Abstracts From the 21st Annual HMO Research Network Conference, March 11–13, 2015, Long Beach, California
Journal of Patient-Centered Research and Reviews
Journal of Patient-Centered Research and Reviews ( JPCRR) is a peerreviewed scientific journal whose mission is to communicate clinical and bench research findings, with the goal of improving the quality of human health, the care of the individual patient, and the care of populations. Recommended Citation Background/Aims: Prostate cancer is the most common and costly cancer among U.S. men and offers a portfolio of case studies and clinical problems. Among these, the documented increases in use
... f robotic radical prostatectomy (RRP) and intensity-modulated radiotherapy (IMRT) are illustrative of rapid diffusion of costly, unproven technological innovations. Evidence of these case studies has been limited to the traditional Medicare fee-for-service (FFS) aged population and may provide a potentially biased view of the universe of cancer care in the United States. To address this issue, we examine treatment patterns among prostate cancer patients enrolled in two integrated health care delivery systems. Methods: The sample included a retrospective cohort of patients diagnosed with nonmetastatic prostate cancer between 2000 and 2008 at the Northwest and Colorado regions of Kaiser Permanente. Choice of radiation therapy was defined as the receipt of external beam radiotherapy (threedimensional [3D] conformal radiation therapy vs. IMRT) as primary treatment within one year of diagnosis. Choice of prostatectomy was defined as the receipt of surgery as primary treatment (open radical prostatectomy vs. minimallyinvasive/robotic radical prostatectomy). A pooled time-series cross-section design was used to analyze the medical care resource use for treating prostate cancer. Results: Both RRP and IMRT were introduced in the two health systems in 2003, and their use has been steadily increasing since then. Use of traditional treatments has either remained stable: open radical prostatectomy; or decreased: 3D conformal radiation therapy. Discussion: We hypothesize that diffusion rates of RRP and IMRT have been slower among HMO enrollees compared with FFS patients. This study improves our understanding of patterns of prostate cancer care in the United States by using a large sample of managed care enrollees across the age spectrum, including men younger than 65 relative to men older than 65 years. Background/Aims: This study evaluated the costeffectiveness of interventions using electronic health records (EHR), automated mailings, and stepped increases in support to improve two-year colorectal cancer screening adherence. Methods: Analyses are based on a parallel-design, randomized trial in which three stepped interventions (EHR-linked mailings ["automated"], automated plus telephone assistance ["assisted"], or automated and assisted plus nurse navigation to testing completion or refusal ["navigated"]) were compared to usual care. Data were collected over August 2008-November 2011 with analysis in 2012-2013. Implementation resources were micro-costed; research and registry development costs were excluded. Incremental cost-effectiveness ratios (ICERs) were based on number of participants current for screening per guidelines over two years. Robustness of results was examined through bootstrapping. Results: Intervention delivery cost per participant current for screening ranged from $21 (automated) to $27 (navigated). When induced testing costs (e.g. screening colonoscopy) were included, automated (ICER: -$159) and assisted (ICER: -$36) were cost-saving relative to usual care. Savings arose from increased fecal occult blood testing, substituting for more expensive colonoscopies in usual care. Results were broadly consistent across demographic subgroups. More intensive interventions were consistently likely to be cost-effective relative to less intensive interventions with willingness to pay values of $600-$1,200 for a 1% increase in the rate of screening adherence yielding probability of cost-effectiveness of at least 80%. Discussion: The cost-effectiveness was demonstrated of a stepped approach to colorectal cancer screening promotion, especially one using automated data systems linked to electronic health records. Background/Aims: Over half of cervical cancers in the United States are diagnosed in unscreened or underscreened women. Innovative strategies that move primary screening out of the clinic could be highly effective in improving screening compliance while maintaining high-quality care. Internationally, there is growing interest in a primary screening strategy of self-collected, in-home high-risk human papillomavirus (hrHPV) screening followed by triage of hrHPV-positive results to clinician-collected cytology or colposcopy. The need for in-clinic screening could be eliminated for a majority of women in organized screening programs if hrHPV screening can be seamlessly integrated with in-clinic follow-up. Methods: In March 2014, we launched a large, pragmatic randomized controlled trial within Group Health to compare effectiveness of two programmatic approaches for increasing cervical cancer screening among 17,000 women aged 30-64 years who are overdue (≥3.4 years since last Pap test or never had a Pap test). The control arm is usual care at Group Health, consisting of an organized program of patient invitation and reminders and electronic medical record (EMR)-based alerts to providers about patients who are not up-to-date with recommended screening. The intervention arm includes usual care plus a mailed in-home hrHPV screening kit. Women mail their screening kits directly to our centralized laboratory, and hrHPV test results are documented in our EMR to notify women's provider teams of appropriate follow-up care. Primary outcomes are early detection and treatment of cervical neoplasia. Secondary outcomes are cervical cancer screening uptake and screening predictors, and patient experiences and attitudes towards in-home hrHPV testing and follow-up of hrHPV-positive results. We are surveying 100 responders and 100 nonresponders to the in-home hrHPV screening invitation, and interviewing 50 women with hrHPV-positive results (including those with timely versus late completion of clinical follow-up). Results: As of September 2014, 6,194 eligible women were identified and randomized. We will highlight the study design and lessons learned from implementing a pragmatic trial. Discussion: This pragmatic trial will generate evidence-based data on the impact of an in-home hrHPV screening program in a real-world clinical setting. Findings will interest organized programs considering implementing in-home screening and low-resource countries caring for populations with limited access to primary care. Background/Aims: The Cancer Research Network (CRN) Scholars Program was launched during the third cycle