Resampling and multiplicity in cost-effectiveness inference

Clinical Outcomes and Cost-Effectiveness of Collaborative Dementia Care: A Secondary Analysis of a Cluster Randomized Clinical Trial

Importance

Long-term evidence for the effectiveness and cost-effectiveness of collaborative dementia care management (CDCM) is lacking.

Objective

To evaluate whether 6 months of CDCM is associated with improved patient clinical outcomes and caregiver burden and is cost-effective compared with usual care over 36 months.

Design, Setting, and Participants

This was a prespecified secondary analysis of a general practitioner (GP)-based, cluster randomized, 2-arm clinical trial conducted in Germany from January 1, 2012, to December 31, 2014, with follow-up until March 31, 2018. Participants were aged 70 years or older, lived at home, and screened positive for dementia. Data were analyzed from March 2011 to March 2018.

Intervention

The intervention group received CDCM, comprising a comprehensive needs assessment and individualized interventions by nurses specifically qualified for dementia care collaborating with GPs and health care stakeholders over 6 months. The control group received usual care.

Main Outcomes and Measures

Main outcomes were neuropsychiatric symptoms (Neuropsychiatric Inventory [NPI]), caregiver burden (Berlin Inventory of Caregivers' Burden in Dementia [BIZA-D]), health-related quality of life (HRQOL, measured by the Quality of Life in Alzheimer Disease scale and 12-Item Short-Form Health Survey [SF-12]), antidementia drug treatment, potentially inappropriate medication, and cost-effectiveness (incremental cost per quality-adjusted life year [QALY]) over 36 months. Outcomes between groups were compared using multivariate regression models adjusted for baseline scores.

Results

A total of 308 patients, of whom 221 (71.8%) received CDCM (mean [SD] age, 80.1 [5.3] years; 142 [64.3%] women) and 87 (28.2%) received usual care (mean [SD] age, 79.2 [4.5] years; 50 [57.5%] women), were included in the clinical effectiveness analyses, and 428 (303 [70.8%] CDCM, 125 [29.2%] usual care) were included in the cost-effectiveness analysis (which included 120 patients who had died). Participants receiving CDCM showed significantly fewer behavioral and psychological symptoms (adjusted mean difference [AMD] in NPI score, -10.26 [95% CI, -16.95 to -3.58]; P = .003; Cohen d, -0.78 [95% CI, -1.09 to -0.46]), better mental health (AMD in SF-12 Mental Component Summary score, 2.26 [95% CI, 0.31-4.21]; P = .02; Cohen d, 0.26 [95% CI, -0.11 to 0.51]), and lower caregiver burden (AMD in BIZA-D score, -0.59 [95% CI, -0.81 to -0.37]; P < .001; Cohen d, -0.71 [95% CI, -1.03 to -0.40]). There was no difference between the CDCM group and usual care group in use of antidementia drugs (adjusted odds ratio, 1.91 [95% CI, 0.96-3.77]; P = .07; Cramér V, 0.12) after 36 months. There was no association with overall HRQOL, physical health, or use of potentially inappropriate medication. The CDCM group gained QALYs (0.137 [95% CI, 0.000 to 0.274]; P = .049; Cohen d, 0.20 [95% CI, -0.09 to 0.40]) but had no significant increase in costs (437€ [-5438€ to 6313€] [US $476 (95% CI, -$5927 to $6881)]; P = .87; Cohen d, 0.07 [95% CI, -0.14 to 0.28]), resulting in a cost-effectiveness ratio of 3186€ (US $3472) per QALY. Cost-effectiveness was significantly better for patients living alone (CDCM dominated, with lower costs and more QALYs gained) than for those living with a caregiver (47 538€ [US $51 816] per QALY).

Conclusions and Relevance

In this secondary analysis of a cluster randomized clinical trial, CDCM was associated with improved patient, caregiver, and health system-relevant outcomes over 36 months beyond the intervention period. Therefore, it should become a health policy priority to initiate translation of CDCM into routine care.

