Treatment of patients diagnosed as having bipolar disorders has evolved rapidly in recent years. A growing number of medicines have received U.S. Food and Drug Administration (FDA) and international regulatory approval for use in treating acute mania or bipolar depression or for long-term maintenance treatment aimed at limiting recurrence risks (1). Despite a growing array of effective and reasonably safe treatments, residual morbidity among patients treated by contemporary community standards remains high. Previous follow-up studies of treated patients with bipolar I disorder who had been ill for several years (2,3,4) or following first episodes (5) have found residual morbidity in 40% of the follow-up period. Approximately three-quarters of the residual morbidity was depressive or dysthymic (2,3,4,5). Unresolved depressive morbidity is probably an important contributor to substance abuse, functional disability, and excess mortality because of high suicide rates in early years and because of medical disorders in later years (6,7,8,9,10,11,12).
Prominent residual depressive morbidity among treated patients with bipolar disorder encourages empirical use of antidepressants in both short- and long-term treatment, despite limited evidence of their efficacy and safety in the treatment of bipolar depression (13,14,15,16). In addition, residual morbidity and the growing number of options for treating such patients encourage use of multiple mood-altering and other psychotropic agents (polytherapy), despite very limited evidence of additional effectiveness or of the relative safety of such empirical interventions (1,17,18). Moreover, polytherapy may have the paradoxical effect of limiting adherence to critically required long-term mood-stabilizing treatment for patients with bipolar disorder (19). A recent analysis of pharmacy benefits records of a national sample of U.S. patients diagnosed as having bipolar disorder considered new treatments with single psychotropic agents (20). Antidepressants were, by far, the most commonly prescribed psychotropics in initial monotherapies, and both lithium and antidepressants were retained longer than other psychotropics, including anticonvulsants, antipsychotics, and sedatives (20).
We report new findings from independent national health care claims data, focusing on initial polytherapy, changes in treatment over one year, and factors associated with long-term treatment adherence.
Study data were obtained from a proprietary research database containing eligibility information and pharmacy and medical claims data from a large commercial U.S. health plan providing coverage for physician, hospital, and prescription drug services, including services for psychiatric disorders. The health plan provides data from both self-insured employers (including retirees with employer-provided Medicare Supplemental Plans) and employer-sponsored commercially insured subscribers representing a geographically diverse sampling of approximately 14 million persons across the United States, with a concentration in the South and Midwest. We derived data from claims submitted by care providers to obtain payment for services rendered. Computerized files used were deidentified and accessed by protocols in compliance with the United States Health Insurance Portability and Accountability Act of 1996 to ensure confidentiality. Data derived from the same source have been used previously for various utilization, safety, and economic analyses (20,21,22,23), but they have not been used for analyses of treatments provided to patients diagnosed as having bipolar disorders.
Medical and pharmacy claims
Medical claims or encounter data were collected from all available sites for all types of services provided and were coded in conformance with insurance industry standards. Claims for ambulatory services submitted by individual providers (usually physicians) use Health Care Financing Agency (HCFA)-1500 forms and claims for hospital-based services use UB-92 forms (www.cms.hhs.gov).
Facility service records contain information on up to nine diagnoses and six procedures, categorized by International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) diagnosis codes (24), and they contain information on procedure codes based on ICD-9-CM, Current Procedural Terminology (CPT), or HCFA Common Procedure Coding System (HCPCS) protocols (www.cms.hhs.gov/home/medi care.asp). Facility claims contain categories of most services but may not account for all drugs administered in the hospital. Provider service records contain information on up to four diagnoses recorded with ICD-9-CM codes and one procedure recorded using ICD-9-CM, CPT, or HCPCS codes. Claims for pharmacy services include drug names, dosage forms, drug strengths, fill dates, and number of days' supply.
Study population and characteristics
Health plan members initiating treatment for bipolar disorders were identified using eligibility criteria and medical and pharmacy claims data acquired between January 1, 2001, and December 31, 2005. To be included in the study, persons were required to be continuously enrolled in the health plan for six months before and for 12 months after a first prescription fill for a psychotropic medication. Persons were considered for the study if they were diagnosed as having bipolar disorder, were aged 17 years or older, were newly initiating pharmacological treatment for bipolar disorder (no psychotropic prescription fills in the preceding six months). Patients were entered in the study between January 1, 2001, and December 31, 2004, and were followed for 12 months after initiating treatment. Persons diagnosed as having schizophrenia, epilepsy, or migraines were excluded.
