care physicians manage patients who are non-persistent or likely not to be persistent, this review aimed to provide an overview of modifiable and non-modifiable factors associated with insulin non-persistence as well as practical strategies to address them. Data were extracted from published studies evaluating factors associated with non-persistence among patients with type 2 diabetes. A targeted literature review was performed using PubMed to identify recent studies (2000-2016) reporting measures of non-persistence with insulin therapy. Practical strategies to identify and prevent non-persistence were based on the authors' direct experience in primary care. Non-modifiable factors associated with non-persistence included gender, age, prior treatments, and cost of therapy. Before/at insulin initiation, modifiable factors included patients' perception of diabetes, preference for oral medication, and concerns/expectations about treatment complexity, inconvenience, or side effects. After initiation, modifiable factors included syringe use, difficulties during the first week of therapy, side effects, and insufficient glycemic control. Open-ended and patient-centered questions and a blame-free environment can help physicians identify, prevent, and reduce non-persistence behaviors. Possible questions to start a conversation with patients are provided. Effective physician-patient communication is essential to the management of diabetes. Primary care physicians should be familiar with the most common reasons for insulin non-persistence.

Value of transfusion independence in severe aplastic anemia from patients' perspectives - a discrete choice experiment

Background

Aplastic anemia is a rare, serious blood disorder due to bone marrow failure to produce blood cells. Transfusions are used to reduce risk of bleeding, infection and relieve anemia symptoms. In severe patients, transfusions may be required more than once/week. It is unclear from the patient perspective the impact that transfusions have on quality of life. This study aimed to elicit patient preferences for attributes associated with severe aplastic anemia (SAA) treatment, including transfusion independence.

Methods

An online discrete choice experiment (DCE) was conducted among patients with SAA who experienced insufficient response to immunosuppressive therapy and transfusion dependence for ≥3 months in the past 2 years. Recruitment occurred through the Aplastic Anemia and Myelodysplastic Syndromes International Foundation and referrals from clinical sites in the US and France. Respondents chose between hypothetical treatment pairs characterized by a common set of attributes: transfusions frequency, fatigue, risk of infection, and risk of serious bleeding. Conditional logit model with effects coding was used to estimate part-worth utilities for different attribute levels and the relative importance of each attribute. Predicted utility scores for transfusion frequency levels were reported.

Results

Thirty patients completed the survey. Most were age ≥ 40 years (73.3%), female (70.0%), and from the US (86.7%). 33.3% underwent bone marrow transplant; 36.7% received iron chelation therapy. Patients largely agreed that transfusion independence would result in less burden on time and costs, greater control and quality of life, less fatigue (86.7% noted each) and less scheduling around medical appointments (83.3%). The DCE found highest relative importance for risk of bleeding (0.30), followed by risk of infection (0.28), fatigue (0.23), and frequency of transfusions (0.20). More frequent transfusions resulted in lower utility, particularly when increasing monthly transfusions frequency from 4 (0.57) to 8 (0.35).

Conclusions

Our study showed that higher utility was associated with fewer transfusions in SAA patients with insufficient response to immunosuppressive therapy. While risk of bleeding, risk of infection, and fatigue were more important for patient treatment preferences, frequency of transfusions was also important.

Basal insulin initiation use and experience among people with type 2 diabetes mellitus with different patterns of persistence: results from a multi-national survey

BACKGROUND AND OBJECTIVE

People with type 2 diabetes mellitus (T2DM) often interrupt basal insulin treatment soon after initiation. This study aimed to describe the experiences during and after basal insulin initiation among people with T2DM with different persistence patterns.

METHODS

Adults with T2DM from France, Germany, Spain, UK, US, Brazil, and Japan were identified from consumer panels for an online survey. Respondents who initiated basal insulin 3-24 months prior to survey date were categorized as continuers (no gaps of ≥7 days in insulin treatment); interrupters (first gap ≥7 days within 6 months of initiation and restarted insulin); and discontinuers (stopped insulin for ≥7 days within 6 months of initiation without restarting).

RESULTS

Among 942 participants, continuers were older than interrupters and discontinuers (46, 37, and 38 years, respectively, p < .01). Continuers reported having fewer concerns before and after insulin initiation than interrupters and discontinuers, while interrupters had the most concerns. Continuers also reported fewer challenges during the first week of insulin use. Continuers were more likely to respond that insulin use had a positive impact on specific aspects of life than interrupters and discontinuers, for example on glycemic control (73.0%, 63.0%, and 61.8%, respectively; p < .01 vs. continuers).

CONCLUSION

Among people with T2DM with different persistence patterns after basal insulin initiation there were significant differences in patient characteristics and experience during and after insulin initiation. Interrupters and discontinuers more frequently reported having concerns and challenges during the initiation process, negative impacts after initiation, and less improvement in glycemic control than continuers.

Oncologist preferences in advanced soft tissue sarcoma: A discrete choice experiment

147

Background: Soft-tissue sarcomas (STS), a rare heterogeneous group of cancers originating in the muscle, fat, blood vessels or other fibrous/connective tissues, account for 1% of all cancers diagnosed annually in the US. The National Comprehensive Cancer Center guidelines include a wide range of regimens that lack supporting randomized trial evidence, and there are few treatment pathways to guide treatment decision making. A discrete choice experiment (DCE) was conducted to quantify the relative value of overall survival (OS), progression-free survival (PFS), tumor response rate (RR), risk of hospitalization due to side effects, and convenience of therapy (days per month to administer treatment) among oncologists. Methods: An online DCE survey was administered to oncologists recruited from an online panel, who were licensed to practice in the US at the time of the survey and prescribed chemotherapy to patients with STS. Oncologists were asked to choose between pairs of hypothetical treatments characterized by a common set of attributes: OS (14, 20 or 26 months), PFS (3, 5 or 7 months), RR (12, 18, or 26%), risk of hospitalization due to side effects (12, 30 or 46%), and days/month to administer treatment (1, 2 or 4 days). A hierarchical Bayes model was used to analyze preferences, the relative importance of treatment attributes (from 0-100%), and trade-offs between attributes. Results: 160 eligible oncologists completed the survey: 74% male; 41% private practice; and 64% affiliated with an academic teaching hospital. OS had the highest relative importance (44.6%, standard deviation, SD, 16.0%), followed by the risk of hospitalization (18.4%, SD 8.3%). PFS, RR, and days to administer treatment had lower relative importance (16.5%, 10.6%, and 9.9%, respectively). For a 1-month increase in OS, oncologists were willing to trade off 8.9 percentage points increase in hospitalization risk, a 2.1 month reduction in PFS, 13.1 percentage points decrease in RR, and an additional 4.6 days/month to administer treatment. Conclusions: Oncologists in the US value maximizing the life of patients with STS while avoiding hospitalizations. Patient preference interviews are ongoing with patients with STS, which will be presented at the meeting.

Basal Insulin Persistence, Associated Factors, and Outcomes After Treatment Initiation: A Retrospective Database Study Among People with Type 2 Diabetes Mellitus in Japan

INTRODUCTION

The objective of this study was to assess basal insulin persistence, associated factors, and economic outcomes for insulin-naïve people with type 2 diabetes mellitus (T2DM) in Japan.