Trial Registration

ClinicalTrials.gov Identifier: NCT01401582.

Translation of Collaborative Dementia Care Management into Different Healthcare Settings: Study Protocol for a Multicenter Implementation Trial (DCM:IMPact)

Background: Collaborative care models for people living with dementia (PwD) have been developed and evaluated, demonstrating safety, efficacy, and cost-effectiveness. However, these studies are based on heterogeneous study populations and primary care settings, limiting the generalizability of the results. Therefore, this study aims to implement and evaluate collaborative care across various healthcare settings and patient populations. Objective: To describe the study design of this multicenter implementation trial. Methods: This single-arm, multicenter, longitudinal implementation study will be conducted in five different healthcare settings, including 1) physicians’ networks, 2) dementia networks, 3) counselling centers, 4) hospitals, and 5) ambulatory care services. Eligibility criteria are: having a formal dementia diagnosis or having been screened positive for dementia and living community-dwelling. The staff of each healthcare setting identifies patients, informs them about the study, and invites them to participate. Participants will receive a baseline assessment followed by collaborative individualized dementia care management, comprising proven safe, effective, and cost-effective modules. Over six months, specially-qualified nurses will assess patients’ unmet needs, transfer them to individualized care plans, and address them, cooperating with various healthcare providers. A follow-up assessment is conducted six months after baseline. Approximately 60–100 PwD per setting per year are expected to participate. Differences across settings will be assessed regarding acceptability, demand, implementation success and barriers, efficacy, and cost-effectiveness. Results: We expect that acceptability, demand, implementation success and barriers, efficacy, and cost-effectiveness will vary by patients’ sociodemographic and clinical characteristics and unmet needs in each setting. Conclusion: The results will provide evidence highlighting differences in the implementation of collaborative care in various healthcare settings and demonstrating the settings with the highest need, best conditions for a successful implementation, and highest (cost-)effectiveness, as well as the population group that benefits most from collaborative care. Trial registration: German Clinical Trials Register: DRKS00025074. Registered 16 April 2021-retrospectively registered.

Behavioral activation with mindfulness in treating subthreshold depression in primary care: A cost-utility and cost-effectiveness analysis alongside a randomized controlled trial

This study aimed to assess the cost-utility and cost-effectiveness of group-based behavioral activation with mindfulness (BAM) versus care as usual (CAU) for treating subthreshold depression in primary care. Adults aged 18 years or older with subthreshold depression were randomized into two arms and were followed up for 12 months. BAM group was provided with eight 2-h weekly treatment by trained allied healthcare workers. CAU group could access to usual medical care but did not receive extra interventions. The health service cost in the past 12 months was self-reported by the participants. Quality-adjusted Life Years (QALYs) and clinical outcome (incidence of major depressive disorder progression) were measured. Willingness-to-pay ratio for cost-utility analysis (CUA) and cost-effectiveness analysis (CEA) was US$50,000 per QALY and US$20,000 per prevented major depression case, respectively. These ratios were used in the cost-effective acceptability curve analyses to estimate the probability of cost-effectiveness of the estimated incremental cost effectiveness ratios (ICER) of BAM versus CAU. A total of 115 and 116 participants were included in the BAM group and CAU respectively. The estimated CUA ICER was US5,979 per QALY and had a probability of 0.93 that BAM was cost-effective when compared to CAU. Furthermore, when compared to CAU, BAM was cost-effective in preventing progression of major depression: the estimated CEA ICER was US$1046 per preventable case of major progression with a probability of 0.99 to be cost-effective. Group-based BAM is considered as a cost-effective alternative treatment for treating subthreshold depression by preventing major depressive disorder.

