An overview and discussion of the Patient-Centered Outcomes Research Institute's decision aid portfolio.

Decision aids (DAs) help patients make informed healthcare decisions in a manner consistent with their values and preferences. Despite their promise, DAs developed with public research dollars are not being implemented and adopted in real-world patient care settings at a rate consistent with which they are being developed. To appraise the sum of the parts of the portfolio and create a strategic imperative surrounding future funding, the Patient-Centered Outcomes Research Institute (PCORI) tasked the Duke Evidence Synthesis Group with evaluating its DA portfolio. This paper describes PCORI's portfolio of DAs according to the Duke Evidence Synthesis Group's analysis in the context of PCORI's mission and the field of decision science. The results revealed a diversity within PCORI's portfolio of funded DA projects. Findings support the movement toward more rigorous DA development, assessment and maintenance. PCORI's funding priorities related to DAs are clarified and comparative questions of interest are posed.

Models in the development of clinical practice guidelines.

Clinical practice guidelines should be based on the best scientific evidence derived from systematic reviews of primary research. However, these studies often do not provide evidence needed by guideline development groups to evaluate the tradeoffs between benefits and harms. In this article, the authors identify 4 areas where models can bridge the gaps between published evidence and the information needed for guideline development applying new or updated information on disease risk, diagnostic test properties, and treatment efficacy; exploring a more complete array of alternative intervention strategies; assessing benefits and harms over a lifetime horizon; and projecting outcomes for the conditions for which the guideline is intended. The use of modeling as an approach to bridge these gaps (provided that the models are high-quality and adequately validated) is considered. Colorectal and breast cancer screening are used as examples to show the utility of models for these purposes. The authors propose that a modeling study is most useful when strong primary evidence is available to inform the model but critical gaps remain between the evidence and the questions that the guideline group must address. In these cases, model results have a place alongside the findings of systematic reviews to inform health care practice and policy.

Comparative effectiveness research in clinical practice.

The Agency for Healthcare Research and Quality (AHRQ) has funded systematic reviews of comparative effectiveness research in 17 areas over the last 10 years as part of a federal mandate. These reviews provide a reliable and unbiased source of comprehensive information about the effectiveness and risks of treatment alternatives for patients and clinicians. This article describes comparative effectiveness research, provides an overview of how physicians can use it in clinical practice, and references important contributions made by the Minnesota Evidence-based Practice Center.

Comparative effectiveness research in practice and policy for radiation oncology.

Interest in comparative effectiveness research (CER) has increased dramatically over the past decade, yet perceptions about what comprises CER varies. CER has several attributes relevant to practice and policy: (1) The goal of CER is to inform decisions about health care. (2) Literature synthesis is used in addition to primary research. (3) CER evaluates not only overall outcomes for the population but also evaluates subgroups that may have heterogeneous outcomes. (4) Research places an emphasis on outcomes in the "real-world" settings. (5) Outcomes studied should be relevant to patients. In radiation oncology, where many of the traditional clinical trials are comparative in nature, the line between CER and "traditional" research may be blurred, but an increased emphasis on CER can help to bridge the research enterprise and clinical practice, helping to inform decision making at the patient, clinician, and policy levels.

Chapter 10: deciding whether to complement a systematic review of medical tests with decision modeling.

Limited by what is reported in the literature, most systematic reviews of medical tests focus on "test accuracy" (or better, test performance), rather than on the impact of testing on patient outcomes. The link between testing, test results and patient outcomes is typically complex: even when testing has high accuracy, there is no guarantee that physicians will act according to test results, that patients will follow their orders, or that the intervention will yield a beneficial endpoint. Therefore, test performance is typically not sufficient for assessing the usefulness of medical tests. Modeling (in the form of decision or economic analysis) is a natural framework for linking test performance data to clinical outcomes. We propose that (some) modeling should be considered to facilitate the interpretation of summary test performance measures by connecting testing and patient outcomes. We discuss a simple algorithm for helping systematic reviewers think through this possibility, and illustrate it by means of an example.

Inventory of data sources for estimating health care costs in the United States.