METHODS

People aged at least 18 years with T2DM with first claim for basal insulin between May 2006 and April 2013 (index date), no insulin use before index date, and continuous insurance coverage for 6 months before (baseline) and 12 months after index date were selected from the Japan Medical Center Database. On the basis of whether there were at least 30-day gaps in basal insulin treatment, patients were classified as continuers (no gap), interrupters (at least one prescription after gap), and discontinuers (no prescription after gap). A multinomial logistic regression model identified factors associated with persistence. Annual healthcare resource use and costs in the year after initiation were compared between continuers and interrupters and between continuers and discontinuers using propensity score-based inverse probability weighting to adjust for baseline differences.

RESULTS

Of the 827 people included (mean age 50 years, ca. 71% male), 36% continued, 42% interrupted, and 22% discontinued basal insulin therapy in the year after initiation. Having at least one inpatient visit and using fewer classes of non-insulin antihyperglycemic medications during baseline were associated with lower likelihoods of continuing therapy. Relative to interrupters and discontinuers, continuers had lower hospitalization rates [continuers, 12.7%; interrupters, 25.4% (p < 0.001); discontinuers, 28.4% (p < 0.001)] and lower inpatient costs [continuers, ¥132,013; interrupters, ¥225,745 (p = 0.054); discontinuers, ¥320,582 (p = 0.036)], but higher pharmacy costs [continuers, ¥158,403; interrupters, ¥134,301 (p = 0.039); discontinuers, ¥121,593 (p = 0.002)] in the year after insulin initiation. Total healthcare costs were similar for the three cohorts.

CONCLUSIONS

Substantial proportions of people with T2DM in Japan interrupt or discontinue basal insulin within the year after initiation, and they have higher rates and costs of hospitalizations than patients who continue with their insulin therapy. Further research is needed to understand reasons behind basal insulin persistence and the implications thereof to help clinicians manage T2DM more effectively.

FUNDING

Eli Lilly and Company, Boehringer Ingelheim.

Basal insulin persistence, associated factors, and outcomes after treatment initiation among people with type 2 diabetes mellitus in the US

OBJECTIVE

To assess basal insulin persistence, associated factors, and economic outcomes for insulin-naïve people with type 2 diabetes mellitus (T2DM) in the US.

RESEARCH DESIGN AND METHODS

People aged ≥18 years diagnosed with T2DM initiating basal insulin between April 2006 and March 2012 (index date), no prior insulin use, and continuous insurance coverage for 6 months before (baseline) and 24 months after index date (follow-up period) were selected using de-identified administrative claims data in the US. Based on whether there were ≥30 day gaps in basal insulin use in the first year post-index, patients were classified as continuers (no gap), interrupters (≥1 prescription after gap), and discontinuers (no prescription after gap).

MAIN OUTCOME MEASURES

Factors associated with persistence - assessed using multinomial logistic regression model; annual healthcare resource use and costs during follow-up period - compared separately between continuers and interrupters, and continuers and discontinuers.

RESULTS

Of the 19,110 people included in the sample (mean age: 59 years, ∼60% male), 20% continued to use basal insulin, 62% had ≥1 interruption, and 18% discontinued therapy in the year after initiation. Older age, multiple antihyperglycemic drug use, and injectable antihyperglycemic use during baseline were associated with significantly higher likelihoods of continuing basal insulin. Relative to interrupters and discontinuers, continuers had fewer emergency department visits, shorter hospital stays, and lower medical costs (continuers: $10,890, interrupters: $13,674, discontinuers: $13,021), but higher pharmacy costs (continuers: $7449, interrupters: $5239, discontinuers: $4857) in the first year post-index (p < 0.05 for all comparisons). Total healthcare costs were similar across the three cohorts. Findings for the second year post-index were similar.

CONCLUSIONS

The majority of people in this study interrupted or discontinued basal insulin treatment in the year after initiation; and incurred higher medical resource use and costs than continuers. The findings are limited to the commercially insured population in the US. In addition, persistence patterns were assessed using administrative claims as opposed to actual medication-taking behavior and did not account for measures of glycemic control. Further research is needed to understand the reasons behind basal insulin persistence and the implications thereof, to help clinicians manage care for T2DM more effectively.

Budget impact model of a 5-grass sublingual immunotherapy tablet for the treatment of grass pollen-induced allergic rhinitis

OBJECTIVE

Allergic rhinitis (AR) is a chronic disease with a substantial clinical and economic burden. This study estimated the potential budget impact (BI) associated with market entry of Sweet Vernal, Orchard, Perennial Rye, Timothy, and Kentucky Blue Grass Mixed Pollens Allergen Extract Tablet for Sublingual Use ('5-grass SLIT tablet') for patients aged 10-65 with grass pollen-induced AR.

METHODS

A budget impact model was constructed to estimate the potential BI from a US payer perspective. The model calculated pharmacy, medical, and total (pharmacy + medical) costs per-member-per-month (PMPM) with and without market entry of the 5-grass SLIT tablet, considering a 3-year time horizon. The target population was determined using an epidemiological approach and existing literature. The treatment market shares without 5-grass SLIT tablet entry were derived from an analysis of de-identified insurance claims data. Pharmacy costs and medical utilization rates and costs were obtained from the claims data analysis and existing literature. One-way sensitivities were conducted for key model inputs.

RESULTS

Using an illustrative example of a hypothetical health plan with one million members, the estimated target population of AR patients aged 10-65 was 26,320. On a PMPM basis, pharmacy costs increased by $0.36, $0.44, and $0.51, while total costs (after medical cost offsets) increased by $0.15, $0.18, and $0.22 in the first, second, and third years following entry of the 5-grass SLIT tablet, respectively. Results were most sensitive to changes in the compliance rate, treatment duration, and price. The BI will vary from the base case example when alternative, payer-specific inputs are used.

CONCLUSIONS

Using base case inputs, use of the 5-grass SLIT tablet to treat grass pollen-induced AR increased the pharmacy budget for a hypothetical third-party payer. Higher pharmacy costs were partially offset by lower medical budget due to reduced resource use compared with existing treatments.

Healthcare resource use and costs of privately insured patients who switch, discontinue, or persist on anti-muscarinic therapy for overactive bladder

OBJECTIVES

To compare the healthcare costs of patients with overactive bladder (OAB) who switch vs persist on anti-muscarinic agents (AMs), describe resource use and costs among OAB patients who discontinue AMs, and assess factors associated with persisting vs switching or discontinuing.

METHODS

OAB patients initiating an AM between January 1, 2007 and March 31, 2012 were identified from a claims database of US privately insured beneficiaries (n ≈ 16 million) and required to have no AM claims in the 12 months before AM initiation (baseline period). Patients were classified as persisters, switchers, or discontinuers, and assigned a study index date based on their AM use in the 6 months following initiation. Baseline characteristics, resource use, and costs were compared between persisters and the other groups. Resource use and costs in the 1 month before and 6 months after the study index date (for switchers, the date of index AM switching; for persisters, a randomly assigned date to reflect the distribution of the time from AM initiation to switching among switchers) were also compared between persisters and switchers in unadjusted and adjusted analyses. Factors associated with persisting vs switching or discontinuing were assessed.

