- Email: [email protected]
Treatment Patterns, HealthCare Resource Utilization and Costs in Patients with Bipolar Disorder, Newly Treated with Extended Release or Immediate Release Quetiapine Fumarate using US Healthcare Administrative Claims Data
Clinical Therapeutics/Volume 35, Number 12, 2013
Treatment Patterns, HealthCare Resource Utilization and Costs in Patients with Bipolar Disorder, Newly Treated with Extended Release or Immediate Release Quetiapine Fumarate using US Healthcare Administrative Claims Data Julie C. Locklear, PharmD, MBA1,*; Berhanu Alemayehu, DrPH1; Robert S. Brody, MPH1; Soheil Chavoshi, MS1; Ozgur Tunceli, PhD2; David Kern, MS2; and Willie R. Earley, MD1,* 1
AstraZeneca Pharmaceuticals LP, Wilmington, Delaware; and 2HealthCore Inc, Wilmington, Delaware
ABSTRACT Background: Differences in treatment patterns, health care resource use, and costs are expected among patients newly treated with quetiapine extended release (XR) or quetiapine immediate release (IR). Objective: To compare treatment patterns, health care resource use, and costs in patients with bipolar disorder newly treated with quetiapine XR or quetiapine IR. Methods: This was an observational, retrospective cohort study that used HealthCore Integrated Research Database–identified patients (age range, 18-64 years) with an International Classification of Disease, Ninth Revision diagnosis of bipolar disorder and Z1 pharmacy claim for quetiapine XR or quetiapine IR between October 2, 2008, and July 31, 2010. Outcomes were as follows: patient characteristics at the index date (first claim for quetiapine XR or quetiapine IR); 12-month preindex clinical characteristics, health care resource use, and costs; and 12-month postindex treatment patterns, health care resource use, and costs, assessed using generalized linear models (adjusted for index date and preindex patient demographic characteristics, clinical characteristics, health care resource use, and costs). Results: In total, 3049 patients with bipolar disorder were analyzed (651 in the quetiapine XR group and 2398 in the quetiapine IR group). Of patients initiating treatment with quetiapine XR, 8.8% had no change in or discontinuation of their index therapy compared with 5.7% of patients treated with quetiapine IR (adjusted odds ratio, 1.44; 95% confidence interval, 1.03-2.00; P ¼ 0.0317). The average daily dose (adjusted mean) of quetiapine XR was higher
December 2013
than quetiapine IR (225 vs 175 mg/d, P o 0.0001). An average daily dose of 300 to 800 mg was reached sooner (15.6 vs 30.8 days, P ¼ 0.0049) and in more patients (44.2% vs 27.2%, P o 0.0001) who were taking quetiapine XR compared with patients taking quetiapine IR. No differences in total health care costs were found between the cohorts; however, patients taking quetiapine XR were less likely to be hospitalized for mental health–related reasons (12.1% vs 18.3%, P ¼ 0.0022) and incurred lower mental health–related costs (US $6686 vs US $7577, P ¼ 0.0063) compared with patients taking quetiapine IR. Conclusions: Treatment patterns and dosing differ in patients with bipolar disorder treated with quetiapine XR compared with those treated with quetiapine IR. Mental health–related hospitalizations and costs may be reduced in the 12 months after patients initiating treatment with quetiapine XR compared with initiating treatment with quetiapine IR. (Clin Ther. 2013;35:1923–1932) & 2013 Elsevier HS Journals, Inc. All rights reserved. Key words: bipolar disorder, health care use, quetiapine IR, quetiapine XR.
INTRODUCTION Bipolar disorder is a lifelong disorder characterized by episodes of hypomania or mania and depression that affects 4.4% of the US population.1,2 Patients with *Former employees of AstraZeneca Accepted for publication October 12, 2013. http://dx.doi.org/10.1016/j.clinthera.2013.10.005 0149-2918/$ - see front matter & 2013 Elsevier HS Journals, Inc. All rights reserved.
1923
Clinical Therapeutics bipolar disorder have high rates of morbidity and disability at all stages of illness, resulting in substantial medical costs.3 Indeed, patients with bipolar disorder use 3 to 4 times as many health care resources as patients without bipolar disorder, primarily because of increased hospitalizations.4 In 2005, the total national expenditure for mental health services in the United States was US $112.8 billion, with US $74.4 billion (66.0%) allocated to mental health service provision, including hospitalization, and US $30.0 billion (26.6%) associated with the cost for retail prescription drugs.5 The estimated direct and indirect costs of bipolar disorder in the United States in 2009 were US $30.7 billion and US $120.3 billion, respectively.6 Randomized controlled trials (RCTs) may be of limited use in measuring health care resource use and costs because of the relatively short exposure to treatment, controlled environment, and limited patient numbers.7 Observational studies, including administrative data, are an important supplement to RCTs, providing data from real-world clinical practice. Quetiapine fumarate is an atypical antipsychotic available in a once-daily extended release (XR) formulation or an immediate release (IR) formulation for the short-term treatment of manic or depressive episodes associated with bipolar disorder and for maintenance treatment of bipolar disorder.8–21 In bipolar mania and maintenance treatment of bipolar disorder, the recommended dosage range of quetiapine is 400 to 800 mg/d. In bipolar depression, the recommended dosage of quetiapine is 300 mg/d.8,9 Recent retrospective analyses of US hospital administrative data suggest that inpatient use of quetiapine XR in patients with bipolar disorder may be associated with reduced length (up to 11%) and cost (up to 10%) of hospitalization.22,23 Therefore, we hypothesized that there were differences in treatment patterns, health care resource use, and costs among newly treated patients who initiated treatment with quetiapine XR or quetiapine IR. The objective of this study was to compare treatment patterns, health care resource use, and costs in patients with bipolar disorder newly treated with quetiapine XR or quetiapine IR.
HealthCore Integrated Research Database, which contains longitudinal claims data from health plans in the Northeastern, Midwestern, Southern, and Western regions of the United States. HealthCore accesses these data in a manner that complies with federal and state laws and regulations, including those related to privacy and security of individually identifiable health information. The same population was used for all analyses, including all patients who met the inclusion and exclusion criteria. Eligible patients were aged 18 to 64 years, were enrolled in a health plan with Z1 pharmacy claim for quetiapine XR or quetiapine IR between October 2, 2008, and July 31, 2010, and had Z1 International Classification of Disease, Ninth Revision (ICD-9) diagnosis code for bipolar disorder (296.0x, 296.1x, 296.4x, 296.5x, 296.6x, 296.7x, 296.8x) on or before the index date (first pharmacy claim for quetiapine XR or quetiapine IR). Patients were excluded if they had o12 months of baseline health plan enrollment before or after the index date; an ICD-9 diagnosis code for schizophrenia (295.0x–295.6x, 295.8x, 295.9x) on the same date as or after the most recent bipolar diagnosis, on or before the index date; a prescription for any atypical antipsychotic, including quetiapine XR or quetiapine IR, during the 6-month preindex period; or a prescription for both quetiapine XR and quetiapine IR on the index date. The treatment cohorts were defined according to initiation of quetiapine XR or quetiapine IR on the index date, regardless of prescription patterns after the initial prescription. Patients were followed up for a 24-month period. All patients had data available for at least 12 months before the index date and 12 months after initiation of quetiapine XR or quetiapine IR treatment, ensuring equal follow-up time for all patients. Treatment patterns (including dosing), health care resource use, and costs were analyzed during the 12-month postindex period. Health care resource use included inpatient visits, emergency department (ED) visits, and office or outpatient visits. Health care costs included plan-paid and patient-paid inpatient visit costs, ED visit costs, outpatient visit costs, pharmacy costs, and costs for other medical services not classified elsewhere.
PATIENTS AND METHODS Study Design and Population
Statistical Analysis
his was an observational, retrospective cohort study that used administrative claims data from the
Logistic regression was used to assess 12-month postindex treatment patterns between the 2 cohorts.
1924
Volume 35 Number 12
J.C. Locklear et al.