Demographic characteristics were assigned according to information reported in index claims for the first psychotropic prescription fill, including sex, current age, and geographic region. If more than one prescription by more than one prescriber was filled on the index date, the provider type was assigned hierarchically in the following order: psychiatrist, general practitioner, other specialist, and unknown.
Diagnosis of bipolar disorder and its subtypes was based on the presence of ICD-9-CM diagnosis codes indicating such illnesses that were then converted to corresponding DSM-IV codes according to guidelines provided by the American Health Information Management Association (AHIMA) (25). For this study, patients who ever had an ICD-9-CM diagnosis of bipolar affective disorder (ICD-9-CM codes 296.4—296.66), corresponding with DSM-IV bipolar I disorder (DSM-IV codes 296.4—296.66), were considered to have type I bipolar disorder; those who did not have bipolar disorder type I but had ever been diagnosed as having ICD-9-CM manic-depressive psychosis, other or unspecified type (296.89), corresponding to DSM-IV bipolar II disorder (296.89), were considered to have type II bipolar disorder; those who did not have type I or II bipolar disorder but had ever been diagnosed with ICD-9-CM affective psychosis, other or unspecified (296.7, 296.80), or with ICD-9-CM atypical manic or depressive disorder (296.81—296.82) were considered to have bipolar disorder not otherwise specified.
Patients also were categorized according to their apparent illness complexity, on the basis of selected clinical factors and presence of higher-severity bipolar disorder indicators or comorbid conditions that may complicate treatment, all as detailed by Solz and Gilbert (26) and applied to other illnesses (26,27). We consulted with clinical and coding experts to identify ICD-9-CM codes associated with relatively high-severity bipolar disorders and complex illness management. Indicators of higher illness complexity included diagnostic indicators for moderate or severe bipolar disorder (296.xx with fifth digit classification as "moderate" or "severe" or with psychotic features), an indication of self-harm (E950.x-E959) or overdose (969.xx), or presence of certain comorbid disorders (including bulimia nervosa [307.51], impulse-control disorder [312.30], or chronic fatigue syndrome [780.71]). Remaining patients with bipolar disorder were considered to represent relatively lower levels of illness complexity, and substance use comorbidity was considered separately.
We also calculated comorbidity scores at baseline and during follow-up by using an algorithm developed by Charlson and collaborators (28) and adapted by Deyo and colleagues (29) for use with administrative claims databases. To better identify patients with specific clinical conditions used to score comorbidity, required diagnosis codes were updated in consultation with clinical and coding experts to reflect recent changes or additions to the codes, without change in the number or identity of clinical conditions used to calculate comorbidity scores. These included diagnosis of the clinical state (depressed, manic or hypomanic, mixed, or unspecified) closest in time to the initial prescription fill was determined.
Psychotropic medicines provided were categorized as mood stabilizers, including lithium and anticonvulsants (carbamazepine or oxcarbazepine, valproate [any salt of valproic acid, including divalproex], and lamotrigine); antipsychotics, including modern antipsychotics (aripiprazole, clozapine, olanzapine, quetiapine, risperidone, and ziprasidone) and older neuroleptics (chlorpromazine, fluphenazine, haloperidol, loxapine, mesoridazine, molindone, perphenazine, pimozide, promazine, thioridazine, thiothixene, trifluoperazine, and triflupromazine); antidepressants, including serotonin reuptake inhibitors (SRIs) (citalopram as its racemate or R-isomer, fluoxetine, fluvoxamine, paroxetine, and sertraline) and serotonin-norepinephrine reuptake inhibitors (SNRIs) (duloxetine and venlafaxine), tricyclic antidepressants (including amitriptyline, amoxapine, clomipramine, desipramine, doxepin, imipramine, maprotiline, protriptyline, nortriptyline, and trimipramine), monoamine oxidase (MAO) inhibitors (isocarboxazid, phenelzine, and tranylcypromine), and modern antidepressants (including bupropion, mirtazapine, nefazodone, and trazodone). We also identified concomitant use of anxiolytic-sedatives, or hypnotics, including benzodiazepines, as well as miscellaneous anticonvulsants of unproved efficacy in bipolar disorders (gabapentin, levetiracetam, tiagabine, topiramate, and zonisamide). Initial treatments and combinations were assigned as psychotropic prescriptions dispensed during the first 30 days of treatment. Final regimens were assigned by psychotropic prescriptions dispensed within the final 90 days of follow-up.