RESULTS

After controlling for baseline characteristics and costs, persisters vs switchers had significantly lower all-cause and OAB-related costs in both the month before (all-cause $1222 vs $1759, OAB-related $142 vs $170) and 6 months after the study index date (all-cause $7017 vs $8806, OAB-related $642 vs $797). Factors associated with switching or discontinuing vs persisting included index AM, younger age, and history of UTI.

CONCLUSION

A large proportion of OAB patients discontinue or switch AMs shortly after initiation, and switching is associated with higher costs.

Duloxetine use in employees with low back pain: treatment patterns and direct and indirect costs

OBJECTIVE

The study aims to examine real-world effects of duloxetine treatment for low back pain (LBP).

METHODS

The study identified employees with ≥1 LBP diagnosis and ≥1 duloxetine prescription within a year after LBP diagnosis from a privately insured claims database (2004-2007). Duloxetine-treated employees were propensity score matched to employees initiating another pharmacological/noninvasive treatment in the same month from LBP diagnosis. Treatment patterns and costs were compared over the 6 months following treatment initiation.

RESULTS

Relative to controls, duloxetine-treated employees (N = 753) had significantly lower rates of other pharmacological/noninvasive therapies and a similar LBP surgery rate (1.7% vs 2.8%, P = 0.1573). Duloxetine-treated employees, despite higher pharmacy costs, had similar direct (health care) costs ($4,935 vs $5,649, P = 0.2662), and significantly lower indirect (workloss) costs ($1,723 vs $2,198, P = 0.0036).

CONCLUSIONS

Duloxetine treatment in LBP employees was associated with reduced rates of many nonsurgical therapies and lower indirect costs. The findings are limited by the observational study design and unmeasured potential confounders.

The prevalence and economic impact of prescription opioid-related side effects among patients with chronic noncancer pain

OBJECTIVES

To estimate the prevalence of opioid-related side effects among patients with chronic noncancer pain (CNCP) who initiated opioids and compare healthcare costs of patients with and without side effects using patient survey, medical charts, and claims data.

PATIENTS, PARTICIPANTS

Patients initiating opioids, who were aged ≥18 years, had ≥1 pain diagnosis, and did not have cancer, were identified through claims data and medical records from a Central Massachusetts medical group practice and mailed surveys between October 2010 and July 2012.

MAIN OUTCOMES MEASURES

Prevalence of opioid-related side effects was estimated from patient surveys, charts, and claims data within 90 days after opioid initiation (study period). Study period healthcare costs were compared between patients with and without side effects (self-reported problematic side effects or side effects recorded in medical charts or claims).

RESULTS

Among patients with CNCP who initiated opioids and completed the survey (N = 167), the average age was 53 years, and 62.9 percent were women. Based on the survey, charts, and claims, 91.6 percent, 15.0 percent, and 19.2 percent of patients, respectively, had ≥1 opioid-related side effect. Overall, 59.3 percent of patients reported having ≥1 problematic side effect or side effect recorded in charts or claims. In the analysis that controlled for baseline characteristics and resource use, patients with versus without side effects had higher mean study period healthcare costs ($3,347 vs $2,521, p = 0.049).

CONCLUSIONS

Prevalence of opioid-related side effects among patients with CNCP who initiated opioids was substantially higher based on patient survey than from charts or claims. Opioid-related side effects were associated with significantly higher healthcare costs.

Duloxetine use in chronic low back pain: treatment patterns and costs

BACKGROUND

Little is known about the real-world treatment patterns and costs of patients with chronic low back pain (CLBP) who are treated with duloxetine compared with those receiving other non-surgical treatments.

OBJECTIVE

Our objective was to compare the real-world treatment patterns and costs between patients with CLBP who initiated duloxetine and matched controls who initiated another non-surgical treatment.

METHODS

The study sample was selected from a US privately insured claims database (2004-8). Selected patients were aged 18-64 years, and had a low back pain (LBP) diagnosis (per Healthcare Effectiveness Data and Information Set [HEDIS] specifications) with a subsequent CLBP-qualifying diagnosis recorded ≥90 days after the initial LBP diagnosis. Duloxetine-treated patients had ≥1 duloxetine prescription within 6 months after CLBP diagnosis, no prior duloxetine claim, and continuous eligibility ≥12 months before first LBP diagnosis and ≥6 months after index duloxetine prescription (study period). Because duloxetine patients had higher rates of co-morbidities, 553 duloxetine-treated patients were matched to 553 control patients who initiated another non-surgical LBP treatment based on propensity score and time from first LBP diagnosis to treatment initiation. A subset (n = 103 each) of matched employees with disability data was also analysed to assess work loss. Main outcomes measures included study period treatment rates and direct (medical and drug) costs from a third-party payer perspective and employee indirect (work-loss) costs. McNemar tests were used to compare LBP treatment rates. Bias-corrected bootstrapping t-tests were used to compare costs.

RESULTS

After matching, the two groups had balanced baseline characteristics including demographics, LBP diagnostic categories, co-morbidity profiles, resource use, treatment patterns and mean direct costs. During the 6-month study period, matched duloxetine-treated patients had significantly lower rates of other pharmacological therapy (e.g. 56.2% vs 64.9% narcotic opioids, p = 0.0024; 34.9% vs 49.5% NSAIDs, p < 0.0001) and non-invasive therapy (28.8% vs 38.5% chiropractic therapy, p = 0.0007; 25.5% vs 35.4% physical therapy, p = 0.0004; 17.5% vs 28.4% exercise therapy, p < 0.0001) than controls. Duloxetine-treated patients versus controls had similar back surgery rates (2.2% vs 3.8%; p = 0.1127) and similar direct costs ($US7658 vs $US7439; p = 0.8119). Among CLBP employees, duloxetine-treated employees versus controls had lower rates of other non-surgical therapy, similar back surgery rates (0.0% vs 3.9%; p = 0.1250), lower total direct and indirect costs ($US5227 vs $US7299; p = 0.0418), and similar indirect costs ($US1806 vs $US2664; p = 0.0528).

CONCLUSIONS

Duloxetine treatment in CLBP patients/employees versus other non-surgical treatment was associated with reduced rates of non-surgical therapies and similar back surgery rates, without increased costs.

Effect of asthma exacerbations on health care costs among asthmatic patients with moderate and severe persistent asthma

BACKGROUND

Health care costs increase in patients with more severe asthma, but the effect of asthma exacerbations on costs among patients with more severe asthma has not been quantified.

OBJECTIVE

This study compared direct health care costs between patients with moderate/severe persistent asthma with and without exacerbations.

METHODS

Patients who had an asthma diagnosis (International Classification of Diseases-ninth revision-Clinical Modification code 493.x), were 12 to 64 years old, and were receiving controller therapy were identified from a large administrative claims database. Patients were categorized as having moderate/severe persistent asthma and were further evaluated for exacerbations during a 12-month exacerbation identification period. Patients with 1 or more exacerbations (asthma-related inpatient or emergency department visit or corticosteroid prescription) were matched to patients without exacerbations on demographic characteristics and asthma severity. Total and asthma-related health care costs during the 1-year study period after the exacerbation index date were calculated.