Table I. Demographic and clinical characteristics of patients with bipolar disorder during the 12-month period before treatment with quetiapine XR or quetiapine IR.* No. (%) of Patients† Quetiapine XR Quetiapine IR (n ¼ 651) (n ¼ 2398)
Characteristic Demographic characteristics Age, mean (SD), y Sex Male Female Physician specialty of psychiatry Geographic location Northeast Midwest South West Common mental health–related comorbidities (incidence 410% patients) Depression Anxiety disorder Drug or alcohol abuse Headache or tension headache Insomnia Adjustment disorder Common non–mental health–related comorbidities (incidence 410% patients) Respiratory disorders Other pain Back pain Hypertension Dyslipidemia Neck pain
39.0 (12.3)
39.3 (12.6)
259 (39.8) 392 (60.2) 332 (51.0)
953 (39.7) 1445 (60.3) 1247 (52.0)
85 235 188 143
(13.1) (36.1) (28.9) (22.0)
407 586 711 694
(17.0) (24.4) (29.6) (28.9)
328 233 181 104 97 83
(50.4) (35.8) (27.8) (16.0) (14.9) (12.7)
1243 748 775 415 313 309
(51.8) (31.2) (32.3) (17.3) (13.1) (12.9)
348 277 180 161 159 67
(53.5) (42.5) (27.6) (24.7) (24.4) (10.3)
1179 1113 656 554 552 336
(49.2) (46.4) (27.4) (23.1) (23.0) (14.0)
*IR ¼ immediate release; XR ¼ extended release. † Data are reported as number (percentage) of patients unless otherwise stated.
Odds ratios (ORs) and 95% CIs were reported. The variables used in the adjusted analyses were age, sex, geographic region, prescriber specialty, preindex comorbidities, and Deyo-Charlson comorbidity index (DCI) score.24 Cox proportional hazard models were used for the survival analyses of time to treatment change during the 12-month postindex treatment period. Hazard ratios and 95% CIs were reported. For the continuous measures of average daily dose (ADD) during the entire postindex period, time from index to optimal ADD (300–800 mg), and the
December 2013
proportion of prescriptions within the optimal ADD, an ANOVA model was used, and mean difference and 95% CI were reported. A generalized linear model with a negative binomial distribution and logarithmic-link function was used to assess 12-month postindex health care resource use. A generalized linear model with gamma distribution was used for the analysis of health care costs. Models were adjusted a priori for multiple confounding variables, including age, sex, geographic region, prescriber specialty, anxiety disorder, transient psychosis, drug
1925
Clinical Therapeutics
Table II. Health care resource use and costs for patients with bipolar disorder during the 12-month period before treatment with quetiapine XR or quetiapine IR.* No. (%) of Patients† Variable Health care resource use Z1 All-cause inpatient visits Z1 Mental health inpatient visits‡ Z1 All-cause ED* visits Z1 Mental health ED* visits§ Z1 All-cause office and outpatient visits Z1 Mental health office and outpatient visits§ Health care costs, mean (SD), US $ All-cause total health care costs Mental health–related total health care costs‡
Quetiapine XR (n ¼ 651) 212 163 235 114 646 616
(32.6) (25.0) (36.1) (17.5) (99.2) (94.6)
14,133 (30,652) 6700 (25,781)
Quetiapine IR (n ¼ 2398) 932 757 908 475 2366 2223
(38.9) (31.6) (37.9) (19.8) (98.7) (92.7)
16,647 (32,311) 8516 (20,999)
*ED ¼ emergency department; IR ¼ immediate release; XR ¼ extended release. † Data are reported as number (percentage) of patients unless otherwise stated. ‡ Mental health–related inpatient visits and costs defined as having a primary diagnosis that is mental health-related. § Mental health–related visits defined as having a mental health–related diagnosis in any position (primary or secondary).
or alcohol abuse, cardiovascular problems, pain, osteoarthritis, obesity, respiratory disorders, DCI index score, anticonvulsants, selective serotonin reuptake inhibitors or serotonin norepinephrine reuptake inhibitors, other central nervous system agents, and the corresponding preindex health care resource use and cost. All health care use and cost analyses were conducted separately for all-cause and mental health–related causes. Mental health–related inpatient visits and costs were defined as those relating to an ICD-9 code for a mental health indication as the primary diagnosis. All statistical analyses were performed with SAS software, version 9.2 (SAS Institute, Cary, North Carolina).
RESULTS Between October 2, 2008, and July 31, 2010, a total of 3049 patients with bipolar disorder were eligible for the current analysis. Index pharmacy claims for 79% of these patients were for quetiapine IR (Table I).
Index Date and Preindex Description of Patient Demographic Characteristics, Clinical Characteristics, Health Care Resource Use, and Costs Demographic characteristics, comorbid conditions, health care use, and costs for patients in the 12 months
1926
before initiating treatment with quetiapine XR or quetiapine IR are presented in Tables I and II. The rate of comorbid anxiety disorder was higher for the group who received quetiapine XR compared with those who received quetiapine IR (35.8% vs 31.2%), whereas drug or alcohol abuse was lower in the quetiapine XR group compared with the quetiapine IR group (27.8% vs 32.3%) (Table I). Other mental health–related comorbid conditions were similar between the 2 cohorts. In addition, no major differences were found between the 2 cohorts with respect to non–mental health–related comorbidities, except for neck pain, which was lower in the quetiapine XR group compared with the group initiating treatment with quetiapine IR (10.3% vs 14%) (Table I). All-cause and mental health–related health care resource use and costs were lower for the quetiapine XR group compared with the quetiapine IR group, respectively (Table II).
Postindex Analyses of Treatment Patterns, Health Care Resource Use, and Costs Treatment Patterns In total, 8.8% of patients treated with quetiapine XR had continuous treatment with no change or discontinuation of their index therapy, compared with 5.7% of those treated with quetiapine IR (adjusted
Volume 35 Number 12
J.C. Locklear et al.
Table III. Treatment patterns in the 12-month postindex period for patients with bipolar disorder treated with quetiapine XR or quetiapine IR.* No. (%) of Patients Treatment Pattern No treatment change or discontinuation (continuous treatment) Discontinuation of index medication Treatment change (nondiscontinuation) Switch from index medication to any atypical antipsychotic or add another atypical antipsychotic Switch from index medication to any conventional antipsychotic or add another conventional antipsychotic Other treatment change§
Quetiapine XR (n ¼ 651)
Quetiapine IR (n ¼ 2398)
Adjusted OR† (95% CI*)
P‡
57
8.8
137
5.7
1.44 (1.03–2.00)
0.0317
431 163 101
66.2 25.0 15.5
1685 576 275
70.3 24.0 11.5
0.82 (0.68–0.99) 1.11 (0.90–1.36) 1.44 (1.12–1.85)
0.0370 0.3235 0.0047
58
8.9
291
12.1
0.75 (0.56–1.02)
0.0651
7
1.1
11
0.5
2.90 (1.09–7.72)
0.0328
IR ¼ immediate release; OR ¼ odds ratio; XR ¼ extended release. *Confidence interval. † Adjusted OR from logistic model controlling for the following covariates: age, sex, geographic region, prescriber specialty, comorbidities (anxiety disorder, transient psychosis, drug or alcohol abuse, or neck pain), and Deyo-Charlson comorbidity score. ‡ P values are not adjusted for multiple comparisons. § Patients who filled multiple types of medications on the same treatment change date. For example, patients filling a conventional antipsychotic and switching from index medication on the same date are captured here.
OR, 1.44; 95% CI, 1.03–2.00; P ¼ 0.0317) (Table III). In total, 66.2% of patients treated with quetiapine XR discontinued their index therapy compared with 70.3% of those treated with quetiapine IR (adjusted OR, 0.82; 95% CI, 0.68–0.99; P ¼ 0.0370) (Table III). Overall, 25.0% and 24.0% of patients experienced a treatment change in the 12 months after initiation of treatment with quetiapine XR or quetiapine IR, respectively. In total, 15.5% of patients treated with quetiapine XR switched to an atypical antipsychotic or added another atypical antipsychotic compared with 11.5% of those treated with quetiapine IR (adjusted OR, 1.44; 95% CI, 1.12–1.85; P ¼ 0.0047), and there was a trend toward patients treated with quetiapine XR being less likely to switch to a conventional antipsychotic or add a conventional antipsychotic compared with quetiapine IR (8.9% vs 12.1%; adjusted OR, 0.75; 95% CI, 0.56–1.02; P ¼ 0.0651) (Table III). Furthermore, the ADD (adjusted mean) of quetiapine XR was higher than that of quetiapine IR (225 vs
December 2013
175 mg/d, P o 0.0001). An ADD of 300 to 800 mg was reached sooner (adjusted mean, 15.6 vs 30.8 days; P ¼ 0.0049) and in more patients (44.2% vs 27.2%, P o 0.0001) treated with quetiapine XR compared with quetiapine IR (Table IV).