OBJECTIVE: To develop an inventory of data sources for estimating health care costs in the United States and provide information to aid researchers in identifying appropriate data sources for their specific research questions. METHODS: We identified data sources for estimating health care costs using 3 approaches: (1) a review of the 18 articles included in this supplement, (2) an evaluation of websites of federal government agencies, non profit foundations, and related societies that support health care research or provide health care services, and (3) a systematic review of the recently published literature. Descriptive information was abstracted from each data source, including sponsor, website, lowest level of data aggregation, type of data source, population included, cross-sectional or longitudinal data capture, source of diagnosis information, and cost of obtaining the data source. Details about the cost elements available in each data source were also abstracted. RESULTS: We identified 88 data sources that can be used to estimate health care costs in the United States. Most data sources were sponsored by government agencies, national or nationally representative, and cross-sectional. About 40% were surveys, followed by administrative or linked administrative data, fee or cost schedules, discharges, and other types of data. Diagnosis information was available in most data sources through procedure or diagnosis codes, self-report, registry, or chart review. Cost elements included inpatient hospitalizations (42.0%), physician and other outpatient services (45.5%), outpatient pharmacy or laboratory (28.4%), out-of-pocket (22.7%), patient time and other direct nonmedical costs (35.2%), and wages (13.6%). About half were freely available for downloading or available for a nominal fee, and the cost of obtaining the remaining data sources varied by the scope of the project. CONCLUSIONS: Available data sources vary in population included, type of data source, scope, and accessibility, and have different strengths and weaknesses for specific research questions.

Economic evaluation alongside a clinical trial of psycho-educational interventions to improve adjustment to survivorship among patients with breast cancer.

PURPOSE: There is little economic research on psychosocial interventions. We aimed to collect data alongside a randomized trial to compare the costs and benefits of three psycho-educational strategies to improve transition to cancer survivorship. METHODS: We evaluated the incremental delivery costs per unit increase in energy (using the Medical Outcomes Study vitality scale) or decrease in distress (from the Revised Impact of Events Scale) in the 6 months postintervention. We also evaluated 1-year post-treatment health care costs. RESULTS: The costs of the control, video, and video plus counseling arms were $11.30, $25.85, and $134.47 per person, respectively. The video costs were $2.22 per unit increase in energy compared with control; among women who were the least prepared for transition, the video was more effective, resulting in even lower costs. The video cost $7,275 per unit change in distress versus control, but costs were lower in the subgroup least prepared for transition ($355). The counseling arm was more expensive and less effective than the video for virtually all end points. However, in one group, women more prepared for transition, counseling cost $1,066 per unit decrease in distress compared with the video. Health care costs tended to increase as intervention intensity increased. CONCLUSION: There are no standards for evaluating cost-effectiveness of trials with psychosocial end points. In this trial, the educational video was the most cost-effective way to improve transition to survivorship. It will be important to confirm whether there is an increased use of services after such interventions and if this represents appropriate use of rehabilitative and supportive care or over-use.

Health limitations and quality of life associated with cancer and other chronic diseases by phase of care.

OBJECTIVE: To estimate health limitations and health-related quality of life (HRQL) associated with cancer and other chronic conditions in a nationally representative sample within a phase-of-care framework. STUDY DESIGN AND SETTING: We used a nested case-control design to assess health limitations and HRQL in individuals reporting a breast, colorectal, prostate, or lung cancer diagnosis, or a diagnosis of arthritis, diabetes, heart disease, or hypertension compared with similar controls without these conditions. All subjects were selected from the 1986-1994 National Health Interview Surveys linked to mortality files in 1995, and classified into the initial, continuing, or last year of life phase of care. Health limitations and HRQL were compared for cases and controls for each condition with 2-sided statistical tests. RESULTS: Across all conditions, individuals in the last year of life phase of care reported greater health limitations and lower HRQL, as measured by the Health Activities and Limitations Index (HALex), than did individuals in the initial and continuing phases of care. Compared with their matched controls, individuals with cancer or other chronic conditions were more likely to report health limitations and lower mean HALex values in the initial, continuing, and last year of life phases of care (P < 0.05). CONCLUSIONS: We observed greater health limitations and lower HRQL associated with cancer and other chronic diseases compared with similar individuals without these conditions. The phase-of-care framework used in this study seems to be applicable to the assessment of HRQL for cancer and other chronic diseases.

Process utility in breast biopsy.