Is the whole larger than the sum of its parts? Impact of missing data imputation in economic evaluation conducted alongside randomized controlled trials

Outcomes in economic evaluations, such as health utilities and costs, are products of multiple variables, often requiring complete item responses to questionnaires. Therefore, missing data are very common in cost-effectiveness analyses. Multiple imputations (MI) are predominately recommended and could be made either for individual items or at the aggregate level. We, therefore, aimed to assess the precision of both MI approaches (the item imputation vs. aggregate imputation) on the cost-effectiveness results. The original data set came from a cluster-randomized, controlled trial and was used to describe the missing data pattern and compare the differences in the cost-effectiveness results between the two imputation approaches. A simulation study with different missing data scenarios generated based on a complete data set was used to assess the precision of both imputation approaches. For health utility and cost, patients more often had a partial (9% vs. 23%, respectively) rather than complete missing (4% vs. 0%). The imputation approaches differed in the cost-effectiveness results (the item imputation: - 61,079€/QALY vs. the aggregate imputation: 15,399€/QALY). Within the simulation study mean relative bias (< 5% vs. < 10%) and range of bias (< 38% vs. < 83%) to the true incremental cost and incremental QALYs were lower for the item imputation compared to the aggregate imputation. Even when 40% of data were missing, relative bias to true cost-effectiveness curves was less than 16% using the item imputation, but up to 39% for the aggregate imputation. Thus, the imputation strategies could have a significant impact on the cost-effectiveness conclusions when more than 20% of data are missing. The item imputation approach has better precision than the imputation at the aggregate level.

Who Benefits Most from Collaborative Dementia Care from a Patient and Payer Perspective? A Subgroup Cost-Effectiveness Analysis

BACKGROUND

Dementia care management (DCM) aims to provide optimal treatment for people with dementia (PwD). Treatment and care needs are dependent on patients' sociodemographic and clinical characteristics and thus, economic outcomes could depend on such characteristics.

OBJECTIVE

To detect important subgroups that benefit most from DCM and for which a significant effect on cost, QALY, and the individual cost-effectiveness could be achieved.

METHODS

The analysis was based on 444 participants of the DelpHi-trial. For each subgroup, the probability of DCM being cost-effective was calculated and visualized using cost-effectiveness acceptability curves. The impact of DCM on individual costs and QALYs was assessed by using multivariate regression models with interaction terms.

RESULTS

The probability of DCM being cost-effective at a willingness-to-pay of 40,000€ /QALY was higher in females (96% versus 16% for males), in those living alone (96% versus 26% for those living not alone), in those being moderately to severely cognitively (100% versus 3% for patients without cognitive impairment) and functionally impaired (97% versus 16% for patients without functional impairment), and in PwD having a high comorbidity (96% versus 26% for patients with a low comorbidity). Multivariate analyses revealed that females (b = -10,873; SE = 4,775, p = 0.023) who received the intervention had significantly lower healthcare cost. DCM significantly improved QALY for PwD with mild and moderate cognitive (b = +0.232, SE = 0.105) and functional deficits (b = +0.200, SE = 0.095).

CONCLUSION

Patients characteristics significantly affect the cost-effectiveness. Females, patients living alone, patients with a high comorbidity, and those being moderately cognitively and functionally impaired benefit most from DCM. For those subgroups, healthcare payers could gain the highest cost savings and the highest effects on QALYs when DCM will be implemented.

Cost-effectiveness of a collaborative dementia care management-Results of a cluster-randomized controlled trial

INTRODUCTION

The purpose of this study was to determine the cost-effectiveness of collaborative dementia care management (DCM).

METHODS

The cost-effectiveness analysis was based on the data of 444 patients of a cluster-randomized, controlled trial, conceptualized to evaluate a collaborative DCM that aimed to optimize treatment and care in dementia. Health-care resource use, costs, quality-adjusted life years (QALYs), and incremental cost per QALY gained were measured over a 24-month time horizon.

RESULTS

DCM increased QALYs (+0.05) and decreased costs (-569€) due to a lower hospitalization and a delayed institutionalization (7 months) compared with usual care. The probability of DCM being cost-effective was 88% at willingness-to-pay thresholds of 40,000€ per QALY gained and higher in patients living alone compared to those not living alone (96% vs. 26%).