Health Care Resource Use and Costs No difference was found in overall health care resource use between cohorts when considering the proportion of patients with inpatient visits (21.7% vs 25.5%, P ¼ 0.1260), ED visits (31.6% vs 33.8%, P ¼ 0.3849), and outpatient or office visits (98.2% vs 98.6%, P ¼ 0.2743) in patients initiating treatment with quetiapine XR or quetiapine IR, respectively (Figure 1). However, patients initiating treatment with quetiapine XR were less likely to be hospitalized for mental health–related reasons (12.1% vs 18.3%, P ¼ 0.0022) compared with patients taking quetiapine IR (Figure 1). The distribution of mental health–related costs in patients treated with quetiapine XR and quetiapine IR
1927
Clinical Therapeutics
Table IV. Dosing of index medication in the 12-month postindex period for patients with bipolar disorder treated with quetiapine XR or quetiapine IR.*
Variable ADD*,∥ during the entire postindex period, mg Time to 300–800 mg optimal ADD, d Total proportion of prescriptions that are 300–800 mg Patients filling at least one prescription with ADD (300–800 mg) at any time, No. (%) Patients whose most frequently prescribed ADD is between 300 and 800 mg, No. (%) Z50% of prescriptions between 300 and 800 mg, No. (%)
Quetiapine XR (n ¼ 651)
Quetiapine IR (n ¼ 2398)
Mean† (SD)
LS* Mean‡
Mean† (SD)
LS Mean‡
Adjusted OR* or Adjusted Mean Difference‡
P§
216 (187)
225
164 (155)
175
50.8
o0.0001
15.2 0.15
0.0049 o0.0001
24.0 (66.0) 15.6 39.6 (79.1) 30.8 0.36 (0.44) 0.39 0.20 (0.36) 0.23 288 (44.2) NA*
652 (27.2)
NA
2.09
o0.0001
251 (38.6)
NA
521 (21.7)
NA
2.22
o0.0001
245 (37.6)
NA
505 (21.1)
NA
2.23
o0.0001
*ADD ¼ average daily dose; IR ¼ immediate release; LS ¼ least squares; NA ¼ not applicable; OR ¼ odds ratio; XR ¼ extended release. † Data are reported as mean (SD) unless otherwise stated. ‡ Difference in LS means (for ADD and time to optimal ADD) and proportion of patients (for all other variables); adjusted OR and adjusted mean differences control for the following covariates: age (continuous), sex, geographic region, prescriber specialty, statistically significant comorbidities, and Deyo-Charlson comorbidity score. § P values are not adjusted for multiple comparisons. ∥ ADD ¼ (quantity dispensed/days’ supply) strength.
is shown in Figure 2. Adjusting for multiple confounding variables, no differences were found in mean overall health care costs between cohorts; however, patients initiating treatment with quetiapine XR incurred lower mental health–related adjusted mean costs (US $6686 vs US $7577, P ¼ 0.0063) compared with those taking quetiapine IR, which may be due to fewer hospitalizations or shorter duration of hospitalizations for mental health–related reasons (Figure 3).
DISCUSSION Retrospective analyses of the HealthCore Integrated Research Database suggest that treatment patterns and dosing differ in patients with bipolar disorder treated with quetiapine XR compared with those treated with quetiapine IR, supporting the need for
1928
physicians to have access to both formulations so that treatment decisions can be made based on individual patient needs. Mental health–related hospitalizations and costs were reduced in the 12 months after patients initiated treatment with quetiapine XR compared with quetiapine IR. A reduction in mental health–related costs associated with quetiapine XR versus quetiapine IR was estimated in this analysis to be a saving of US $891 (12%) per patient. Bipolar disorder is associated with high medical costs, partially due to costly hospitalizations.4 A study comparing the relative risk for hospitalization in patients with bipolar disorder receiving antipsychotic medication suggested that 7% to 10% of treatment episodes require hospitalization, with a mean cost of US $14,500 per inpatient stay.25 Previous studies
Volume 35 Number 12
J.C. Locklear et al.
P = 0.2743 100
98.2 98.6
Quetiapine XR Quetiapine IR
Patients (%)
80
comparing hospitalization in patients treated with antipsychotics have reported lower annual risk and reduced length of hospitalization in patients treated with quetiapine IR for bipolar disorder or schizophrenia, compared with risperidone, olanzapine, ziprasidone, aripiprazole, and haloperidol.25,26
60 40 20
P = 0.1260 25.5 21.7
P = 0.3849 31.6 33.8
40
Quetiapine XR
35
All-cause inpatient visits
All-cause ED visits
All-cause office/outpatient visits P = 0.4608
15
5
80
00 ,0 90
00 ,0 75
00 ,0 60
00 ,0 45
,0
,0 15
30
0 60
00
0
Mental health-related inpatient costs (US$) 40 P = 0.7397 17.4 18.6
Mental healthrelated* inpatient visits
Mental healthrelated† ED visits
35 30 25 20 15 10 5
,0 00 90
,0 00 75
,0 00 60
,0 00 45
30
,0 00
0 ,0 00
Mental healthrelated† office/ outpatient visits
Figure 1. Postindex all-cause (A) and mental health–related (B) adjusted health care resource use in patients with bipolar disorder treated with quetiapine extended release (XR) or quetiapine immediate release (IR). All-cause and mental health–related resource use (inpatient, emergency department [ED], or office or outpatient visits) are adjusted for baseline demographic, clinical, and use variables as outlined in the Methods section. *Mental health–related visits defined as having a primary diagnosis that is mental health related. †Mental health–related visits defined as having a mental health–related diagnosis in any position (primary or secondary).
December 2013
Quetiapine IR
40
15
0
P = 0.0022 18.3 12.1
0
20
Patients (%)
Patients (%)
20
10
94.5 94.5
Quetiapine XR Quetiapine IR
25
00
100
Patients (%)
30 0
Mental health-related inpatient costs (US$)
Figure 2. Distribution of mental health–related inpatient costs in patients with bipolar disorder treated with quetiapine extended release (XR) (A) or quetiapine immediate release (IR) (B). Raw data excluding US $0 and capping mental health–related cost at US $100,000 (1 patient treated with quetiapine XR and 8 patients treated with quetiapine IR were excluded).
1929
Clinical Therapeutics
15,000
P = 0.6059 13,897 14,205
Quetiapine XR Quetiapine IR
Mean cost (US$)
12,000 P = 0.0063 7577 6686
9000 6000 3000 0
All-cause costs
Mental health-related costs
Figure 3. Adjusted all-cause and mental health– related health care costs during the 12-month postindex date period in patients with bipolar disorder treated with quetiapine extended release (XR) or quetiapine immediate release (IR). All-cause and mental health–related resource use (inpatient, emergency department, or office/outpatient visits) are adjusted for baseline demographic, clinical, and use variables as outlined in the Methods section.
Differences in the patient populations treated with quetiapine XR or quetiapine IR are to be expected in a real-world setting and highlight that access to a broad range of therapeutic agents is essential to treat patients with bipolar disorder effectively and according to individual patient needs.27 Retrospective analysis of a Finnish hospital database reported that treatment patterns differed between those patients with bipolar disorder treated with quetiapine XR and those treated with quetiapine IR.28 The mean daily dose of quetiapine XR was significantly higher than that of quetiapine IR (466 vs 308 mg, P ¼ 0.009), and quetiapine XR was used with antiepileptic medication to a greater extent than quetiapine IR (53.3% vs 14.3%, P ¼ 0.03).28 Generic versions of quetiapine IR became available in 2012, which will result in lower drug costs in patients for whom treatment with quetiapine IR is appropriate. However, because treatment patterns and dosing differ in patients with bipolar disorder treated with quetiapine XR compared with quetiapine IR, quetiapine XR may provide an alternative
1930
treatment option to quetiapine IR and its generic formulations at an acceptable cost, enabling treatment decisions to be based on individual patient needs. Clinical practice data are an important supplement to RCTs. This study uses real-world data from the HealthCore database, which represents claims information from the largest commercially insured population in the United States. The study analyzed claims data from a large (n 4 3000) population of patients with bipolar disorder and provides information on end points that are important to payers, including health care use, cost, and compliance information. Because this study was a retrospective database analysis, it is potentially associated with a number of methodologic limitations related to nonrandomization, and because statistical tests were not adjusted for multiple comparisons, P values should be viewed with caution. Despite these limitations, the generalized linear model analyses controlled for known potential confounding effects collected during this study; however, this study did not adjust for unknown and uncollected variables. Far fewer patients were treated with quetiapine XR (n ¼ 651) than quetiapine IR (n ¼ 2398), and the study does not account for sample imbalances. An additional limitation of the study is that multiple statistical tests were performed without adjusting for multiple comparisons. Furthermore, in the 12 months after initiation of their index therapy, only 8.8% of patients had continuous treatment with quetiapine XR, and 5.7% of patients had continuous treatment with quetiapine IR. Therefore, it is likely that most antipsychotic treatment costs in the 12 months after initiation of treatment with quetiapine XR or quetiapine IR are not attributable to the index therapy. There are also limitations to analyzing cost data where the distributions have large zero values and outliers. This is the case for mental health–related costs as shown in Figure 2. The methods prespecified and presented in this study were selected to manage the zero-cost values and skewed distribution when calculating mean costs. However, we acknowledge that alternative methods are available, and sensitivity analyses would need to be performed to increase the level of confidence in the model used in this study and the robustness of the results. Therefore, this is a potential limitation with regards to interpreting the cost data analyses.
