Initiating Drug Therapy in Multiple Sclerosis Patients: Effect on Healthcare Costs

Multiple sclerosis (MS) is a disease of inflammatory demyelination of axons of the central nervous system that results in episodes of transient neurologic deficits, progressing to permanent neurologic deterioration over time.1 The cause of MS remains unclear, but it is thought to involve both genetic susceptibility and environmental triggers, possibly viral, which result in self-sustaining autoimmune dysfunction.

The resulting demyelination of the nerves, along with some axonal injury, causes a variety of symptoms through disruption of the nerve’s ability to conduct electrical impulses. Symptoms depend on the nerves affected, but include both motor and sensory deficits such as fatigue, visual disturbances, limb and bladder spasticity, impotence, depression, and other emotional and intellectual changes.1 Different forms of MS are seen based on clinical presentation and symptoms: about 80% to 85% of patients have relapsing-remitting MS (RRMS), with progressive forms being the next phase for most patients with RRMS. In RRMS, which often is the initial type at diagnosis, symptoms occur in discrete attacks of varying severity. As the disease advances, symptoms become more permanent and get progressively worse (secondary progressive MS). Some patients initially present with a progressive form (primary progressive MS or progressive relapsing MS) of the disease.2-5

Multiple sclerosis affects all races and socioeconomic groups, but is mainly seen in females, affecting from 1.4 to 3.1 times as many women than men. Onset of the disease generally occurs in people in their 20s and 30s.2 In a recent study from the National Institutes of Health, it was estimated that the annual incidence of MS in the United States is approximately 4.2 new cases per 100,000 persons, and the estimated prevalence is 266,000 for a rate of 100 per 100,000 population, although the National Multiple Sclerosis Society of America states the prevalence as high as 400,000 in the United States and as many as 2.5 million worldwide.6,7

The direct and indirect costs associated with MS are substantial. Patients experience considerable disability for a long period of time, as MS generally does not significantly shorten life spans.8 More than 50% of those with MS are unemployed within 10 years of diagnosis, when most people are at the height of their careers.9 The total annual cost attributed to MS in the United States exceeds $6.8 billion, of which 75% to 80% are indirect costs, such as lost earnings for both the patient and caregivers, as well as lost taxation and state benefits.10-12 This amounts to $2.2 million per patient over their lifetime.10

The direct costs for treating MS are only expected to increase with early initiation and prolonged utilization of disease-modifying drug (DMD) therapy. Furthermore, in addition to the DMDs, management includes other pharmacotherapeutic classes, physical therapy, and the involvement of multiple healthcare specialties. In a study of MS patients from a commercial administrative claims database, approximately 56% were found to have been prescribed at least 1 DMD, and 6% took 2 or more drugs.13 Patients who take MS-specific medications also have been found to be more likely to take medications for specific symptoms such as depression (43%), spasticity (31%), pain/dysesthesias (28%; feelings of burning, wetness, itching, pins/needles under the skin),13 bladder problems (25%), and fatigue (19%). In addition, healthcare utilization increased for patients with MS: 12% were hospitalized at some point during the study, 11% visited the emergency department, 2.4% had stays in the intensive care unit, and 1.8% had skilled nursing facility admissions.13 A 2004 analysis that evaluated direct costs associated with MS in more than 10,000 patients from a national claims database revealed total annual costs of $12,879 per person with 26% of costs attributable to outpatient care and 65% to pharmacotherapy, of which 95% was directly attributable to DMD therapy.14

Considerable direct and indirect costs, combined with increasing numbers of patients diagnosed with MS, highlight the need for careful examination of the therapies used in the management of this disease. It is important for managed care professionals to understand both the clinical and financial impact DMDs have on the disease so that they can optimally manage patients with the most cost-effective treatment options through the provision of pharmacy benefits.

This study utilized a large national administrative claims database to observe and report information including charges and utilization of healthcare services for patients newly treated with self-injectable DMDs for a total of 2 calendar years (the first year when treatment was begun and the subsequent year). The analysis takes a broad perspective on this information, putting the data in a format that could be easily reproducible, expanded, and refined.