RESULTS

Patients with exacerbations had significantly higher total health care costs ($9223 vs $5011, P < .0001) and asthma-related costs ($1740 vs $847, P < .0001). The cost differences remained significant after controlling for patient differences by using multivariate models. Patients with exacerbations (n = 3830) had higher rates of sinusitis, allergy-related diagnoses or medications, pneumonia, and mental disorders and higher average Charlson Comorbidity Index scores at baseline. Patients with exacerbations filled their prescriptions for controllers more often and had higher asthma-related drug costs.

CONCLUSIONS

Patients with moderate/severe persistent asthma who had exacerbations had higher total and asthma-related health care costs than those without exacerbations. Moreover, controller medication use was higher in patients with exacerbations.

Impact of medication adherence to disease-modifying drugs on severe relapse, and direct and indirect costs among employees with multiple sclerosis in the US

OBJECTIVE

To compare rates of severe relapse and total direct and indirect costs over a 2-year period between US-based employees with multiple sclerosis (MS) who were adherent and non-adherent to disease-modifying drugs (DMDs).

METHODS

Employees with ≥1 MS diagnosis (ICD-9-CM: 340.x) and ≥1 DMD pharmacy claim between 1/1/2002-12/31/2007 were identified from a large US administrative claims database. Patients had continuous coverage ≥6 months before (baseline) and ≥24 months after (study period) their index date (first DMD claim). Adherence was measured using medication possession ratio (MPR) over the study period. Patients with MPR ≥80% were considered adherent (n = 448) and those with MPR <80% as non-adherent (n = 200). Multivariate analyses were used to compare rates of severe relapse (inpatient or Emergency Department visit with MS diagnosis) and costs in 2007 dollars between DMD adherent and non-adherent patients. Direct costs were calculated as reimbursements to providers for medical services and prescription drugs excluding DMDs. Indirect costs included disability and medically-related absenteeism costs.

RESULTS

DMD adherent patients were on average older (43.5 vs 41.8 years, p = 0.015) and more likely to be male (38.6% vs 26.0%, p = 0.002) compared with non-adherent patients. Adherent patients had lower rates of depression, higher rates of previous DMD use, and higher baseline MS-related costs. After adjusting for differences in baseline characteristics, DMD adherent patients had a lower rate of severe relapse (12.4% vs 19.9%, p = 0.013) and lower total (direct and indirect) costs ($14,095 vs $16,638, p = 0.048) over the 2-year study period.

CONCLUSIONS

In this study, DMD adherence was associated with a significantly lower rate of severe relapse and lower total costs over 2 years. Causality cannot be inferred because adherence and outcomes were measured over the same period. The study was subject to limitations associated with use of claims data and the absence of clinical measures.

Excess costs associated with patients with chronic thromboembolic pulmonary hypertension in a US privately insured population

BACKGROUND

Chronic thromboembolic pulmonary hypertension (CTEPH) is a rare and potentially fatal disease. Little is known about the economic burden associated with CTEPH patients in the US.

OBJECTIVES

The objective of this study was to estimate excess direct costs associated with privately insured patients with CTEPH in the US.

METHODS

From a privately insured claims database (>8 million beneficiaries, 2002-7), 289 CTEPH patients were identified using the criteria: two or more claims for pulmonary hypertension (PH), International Classification of Diseases, ninth edition, clinical modification (ICD-9-CM) code 416.0 or 416.8; one or more claim for pulmonary embolism (ICD-9-CM: 415.1, V12.51; ICD-9 procedure: 38.7; Current Procedural Terminology [CPT]-4 code: 36010, 37620, 75825, 75940; Healthcare Common Procedure Coding System [HCPCS] code: C1880) within 12 months prior or 1 month after the initial PH claim (index date); one or more claim for right heart catheterization (RHC) within 6 months prior to any PH claim or one or more claim for echocardiogram within 6 months prior to a specialist-diagnosed PH claim; aged 18-64 years. Patients with CTEPH were matched demographically to controls without PH. Patients were followed as long as continuously eligible; mean follow-up in CTEPH patients was 21.5 months. Chi-squared tests were used to compare baseline co-morbidities. Wilcoxon rank-sum tests were used to compare direct (medical and pharmaceutical) patient-month costs to insurers.

RESULTS

The average age for CTEPH patients was 52.2 years, and 57.1% were women. Compared with controls, CTEPH patients had significantly higher baseline rates of co-morbidities (e.g. essential hypertension, congestive heart failure and chronic pulmonary disease) and a higher mean Charlson Co-morbidity Index score. Mean direct patient-month costs (year 2007 values) were $US4782 for CTEPH patients and $US511 for controls (p < 0.0001). Sensitivity analysis restricting the sample to patients diagnosed following RHC yielded a 15% increase in excess costs relative to the original sample. Regarding cost drivers, inpatient services accounted for 54%, outpatient and other services for 33% and prescription drugs for 11% of total direct healthcare costs per patient-month in CTEPH patients. Circulatory-/respiratory-related patient-month costs were $US2496 among CTEPH patients and $US128 among controls (p < 0.0001).

CONCLUSIONS

CTEPH patients had substantially higher costs and co-morbidity than matched controls, with circulatory-/respiratory-related costs accounting for 55% of excess costs. The high burden of illness suggests opportunities for savings from improved management.

Excess costs associated with patients with pulmonary arterial hypertension in a US privately insured population

BACKGROUND

Pulmonary arterial hypertension (PAH) is a rare but fatal disease. Little is known about the economic burden associated with PAH patients in the US.

OBJECTIVES

The objective of this study was to estimate excess direct costs associated with privately insured PAH patients in the US.

METHODS

From a privately insured claims database (>8 million beneficiaries, 2002-7), 471 patients with PAH were identified using the criteria: two or more claims for primary pulmonary hypertension (PH), International Classification of Diseases, ninth edition, clinical modification (ICD-9-CM) code 416.0; no left heart disease, lung diseases, chronic thromboembolic PH or miscellaneous PH diagnoses within 12 months prior or 1 month after the initial PH claim (index date); one or more claim for right heart catheterization (RHC) within 6 months prior to any PH claim or one or more claim for echocardiogram within 6 months prior to a specialist-diagnosed PH claim; aged 18-64 years. Patients with PAH were matched demographically to controls without PH. Patients were followed as long as continuously eligible; mean follow-up of PAH patients was 24.8 months. Chi-squared tests were used to compare baseline co-morbidities. Wilcoxon rank-sum tests were used to compare direct (medical and pharmaceutical) patient-month costs to insurers.