PURPOSE: To determine whether the waiting trade-off (WTO) is feasible for differentiating short-term biopsy preferences in an acute situation where anxiety is the symptomatic disease state. METHODS: 75 women with past experience of either breast core-needle biopsy (CNB), more invasive excisional surgical biopsy (EXB), or both, had telephone WTO assessments. Patients' baseline and test-related anxiety were valued by time trade-off (TTO) used to scale the WTO. Rating scales (RS) were obtained for convergent validity assessment with WTO and TTO. RESULTS: Data were obtained in 38 women who had both CNB and EXB ("paired") and 20 who had CNB only and 16 who had EXB only ("unpaired"). Patients rated only the procedure(s) they experienced. Median paired and mean unpaired WTO scores indicated patients were willing to wait significantly longer to avoid EXB (P = 0.0003, P = 0.0002, respectively). The waiting time difference between EXB and CNB was 2.1 weeks greater in unpaired data than paired data. RS scores comparing the procedures were significantly different only for paired data (P < 0.05). Median TTO preferences for baseline (1.00) and test anxiety (0.93) obtained in 74 patients were significantly different (P < 0.0001) and consistent with RS. Correlation was noted between WTO and RS (-0.307 to -0.453, P = 0.0205 to 0.0001). The median EXB quality-adjusted life years toll (1.5 quality-adjusted life days) calculated from pooled WTO data (paired and unpaired) from 54 patients is near a threshold in a published model. CONCLUSION: The WTO is feasible for discriminating preferences for short-term health states in an acute medical scenario where it might have been expected to be impracticable.

Report of nationally representative values for the noninstitutionalized US adult population for 7 health-related quality-of-life scores.

BACKGROUND: Despite widespread use of generic health-related quality-of-life (HRQoL) scores, few have publicly published nationally representative US values. PURPOSE: To create current nationally representative values for 7 of the most common HRQoL scores, stratified by age and sex. METHODS: The authors used data from the 2001 Medical Expenditures Panel Survey (MEPS) and the 2001 National Health Interview Survey (NHIS), nationally representative surveys of the US noninstitutionalized civilian population: The MEPS was used to calculate 6 HRQoL scores: categorical self-rated health, EuroQoL-5D with US scoring, EuroQoL-5D with UK scoring, EuroQol Visual Analog Scale, mental and physical component summaries from the SF-12, and the SF-6D. The authors estimated Quality of Well-being scale scores from the NHIS. RESULTS: They included 22,523 subjects from MEPS 2001 and 32,472 subjects from NHIS 2001. Most age and sex categories had instrument completion rates above 85%. Females reported lower scores than males across all ages and instruments. In general, those in older age groups reported lower scores than younger age groups, with the exception of the mental component summary from the SF-12. CONCLUSION: This is one of the first sets of publicly available, nationally representative US values for any standardized HRQoL measure. These values are important for use in both generalized comparisons of health status and in cost-effectiveness analyses.

Cost-effectiveness analysis based on the atypical squamous cells of undetermined significance/low-grade squamous intraepithelial lesion Triage Study (ALTS).

BACKGROUND: The ALTS (atypical squamous cells of undetermined significance [ASCUS] and low-grade squamous intraepithelial lesion [LSIL] Triage Study) suggests that, for women diagnosed with ASCUS, human papillomavirus (HPV) DNA testing followed by referral to colposcopy of only those women with oncogenic HPV (i.e., HPV DNA testing) is as effective at detecting cervical intraepithelial neoplasia (CIN) 3 or cancer (CIN3+) as referring all women with ASCUS for immediate colposcopy. We conducted a cost-effectiveness analysis of the ALTS trial to determine whether HPV DNA testing is a cost-effective alternative to immediate colposcopy or conservative management with up to three cytology examinations. METHODS: Data from the ALTS trial were used in conjunction with medical care costs in a short-term decision model. The model compared the incremental costs per case of CIN3+ detected as measured by the incremental cost-effectiveness ratio (ICER) for the following management strategies for women with ASCUS: immediate colposcopy, HPV DNA testing, and conservative management with up to three cytology examinations. RESULTS: The least costly and least sensitive strategy was conservative management with one repeat cytology examination using a threshold of high-grade squamous intraepithelial lesion (HSIL) for referral to colposcopy. Compared with this strategy, triage to colposcopy based on a positive HPV DNA test result had an ICER of 3517 dollars per case of CIN3+ detected. Immediate colposcopy and conservative management with up to three repeat cytology visits detected fewer cases of CIN3+ and were more costly than HPV DNA testing. Immediate colposcopy became cost-effective at 20,370 dollars compared with HPV DNA testing only if colposcopy and biopsy were assumed to be 100% sensitive. CONCLUSIONS: HPV DNA testing is an economically viable strategy for triage of ASCUS cytology. The less than perfect sensitivity of colposcopy and biopsy needs to be accounted for in future clinical guidelines and policy analyses.