Volume 35 Number 12
J.C. Locklear et al.
CONCLUSION This study provides evidence that in the real-world management of bipolar disorder, treatment patterns and dosing during a 12-month period differ with quetiapine XR compared with quetiapine IR. Taken together, these results support the need for physicians to have access to both formulations so that they can make treatment decisions based on individual patients’ needs.
ACKNOWLEDGEMENTS This study was sponsored by AstraZeneca Pharmaceuticals. The authors would like to thank Claire Chadwick, of Complete Clarity, who provided medical writing support funded by AstraZeneca. All authors were involved in analysis and interpretation of data. All authors contributed to and critically reviewed the manuscript at all stages of development, and all authors approved the final manuscript.
CONFLICTS OF INTEREST Drs Locklear, Alemayehu, and Earley were employees of AstraZeneca when this analysis was conceived, conducted, and completed. Messrs Brody and Chavoshi are employees of AstraZeneca. Dr Tunceli and Mr Kern are employees of HealthCore Inc, which received funding from AstraZeneca for this research. The authors have indicated that they have no other conflicts of interest regarding the content of this article.
REFERENCES 1. Merikangas KR, Akiskal HS, Angst J, et al. Lifetime and 12-month prevalence of bipolar spectrum disorder in the national comorbidity survey replication. Arch Gen Psychiatry. 2007;64:543–552. 2. Saunders KE, Goodwin GM. New approaches in the treatment of bipolar depression. Curr Top Behav Neurosci. 2013;14:291–307. 3. Baldessarini RJ, Leahy L, Arcona S, et al. Patterns of psychotropic drug prescription for U.S. patients with diagnoses of bipolar disorders. Psychiatr Serv. 2007;58:85–91. 4. Bryant-Comstock L, Stender M, Devercelli G. Health care utilization and costs among privately insured patients with bipolar I disorder. Bipolar Disord. 2002;4:398–405. 5. National Center for Health Statistics. Health, United States, 2010. http://www.cdc.gov/nchs/data/hus/2010/ 128.pdf. Accessed November 16, 2012.
December 2013
6. Dilsaver SC. An estimate of the minimum economic burden of bipolar I and II disorders in the United States: 2009. J Affect Disord. 2011;129:79–83. 7. Garrison LP Jr, Neumann PJ, Erickson P, et al. Using realworld data for coverage and payment decisions: the ISPOR Real-World Data Task Force report. Value Health. 2007;10:326–335. 8. AstraZeneca. Seroquel (quetiapine fumarate) - US prescribing information, 2011. http://www1.astrazeneca-us. com/pi/seroquel.pdf. Accessed July 9, 2012. 9. AstraZeneca. Seroquel XR (quetiapine fumarate) extendedrelease tablets/US prescribing information, 2012. http:// www1.astrazeneca-us.com/pi/seroquelxr.pdf. Accessed July 9, 2012. 10. Calabrese JR, Keck PE Jr, Macfadden W, et al. A randomized, double-blind, placebo-controlled trial of quetiapine in the treatment of bipolar I or II depression. Am J Psychiatry. 2005;162:1351–1360. 11. Cutler AJ, Datto C, Nordenhem A, et al. Extended-release quetiapine as monotherapy for the treatment of adults with acute mania: a randomized, double-blind, 3-week trial. Clin Ther. 2011;33:1643–1658. 12. McElroy SL, Weisler RH, Chang W, et al. A double-blind, placebo-controlled study of quetiapine and paroxetine as monotherapy in adults with bipolar depression (EMBOLDEN II). J Clin Psychiatry. 2010;71:163–174. 13. Sachs G, Chengappa KN, Suppes T, et al. Quetiapine with lithium or divalproex for the treatment of bipolar mania: a randomized, double-blind, placebo-controlled study. Bipolar Disord. 2004;6:213–223. 14. Suppes T, Vieta E, Liu S, et al. Maintenance treatment for patients with bipolar I disorder: results from a North American study of quetiapine in combination with lithium or divalproex (trial 127). Am J Psychiatry. 2009;166:476– 488. 15. Suppes T, Datto C, Minkwitz M, et al. Effectiveness of the extended release formulation of quetiapine as monotherapy for the treatment of acute bipolar depression. J Affect Disord. 2010;121:106–115. 16. Thase M, Macfadden W, McCoy R, et al. Efficacy of quetiapine monotherapy in bipolar depression: a confirmatory double-blind, placebo-controlled study (The BOLDER II Study). Int J Neuropsychopharmacol. 2006;9 (Suppl 1):S237. Abstract P03.035. 17. Vieta E, Mullen J, Brecher M, et al. Quetiapine monotherapy for mania associated with bipolar disorder: combined analysis of two international, double-blind, randomised, placebo-controlled studies. Curr Med Res Opin. 2005;21:923–934. 18. Vieta E, Suppes T, Eggens I, et al. Efficacy and safety of quetiapine in combination with lithium or divalproex for maintenance of patients with bipolar I disorder (international trial 126). J Affect Disord. 2008;109:251–263.
1931
Clinical Therapeutics 19. Weisler RH, Nolen WA, Neijber A, et al. Continuation of quetiapine versus switching to placebo or lithium for maintenance treatment of bipolar I disorder (Trial 144: a randomized controlled study). J Clin Psychiatry. 2011;72:1452–1464. 20. Yatham LN, Paulsson B, Mullen J, Vågerö M. Quetiapine versus placebo in combination with lithium or divalproex for the treatment of bipolar mania. J Clin Psychopharmacol. 2004;24:599–606. 21. Young AH, McElroy SL, Bauer M, et al. A double-blind, placebocontrolled study of quetiapine and lithium monotherapy in adults in the acute phase of bipolar depression (EMBOLDEN I). J Clin Psychiatry. 2010;71:150–162. 22. Bonafede M, Locklear J, Wahlqvist P, et al. The impact of once-daily extended-release quetiapine fumarate (quetiapine XR) on length of hospitalisation of patients with acute bipolar mania. Value Health. 2011:14. Abstract A300. 23. Locklear J, Wahlqvist P, Gustafsson U, et al. The impact of once-daily extended release quetiapine fumarate (quetiapine XR) on length and costs of hospitalisation of patients with schizophrenia or bipolar disorder. Value Health. 2011:14. Abstract A299. 24. Deyo RA, Cherkin DC, Ciol MA. Adapting a clinical comorbidity index for use with ICD-9-CM administrative databases. J Clin Epidemiol. 1992;45:613–619. 25. Gianfrancesco F, Rajagopalan K, Goldberg JF, Wang RH. Hospitalization risks in the treatment of bipolar disorder: comparison of antipsychotic medications. Bipolar Disord. 2007;9:252–261. 26. Strutton DR, Gutierrez B, Blanchette CM. Influence of medication treatment patterns on resource use for patients hospitalized for schizophrenia. Am J Health Syst Pharm. 2008;65: 42–48. 27. Altamura AC, Armadoros D, Jaeger M, et al. Importance of open access
1932
to atypical antipsychotics for the treatment of schizophrenia and bipolar disorder: a European perspective. Curr Med Res Opin. 2008;24:2271–2282. 28. Hallinen T, Soini EJ, Granström O, et al. Differential use of extended and immediate release quetiapine: a retrospective registry study of Finnish inpatients with schizophrenia spectrum and bipolar disorders. BMJ Open. 2012;2:e000915.