METHODS

Data Source

Data were obtained from the IMS/PharMetrics Integrated Patient-Centric Database, which at the time of this evaluation was composed of fully adjudicated (ie, final paid claims, subsequent to reversals and adjustments) medical and pharmaceutical claims for more than 55 million unique patients from 75 health plans across the United States. The database includes both inpatient and outpatient diagnoses (in International Classification of Diseases, Ninth Revision, Clinical Modification [ICD-9-CM ] format) and procedures (in ICD-9-CM, Current Procedural Terminology, and Health Care Financing Administration Common Procedure Coding System formats), as well as both community and mail-order pharmacy claims; available data on pharmacy claims are based primarily on National Drug Code and Common Procedure Coding System drug codes. Both paid and charged amounts are available for all services rendered, as well as the date of service for all claims. Additional data elements include demographic variables (age, sex, geographic region), provider specialty, and start and stop dates for plan enrollment.

Records in the database are representative of the national commercially insured population on a variety of demographic measures, including geographic region, age, sex, and product type (eg, health maintenance organization, preferred provider organization). The data also are longitudinal, with an average member enrollment time of 2 years. Only health plans submitting data for all of their members are included in the database. Data submissions are subjected to a series of data quality checks to ensure a standardized format and minimal error rates—for example, presence of claim reversals or adjustments, and errors in age and sex coding.

Patient Selection

From the source database, patients were selected for observation based on the presence of a diagnosis code on medical claims for MS (ICD-9-CM code 340) during the 2005 pharmacy or other carve-outs). Of the remaining patients, only those MS patients new to treatment (incident) in 2005 were observed. This status was determined by the absence of either medical claims with a diagnosis for MS, use of self-injectable DMDs, or receipt of infused agents (ie, Novantrone, natalizumab) during the 2004 calendar year, thus providing a “clean” period of at least 365 days.

Of the incident patients, only those who received a selfinjectable DMD (ie, interferon [IFN] products, glatiramer acetate) in 2005 were observed. Mutually exclusive cohorts were formed based on the initial self-injectable DMD used, on an intent-to-treat basis.

Reporting Metrics

Charges and utilization of medical services and prescription drugs were identified and captured using the Episode Treatment Groups (ETG) software.15 Data were collected specifically for ETG numbers 149 and 150 (“Inflammation of the central nervous system, with or without surgery”) during the study period. These data were broken down by the service category (inpatient, outpatient, emergency department, and pharmacy), depicting at which point in the healthcare continuum the service was received. All dollar amounts presented in this study represent submitted or billed charges by a practitioner or institution to the health plan or insurer for payment.

These economic data as well as the presence of specific conditions of interest and drug therapies related to MS were identified, collected, and presented for both the entire year when self-injectable DMD therapy was started (2005) and the following calendar year (2006). To create these markers, patients needed to have 1 or more medical claims or pharmacy claims (using ICD-9- CM and National Drug Code codes, respectively) during a calendar year.

Statistics

Statistical testing was performed using SPSS for Windows version 15 (SPSS Inc, Chicago, IL).

RESULTS

Multiple sclerosis patients (n = 21,517) were identified from the source database in 2005. A total of 4262 incident patients met the eligibility criteria; of these patients, 1840 (43.4%) had received a self-injectable DMD in 2005. Patients most commonly received glatiramer acetate (33.2%) or intramuscular (IM) IFN beta-1a (31.9%) as their initial therapy. See eAppendix A (available at www.ajpblive.com).

Conditions of Interest and Use of Additional Drugs

Moving from 2005 into 2006, there was a decrease in medical claims for many common signs and symptoms (eg, abnormality of gait, burning/numbness/tingling, malaise/ fatigue) associated with MS. Claims for urinary incontinence and trigeminal neuralgia increased, but not significantly. See eAppendix B (available at www.ajpblive.com).

As reflected in eAppendix C (available at www. ajpblive.com), there was statistically significantly less utilization of specific pharmacotherapy commonly utilized in the management of MS including corticosteroids (both injectable and oral), benzodiazepines, and nonsteroidal anti-inflammatory drugs.

Healthcare Charges and Utilization

For MS patients treated with any self-injectable DMD, total (medical and pharmacy) annual charges in 2005 were $19,247 and increased to $22,035 in 2006. However, although total charges increased, there also was a 68% decrease in costs associated with inpatient care, a near 17% decrease in charges for outpatient care, and a 39% decrease in emergency department care charges. As could be expected, there was a 46% increase in the use of selfinjectable DMDs, corresponding to initiation and continuation of therapy (Table 1).