Patterns of care in early-stage breast cancer survivors in the first year after cessation of active treatment.

PURPOSE: Patterns of health care use have not been well described for breast cancer survivors. The purpose of this study was to describe the health service use in a survivor cohort. PATIENTS AND METHODS: Women with stage I or II breast cancer were recruited (n = 558) after primary treatment for a multicenter, randomized trial of psychoeducational interventions for facilitating transition to survivorship; 418 women completed the study. Participants completed calendar diaries detailing health care use for 1 year after treatment. Services were coded using Current Procedural Terminology-Fourth Edition codes; costs were estimated using year 2000 Medicare reimbursements. RESULTS: Health care use diary data were available for 391 women (70% of the sample). On average, these survivors reported 30 episodes of health service use in the year after treatment. Total annual costs of care averaged more than 1,800 dollars per survivor; medical office visits were the major component of costs. Type of cancer treatment, depression, and physical function and comorbid illness were independent predictors of the costs of services. There were geographic variations in initial local treatment patterns and in post-treatment costs. Notably, all women should have received surveillance mammography in the time period, but only 61.9% did so; the odds of mammogram receipt were higher for women who had a lumpectomy (v mastectomy) and women who were white (v nonwhite). CONCLUSION: Use of health services is frequent and intensive in the first year after treatment for breast cancer. Despite frequent contact with the health care system, there is room for improvement in providing guideline-suggested surveillance mammography for survivors.

A national catalog of preference-based scores for chronic conditions in the United States.

BACKGROUND: The variability in preferences used in quality-adjusted life-years estimation jeopardizes the comparability of cost-effectiveness analyses and has led the Panel on Cost-Effectiveness in Health and Medicine (the PCEHM) to call for a catalog of "off-the-shelf" preference weights associated with conditions that can be used by health researchers without the burden of collecting primary data. OBJECTIVE: The current research responds to the call by developing a nationally representative catalog of preference-based scores for chronic conditions and associated sociodemographic characteristics. METHODS: The authors report the EQ-5Dindex scores of chronic conditions and associated sociodemographic characteristics in the nationally representative Medical Expenditure Panel Survey (MEPS). Chronic conditions were coded using "quality priority conditions" (QPC) and clinical classification categories (CCC). OLS, Tobit, and censored least absolute deviations (CLAD) regression models were used to provide condition estimates adjusted for age, comorbidity, gender, race, ethnicity, income, and education. RESULTS: Unadjusted and adjusted EQ-5Dindex scores for each QPC and CCC code are presented. EQ-5Dindex scores for older age categories were lower than younger categories, female scores were lower than males, certain racial groups had lower scores than others, and EQ-5Dindex scores were higher for individuals with higher education and income levels. CONCLUSION: The preference-based chronic condition scores reported in this research are nationally representative and may be useful to researchers to calculate quality-adjusted life-years for cost-effectiveness analyses and population-based burden of illness studies without the difficulty of primary data collection. Further research is necessary to validate these scores in condition-specific studies.

Federal sponsorship of cost-effectiveness and related research in health care: 1997-2001.

OBJECTIVES: To describe recent federal sponsorship of cost-effectiveness and related health economics research to provide insight into the functioning of existing research support systems and assess the roles of federal health agencies. METHODS: Using the PubMed database, we identified cost-effectiveness and related publications citing support from a US government entity and published during the period of 1997 through 2001, and audited them for information on funding sources, study type, and content focus. RESULTS: Five Department of Health and Human Services agencies and centers and the Veterans Administration are cited as funders in 74% of 520 federally supported health economics publications we identified. Three-fourths of federally supported publications address five areas of high disease burden: infections, cancer, HIV/AIDS, cardiovascular disease, and substance abuse. Other high burden diseases, including mental health, diabetes, and injuries, receive less attention. Federal support of health economics studies of health education and care delivery-intervention types underexamined in the field-is relatively strong but most often focuses on substance abuse or mental health services. Each of the top federal funders has a distinct funding pattern, but there are substantial areas of overlap within which we could not identify content domains specific to one funder or another. CONCLUSIONS: Federal support of health economics research has paralleled growth in the field. Federal funders support projects consistent with their mission and focus on high-burden disease areas. However, overlapping funding areas, ambiguity concerning agency interests within overlapping content areas, and gaps in some disease and intervention areas suggest that the coordination of health economics research funding could be improved.