We also examined overall health care utilization during follow-up and recorded office or emergency service visits, hospitalizations, and hospital days per patient, based on services rendered, as identified in the claims database.
To evaluate mood stabilizer adherence, specifically, we identified a subgroup of patients who took a single mood stabilizer and initiated treatment with lithium or a mood-stabilizing anticonvulsant (carbamazepine, lamotrigine, oxcarbazepine, or valproate) and neither switched nor augmented therapy with any other mood stabilizer during follow-up. We measured mood stabilizer adherence by using a medication possession ratio (MPR). MPR is the percentage of the past 365 days with apparent access to an initial mood stabilizer. Also, a lower MPR would capture both treatment nonadherence and interruptions in treatment.
We analyzed patient demographic and clinical characteristics, treatment regimens, and health care utilization in the study sample. We considered initial treatment patterns and examined changes in treatment during a year of follow-up, comparing all single and multiple psychotropic prescriptions dispensed during the first 30 days (initial treatment) to those during the final 90 days of follow-up (final treatment).
To identify characteristics associated with treatment adherence in the subsample of patients treated with a single mood stabilizer for up to a year, we compared factors of potential interest (including sex, age, diagnostic type, comorbidities, health care service use, prescriber type, geographic regions, mood-stabilizer treatment, and psychotropic co-treatments), contrasting adherent patients (MPR ≥80%) and nonadherent patients (MPR <80%) in preliminary bivariate comparisons. Factors differing between these adherence subgroups were further evaluated by multivariate least-squares regression modeling. Modeling included days hospitalized per patient to control for potential effects of prolonged hospitalizations with incompletely recorded inpatient pharmacy claims. Index year of starting treatment was also included to adjust for possible changes in prescribing patterns over time. When multiple items were considered in some comparisons, one was selected as a comparator against which effects of other factors on treatment adherence were related.
Data are reported as means and standard deviations (SDs) unless stated otherwise, and statistical tests required two-tailed p<.05 for significance. Analyses used commercial statistical software (SAS 9.1.3).
A total of 7,406 treated patients met study inclusion criteria; 55% were identified as having bipolar I disorder, 15% as having bipolar II disorder, and 30% as having bipolar disorder not otherwise specified (Table 1). Mean age was 35.4±12.4 years, and a slight majority (57%) were women. More cases were located in the South (42%) and Midwest (40%) than in the West (10%) or Northeast (8%) of the United States, closely paralleling the overall population distribution of the health plan. Most patients (67%) were treated initially by psychiatrists (37%) or primary care physicians (30%).
Among all 7,406 patients, 67% had an initial prescription fill for a single primary mood-altering drug (monotherapy), and 33% (2,444) had initial prescription fills for two or more major psychotropic drugs (polytherapy; major psychotropic drugs include anticonvulsants with mood-stabilizing or antimanic effects, lithium salts, antipsychotics, or antidepressants) (Figure 1). If sedatives and miscellaneous anticonvulsants are included, 43% (N=3,191) would have had polytherapy (Table 2). Initial prescription fills for primary monotherapies ranked as follows: antidepressants > anticonvulsants > antipsychotics > lithium. With initial prescription fills for polytherapy, antidepressants (79% of patients) were often used in combination with other agents more than were anticonvulsants (49%), antipsychotics (50%), or lithium (16% of patients). Overall, initial utilization rates ranked as follows: antidepressants > anticonvulsants ≥ antipsychotics > sedatives (anxiolytics or hypnotics) > lithium > miscellaneous anticonvulsants (Figure 1). Almost all antidepressants and antipsychotics given were modern agents.