Among the individual self-injected DMDs, the increase in total episode cost between 2005 and 2006 was lowest in the IFN beta-1a IM group, with a year-over-year increase of 8.1% compared with 18.1% for subcutaneous IFN beta- 1a, 21.9% for IFN beta-1b, and 15.5% for glatiramer acetate (Tables 2-5). Additionally, the IFN beta-1a IM group had the most significant decrease in inpatient charges (86.6%) and the lowest increase in pharmacy charges (38.5%) compared with the other self-injected DMDs. Glatiramer acetate was associated with the smallest decrease in inpatient charges (48.3%).

During year 2 of the analysis, patients receiving IFN beta-1a IM therapy had evidence of more prescriptions for that agent compared with patients receiving the other selfinjectable DMDs (8.13 prescriptions for IFN beta-1a IM, 7.23 for subcutaneous IFN beta-1a, 7.96 for IFN beta-1b, and 7.65 for glatiramer acetate; Tables 2-5).

DISCUSSION

This analysis represents a high-level, populationbased look at the impact that self-injectable DMD treatment has on some of the economic measures associated with the treatment of MS in a commercial population. Such analyses, though they may not be as detailed as other pharmacoeconomic studies, serve an important role in evaluating readily available claims data for observational reporting. Although there was an overall increase in total annual charges between 2005 and 2006 of about 14.5%, there was evidence to suggest that the increased utilization of DMDs in year 2 may have resulted in the decreased utilization and charges associated with inpatient (68.3% decrease in charges), outpatient (16.9% decrease in charges), and emergency department (39.2% decrease in charges) care. These decreases could translate into clinical improvement, as other studies have shown decreased healthcare costs to be associated with increased functional capacity.16 Following similar cohorts of patients for longer periods of time may yield data that suggest ongoing financial benefits for the payer with long-term self-injectable DMD therapy.

Although differences (eg, charge data, utilization data) did exist among the individual self-injectable DMDs, it is not possible to establish explicit cause-and-effect relationships from this type of analysis, nor to evaluate the impact of disease severity because of the lack of clinical information (even including the current form of MS)available from claims data. Because of this, direct comparison of cohorts is problematic, as not all patients may have been at a similar disease stage when initiating therapy. 16-18 Kobelt et al reported that IFN beta-1a IM and glatiramer acetate are commonly used in patients with milder disease.16 Overlaying that information onto the current analysis could explain the lower charges observed during the initial year of the study for those cohorts compared with the remaining groups, although inpatient charges were among the highest in the IFN beta-1a IM group.

Other published trials evaluating annual direct costs of care for MS have ranged from $7677 to $12,879 per patient.14,18 Although these costs are lower than those in the current analysis, the present study sought to evaluate patients new to self-injectable DMD therapy to observe the costs potentially associated with the disease at onset and the impact that self-injectable DMDs have on these patients from an economic standpoint. We are not aware of any similar published analyses utilizing integrated managed care claims data focusing on MS patients new to selfinjectable DMD therapy.

Limitations

Administrative claims data do not include clinical information such as laboratory values and other measures (eg, disease severity); therefore, this study could not aggregate data by MS disease type (eg, primary progressive MS vs RRMS or secondary progressive MS). This is a descriptive analysis and cannot provide insight into cause-and-effect relationships. It was not possible to determine how economic outcomes were related to clinical status, or the effect of disease severity on economic outcomes. Episodes were captured based on their occurrence within the calendar year of analysis; they were not specifically associated with a defining event (which would provide evidence of the potential severity of the disease for a specific patient). Finally, this study relied on third-party software (ie, ETG software) to aggregate administrative claims data. This method has 2 implications. First, although MS-related costs aggregated in this manner may identify the direct medical costs to treat MS, the total cost burden that an MS patient places on the healthcare system (ie, the impact that MS may have on all other conditions a patient may have) may not have been captured. Second, using a commercially available and widely used tool such as ETG software enables others to perform an analysis similar to this study in their own environment and to use the data from the present study for comparison. Although having its limitations, this analysis was undertaken to bridge the gap between true pharmacoeconomic analyses and actual reporting of medical and pharmaceutical claims data—–resulting in a less sophisticated type of analysis that is directional in nature.

CONCLUSION

These MS-specific, population-based measurements represent an exercise in developing data that are easily reproducible within a managed care environment. The information from this analysis offers a starting point and comparator to further investigate the impact that selfinjectable DMDs have on the overall economic landscape of direct treatment costs for MS, both in the short term and the long term.