Address correspondence to: Robert S. Brody, MPH, AstraZeneca LP, 1800 Concord Pike, PO Box 15437, Wilmington, DE, 19807-5437. E-mail: Bob. [email protected]
Volume 35 Number 12
Treatment Patterns, HealthCare Resource Utilization and Costs in Patients with Bipolar Disorder, Newly Treated with Extended Release or Immediate Release Quetiapine Fumarate using US Healthcare Administrative Claims Data Julie C. Locklear, PharmD, MBA1,*; Berhanu Alemayehu, DrPH1; Robert S. Brody, MPH1; Soheil Chavoshi, MS1; Ozgur Tunceli, PhD2; David Kern, MS2; and Willie R. Earley, MD1,* 1
AstraZeneca Pharmaceuticals LP, Wilmington, Delaware; and 2HealthCore Inc, Wilmington, Delaware
ABSTRACT Background: Differences in treatment patterns, health care resource use, and costs are expected among patients newly treated with quetiapine extended release (XR) or quetiapine immediate release (IR). Objective: To compare treatment patterns, health care resource use, and costs in patients with bipolar disorder newly treated with quetiapine XR or quetiapine IR. Methods: This was an observational, retrospective cohort study that used HealthCore Integrated Research Database–identified patients (age range, 18-64 years) with an International Classification of Disease, Ninth Revision diagnosis of bipolar disorder and Z1 pharmacy claim for quetiapine XR or quetiapine IR between October 2, 2008, and July 31, 2010. Outcomes were as follows: patient characteristics at the index date (first claim for quetiapine XR or quetiapine IR); 12-month preindex clinical characteristics, health care resource use, and costs; and 12-month postindex treatment patterns, health care resource use, and costs, assessed using generalized linear models (adjusted for index date and preindex patient demographic characteristics, clinical characteristics, health care resource use, and costs). Results: In total, 3049 patients with bipolar disorder were analyzed (651 in the quetiapine XR group and 2398 in the quetiapine IR group). Of patients initiating treatment with quetiapine XR, 8.8% had no change in or discontinuation of their index therapy compared with 5.7% of patients treated with quetiapine IR (adjusted odds ratio, 1.44; 95% confidence interval, 1.03-2.00; P ¼ 0.0317). The average daily dose (adjusted mean) of quetiapine XR was higher
December 2013
than quetiapine IR (225 vs 175 mg/d, P o 0.0001). An average daily dose of 300 to 800 mg was reached sooner (15.6 vs 30.8 days, P ¼ 0.0049) and in more patients (44.2% vs 27.2%, P o 0.0001) who were taking quetiapine XR compared with patients taking quetiapine IR. No differences in total health care costs were found between the cohorts; however, patients taking quetiapine XR were less likely to be hospitalized for mental health–related reasons (12.1% vs 18.3%, P ¼ 0.0022) and incurred lower mental health–related costs (US $6686 vs US $7577, P ¼ 0.0063) compared with patients taking quetiapine IR. Conclusions: Treatment patterns and dosing differ in patients with bipolar disorder treated with quetiapine XR compared with those treated with quetiapine IR. Mental health–related hospitalizations and costs may be reduced in the 12 months after patients initiating treatment with quetiapine XR compared with initiating treatment with quetiapine IR. (Clin Ther. 2013;35:1923–1932) & 2013 Elsevier HS Journals, Inc. All rights reserved. Key words: bipolar disorder, health care use, quetiapine IR, quetiapine XR.
INTRODUCTION Bipolar disorder is a lifelong disorder characterized by episodes of hypomania or mania and depression that affects 4.4% of the US population.1,2 Patients with *Former employees of AstraZeneca Accepted for publication October 12, 2013. http://dx.doi.org/10.1016/j.clinthera.2013.10.005 0149-2918/$ - see front matter & 2013 Elsevier HS Journals, Inc. All rights reserved.
1923
Clinical Therapeutics bipolar disorder have high rates of morbidity and disability at all stages of illness, resulting in substantial medical costs.3 Indeed, patients with bipolar disorder use 3 to 4 times as many health care resources as patients without bipolar disorder, primarily because of increased hospitalizations.4 In 2005, the total national expenditure for mental health services in the United States was US $112.8 billion, with US $74.4 billion (66.0%) allocated to mental health service provision, including hospitalization, and US $30.0 billion (26.6%) associated with the cost for retail prescription drugs.5 The estimated direct and indirect costs of bipolar disorder in the United States in 2009 were US $30.7 billion and US $120.3 billion, respectively.6 Randomized controlled trials (RCTs) may be of limited use in measuring health care resource use and costs because of the relatively short exposure to treatment, controlled environment, and limited patient numbers.7 Observational studies, including administrative data, are an important supplement to RCTs, providing data from real-world clinical practice. Quetiapine fumarate is an atypical antipsychotic available in a once-daily extended release (XR) formulation or an immediate release (IR) formulation for the short-term treatment of manic or depressive episodes associated with bipolar disorder and for maintenance treatment of bipolar disorder.8–21 In bipolar mania and maintenance treatment of bipolar disorder, the recommended dosage range of quetiapine is 400 to 800 mg/d. In bipolar depression, the recommended dosage of quetiapine is 300 mg/d.8,9 Recent retrospective analyses of US hospital administrative data suggest that inpatient use of quetiapine XR in patients with bipolar disorder may be associated with reduced length (up to 11%) and cost (up to 10%) of hospitalization.22,23 Therefore, we hypothesized that there were differences in treatment patterns, health care resource use, and costs among newly treated patients who initiated treatment with quetiapine XR or quetiapine IR. The objective of this study was to compare treatment patterns, health care resource use, and costs in patients with bipolar disorder newly treated with quetiapine XR or quetiapine IR.
HealthCore Integrated Research Database, which contains longitudinal claims data from health plans in the Northeastern, Midwestern, Southern, and Western regions of the United States. HealthCore accesses these data in a manner that complies with federal and state laws and regulations, including those related to privacy and security of individually identifiable health information. The same population was used for all analyses, including all patients who met the inclusion and exclusion criteria. Eligible patients were aged 18 to 64 years, were enrolled in a health plan with Z1 pharmacy claim for quetiapine XR or quetiapine IR between October 2, 2008, and July 31, 2010, and had Z1 International Classification of Disease, Ninth Revision (ICD-9) diagnosis code for bipolar disorder (296.0x, 296.1x, 296.4x, 296.5x, 296.6x, 296.7x, 296.8x) on or before the index date (first pharmacy claim for quetiapine XR or quetiapine IR). Patients were excluded if they had o12 months of baseline health plan enrollment before or after the index date; an ICD-9 diagnosis code for schizophrenia (295.0x–295.6x, 295.8x, 295.9x) on the same date as or after the most recent bipolar diagnosis, on or before the index date; a prescription for any atypical antipsychotic, including quetiapine XR or quetiapine IR, during the 6-month preindex period; or a prescription for both quetiapine XR and quetiapine IR on the index date. The treatment cohorts were defined according to initiation of quetiapine XR or quetiapine IR on the index date, regardless of prescription patterns after the initial prescription. Patients were followed up for a 24-month period. All patients had data available for at least 12 months before the index date and 12 months after initiation of quetiapine XR or quetiapine IR treatment, ensuring equal follow-up time for all patients. Treatment patterns (including dosing), health care resource use, and costs were analyzed during the 12-month postindex period. Health care resource use included inpatient visits, emergency department (ED) visits, and office or outpatient visits. Health care costs included plan-paid and patient-paid inpatient visit costs, ED visit costs, outpatient visit costs, pharmacy costs, and costs for other medical services not classified elsewhere.
PATIENTS AND METHODS Study Design and Population
Statistical Analysis
his was an observational, retrospective cohort study that used administrative claims data from the
Logistic regression was used to assess 12-month postindex treatment patterns between the 2 cohorts.
1924
Volume 35 Number 12
J.C. Locklear et al.