RESULTS

The average age for PAH patients was 52.2 years, and 55.8% were women. Compared with controls, PAH patients had significantly higher baseline rates of co-morbidities (e.g. essential hypertension, diabetes mellitus and congestive heart failure) and a higher mean Charlson Co-morbidity Index score. Mean direct patient-month costs (year 2007 values) were $US2023 for PAH patients and $US498 for controls (p < 0.0001), yielding excess costs of $US1525. Sensitivity analysis restricting the sample to patients diagnosed following RHC yielded a 64% increase in excess costs relative to the original sample. Regarding cost drivers, inpatient services accounted for 45%, outpatient and other services for 38% and prescription drugs for 15% of total direct healthcare costs per patient-month in PAH patients. Circulatory/respiratory system-related patient-month costs were $US724 among PAH patients and $US114 among controls (p < 0.0001).

CONCLUSIONS

Patients with PAH had substantially higher costs and co-morbidity than controls, with circulatory/respiratory system-related costs accounting for 40% of excess costs. The high burden of illness suggests opportunities for savings from improved management.

Prevalence of pulmonary arterial hypertension and chronic thromboembolic pulmonary hypertension in the United States

BACKGROUND

The prevalence of pulmonary arterial hypertension (PAH) and chronic thromboembolic pulmonary hypertension (CTEPH) in the US is largely unknown. Prior research has estimated PAH prevalence in Europe at ∼15-52 per million.

METHODS

Using a privately insured claims database (1999-2007) for the under age 65 population and a Medicare claims database for the 65+ population, and following the current clinical classification of PH, CTEPH patients were identified as having: ≥2 claims for pulmonary hypertension (PH) [ICD-9-CM: 416.0, 416.8]; ≥1 claim for pulmonary embolism (PE) ≤12 months prior or 1 month after the initial PH claim (index date). PAH patients were identified: ≥2 claims for primary PH [416.0]; no left heart disease, lung diseases, CTEPH, or miscellaneous PH diagnoses ≤12 months prior or 1 month after the index date. Both cohorts were required to have ≥1 claim for right heart catheterization ≤6 months prior to any PH claim, or ≥1 claim for echocardiogram ≤6 months prior to a specialist-diagnosed PH claim. Age- and gender-standardized prevalence rates per million individuals (PMI) were calculated using appropriate population weights.

RESULTS

Prevalence rates (95% CI) of CTEPH were estimated at 63 (34-91) PMI among the privately insured (<65), and 1007 (904-1111) PMI among the Medicare population (≥65). The corresponding estimates for PAH were 109 (71-146) PMI among the <65 population, and 451 (384-519) PMI for Medicare.

LIMITATIONS

Identification of PAH and CTEPH patients in administrative claims data is challenging, due to lack of specific ICD-9-CM codes for the conditions and risk of misdiagnosis.

CONCLUSIONS

Prevalence rates of CTEPH and PAH increase with age, and are higher among women. The increased risk of PE may explain the sharp age gradient for CTEPH prevalence. The estimated US prevalence of PAH is higher than existing estimates.

Cost of post-traumatic stress disorder vs major depressive disorder among patients covered by medicaid or private insurance

OBJECTIVE

To compare healthcare costs and resource utilization among patients with post-traumatic stress disorder (PTSD) vs control subjects with major depressive disorder (MDD) in populations covered by Medicaid or private insurance.

STUDY DESIGN

Retrospective analysis of Medicaid and private insurance administrative claims data.

METHODS

Patients with at least 2 PTSD diagnoses during or after 1999, and at least 1 PTSD diagnosis during or after 2003, were identified from deidentified Medicaid claims from Florida, Missouri, and New Jersey (1999-2007) and from a privately insured claims database (1999-2008). Patients had continuous eligibility 6 months before (baseline) and 12 months after (study period) the index date and were aged 18 to 64 years. Potential control subjects having MDD without PTSD diagnosis were identified using similar selection criteria. Control subjects with MDD were matched to patients with PTSD on age, sex, state or region, employment status (private insurance only), index year, and race/ethnicity (Medicaid only). Study period per-patient utilization and costs, calculated as reimbursements to providers for medical services and prescription drugs, were compared using univariate and multivariate analyses.

RESULTS

Patients with PTSD had higher rates of other mental health disorders (eg, anxiety and bipolar disorder) and higher mental health-related resource use and costs than control subjects with MDD in both Medicaid and privately insured populations. The mean study period total direct healthcare costs were higher for patients with PTSD than for control subjects with MDD ($18,753 vs $17,990 for Medicaid and $10,960 vs $10,024 for private insurance, P <.05 for both). The difference in total direct costs was driven by higher mental health-related resource use for patients with PTSD.

CONCLUSION

Patients having PTSD had 4.2% to 9.3% higher mean annual per-patient healthcare costs compared with matched control subjects having MDD among patients covered by Medicaid or private insurance.

Real-world practice patterns, health-care utilization, and costs in patients with low back pain: the long road to guideline-concordant care

BACKGROUND CONTEXT

Treatment guidelines suggest that most acute low back pain (LBP) episodes substantially improve within a few weeks and that immediate use of imaging and aggressive therapies should be avoided.

PURPOSE

Assess the actual practice patterns of imaging, noninvasive therapy, medication use, and surgery in patients with LBP, and compare their costs to those of matched controls without LBP.

STUDY DESIGN

A retrospective analysis of claims data from 40 self-insured employers in the United States.

PATIENT SAMPLE

The study sample included 211,551 patients, aged 18 to 64 years, with one LBP diagnosis or more (per Healthcare Effectiveness Data and Information Set specification) during 2004 to 2006, identified from a claims database. Patients had continuous eligibility for 12 months or more after their index LBP diagnosis (study period), for 6 months or more before their index diagnosis (baseline period), and no other LBP diagnosis during the baseline period. Patients with LBP were matched to a random cohort of patients without LBP by age, gender, employment status, and index year.

OUTCOMES MEASURES

Physiological measures (eg, imaging and diagnostic tests), functional measures (eg, pharmacologic and nonpharmacologic treatment for LBP, health-care resource use), and direct (medical and prescription drug) and indirect (disability and medically related absenteeism) costs were assessed within the year after the LBP diagnosis.

METHODS

Univariate analyses described treatment patterns and compared baseline characteristics and study period costs.

RESULTS

Patients with LBP had significantly higher rates of baseline comorbidities and resource use compared with controls. Of patients with LBP, 41.6% had imaging mean (median) [standard deviation] 34.3 (0) [78.6] days after the LBP diagnosis. Most patients with LBP (69.4%) used medications starting 51.9 (8) [86.2] days after the diagnosis. Opioids were commonly prescribed early (41.6% of patients; after 82.8 (25) [105.9] days). Of patients with LBP, 2.05% had surgery during the study period. Patients with LBP were likely to have chiropractic treatment first, followed by pharmacotherapy with muscle relaxants and nonsteroidal anti-inflammatory drugs. Except for less surgery, these findings also held for patients with only nonspecific LBP. Patients with LBP had higher mean direct costs compared with controls ($7,211 vs. $2,382, respectively; p<.0001), with surgery patients having mean direct costs of $33,931.

CONCLUSIONS

Contrary to clinical guidelines, many patients with LBP start incurring significant resource use and associated expenses soon after the index diagnosis. Achieving guideline-concordant care will require substantial changes in LBP practice patterns.