DISCUSSION

DCM is likely to be a cost-effective strategy in treating dementia and thus beneficial for public health-care payers and patients, especially for those living alone.

Treatment patterns and cost-effectiveness of first line treatment of advanced non-squamous non-small cell lung cancer in Medicare patients

AIM

To assess the cost-effectiveness of first-line pemetrexed/platinum and other commonly administered regimens in a representative US elderly population with advanced non-squamous non-small cell lung cancer (NSCLC).

MATERIALS AND METHODS

This study utilized the Surveillance Epidemiology and End Results (SEER) cancer registry linked to Medicare claims records. The study population included all SEER-Medicare patients diagnosed in 2008-2009 with advanced non-squamous NSCLC (stages IIIB-IV) as their only primary cancer and who started chemotherapy within 90 days of diagnosis. The study evaluated the four most commonly observed first-line regimens: paclitaxel/carboplatin, platinum monotherapy, pemetrexed/platinum, and paclitaxel/carboplatin/bevacizumab. Overall survival and total healthcare cost comparisons as well as incremental cost-effectiveness ratios (ICERs) were calculated for pemetrexed/platinum vs each of the other three. Unstratified analyses and analyses stratified by initial disease stage were conducted.

RESULTS

The final study population consisted of 2,461 patients. Greater administrative censorship of pemetrexed recipients at the end of the study period disproportionately reduced the observed mean survival for pemetrexed/platinum recipients. The disease stage-stratified ICER analysis found that the pemetrexed/platinum incurred total Medicare costs of $536,424 and $283,560 per observed additional year of life relative to platinum monotherapy and paclitaxel/carboplatin, respectively. The pemetrexed/platinum vs triplet comparator analysis indicated that pemetrexed/platinum was associated with considerably lower total Medicare costs, with no appreciable survival difference.

LIMITATIONS

Limitations included differential censorship of the study regimen recipients and differential administration of radiotherapy.

CONCLUSIONS

Pemetrexed/platinum yielded either improved survival at increased cost or similar survival at reduced cost relative to comparator regimens in the treatment of advanced non-squamous NSCLC. Limitations in the study methodology suggest that the observed pemetrexed survival benefit was likely conservative.

Median-based incremental cost-effectiveness ratios with censored data

Cost-effectiveness is an essential part of treatment evaluation, in addition to effectiveness. In the cost-effectiveness analysis, a measure called the incremental cost-effectiveness ratio (ICER) is widely utilized, and the mean cost and the mean (quality-adjusted) life years have served as norms to summarize cost and effectiveness for a study population. Recently, the median-based ICER was proposed for complementary or sensitivity analysis purposes. In this article, we extend this method when some data are censored.

The cost of the inpatient management of febrile neutropenia in cancer patients--a micro-costing study in the Irish healthcare setting

The objective was to evaluate the resource use and cost of hospitalisation for febrile neutropenia (FN) from the health-payer's perspective. This was a single centre study. Adults undergoing chemotherapy, who were admitted for FN, were identified prospectively. Patient medical records were reviewed retrospectively. Demographics and resource utilisation data were obtained from a cohort of 32 patients (69% female, mean age = 58.8 years). Twenty-five per cent of patients had more than one FN episode. In total, 42 FN episodes were captured; 60% of episodes had occurred within the first two cycles of chemotherapy. The bootstrap estimation was used to determine mean hospital length of stay (LOS) with standard deviation (±SD) and mean costs ± SD. The mean LOS was 7.3 ± 0.5 days. The mean cost per FN episode was €8915 ± 718. The major cost driver was hospital bed-stay (mean cost of €6851 ± 549). Other cost drivers included antibacterial treatment at €760 ± 156, laboratory investigations at €538 ± 47 and the requirement for blood bank products at €525 ± 189. To our knowledge, this is the first investigation of the cost of chemotherapy induced FN within the context of the Irish healthcare setting.