Table I. Demographic and clinical characteristics of patients with bipolar disorder during the 12-month period before treatment with quetiapine XR or quetiapine IR.* No. (%) of Patients† Quetiapine XR Quetiapine IR (n ¼ 651) (n ¼ 2398)
Characteristic Demographic characteristics Age, mean (SD), y Sex Male Female Physician specialty of psychiatry Geographic location Northeast Midwest South West Common mental health–related comorbidities (incidence 410% patients) Depression Anxiety disorder Drug or alcohol abuse Headache or tension headache Insomnia Adjustment disorder Common non–mental health–related comorbidities (incidence 410% patients) Respiratory disorders Other pain Back pain Hypertension Dyslipidemia Neck pain
39.0 (12.3)
39.3 (12.6)
259 (39.8) 392 (60.2) 332 (51.0)
953 (39.7) 1445 (60.3) 1247 (52.0)
85 235 188 143
(13.1) (36.1) (28.9) (22.0)
407 586 711 694
(17.0) (24.4) (29.6) (28.9)
328 233 181 104 97 83
(50.4) (35.8) (27.8) (16.0) (14.9) (12.7)
1243 748 775 415 313 309
(51.8) (31.2) (32.3) (17.3) (13.1) (12.9)
348 277 180 161 159 67
(53.5) (42.5) (27.6) (24.7) (24.4) (10.3)
1179 1113 656 554 552 336
(49.2) (46.4) (27.4) (23.1) (23.0) (14.0)
*IR ¼ immediate release; XR ¼ extended release. † Data are reported as number (percentage) of patients unless otherwise stated.
Odds ratios (ORs) and 95% CIs were reported. The variables used in the adjusted analyses were age, sex, geographic region, prescriber specialty, preindex comorbidities, and Deyo-Charlson comorbidity index (DCI) score.24 Cox proportional hazard models were used for the survival analyses of time to treatment change during the 12-month postindex treatment period. Hazard ratios and 95% CIs were reported. For the continuous measures of average daily dose (ADD) during the entire postindex period, time from index to optimal ADD (300–800 mg), and the
December 2013
proportion of prescriptions within the optimal ADD, an ANOVA model was used, and mean difference and 95% CI were reported. A generalized linear model with a negative binomial distribution and logarithmic-link function was used to assess 12-month postindex health care resource use. A generalized linear model with gamma distribution was used for the analysis of health care costs. Models were adjusted a priori for multiple confounding variables, including age, sex, geographic region, prescriber specialty, anxiety disorder, transient psychosis, drug
1925
Clinical Therapeutics
Table II. Health care resource use and costs for patients with bipolar disorder during the 12-month period before treatment with quetiapine XR or quetiapine IR.* No. (%) of Patients† Variable Health care resource use Z1 All-cause inpatient visits Z1 Mental health inpatient visits‡ Z1 All-cause ED* visits Z1 Mental health ED* visits§ Z1 All-cause office and outpatient visits Z1 Mental health office and outpatient visits§ Health care costs, mean (SD), US $ All-cause total health care costs Mental health–related total health care costs‡
Quetiapine XR (n ¼ 651) 212 163 235 114 646 616
(32.6) (25.0) (36.1) (17.5) (99.2) (94.6)
14,133 (30,652) 6700 (25,781)
Quetiapine IR (n ¼ 2398) 932 757 908 475 2366 2223
(38.9) (31.6) (37.9) (19.8) (98.7) (92.7)
16,647 (32,311) 8516 (20,999)
*ED ¼ emergency department; IR ¼ immediate release; XR ¼ extended release. † Data are reported as number (percentage) of patients unless otherwise stated. ‡ Mental health–related inpatient visits and costs defined as having a primary diagnosis that is mental health-related. § Mental health–related visits defined as having a mental health–related diagnosis in any position (primary or secondary).
or alcohol abuse, cardiovascular problems, pain, osteoarthritis, obesity, respiratory disorders, DCI index score, anticonvulsants, selective serotonin reuptake inhibitors or serotonin norepinephrine reuptake inhibitors, other central nervous system agents, and the corresponding preindex health care resource use and cost. All health care use and cost analyses were conducted separately for all-cause and mental health–related causes. Mental health–related inpatient visits and costs were defined as those relating to an ICD-9 code for a mental health indication as the primary diagnosis. All statistical analyses were performed with SAS software, version 9.2 (SAS Institute, Cary, North Carolina).
RESULTS Between October 2, 2008, and July 31, 2010, a total of 3049 patients with bipolar disorder were eligible for the current analysis. Index pharmacy claims for 79% of these patients were for quetiapine IR (Table I).
Index Date and Preindex Description of Patient Demographic Characteristics, Clinical Characteristics, Health Care Resource Use, and Costs Demographic characteristics, comorbid conditions, health care use, and costs for patients in the 12 months
1926
before initiating treatment with quetiapine XR or quetiapine IR are presented in Tables I and II. The rate of comorbid anxiety disorder was higher for the group who received quetiapine XR compared with those who received quetiapine IR (35.8% vs 31.2%), whereas drug or alcohol abuse was lower in the quetiapine XR group compared with the quetiapine IR group (27.8% vs 32.3%) (Table I). Other mental health–related comorbid conditions were similar between the 2 cohorts. In addition, no major differences were found between the 2 cohorts with respect to non–mental health–related comorbidities, except for neck pain, which was lower in the quetiapine XR group compared with the group initiating treatment with quetiapine IR (10.3% vs 14%) (Table I). All-cause and mental health–related health care resource use and costs were lower for the quetiapine XR group compared with the quetiapine IR group, respectively (Table II).
Postindex Analyses of Treatment Patterns, Health Care Resource Use, and Costs Treatment Patterns In total, 8.8% of patients treated with quetiapine XR had continuous treatment with no change or discontinuation of their index therapy, compared with 5.7% of those treated with quetiapine IR (adjusted
Volume 35 Number 12
J.C. Locklear et al.
Table III. Treatment patterns in the 12-month postindex period for patients with bipolar disorder treated with quetiapine XR or quetiapine IR.* No. (%) of Patients Treatment Pattern No treatment change or discontinuation (continuous treatment) Discontinuation of index medication Treatment change (nondiscontinuation) Switch from index medication to any atypical antipsychotic or add another atypical antipsychotic Switch from index medication to any conventional antipsychotic or add another conventional antipsychotic Other treatment change§
Quetiapine XR (n ¼ 651)
Quetiapine IR (n ¼ 2398)
Adjusted OR† (95% CI*)
P‡
57
8.8
137
5.7
1.44 (1.03–2.00)
0.0317
431 163 101
66.2 25.0 15.5
1685 576 275
70.3 24.0 11.5
0.82 (0.68–0.99) 1.11 (0.90–1.36) 1.44 (1.12–1.85)
0.0370 0.3235 0.0047
58
8.9
291
12.1
0.75 (0.56–1.02)
0.0651
7
1.1
11
0.5
2.90 (1.09–7.72)
0.0328
IR ¼ immediate release; OR ¼ odds ratio; XR ¼ extended release. *Confidence interval. † Adjusted OR from logistic model controlling for the following covariates: age, sex, geographic region, prescriber specialty, comorbidities (anxiety disorder, transient psychosis, drug or alcohol abuse, or neck pain), and Deyo-Charlson comorbidity score. ‡ P values are not adjusted for multiple comparisons. § Patients who filled multiple types of medications on the same treatment change date. For example, patients filling a conventional antipsychotic and switching from index medication on the same date are captured here.
OR, 1.44; 95% CI, 1.03–2.00; P ¼ 0.0317) (Table III). In total, 66.2% of patients treated with quetiapine XR discontinued their index therapy compared with 70.3% of those treated with quetiapine IR (adjusted OR, 0.82; 95% CI, 0.68–0.99; P ¼ 0.0370) (Table III). Overall, 25.0% and 24.0% of patients experienced a treatment change in the 12 months after initiation of treatment with quetiapine XR or quetiapine IR, respectively. In total, 15.5% of patients treated with quetiapine XR switched to an atypical antipsychotic or added another atypical antipsychotic compared with 11.5% of those treated with quetiapine IR (adjusted OR, 1.44; 95% CI, 1.12–1.85; P ¼ 0.0047), and there was a trend toward patients treated with quetiapine XR being less likely to switch to a conventional antipsychotic or add a conventional antipsychotic compared with quetiapine IR (8.9% vs 12.1%; adjusted OR, 0.75; 95% CI, 0.56–1.02; P ¼ 0.0651) (Table III). Furthermore, the ADD (adjusted mean) of quetiapine XR was higher than that of quetiapine IR (225 vs
December 2013
175 mg/d, P o 0.0001). An ADD of 300 to 800 mg was reached sooner (adjusted mean, 15.6 vs 30.8 days; P ¼ 0.0049) and in more patients (44.2% vs 27.2%, P o 0.0001) treated with quetiapine XR compared with quetiapine IR (Table IV).