Direct and indirect costs of employees with treatment-resistant and non-treatment-resistant major depressive disorder

OBJECTIVE

Treatment-resistant depression (TRD) imposes substantial cost from the perspective of employers. The objective of this study was to assess direct healthcare costs and indirect (disability and medical-related absenteeism) costs associated with TRD compared with non-treatment-resistant major depressive disorder (MDD).

METHODS

Employees with one or more inpatient, or two or more outpatient/other MDD diagnoses (ICD-9-CM: 296.2x, 296.3x) from 2004 through 2007, ages 18-63 years, were selected from a claims database. Employees who initiated a third antidepressant following two antidepressant treatments of adequate dose and duration or who met published TRD criteria were classified as TRD likely (N = 2312). The index date was the date of the first antidepressant, starting 1/1/2004. The control group was an age- and sex-matched cohort of employees with MDD but without TRD. All had continuous eligibility during the 6-month pre-index (baseline) and 24-month post-index (study) period. McNemar tests were used to compare baseline comorbidities. Wilcoxon signed-rank tests were used to compare costs from employer perspective.

RESULTS

TRD-likely employees were on average 48 years old, and 64.8% were women. Compared with MDD controls, TRD-likely employees had significantly higher rates of mental-health disorders, chronic pain, fibromyalgia, and higher Charlson Comorbidity Index. Average direct 2-year costs were significantly higher for TRD-likely employees ($22,784) compared with MDD controls ($11,733), p < 0.0001. Average indirect costs were also higher among TRD-likely employees ($12,765) compared with MDD controls ($6885), p < 0.0001.

LIMITATIONS

Limitations of claims data related to accuracy of diagnosis coding and lack of clinical information apply to this study.

CONCLUSIONS

Based on comorbidities and healthcare resources used, patients with TRD appeared to represent a clinically complex subgroup of individuals with MDD. TRD was associated with significant cost burden.

Descriptive analysis of Medicaid patients with postherpetic neuralgia treated with lidocaine patch 5%

OBJECTIVES

To compare demographic and comorbidity profiles and healthcare costs of Medicaid patients with postherpetic neuralgia (PHN) treated with lidocaine patch 5% (lidocaine patch) versus patients not treated with the lidocaine patch. Repeat comparison for the subset of patients treated in long-term care (LTC) settings.

METHODS

Patients, age≥18 years, with PHN diagnosis, or PHN-likely patients with herpes zoster diagnosis and ≥30 days of PHN-recommended treatment, were identified in Medicaid claims from Florida, Iowa, Missouri, and New Jersey (1999-2007). Patients had continuous eligibility 6 months before (baseline) and 12 months after (study period) the PHN index date. Patients with ≥1 claim for a lidocaine patch during the study period (n=872) were compared to patients without a lidocaine patch claim (comparison group). Baseline characteristics, study period treatment and healthcare costs (reimbursements by Medicaid for medical services and prescription drugs) were compared between groups using univariate analyses.

RESULTS

PHN patients in the lidocaine patch group were older (64.5 vs. 62.2 years; p=0.002) and had higher rates of pain-related comorbidities (e.g., back/neck pain, osteoarthritis) than comparison patients. Average PHN-related drug costs per patient were higher ($1994 vs. 1137; p<0.0001) among lidocaine patch patients, with lidocaine patch accounting for $505 of the difference. PHN-related medical costs appeared lower in the lidocaine patch group, although not statistically significant ($983 vs. 1294; p=0.1348). No significant differences were found in total healthcare costs ($20,175 vs. 19,124; p=0.3720) across groups, despite higher total prescription drug costs among lidocaine patch patients. A similar pattern was observed among LTC patients.

CONCLUSIONS

Despite higher rates of comorbidities and prescription drug costs, lidocaine patch patients had similar study period healthcare costs as comparison patients. The cost of the lidocaine patch represented a small fraction of overall costs incurred over the study period.

LIMITATIONS

Findings are based on a Medicaid sample and may not be generalizable to all PHN patients.

Comparing healthcare costs of Medicaid patients with postherpetic neuralgia (PHN) treated with lidocaine patch 5% versus gabapentin or pregabalin

OBJECTIVE

To compare healthcare resource utilization and costs of postherpetic neuralgia (PHN) patients initiating lidocaine patch 5% (lidocaine patch) or oral gabapentin/pregabalin.

METHODS

Patients with PHN diagnosis, or herpes zoster diagnosis and ≥30 days PHN-recommended treatment were selected from de-identified Medicaid claims data from Florida, Iowa, Missouri, and New Jersey, 1999-2007. Patients initiated monotherapy with lidocaine patch or gabapentin/pregabalin after PHN diagnosis, had continuous eligibility 6 months before (baseline) and 6 months after (study period) medication index date, and were ≥18 years old. Lidocaine patch patients were matched to gabapentin/pregabalin patients based on their propensity to initiate treatment. Study period resource utilization and costs from a Medicaid perspective were compared between treatment groups using univariate analysis.

RESULTS

Matched patients were on average 61.3 years old, approximately 73% were women, and 55% had other painful conditions during the baseline period. 6-month per patient PHN-related prescription drug costs were similar for matched lidocaine patch (n=312) and gabapentin/pregabalin (n=312) patients ($854 vs. 820, p=0.75), while PHN-related medical costs appeared lower in the lidocaine patch group ($145 vs. 353, p=0.12). Furthermore, there were no statistically significant differences between treatment groups during the observation period in overall resource utilization, total prescription drug costs, and total medical costs per patient.

CONCLUSIONS

In spite of higher list prices, PHN patients treated with lidocaine patch cost no more than patients treated with gabapentin or pregabalin in terms of overall healthcare costs over the 6-month study period. The study suggests that PHN-related medical costs may be lower among lidocaine patch patients.

LIMITATIONS

Findings are based on a Medicaid sample and may not be generalizable to all PHN patients.

Economic burden of epilepsy among the privately insured in the US

BACKGROUND

The direct cost burden of epilepsy in the US from a third-party payer perspective has not been evaluated. Furthermore, no study has quantified the indirect (work-loss) cost burden of epilepsy from an employer perspective in the US.

OBJECTIVE

To assess the annual direct costs for privately insured US patients diagnosed with epilepsy, and indirect costs for a subset of employees from an employer perspective.

METHODS

A retrospective analysis of a claims database for the privately insured, including employee disability claims from 1999 through 2005 and comprising 17 US companies, was conducted. A total of 4323 patients aged 16-64 years (including 1886 employees) with at least one epilepsy diagnosis (International Classification of Diseases, 9th edition, Clinical Modification [ICD-9-CM] code 345.x) over the period 1999-2004 were included. The control group was a demographically matched cohort of randomly chosen beneficiaries without an epilepsy diagnosis. All had continuous health coverage during 2004 (baseline) and 2005 (study period). Main outcome measures included annual direct (medical and pharmaceutical) costs and, for employees, indirect (disability and medically related absenteeism) and total costs for the study period. Wilcoxon rank-sum tests were used for univariate comparisons of annual direct costs, indirect costs (costs for the subset of employees with these data), and total (direct and indirect) costs during the study period. Two-part multivariate models that adjusted for patient characteristics were also used to compare costs between the study and control groups.