Cost-effectiveness analysis: a proposal of new reporting standards in statistical analysis

Cost-effectiveness analysis (CEA) is a method for evaluating the outcomes and costs of competing strategies designed to improve health, and has been applied to a variety of different scientific fields. Yet there are inherent complexities in cost estimation and CEA from statistical perspectives (e.g., skewness, bidimensionality, and censoring). The incremental cost-effectiveness ratio that represents the additional cost per unit of outcome gained by a new strategy has served as the most widely accepted methodology in the CEA. In this article, we call for expanded perspectives and reporting standards reflecting a more comprehensive analysis that can elucidate different aspects of available data. Specifically, we propose that mean- and median-based incremental cost-effectiveness ratios and average cost-effectiveness ratios be reported together, along with relevant summary and inferential statistics, as complementary measures for informed decision making.

Selective decontamination of the digestive tract and selective oropharyngeal decontamination in intensive care unit patients: a cost-effectiveness analysis

OBJECTIVE

To determine costs and effects of selective digestive tract decontamination (SDD) and selective oropharyngeal decontamination (SOD) as compared with standard care (ie, no SDD/SOD (SC)) from a healthcare perspective in Dutch Intensive Care Units (ICUs).

DESIGN

A post hoc analysis of a previously performed cluster-randomised trial (NEJM 2009;360:20).

SETTING

13 Dutch ICUs.

PARTICIPANTS

Patients with ICU-stay of >48 h that received SDD (n=2045), SOD (n=1904) or SC (n=1990).

INTERVENTIONS

SDD or SOD.

PRIMARY AND SECONDARY OUTCOME MEASURES

Effects were based on hospital survival, expressed as crude Life Years Gained (cLYG). The incremental cost-effectiveness ratio (ICER) was calculated, with corresponding cost acceptability curves. Sensitivity analyses were performed for discount rates, costs of SDD, SOD and mechanical ventilation.

RESULTS

Total costs per patient were €41 941 for SC (95% CI €40 184 to €43 698), €40 433 for SOD (95% CI €38 838 to €42 029) and €41 183 for SOD (95% CI €39 408 to €42 958). SOD and SDD resulted in crude LYG of +0.04 and +0.25, respectively, as compared with SC, implying that both SDD and SOD are dominant (ie, cheaper and more beneficial) over SC. In cost-effectiveness acceptability curves probabilities for cost-effectiveness, compared with standard care, ranged from 89% to 93% for SOD and from 63% to 72% for SDD, for acceptable costs for 1 LYG ranging from €0 to €20 000. Sensitivity analysis for mechanical ventilation and discount rates did not change interpretation. Yet, if costs of the topical component of SDD and SOD would increase 40-fold to €400/day and €40/day (maximum values based on free market prices in 2012), the estimated ICER as compared with SC for SDD would be €21 590 per LYG. SOD would remain cost-saving.

CONCLUSIONS

SDD and SOD were both effective and cost-saving in Dutch ICUs.

Median-Based Incremental Cost-Effectiveness Ratio (ICER)

Cost-effectiveness analysis (CEA) is a type of economic evaluation that examines the costs and health outcomes of alternative strategies and has been extensively applied in health sciences. The incremental cost-effectiveness ratio (ICER), which represents the additional cost of one unit of outcome gained by one strategy compared with another, has become a popular methodology in CEA. Despite its popularity, limited attention has been paid to summary measures other than the mean for summarizing cost as well as effectiveness in the context of CEA. Although some apparent advantages of other central tendency measures such as median for cost data that are often highly skewed are well understood, thus far, the median has rarely been considered in the ICER. In this paper, we propose the median-based ICER, along with inferential procedures, and suggest that mean and median-based ICERs be considered together as complementary tools in CEA for informed decision making, acknowledging the pros and cons of each. If the mean and median-based CEAs are concordant, we may feel reasonably confident about the cost-effectiveness of an intervention, but if they provide different results, our confidence may need to be adjusted accordingly, pending further evidence.