Health Care Resource Use and Costs No difference was found in overall health care resource use between cohorts when considering the proportion of patients with inpatient visits (21.7% vs 25.5%, P ¼ 0.1260), ED visits (31.6% vs 33.8%, P ¼ 0.3849), and outpatient or office visits (98.2% vs 98.6%, P ¼ 0.2743) in patients initiating treatment with quetiapine XR or quetiapine IR, respectively (Figure 1). However, patients initiating treatment with quetiapine XR were less likely to be hospitalized for mental health–related reasons (12.1% vs 18.3%, P ¼ 0.0022) compared with patients taking quetiapine IR (Figure 1). The distribution of mental health–related costs in patients treated with quetiapine XR and quetiapine IR
1927
Clinical Therapeutics
Table IV. Dosing of index medication in the 12-month postindex period for patients with bipolar disorder treated with quetiapine XR or quetiapine IR.*
Variable ADD*,∥ during the entire postindex period, mg Time to 300–800 mg optimal ADD, d Total proportion of prescriptions that are 300–800 mg Patients filling at least one prescription with ADD (300–800 mg) at any time, No. (%) Patients whose most frequently prescribed ADD is between 300 and 800 mg, No. (%) Z50% of prescriptions between 300 and 800 mg, No. (%)
Quetiapine XR (n ¼ 651)
Quetiapine IR (n ¼ 2398)
Mean† (SD)
LS* Mean‡
Mean† (SD)
LS Mean‡
Adjusted OR* or Adjusted Mean Difference‡
P§
216 (187)
225
164 (155)
175
50.8
o0.0001
15.2 0.15
0.0049 o0.0001
24.0 (66.0) 15.6 39.6 (79.1) 30.8 0.36 (0.44) 0.39 0.20 (0.36) 0.23 288 (44.2) NA*
652 (27.2)
NA
2.09
o0.0001
251 (38.6)
NA
521 (21.7)
NA
2.22
o0.0001
245 (37.6)
NA
505 (21.1)
NA
2.23
o0.0001
*ADD ¼ average daily dose; IR ¼ immediate release; LS ¼ least squares; NA ¼ not applicable; OR ¼ odds ratio; XR ¼ extended release. † Data are reported as mean (SD) unless otherwise stated. ‡ Difference in LS means (for ADD and time to optimal ADD) and proportion of patients (for all other variables); adjusted OR and adjusted mean differences control for the following covariates: age (continuous), sex, geographic region, prescriber specialty, statistically significant comorbidities, and Deyo-Charlson comorbidity score. § P values are not adjusted for multiple comparisons. ∥ ADD ¼ (quantity dispensed/days’ supply) strength.
is shown in Figure 2. Adjusting for multiple confounding variables, no differences were found in mean overall health care costs between cohorts; however, patients initiating treatment with quetiapine XR incurred lower mental health–related adjusted mean costs (US $6686 vs US $7577, P ¼ 0.0063) compared with those taking quetiapine IR, which may be due to fewer hospitalizations or shorter duration of hospitalizations for mental health–related reasons (Figure 3).
DISCUSSION Retrospective analyses of the HealthCore Integrated Research Database suggest that treatment patterns and dosing differ in patients with bipolar disorder treated with quetiapine XR compared with those treated with quetiapine IR, supporting the need for
1928
physicians to have access to both formulations so that treatment decisions can be made based on individual patient needs. Mental health–related hospitalizations and costs were reduced in the 12 months after patients initiated treatment with quetiapine XR compared with quetiapine IR. A reduction in mental health–related costs associated with quetiapine XR versus quetiapine IR was estimated in this analysis to be a saving of US $891 (12%) per patient. Bipolar disorder is associated with high medical costs, partially due to costly hospitalizations.4 A study comparing the relative risk for hospitalization in patients with bipolar disorder receiving antipsychotic medication suggested that 7% to 10% of treatment episodes require hospitalization, with a mean cost of US $14,500 per inpatient stay.25 Previous studies
Volume 35 Number 12
J.C. Locklear et al.
P = 0.2743 100
98.2 98.6
Quetiapine XR Quetiapine IR
Patients (%)
80
comparing hospitalization in patients treated with antipsychotics have reported lower annual risk and reduced length of hospitalization in patients treated with quetiapine IR for bipolar disorder or schizophrenia, compared with risperidone, olanzapine, ziprasidone, aripiprazole, and haloperidol.25,26
60 40 20
P = 0.1260 25.5 21.7
P = 0.3849 31.6 33.8
40
Quetiapine XR
35
All-cause inpatient visits
All-cause ED visits
All-cause office/outpatient visits P = 0.4608
15
5
80
00 ,0 90
00 ,0 75
00 ,0 60
00 ,0 45
,0
,0 15
30
0 60
00
0
Mental health-related inpatient costs (US$) 40 P = 0.7397 17.4 18.6
Mental healthrelated* inpatient visits
Mental healthrelated† ED visits
35 30 25 20 15 10 5
,0 00 90
,0 00 75
,0 00 60
,0 00 45
30
,0 00
0 ,0 00
Mental healthrelated† office/ outpatient visits
Figure 1. Postindex all-cause (A) and mental health–related (B) adjusted health care resource use in patients with bipolar disorder treated with quetiapine extended release (XR) or quetiapine immediate release (IR). All-cause and mental health–related resource use (inpatient, emergency department [ED], or office or outpatient visits) are adjusted for baseline demographic, clinical, and use variables as outlined in the Methods section. *Mental health–related visits defined as having a primary diagnosis that is mental health related. †Mental health–related visits defined as having a mental health–related diagnosis in any position (primary or secondary).
December 2013
Quetiapine IR
40
15
0
P = 0.0022 18.3 12.1
0
20
Patients (%)
Patients (%)
20
10
94.5 94.5
Quetiapine XR Quetiapine IR
25
00
100
Patients (%)
30 0
Mental health-related inpatient costs (US$)
Figure 2. Distribution of mental health–related inpatient costs in patients with bipolar disorder treated with quetiapine extended release (XR) (A) or quetiapine immediate release (IR) (B). Raw data excluding US $0 and capping mental health–related cost at US $100,000 (1 patient treated with quetiapine XR and 8 patients treated with quetiapine IR were excluded).
1929
Clinical Therapeutics
15,000
P = 0.6059 13,897 14,205
Quetiapine XR Quetiapine IR
Mean cost (US$)
12,000 P = 0.0063 7577 6686
9000 6000 3000 0
All-cause costs
Mental health-related costs
Figure 3. Adjusted all-cause and mental health– related health care costs during the 12-month postindex date period in patients with bipolar disorder treated with quetiapine extended release (XR) or quetiapine immediate release (IR). All-cause and mental health–related resource use (inpatient, emergency department, or office/outpatient visits) are adjusted for baseline demographic, clinical, and use variables as outlined in the Methods section.
Differences in the patient populations treated with quetiapine XR or quetiapine IR are to be expected in a real-world setting and highlight that access to a broad range of therapeutic agents is essential to treat patients with bipolar disorder effectively and according to individual patient needs.27 Retrospective analysis of a Finnish hospital database reported that treatment patterns differed between those patients with bipolar disorder treated with quetiapine XR and those treated with quetiapine IR.28 The mean daily dose of quetiapine XR was significantly higher than that of quetiapine IR (466 vs 308 mg, P ¼ 0.009), and quetiapine XR was used with antiepileptic medication to a greater extent than quetiapine IR (53.3% vs 14.3%, P ¼ 0.03).28 Generic versions of quetiapine IR became available in 2012, which will result in lower drug costs in patients for whom treatment with quetiapine IR is appropriate. However, because treatment patterns and dosing differ in patients with bipolar disorder treated with quetiapine XR compared with quetiapine IR, quetiapine XR may provide an alternative
1930
treatment option to quetiapine IR and its generic formulations at an acceptable cost, enabling treatment decisions to be based on individual patient needs. Clinical practice data are an important supplement to RCTs. This study uses real-world data from the HealthCore database, which represents claims information from the largest commercially insured population in the United States. The study analyzed claims data from a large (n 4 3000) population of patients with bipolar disorder and provides information on end points that are important to payers, including health care use, cost, and compliance information. Because this study was a retrospective database analysis, it is potentially associated with a number of methodologic limitations related to nonrandomization, and because statistical tests were not adjusted for multiple comparisons, P values should be viewed with caution. Despite these limitations, the generalized linear model analyses controlled for known potential confounding effects collected during this study; however, this study did not adjust for unknown and uncollected variables. Far fewer patients were treated with quetiapine XR (n ¼ 651) than quetiapine IR (n ¼ 2398), and the study does not account for sample imbalances. An additional limitation of the study is that multiple statistical tests were performed without adjusting for multiple comparisons. Furthermore, in the 12 months after initiation of their index therapy, only 8.8% of patients had continuous treatment with quetiapine XR, and 5.7% of patients had continuous treatment with quetiapine IR. Therefore, it is likely that most antipsychotic treatment costs in the 12 months after initiation of treatment with quetiapine XR or quetiapine IR are not attributable to the index therapy. There are also limitations to analyzing cost data where the distributions have large zero values and outliers. This is the case for mental health–related costs as shown in Figure 2. The methods prespecified and presented in this study were selected to manage the zero-cost values and skewed distribution when calculating mean costs. However, we acknowledge that alternative methods are available, and sensitivity analyses would need to be performed to increase the level of confidence in the model used in this study and the robustness of the results. Therefore, this is a potential limitation with regards to interpreting the cost data analyses.