At one year, 2.2-fold fewer (31%) patients received monotherapy, 32% received two or more primary psychotropics (mood stabilizers, antipsychotics, or antidepressants), 50% of those still being treated received two or more major psychotropic drugs that act on the central nervous system and are used with the intention of mood stabilization (not including sedatives and miscellaneous anticonvulsants), and 37% had no evidence of receiving a primary mood-altering agent in the preceding 90 days (Table 2 and Figure 1). Although adequate clarification of the basis of being untreated at one year was not feasible, this untreated subgroup was not less likely to be in the care of a psychiatrist than other types of providers (data not shown).
Major changes from baseline to one year were for patients moving from monotherapy (67%) to no treatment (37%). Among monotherapies at one-year, antidepressants again dominated, followed by anticonvulsants, lithium, and antipsychotics. For polytherapy during follow-up, most regimens also included antidepressants, as well as prevalent use of anticonvulsants and antipsychotics and a slightly greater use of lithium. Overall utilization rates among those being treated at follow-up ranked as follows: antidepressants (72%) > other mood-altering anticonvulsants (38%) > antipsychotics (30%) > sedatives (24%) > lithium (15%) > miscellaneous anticonvulsants (12%) (Table 2 and Figure 1).
Adherence to mood stabilizer treatment
Among the 7,406 study patients, 2,197 (30%) were given a prescription for a single mood stabilizer at baseline and had no evidence of switching or augmenting this basic treatment with an additional mood stabilizer (anticonvulsant or lithium) during follow-up. In this subsample of patients given a single mood stabilizer, valproate (N=1,125, or 51%) was the most commonly used agent, followed by lithium (N=517, or 24%), carbamazepine or oxcarbazepine (N=308, or 14%), and lamotrigine (N=247, or 11%). However, many of the patients in this subsample were also given an antidepressant (N=1,213, or 55%), sedative (N=791, or 36%), or antipsychotic drug (N=763, or 35%), and a minority (N=306, or 14%) received a miscellaneous anticonvulsant at some time, with an average of 2.85±1.72 psychotropics per person (median=2, range=1—12), even with nominally simplified treatment. Only 28% (N= 620) of the subsample were considered adherent (MPR ≥80%) to their single mood stabilizer therapy, and 72% (N=1,577) were considered nonadherent (MPR <80%).
Of note, patients given lithium as the single mood stabilizer were much less likely to receive adjunctive psychotropic agents during the following year than were those whose single mood stabilizer was an anticonvulsant. This difference was significant for all types of psychotropic drugs except anxiolytics, and it was particularly striking with antidepressants (Table 3).
Factors associated with mood stabilizer adherence
Preliminary, unadjusted, bivariate comparisons of the treatment-adherent subgroup (N=620) and the nonadherent subgroup (N=1,577) indicated minor differences between each other or versus the overall cohort of 7,406 patients with respect to geographic distribution, year of study entry, sex, entry age, diagnosis, illness complexity, comorbidity index score, and prescriber type (data not shown). Moreover, the subgroups were very similar in utilization rates for adjunctive psychotropics other than the identified primary mood stabilizer. A larger proportion of adherent patients received lithium or lamotrigine, whereas nonadherent patients were more often given valproate, carbamazepine, or oxcarbazepine. Adherent patients also were less likely to abuse alcohol or drugs, compared with nonadherent patients (N=24, or 3.9%, versus N=120, or 7.6%), but they were somewhat more likely to have a comorbid anxiety disorder (N=6, or 1.0%, versus N=7, or .4%). Also of interest, patients adherent to mood stabilizers had 20% more total outpatient visits per person per year (19.0 versus 15.8) and 55% more ambulatory visits related to bipolar disorder (6.8 versus 4.4); they also had 44% more emergency service visits related to bipolar disorder (.52 versus .36 per person per year), but they had 27% fewer hospitalizations for all reasons (.30 versus .41 per person per year, mainly for nonpsychiatric indications) and 37% fewer hospitalization days (1.7 versus 2.7 per person per year) (data not shown).
Multivariate analyses based on the preceding observations and inclusion of factors considered clinically plausible indicated several measures that were independently and significantly associated with mood stabilizer adherence, based on MPR as the outcome measure. MPR was higher with older age (p<.001), lack of substance abuse (p=.001), treatment by a psychiatrist versus a primary care physician (p=.008), and lower illness complexity (p=.013) (Table 4). Adherence to specific mood stabilizers was ranked as follows: lamotrigine ≥ lithium ≥ valproate or carbamazepine or oxcarbazepine (p<.001). Treatment adherence was lower if co-treatment with miscellaneous anticonvulsants was used (p=.004). However, no significant associations were found between MPR and co-treatment with antidepressants or antipsychotics, nor with sex, diagnostic subtype, comorbidity index, or geographical region.