Average cost-effectiveness ratio with censored data

In cost-effectiveness analysis, interest could lie foremost in the incremental cost-effectiveness ratio (ICER), which is the ratio of the incremental cost to the incremental benefit of two competing interventions. The average cost-effectiveness ratio (ACER) is the ratio of the cost to benefit of an intervention without reference to a comparator. A vast literature is available for statistical inference of the ICERs, but limited methods have been developed for the ACERs, particularly in the presence of censoring. Censoring is a common feature in prospective studies, and valid analyses should properly adjust for censoring in cost as well as in effectiveness. In this article, we propose statistical methods for constructing a confidence interval for the ACER from censored data. Different methods-Fieller, Taylor, bootstrap-are proposed, and through simulation studies and data analysis, we address the performance characteristics of these methods.

Cost-Effectiveness Inferences from Bootstrap Quadrant Confidence Levels: Three Degrees of Dominance

When with at least 95% confidence a new treatment is shown to be not only less costly (LC), but also more effective (ME), than a current treatment, that new treatment can be said to "strictly dominate" the current treatment statistically. But what can be said when head-to-head treatment comparisons turn out to be less clear-cut than this? Here, we propose two additional sets of specific LC and/or ME confidence thresholds to define the concepts of "some dominance" and "much dominance." Confidence levels associated with entire quadrants of the incremental cost-effectiveness (ICE) plane are easily computed using the same bootstrapping techniques used to estimate an "acceptability curve." Our two proposed additional "degrees" of dominance, although less stringent than strict dominance, are nevertheless more stringent than commonly accepted approaches using ICE ratio or net benefit calculations. To illustrate analysis concepts, we use data from a randomized, double-blind, placebo- and active comparator-controlled clinical registration trial for treatment of major depressive disorder (MDD). As is typical, our case study is rather small and short term, providing outcome information for a total of only 264 patients during their initial 8 weeks of acute-phase MDD treatment. Thus, we focus attention on sensitivity analyses, showing that the bootstrap distribution of cost-effectiveness uncertainty is robust across two alternative ways of measuring overall effectiveness and three alternative ways of imputing missing values. Evaluation of the balance between cost and benefit is particularly difficult when a new pharmacological treatment is first introduced, yet information of this sort is highly desired by decision makers. We show that, even with only a relatively modest amount of clinical trial information, sensitivity analyses can still confirm that cost-effectiveness comparisons are being made in a consistent fashion. In contrast, extensive follow-up comparisons using data from actual clinical practice will almost always ultimately be needed to better inform health policy makers.

Cost-Effectiveness of Duloxetine Versus Routine Treatment for U.S. Patients With Diabetic Peripheral Neuropathic Pain

The purpose of this study was to compare the cost-effectiveness of duloxetine versus routine treatment in management of diabetic peripheral neuropathic pain (DPNP). Two hundred thirty-three patients with DPNP who completed a 12-week, double-blind, placebo-controlled, randomized, multicenter duloxetine trial were re-randomized into a 52-week, open-label trial of duloxetine 60 mg twice daily versus routine treatment. Routine treatment included pain management therapies. Effectiveness was measured by using the bodily pain domain (BP) of the Medical Outcomes Study Short Form 36 (SF-36). Costs were analyzed from 3 perspectives: third party payer (direct medical costs), employer (direct and indirect medical costs), and societal (patient's out-of-pocket costs and total medical costs). Costs of study medications were not included because of limited data. Bootstrap method was applied to calculate statistical inference of the incremental cost-effectiveness ratio (ICER). Routine treatment most frequently used included gabapentin (56%), venlafaxine (36%), and amitripytline (15%). From employer and societal perspectives, duloxetine was cost-effective (ICER= -342 dollars and -429 dollars, respectively, per unit of SF-36 BP; both P <or= .03) and the dominant therapy compared with routine DPNP treatment (both P < .05). From payer perspective, duloxetine trended toward cost-effectiveness (ICER= -249 dollars per unit of SF-36 BP; P <or= .06). These results, however, reflect the controlled environment of a clinical trial. An analysis of real-world data would be beneficial.