RESULTS

Patients with epilepsy were an average age of 43 years and 57% were female. They had more co-morbidities than controls. On average, direct annual costs were significantly higher per patient with epilepsy than per control ($US10 258 vs $US3862, respectively; p < 0.0001) [year 2005 values], with an annual per-patient difference of $US6396. Epilepsy-related costs ($US2057) accounted for 20% of direct costs for patients with epilepsy. Annual indirect costs were significantly higher for employees with epilepsy than for employed controls ($US3192 vs $US1242, respectively; p < 0.0001), with a difference of $US1950. Total direct plus indirect costs for employees with epilepsy were also higher than those for employed controls ($US13 595 vs $US5338, respectively; p < 0.0001), with a difference of $US8257.

CONCLUSIONS

Epilepsy was associated with significant economic burden. The excess direct costs in patients with epilepsy are underestimated when only epilepsy-related costs are considered.

Economic impact of multiple sclerosis disease-modifying drugs in an employed population: direct and indirect costs

OBJECTIVE

The study objective is to compare the annual total medical and indirect costs of newly treated and untreated employees with multiple sclerosis (MS).

RESEARCH DESIGN AND METHODS

A retrospective database analysis of employer medical, drug, and disability claims database (Ingenix Employer database, 1999-2005; 17 large US companies) was conducted for employees 18-64 years of age with > or =1 MS diagnosis after January 1, 2002. Employees with > or =1 MS disease-modifying drug (DMD) claim comprised the newly treated group; employees with MS but no DMD at any time comprised the untreated, comparison group. Index date was the day after the most recent claim (treated, DMD claim; untreated, MS claim) meeting the following requirements: continuous health coverage for 3 months before (baseline period) and 12 months after the index date (study period) and actively employed during baseline.

MAIN OUTCOME MEASURES

Total medical costs and indirect (work loss) costs over the 1-year study period (2006 $US) were compared for DMD-treated and untreated MS employees, adjusting for baseline characteristics, including comorbidities.

RESULTS

During the baseline, MS employees who became treated (n = 258) were younger (40.9 vs. 44.4 years, p < 0.0001) and had a higher proportion of women (72 vs. 62%, p = 0.007) than the untreated group of MS employees who never received DMD treatment (n = 322). The 3-month baseline MS-related medical costs were higher among treated MS employees ($2520 vs. $1012, p < 0.0001). There was a nonsignificant trend toward higher baseline non-MS-related medical costs in untreated versus treated MS employees. Risk-adjusted total annual medical costs ($4393 vs. $6187, p < 0.0001) and indirect costs ($2252 vs. $3053, p < 0.0001) were significantly lower for treated MS employees than for untreated MS employees.

CONCLUSIONS

Initiation of MS disease-modifying drugs was associated with substantial significant medical and indirect savings for employees with MS. Study findings should be considered in the context of the study limitations (e.g., analytic focus on employees with at least 12-month follow-up; lack of clinical detail on MS severity).

Asthma severity categorization using a claims-based algorithm or pulmonary function testing

OBJECTIVES

This study was performed to determine whether pulmonary function test results would appreciably alter asthma severity categorization determined by an algorithm using information readily available in administrative databases.

METHODS

Patients 6 to 64 years of age with asthma diagnosed from 1999-2005, who had at least one pulmonary function test, were identified from a claims database of a medical group practice located in central Massachusetts. Asthma severity for these patients was categorized using information available in an administrative database (claims-based algorithm) and by percent predicted forced expiratory volume in 1 second (FEV(1)) or peak expiratory flow (PEF) abstracted from medical charts (pulmonary function test method). Gamma rank correlation index was used to measure the association between the two severity categorization methods. Total and asthma-related healthcare costs for each severity category were compared between the two different approaches.

RESULTS

There was a significant ordinal association between severity categorization with the two classification approaches (p = 0.0002). The pulmonary function test method resulted in more frequent mild categorizations and less frequent moderate and severe categorizations than the claims-based algorithm. In only 10.9% of patients did the pulmonary function test method result in a more severe asthma category than the claims-based algorithm. Patients with more severe asthma, determined by both methods, had higher total and asthma-related health care costs. Total and asthma-related health care costs were similar for each asthma severity categorization for the two classification approaches, except for asthma-related costs in the moderate severity categories.

CONCLUSION

The claims-based algorithm generally categorized patients as having more severe asthma than the approach using pulmonary function test results. Pulmonary function test results would have appreciably changed asthma severity categorization in only a small percent of patients. These findings add further support to the use of administrative database analyses for the evaluation of asthma care in large populations.

The cost of disability and medically related absenteeism among employees with multiple sclerosis in the US

Studies have not previously reported the indirect cost burden of multiple sclerosis (MS) from an employer perspective. To compare annual indirect costs between privately insured US employees with MS and matched employee controls. A retrospective analysis of a privately insured claims database containing disability data from 17 US companies was conducted. Employees with >/=1 MS diagnosis (ICD-9-CM: 340.x) after 1 January 2002, aged 18-64 years, were selected. Employees with MS were matched by age and sex to employee controls without MS. All were required to have continuous health coverage 3 months before MS diagnosis (baseline) and 12 months after (study period). Main outcomes measures included study period annual indirect (disability and medically related absenteeism) costs. For completeness, we also included measures of direct (medical and drug) costs. Chi-squared tests were used to compare baseline co-morbidities and differences in indirect resource use (disability and medically related absenteeism) between employees with MS and controls. Wilcoxon rank-sum tests were used for univariate comparisons of disability and medically related absenteeism days and associated annual indirect and direct costs between employees with MS and controls. Generalized linear models, controlling for differences in baseline characteristics, were used to estimate risk-adjusted annual costs for employees with MS and controls. Employees with MS (n = 989) averaged 44 years of age, and 66% were female. Compared with employee controls, employees with MS had significantly higher rates of mental health disorders, other neurological disorders and physical disorders measured by the Charlson Co-morbidity Index. Employees with MS were more likely to have short-term or long-term disability than employee controls (21.4% vs 5.2%, respectively; p < 0.0001), resulting in a higher mean number of disability days per year (29.8 vs 4.5; p < 0.0001). Employees with MS also had a higher rate of medically related absenteeism and associated absenteeism days than employee controls. On average, annual costs (year 2006 values) for disability were significantly higher for employees with MS ($US3868) than employee controls ($US414; p < 0.0001). Annual medically related absenteeism costs were also higher for employees with MS than for controls ($US1901 vs $US1003, respectively; p < 0.0001). On average, total annual indirect costs for employees with MS were $US5769 compared with $US1417 for controls (p < 0.0001). MS is a chronic and debilitating disease that poses a substantial employer burden in terms of medically related absenteeism and disability costs. Indirect costs of employees with MS were >4 times those of employee controls.

Adherence to inhaled corticosteroid use and local adverse events in persistent asthma

OBJECTIVES

To measure adherence to inhaled corticosteroid (ICS) therapy using prescription claims and a patient survey, to identify local adverse events (LAEs) from the patient perspective and from medical records, and to evaluate the association between LAEs and adherence to ICS therapy.

STUDY DESIGN

Survey administration and claims-based and medical record-abstracted data.