Finally, in multivariate modeling, greater mood stabilizer adherence was associated significantly with more bipolar disorder-related outpatient and emergency service visits (both p<.001) but much lower utilization of emergency services for indications not related to bipolar disorder (p<.001). These variations in service utilization rates (Table 4) suggest that greater adherence may have been associated with closer outpatient follow-up, with complex implications for the economics of treatment.
This study examined all initial psychotropic treatments for U.S. patients who were diagnosed as having bipolar disorder who had not received medication for bipolar disorder for at least six months at baseline. The study also reassessed treatments a year later and examined treatment adherence in some detail by using a national claims database not previously employed for the study of treatment of such patients. New findings included the following: one-third of newly treated patients were started on polytherapy (two or more major psychotropic drugs), and even more treated patients received multiple psychotropic drugs a year later. However, at the one-year follow-up, 37% were not receiving any psychotropic medicine (a mood stabilizer, antipsychotic, or antidepressant). As noted previously (20), antidepressants were by far the most commonly prescribed psychotropic drugs for American patients diagnosed as having bipolar disorders (Table 2).
Health care insurance claims data are valuable for efficient and effective examination of clinical outcomes, treatment patterns, resource utilization, and costs. However, such data are collected to guide reimbursements and are limited when used for research purposes. Notably, a claim for a filled prescription does not prove medication was taken as prescribed. Also, medications may be provided without a prescription (for example, samples from a clinician or medications administered in a hospital), which typically are not captured in claims data. Similarly, the presence or absence of a diagnosis code in a claim does not guarantee presence or absence of a specific disorder, nor does it guarantee the accuracy of clinical assessment and documentation. Important clinical information—including specific, current, symptomatic indications for a treatment; details of illness history; test results; and physical findings—often is not available in claims files. Finally, the population studied may not be representative of all patients with bipolar disorder, and there may be a risk of underdiagnosis of bipolar disorder among insured patients, for example, to avoid potential stigmatization. The study patients represent a largely working population with geographic distribution similar to the enrollment patterns of the health plan. Despite their limitations, claims data, including data from the source used in this study, have been widely employed in research and can provide indications of clinical practice patterns and their changes over time (20,21,22,23).
Prevalent use of polytherapy initially as well as later in treatment of U.S. patients with bipolar disorder probably reflects the growing availability of both FDA-approved and off-label treatment options, as well as incomplete therapeutic responses among most patients with bipolar disorders (2,3,4,5). Available data were not adequate to specify the basis of the striking rate of treatment discontinuation (37%) found by one year (Table 2). Nontreatment is likely to reflect dissatisfaction with treatment, intolerance of it, or a perceived lack of active symptoms requiring treatment, arising from misunderstanding among patients or clinicians of the crucial role of long-term prophylaxis in the treatment of bipolar disorder even through periods of euthymia (1,30). A particularly striking finding was that polytherapy, particularly involving co-treatment with antidepressants, was much less prevalent among patients using lithium than among those using an anticonvulsant as a primary mood stabilizer. This finding is consistent with previous reports of relatively high proportions of patients maintained on mood-stabilizing monotherapy among those treated with lithium versus an anticonvulsant (3,31). These trends may suggest superior efficacy of lithium as a primary mood-stabilizing agent, including its effect on suicidal risk (32,33).
In accord with a previous analysis (20), we found again that antidepressants were by far the most commonly prescribed psychotropic medications for patients with bipolar disorder in the United States, both initially and during long-term follow-up (70%—72% of treated patients). Their popularity probably reflects both the clinical perception that such treatments are relatively well tolerated despite indications that treatment response in depressive phases of the disorder is limited (2,3,4,5,14,15,16,17,18,19,20). In addition, adverse behavioral effects of antidepressants—such as mixed states, mild hypomania, and moderately increased cycling—may be overlooked or misdiagnosed (1). Intensive reliance on antidepressants is all the more remarkable given limited evidence of their short- and especially long-term efficacy and psychiatric safety in bipolar disorder (16,18,34). Far less commonly employed treatments were FDA-approved and research-supported mood stabilizers (1), which were prescribed for only a minority of the patients initially (36%) and at follow-up (50% of those treated and only 31% of all patients) (Table 2). Anticonvulsants (28% of patients initially and 24% of those treated at follow-up, most often valproate) were prescribed much less frequently than antidepressants (≥70% of patients) but more often than lithium (9% of patients on average) (Table 2).