Geographic disparities in cervical cancer mortality: what are the roles of risk factor prevalence, screening, and use of recommended treatment?

CONTEXT: Despite advances in early detection and prevention of cervical cancer, women living in rural areas, and particularly in Appalachia, the rural South, the Texas/Mexico border, and the central valley of California, have had consistently higher rates of cervical cancer mortality than their counterparts in other areas during the past several decades. METHODS: This paper reviews the published literature from 1966 to July 2002 to assess three potential pathways underlying this excess mortality--high human papilloma virus (HPV) prevalence, lack of or infrequent screening and advanced disease at diagnosis, and under-use of recommended treatment and shorter survival. FINDINGS: Living in rural areas may impose barriers to cervical cancer control, including lack of transportation and medical care infrastructures. Population characteristics that place women at greater risk for developing and dying from cervical cancer, such as low income, lack of health insurance, and physician availability, are concentrated in rural areas. Published data, however, are insufficient to identify the key reasons for the observed mortality patterns. CONCLUSIONS: At this time, given the lack of definitive evidence in the published literature, decisions about priorities in areas with high rates of cervical cancer mortality will depend on knowledge of current levels of screening, incidence, and stage distribution; and service delivery infrastructures, resources, and acceptability of interventions to the target population.

Burden of illness in cancer survivors: findings from a population-based national sample.

BACKGROUND: Population trends in aging and improved cancer survival are likely to result in increased cancer prevalence in the United States, but few estimates of the burden of illness among cancer survivors are currently available. The purpose of this study was to estimate the burden of illness in cancer survivors in a national, population-based sample. METHODS: A total of 1823 cancer survivors and 5469 age-, sex-, and educational attainment-matched control subjects were identified from the 2000 National Health Interview Survey. Multiple measures of burden, including utility, a summary measure of health, and days lost from work, were compared using two-sided tests of statistical significance for the two groups overall and for subgroups stratified by tumor site and time since diagnosis. RESULTS: Compared with matched control subjects, cancer survivors had poorer outcomes across all burden measures (P<.01). Cancer survivors had lower utility values (0.74 versus 0.80; P<.001) and higher levels of lost productivity and were more likely to report their health as fair or poor (31.0% versus 17.9%; P<.001) than matched control subjects. Cancer survivors reported statistically significantly higher burden than did control subjects across tumor sites and across time since diagnosis (i.e., within the past year, 2-5 years, 6-10 years, and > or =11 years for the majority of measures. CONCLUSIONS: Cancer survivors have poorer health outcomes than do similar individuals without cancer across multiple burden measures. These decrements are consistent across tumor sites and are found in patients many years following reported diagnosis. Improved measurement of long-term burden of illness will be important for future prospective research.

Predicting EuroQoL EQ-5D preference scores from the SF-12 Health Survey in a nationally representative sample.

PURPOSE: To predict the EuroQoL EQ-5D utility index from the SF-12 Health Survey for a US national sample of adults. METHODS: The authors used the 2000 Medical Expenditure Panel Survey to examine the relationship between instruments. Linear regression was used to predict EQ-5D scores from Physical Component Summary (PCS) and Mental Component Summary (MCS) scores of the SF-12. A prediction model was derived in one half of the sample and validated in the other half. RESULTS: Complete responses to both measures were available for 14,580 adults; 7313 (50.2%) surveys were used for the derivation set. The 2-variable model predicted 61% of the variance in EQ-5D scores and provided reasonable ability to predict mean EQ-5D scores from mean PCS and MCS scores. Confidence intervals are dependent on sample size and variance of PCS and MCS scores. CONCLUSIONS: EQ-5D scores can be reasonably predicted from the SF-12. This model allows researchers to estimate utility data for use in decision and cost-utility analyses.