PERSPECTIVE

Evaluation of the cost and benefit of new pharmacologic treatments is highly desired by decision makers. From both employer perspective and societal perspective (including patient's out-of-pocket costs), this study demonstrated that duloxetine was more cost-effective than routine treatment in management of DPNP.

Cost-effectiveness of olanzapine as first-line treatment for schizophrenia: results from a randomized, open-label, 1-year trial

OBJECTIVES

This randomized, open-label trial was designed to help inform antipsychotic treatment policies. It compared the 1-year cost-effectiveness of initial treatment with olanzapine (OLZ) (n = 229) versus a "fail-first" algorithm on conventional antipsychotics (then olanzapine if indicated) (CON) (n = 214); and versus initial treatment with risperidone (RIS) (n = 221).

METHODS

Individuals with schizophrenia or schizoaffective disorder were recruited from May 1998 to September 2001. Clinical, functioning, and resource utilization data were collected at baseline and five postbaseline visits. Brief Psychiatric Rating Scale scores defined "clinical effectiveness;" Lehman Quality of Life Scale social relations scores defined "social effectiveness."

RESULTS

Requiring failure on less expensive antipsychotics before use of olanzapine did not result in total cost savings, despite significantly higher antipsychotic costs with OLZ. Total 1-year mean costs were 21,283 dollars for CON; 20,891 dollars for OLZ; and 21,347 dollars for RIS (pair-wise comparisons nonsignificant). Intent-to-treat effectiveness comparisons (nonsignificant) were augmented by analyses that adjusted for duration on initial antipsychotic treatment, and by comparisons of patients remaining on initial antipsychotic treatment versus those who required switching. When accounting for differential switching rates (OLZ 0.14 vs. CON 0.53, P < 0.0001; vs. RIS 0.31, P < 0.0001), OLZ was significantly more effective than CON on clinical (P = 0.025) and social (P = 0.043) measures, and significantly more effective than RIS on the social (P = 0.002) measure. Further, patients initiated on an antipsychotic from which they needed to switch required additional resources for hospitalization (P = 0.036) and crisis services (P = 0.029).

CONCLUSIONS

Approaches that integrate costs, effectiveness, and treatment patterns are important for providing optimal information regarding the value of first-line antipsychotic options for schizophrenia.

Evaluation and review of pharmacoeconomic models

Difficult decisions as to whether to provide or withhold drug therapy to patients are needed to be made on a daily basis. These decisions should be based on carefully designed and constructed pharmacoeconomic models, with explicit and justifiable parameter values, validated by publication in peer-reviewed literature. This review describes and evaluates the common types of pharmacoeconomic models, modelling approaches and methods. It also discusses model quality, validity and usefulness.

Dealing with skewed data: an example using asthma-related costs of medicaid clients

BACKGROUND

Cost data often are nonnormally distributed due to a few very high cost values that may not necessarily be dismissed as outliers. Researchers have not reached agreement on how to appropriately deal with skewed cost data.

OBJECTIVES

This study presents an example of skewed cost data that were collected retrospectively from the Texas Medicaid database. Common methods of dealing with skewed cost distributions are discussed. Data were analyzed using various methods, and the statistical results of each test were compared.

METHODS

Prescription and medical claims data extracted from the Texas Medicaid database were analyzed using the Mann-Whitney U test and t tests of untransformed, log-transformed, and bootstrapped data.

RESULTS

All distributions of the untransformed cost data were nonnormally distributed, and comparison groups had unequal variances. The Mann-Whitney U test negated the effect of the high-cost patients and gave a significant result for overall cost differences between groups, but in the opposite direction of the mean. The t tests on raw data and log-transformed data may not have been optimal because distributions of both raw costs and log-costs were nonnormal.

CONCLUSIONS

The bootstrap method does not need to meet the assumptions of normality and equal variances. In analyses of small sample sizes with skewed cost data, the bootstrap method may offer an alternative to the more traditional nonparametric or log-transformation techniques.