Additional new findings included identification of factors independently and significantly associated with long-term adherence to an initial mood-stabilizing treatment (Table 3). Perhaps not surprisingly (1,19,20), only a minority (30%) of U.S. patients diagnosed as having bipolar disorder were nominally continued for a year on an initial mood stabilizer, and only 28% of this subsample were considered to be treatment adherent, on the basis of an MPR ≥80% averaged over 12 months. Factors associated with greater treatment adherence included being older, use of lamotrigine or lithium, lack of substance abuse, and treatment by a psychiatrist rather than a primary care physician. Inferior adherence was associated with use of valproate (the most commonly prescribed anticonvulsant mood stabilizer), use of carbamazepine or oxcarbazepine, use of supplemental anticonvulsants that lack FDA-approval for use in bipolar disorder, alcohol or drug abuse, and greater illness complexity.
Finally, we found complex associations between treatment adherence and utilization of health services. Office, and especially emergency service, visits for bipolar disorder-related care were more frequent in association with greater treatment adherence, whereas emergency service utilization for other indications and days per year of hospitalization for any reason were lower with greater adherence to mood stabilizer treatment (Table 3). The lesser utilization of emergency and inpatient services suggests potential cost savings with greater adherence to mood-stabilizing treatments. On the other hand, greater use of ambulatory and emergency services for bipolar disorder-related indications suggests that greater treatment adherence may reflect relatively close monitoring, with more frequent clinician contacts. However, the available data do not permit clarification of cause-effect relationships between treatment adherence and utilization of clinical services.
Whatever their interpretation, the findings presented here underscore the clinical impression that polytherapy, discontinuous long-term treatment, and heavy reliance on largely unproved antidepressant treatment all are prominent characteristics of contemporary treatment of patients diagnosed as having bipolar disorders in the United States. Use of antidepressants was highly prevalent and sustained, despite limited evidence of their efficacy or safety in bipolar depression. Mood-altering polytherapy was used by about one-third of patients at the start of new treatment and at follow-up a year later, but was less prevalent when lithium was the primary mood stabilizer, rather than an anticonvulsant or antipsychotic agent. At one year of follow-up, 62% of patients still being treated were taking two or more major centrally active drugs of all types (polytherapy). A remarkable new finding was that at one year the proportion of patients not receiving psychotropic treatment was about as prevalent as the proportions receiving mood-stabilizing monotherapy or polytherapy, but the reason for this finding requires further clarification. Adherence to long-term mood stabilizer treatment, although uncommon, was associated with several plausible clinical factors. Our findings of heavy reliance on antidepressants and polytherapy, low mood stabilizer utilization and adherence rates, and high rates of dropout from long-term mood-stabilizing treatment strongly suggest that more effective and better-tolerated mood-stabilizing treatments are required for patients with bipolar disorders, along with redoubled educational efforts to underscore the importance of sustained, long-term prophylactic treatment of such patients, even through periods of relative euthymia.
This study was supported by Novartis Corporation, the Bruce J. Anderson Foundation, and the McLean Private Donors Psychopharmacology Research Fund (to RJB). The authors thank Jessica Goren, Pharm.D., for providing helpful suggestions, including the need to compare polytherapy with types of primary mood stabilizer.
Dr. Baldessarini has recently served as a consultant for, or carried out investigator-initiated research supported with funds from, Auritec Pharmaceuticals, Biotrofix, Luitpold Pharmaceuticals, Eli Lilly and Company, IFI SpA, Janssen Pharmaceuticals, JDS-Noven Pharmaceuticals, NeuroHealing Pharmaceuticals, Novartis Pharmaceuticals, and Solvay Pharmaceuticals. The other authors report no competing interests.