Volume 35 Number 12
J.C. Locklear et al.
CONCLUSION This study provides evidence that in the real-world management of bipolar disorder, treatment patterns and dosing during a 12-month period differ with quetiapine XR compared with quetiapine IR. Taken together, these results support the need for physicians to have access to both formulations so that they can make treatment decisions based on individual patients’ needs.
ACKNOWLEDGEMENTS This study was sponsored by AstraZeneca Pharmaceuticals. The authors would like to thank Claire Chadwick, of Complete Clarity, who provided medical writing support funded by AstraZeneca. All authors were involved in analysis and interpretation of data. All authors contributed to and critically reviewed the manuscript at all stages of development, and all authors approved the final manuscript.
CONFLICTS OF INTEREST Drs Locklear, Alemayehu, and Earley were employees of AstraZeneca when this analysis was conceived, conducted, and completed. Messrs Brody and Chavoshi are employees of AstraZeneca. Dr Tunceli and Mr Kern are employees of HealthCore Inc, which received funding from AstraZeneca for this research. The authors have indicated that they have no other conflicts of interest regarding the content of this article.
REFERENCES 1. Merikangas KR, Akiskal HS, Angst J, et al. Lifetime and 12-month prevalence of bipolar spectrum disorder in the national comorbidity survey replication. Arch Gen Psychiatry. 2007;64:543–552. 2. Saunders KE, Goodwin GM. New approaches in the treatment of bipolar depression. Curr Top Behav Neurosci. 2013;14:291–307. 3. Baldessarini RJ, Leahy L, Arcona S, et al. Patterns of psychotropic drug prescription for U.S. patients with diagnoses of bipolar disorders. Psychiatr Serv. 2007;58:85–91. 4. Bryant-Comstock L, Stender M, Devercelli G. Health care utilization and costs among privately insured patients with bipolar I disorder. Bipolar Disord. 2002;4:398–405. 5. National Center for Health Statistics. Health, United States, 2010. http://www.cdc.gov/nchs/data/hus/2010/ 128.pdf. Accessed November 16, 2012.
December 2013
6. Dilsaver SC. An estimate of the minimum economic burden of bipolar I and II disorders in the United States: 2009. J Affect Disord. 2011;129:79–83. 7. Garrison LP Jr, Neumann PJ, Erickson P, et al. Using realworld data for coverage and payment decisions: the ISPOR Real-World Data Task Force report. Value Health. 2007;10:326–335. 8. AstraZeneca. Seroquel (quetiapine fumarate) - US prescribing information, 2011. http://www1.astrazeneca-us. com/pi/seroquel.pdf. Accessed July 9, 2012. 9. AstraZeneca. Seroquel XR (quetiapine fumarate) extendedrelease tablets/US prescribing information, 2012. http:// www1.astrazeneca-us.com/pi/seroquelxr.pdf. Accessed July 9, 2012. 10. Calabrese JR, Keck PE Jr, Macfadden W, et al. A randomized, double-blind, placebo-controlled trial of quetiapine in the treatment of bipolar I or II depression. Am J Psychiatry. 2005;162:1351–1360. 11. Cutler AJ, Datto C, Nordenhem A, et al. Extended-release quetiapine as monotherapy for the treatment of adults with acute mania: a randomized, double-blind, 3-week trial. Clin Ther. 2011;33:1643–1658. 12. McElroy SL, Weisler RH, Chang W, et al. A double-blind, placebo-controlled study of quetiapine and paroxetine as monotherapy in adults with bipolar depression (EMBOLDEN II). J Clin Psychiatry. 2010;71:163–174. 13. Sachs G, Chengappa KN, Suppes T, et al. Quetiapine with lithium or divalproex for the treatment of bipolar mania: a randomized, double-blind, placebo-controlled study. Bipolar Disord. 2004;6:213–223. 14. Suppes T, Vieta E, Liu S, et al. Maintenance treatment for patients with bipolar I disorder: results from a North American study of quetiapine in combination with lithium or divalproex (trial 127). Am J Psychiatry. 2009;166:476– 488. 15. Suppes T, Datto C, Minkwitz M, et al. Effectiveness of the extended release formulation of quetiapine as monotherapy for the treatment of acute bipolar depression. J Affect Disord. 2010;121:106–115. 16. Thase M, Macfadden W, McCoy R, et al. Efficacy of quetiapine monotherapy in bipolar depression: a confirmatory double-blind, placebo-controlled study (The BOLDER II Study). Int J Neuropsychopharmacol. 2006;9 (Suppl 1):S237. Abstract P03.035. 17. Vieta E, Mullen J, Brecher M, et al. Quetiapine monotherapy for mania associated with bipolar disorder: combined analysis of two international, double-blind, randomised, placebo-controlled studies. Curr Med Res Opin. 2005;21:923–934. 18. Vieta E, Suppes T, Eggens I, et al. Efficacy and safety of quetiapine in combination with lithium or divalproex for maintenance of patients with bipolar I disorder (international trial 126). J Affect Disord. 2008;109:251–263.
1931
Clinical Therapeutics 19. Weisler RH, Nolen WA, Neijber A, et al. Continuation of quetiapine versus switching to placebo or lithium for maintenance treatment of bipolar I disorder (Trial 144: a randomized controlled study). J Clin Psychiatry. 2011;72:1452–1464. 20. Yatham LN, Paulsson B, Mullen J, Vågerö M. Quetiapine versus placebo in combination with lithium or divalproex for the treatment of bipolar mania. J Clin Psychopharmacol. 2004;24:599–606. 21. Young AH, McElroy SL, Bauer M, et al. A double-blind, placebocontrolled study of quetiapine and lithium monotherapy in adults in the acute phase of bipolar depression (EMBOLDEN I). J Clin Psychiatry. 2010;71:150–162. 22. Bonafede M, Locklear J, Wahlqvist P, et al. The impact of once-daily extended-release quetiapine fumarate (quetiapine XR) on length of hospitalisation of patients with acute bipolar mania. Value Health. 2011:14. Abstract A300. 23. Locklear J, Wahlqvist P, Gustafsson U, et al. The impact of once-daily extended release quetiapine fumarate (quetiapine XR) on length and costs of hospitalisation of patients with schizophrenia or bipolar disorder. Value Health. 2011:14. Abstract A299. 24. Deyo RA, Cherkin DC, Ciol MA. Adapting a clinical comorbidity index for use with ICD-9-CM administrative databases. J Clin Epidemiol. 1992;45:613–619. 25. Gianfrancesco F, Rajagopalan K, Goldberg JF, Wang RH. Hospitalization risks in the treatment of bipolar disorder: comparison of antipsychotic medications. Bipolar Disord. 2007;9:252–261. 26. Strutton DR, Gutierrez B, Blanchette CM. Influence of medication treatment patterns on resource use for patients hospitalized for schizophrenia. Am J Health Syst Pharm. 2008;65: 42–48. 27. Altamura AC, Armadoros D, Jaeger M, et al. Importance of open access
1932
to atypical antipsychotics for the treatment of schizophrenia and bipolar disorder: a European perspective. Curr Med Res Opin. 2008;24:2271–2282. 28. Hallinen T, Soini EJ, Granström O, et al. Differential use of extended and immediate release quetiapine: a retrospective registry study of Finnish inpatients with schizophrenia spectrum and bipolar disorders. BMJ Open. 2012;2:e000915.
Address correspondence to: Robert S. Brody, MPH, AstraZeneca LP, 1800 Concord Pike, PO Box 15437, Wilmington, DE, 19807-5437. E-mail: Bob. [email protected]
Volume 35 Number 12