Short-term impact of cancer prevention and screening activities on quality of life.

PURPOSE: There are few data on the short-term effects of participating in cancer prevention activities, undergoing genetic risk assessment, or having routine screening. The objective of this article is to systematically review existing research on short-term effects of prevention, genetic counseling and testing, and screening activities on quality of life. METHODS: We conducted a MEDLINE search for original research studies that were published between January 1, 1985, and December 31, 2002, and conducted in North America or Western Europe. Data were abstracted and summarized using a standardized format. RESULTS: We reviewed 210 publications. Most studies focused on psychological states (anxiety, depression), symptoms, or general health status. One hundred thirty-one studies used 51 previously validated noncancer instruments. Many researchers (12.6%) also added cancer-specific measures, such as perceived cancer risk or symptom indices. Only one study measured satisfaction or quality of provider-client communication. While one report examined lost workdays, no other economic consequences of short-term outcomes were evaluated. Among seven studies that assessed short-term outcomes preferences, only four specifically used time trade-off or linear rating scale methods. No study used standard gamble or willingness-to-pay methods. The overwhelming majority of research indicated that short-term effects were transient. Only two studies linked short-term effects to long-term cancer-related health behaviors such as repeat screening. CONCLUSION: There is considerable heterogeneity in short-term outcome measurement. Clinicians need to be aware of potential for short-term, transient adverse effects. The impact of short-term experiences should to be linked to long-term health status and use of services.

Demographic variation in SF-12 scores: true differences or differential item functioning?

BACKGROUND: Demographic differences have been reported in summary measures of physical and mental health based on the SF-12 instrument. OBJECTIVES: This study examines the extent to which differential item functioning (DIF) contributes to observed subgroup differences in health status. DIF refers to situations in which the psychometric properties of items are not invariant across different groups. The presence of DIF confounds interpretation of subgroup differences. SUBJECTS: A national sample of 11,626 adult respondents in the 2000 Medical Expenditure Panel Survey who completed a self-administered questionnaire. MEASURES: In addition to the SF-12, we collected data on demographic characteristics (age, gender, education, and race/ethnicity) and whether the person had ever been diagnosed with six chronic medical conditions. RESULTS: Multiple-indicator multiple-cause latent variable models showed significant differences in physical health by gender, age, and education. Adjusting for DIF reduced but did not eliminate age and education differences. However, for mental health, adjusting for DIF resulted in Black-White differences becoming nonsignificant, and the effect for the oldest age group switched from positive to negative. Race/ethnicity was not associated with physical health status. CONCLUSIONS: Age group comparisons of mental health may be particularly affected by DIF. Differences in education, as well as age and gender, need to be controlled when making group comparisons. Additional work is needed to understand factors that give rise to demographic differences in reported health status.

Acceptability of diagnostic tests for breast cancer.

PURPOSE: To assess the acceptability of new non-invasive breast cancer diagnostic tests intended to triage women in need of biopsy. METHODS: Women who had abnormal screening tests and had been recommended to have a biopsy were invited to receive digital mammography, magnetic resonance imaging (MRI), and nuclear medicine evaluation (Tc-99m-sestamibi scanning) before biopsy. Participants completed a questionnaire about satisfaction and acceptability of the procedures. Satisfaction measured women's overall and test-specific satisfaction. Acceptability was measured by self-reported discomfort, embarrassment and women's preference in terms of willingness to pay to avoid a biopsy. RESULTS: Women were satisfied with all of the potential diagnostic triage procedures. Most found the tests more comfortable than a routine mammogram (47, 50, and 66% undergoing MRI, digital mammography, and sestamibi scanning, respectively). Women who provided a response to willingness to pay questions (N = 43) were willing to pay an average of 611 dollars to have a test instead of a biopsy, if the test was as accurate as biopsy. The willingness to pay significantly decreased to 308 dollars if the test only had 95% accuracy. Those who had prior benign breast disease were less willing to pay for a test with 95% accuracy than those without this history. CONCLUSION: Instead of immediate biopsy after an abnormal screening, these results suggest that women would find non-invasive triage tests acceptable, or preferable to biopsy if they were equally accurate or nearly equally accurate as a biopsy. New technologies to diagnose breast cancer should focus on decreasing discomfort as well as increasing test accuracy.