METHODS

Patients aged 6 to 64 years with persistent asthma (defined using an established algorithm) and at least 2 ICS prescriptions were selected from a claims database (1999-2006) of a central Massachusetts medical group practice. Prescription claims were used to calculate the ICS medication possession ratio (MPR). A survey obtained information about patient-reported adherence to ICSs using the Morisky scale and a visual analog scale (VAS) and about LAEs using the validated Inhaled Corticosteroid Questionnaire. Medical records of survey respondents were abstracted for LAEs.

RESULTS

Among 372 survey respondents, 2.7% met the claims-based measure of good adherence (MPR, > or =80%). Patient-reported adherence was much higher; 20.7% of patients were highly adherent based on the Morisky scale (score, 0) and 55.4% based on the VAS (score, > or =80%). Medical record review identified 27.2% of patients having at least 1 LAE within 1 year after the ICS index date, but 47.3% of patients reported being bothered at least "quite a lot" by LAEs. Multivariate analysis indicated that unpleasant taste was significantly related to lower adherence based on the Morisky scale (P = .02).

CONCLUSIONS

Patient-reported adherence and LAEs were higher than those measured from claims or medical records. Unpleasant taste seems to be associated with lower adherence based on the Morisky scale.

Costs and resource use of mild persistent asthma patients initiated on controller therapy

BACKGROUND

The treatment of mild persistent asthma is controversial.

OBJECTIVES

A retrospective database approach was used to evaluate different alternatives to treating mild persistent asthma. We hypothesized that treatment with inhaled corticosteroids (ICS) would result in lowest costs than treatment with leukotriene modifiers (LM) and combination therapy with ICS long-acting inhaled beta(2)-agonists (LABA) because it would be associated with fewer acute care visits and hospitalizations than LM and it would have lower drug acquisition costs than both ICS+LABA and LM.

METHODS

Costs and resource utilization were compared in 1,283 mild persistent asthma patients initiating regular use of either ICS, ICS+LABA, or LM. Mild persistent asthma patients were identified from a privately insured claims database (1999-2005) using an established algorithm. Wilcoxon rank-sum tests and generalized linear models were used to compare costs.

RESULTS

Of the total patients who met study criteria, 319 patients (24.9%) initiated regular ICS use, 414 (32.3%) ICS+LABA use, and 550 (42.9%) LM use. Over the 1 year after controller therapy initiation, asthma-related direct costs were significantly lower with ICS compared with ICS+LABA or LM ($819 for ICS, $1,094 for ICS+LABA, and $869 for LM, p < 0.001 for all comparisons). There were no significant differences in resource use.

CONCLUSION

In this analysis, physicians, despite guideline recommendations, chose to treat patients with mild persistent asthma more often with LM and ICS+LABA than with ICS. However, therapy with ICS was less costly than treatment with either LM or ICS+LABA, primarily due to differences in drug costs, and provided similar outcomes.

Health care costs of adults treated for attention-deficit/hyperactivity disorder who received alternative drug therapies

BACKGROUND

Many therapies exist for treating adult attention-deficit/hyperactivity disorder (ADHD), also referred to as attention-deficit disorder (ADD), but there is no research regarding cost differences associated with initiating alternative ADD/ADHD drug therapies in adults.

OBJECTIVE

To compare from the perspective of a large self-insured employer the risk-adjusted direct health care costs associated with 3 alternative drug therapies for ADD in newly treated patients: extended-release methylphenidate (osmotic release oral system-MPH), mixed amphetamine salts extended release (MAS-XR), or atomoxetine.

METHODS

We analyzed data from a US claims database of 5 million beneficiaries from 31 large self-insured employers (1999-2004). Analysis was restricted to adults aged 18 to 64 years with at least 1 diagnosis of ADD/ADHD (International Classification of Diseases, Ninth Revision, Clinical Modification [ICD-9-CM] codes 314.0x--attention deficit disorder; 314.00--attention deficit disorder without hyperactivity; or 314.01--attention-deficit disorder with hyperactivity) and at least 1 pharmacy claim for OROS-MPH, MAS-XR, or atomoxetine identified using National Drug Codes. In preliminary analysis, we calculated the duration of index ADHD drug therapy as time from index therapy initiation to a minimum 60-day gap. Because the median duration of index ADHD drug therapy was found to be approximately 90 days, the primary measures were total direct medical plus drug costs and medical-only costs computed over 6 months following therapy initiation. Adults were required to have continuous eligibility 6 months before and 6 months after their latest drug therapy initiation and no ADHD therapy during the previous 6 months. Cost was measured as the payment amount made by the health plan to the provider rather than billed charges, and it excluded patient copayments and deductibles. Medical costs included costs incurred for all-cause inpatient and outpatient/other services. Costs were adjusted for inflation to 2004 U.S. dollars using the consumer price index for medical care. T tests were used for descriptive cost comparisons. Generalized linear models (GLMs) were used to compare costs of adults receiving alternative therapies, adjusting for demographic characteristics, substance abuse, depression, and the Charlson Comorbidity Index.

RESULTS

Of the 4,569 patients who received 1 of these 3 drug therapies for ADHD, 31.8% received OROS-MPH for a median duration of 99 days of therapy, 34.0% received MAS-XR for a median 128 days, and 34.2% received atomoxetine for a median 86 days. In the 6-month follow-up period, the mean (standard deviation) total medical and drug costs were $2,008 ($3,231) for OROS-MPH, $2,169 ($4,828) for MAS-XR, and $2,540 ($4,269) for atomoxetine-treated adults. The GLM for patient characteristics suggested that 6-month, risk-adjusted mean medical costs, excluding drug costs, for adults treated with OROS-MPH were $142 less (10.4%, $1,220 vs. $1,362) compared with MAS-XR (P =0.022) and $132 less (9.8%, $1,220 vs. $1,352) compared with atomoxetine (P =0.033); risk-adjusted mean medical costs were not significantly different between MAS-XR and atomoxetine. The GLM comparison of risk-adjusted total direct costs, including drug cost, was on average $156 less (8.0%, $1,782 vs. $1,938) for OROS-MPH compared with MAS-XR (P = 0.017) and $226 less (11.3%, $1,782 vs. $2,008) compared with atomoxetine (P <0.001); the risk-adjusted total direct costs were not significantly different between MAS-XR and atomoxetine. Two high-cost outliers (greater than 99.96th percentile, 1 each for OROS-MPH and atomoxetine) accounted for $47 (30%) of the $156 cost difference between OROS-MPH and MAS-XR and $11 (5%) of the $226 cost difference between OROS-MPH and atomoxetine, and the medical diagnoses for the highest-cost claims for these 2 outlier patients were unrelated to ADHD.

CONCLUSIONS

After adjusting for patient characteristics including substance abuse, depression, and the Charlson Comorbidity Index, adults treated with OROS-MPH had, on average, slightly lower medical and total medical and drug costs than those treated with MAS-XR or atomoxetine over the 6-month period after drug therapy initiation. Approximately 30% of the cost difference compared with MAS-XR was attributable to 1 high-cost outlier with medical diagnoses for the highest-cost claim that were unrelated to ADHD.