of the collaborative grant to increase research capacity within the CRN. Specific goals of the program include: (1) becoming a principal investigator on a successful investigator-initiated grant within the CRN, and (2) primary author on peerreviewed, published manuscripts reporting original research from the CRN. In addition to developing research expertise utilizing CRN resources, the CRN Scholars Program fosters development of skills difficult to acquire in other training environments, including the initiation of collaborative research proposals involving stakeholders in integrated health care delivery systems, use and analysis of complex clinical, claims and pharmaceutical data, and strategic grant proposal development tailored to the unique resources and opportunities afforded by the CRN. Methods: The 26-month program facilitates individualized mentorship, creation of research advisory teams for each scholar comprised of investigators from their home institution and collaborating CRN sites. Monthly one-on-one mentoring meetings and quarterly mentoring team meetings enable scholars to gain familiarity with the CRN landscape to develop novel, investigator-initiated projects with CRN collaborators. Additionally, scholars attend semiannual in-person meetings and conduct bimonthly group conference calls examining academic career development, writing and review of research proposals, analytic design, discussing available funding opportunities, execution of multisite research studies, and media training. Results: The first two Scholars Program cohorts included CRN-affiliated junior investigators, while the current iteration of the Scholars Program (CRN4) has shifted focus to include external junior investigator researchers pursuing training in the conduct of multisite, multidisciplinary population-based studies. Members of the CRN4 cohort have successfully received career development awards and independent grants to support their CRN work, published several first-authored manuscripts and received awards honoring their research. Discussion: The CRN Scholars Program is working to expand the scope and scale of investigators conducting cancer research within integrated health care delivery systems, a goal of significant importance given nationwide shifts in policy and practice in the organization of medical care and emphasis on population health. Accordingly, the CRN Scholars Program is well positioned to produce future leaders in cancer research and, ultimately, improve the lives of patients. Background/Aims: The development of a portable, automated method for identifying individuals with lung nodules will facilitate the efficient conduct of populationbased studies of nodule care and associated outcomes. We evaluated the performance of a previously developed natural language processing (NLP) algorithm for identifying health plan beneficiaries with pulmonary nodules. Methods: A cross-sectional study was performed of 500 randomly selected adult, in-network health plan beneficiaries with continuous enrollment at Group Health Cooperative who underwent a computed tomography (CT) of the chest in 2012, had no history of lung cancer and had not undergone a CT between 2009 and 2011. An NLP algorithm originally developed at Kaiser Permanente Southern California assessed electronic radiology reports using keywords and qualifiers relating to pulmonary nodules ranging in size from 5 to 30 mm among individuals who had undergone CT and had an International Classification of Diseases (ICD-9-CM) diagnostic code for a lung nodule. This algorithm was applied to our patient population and modified to identify pulmonary nodules regardless of size. A trained chart abstractor reviewed radiology reports to determine whether the radiologist reported a lung nodule. An experienced, board-certified thoracic surgeon adjudicated radiology reports with unclear documentation of a nodule. Results: The true prevalence of pulmonary nodules among individuals undergoing CT in 2012 -median age 65 years, 43% men, 84% white, 51% smokers -was 34%. Median nodule size was 6 mm (range 2-87 mm). NLP identified 218 (44%) individuals with a nodule. The accuracy of NLP was as follows: sensitivity 91%, specificity 81%, positive predictive value 72% and negative predictive value 95%. Discussion: An automated method of using NLP and electronic radiology text reports -originally developed at one Cancer Research Network (CRN) site -reasonably identifies health plan members with pulmonary nodules at another CRN site. This finding supports the notion that automated methods are portable across integrated health systems and institutions using electronic medical records. Ongoing work seeks to determine whether modifications to the NLP algorithm can improve performance. Given its current performance characterized by a high negative predictive value, NLP could be used to decrease the burden of chart abstraction in population-based studies of nodule care. Background/Aims: The rising economic burden of cancer in the United States has become a significant public policy issue. Current knowledge of cancer care costs is largely limited to Medicare fee-for-service patients age ≥ 65 years, based on linked Surveillance, Epidemiology, and End Results (SEER) Medicare data. Accordingly, estimates that reflect care costs across the cancer experience, from diagnosis to end-of-life and among individuals < 65 years of age, are needed. This study estimated total and incremental long-term costs of care for individuals age ≥ 18 years diagnosed with breast, colorectal, lung and prostate cancer in integrated health care settings. Methods: We used administrative and electronic medical record data linked with tumor registry information for individuals who were enrolled in one of four nonprofit integrated health care systems and diagnosed with primary breast, colorectal, lung or prostate cancer between 1/1/2000-12/31/2008 (cases). We also identified age-/sex-matched cancer-free individuals (controls). Participating health systems included Group Health Cooperative, Henry Ford Health System, Kaiser Permanente Northwest and Kaiser Permanent Colorado. Long-term costs (at 5, 10 and 20 years) of care were based on the Phases-of-Care approach (Initial, Continuing and End-of-life phases) and measured using the Kaplan-Meier Sample Average estimator, incorporating cost data from the standardized relative resource cost algorithm and cancer site-, age-and stage-specific survival probabilities from SEER. Costs were calculated for both cases and controls to estimate total and incremental care costs, stratified by cancer site, age group (<65 vs. ≥65 years) and stage at diagnosis (advanced vs. not advanced). Results: Total and incremental long-term costs of cancer care were significantly higher among individuals age < 65 years at diagnosis compared to those ≥ 65 years. There was considerable variability in total and incremental long-term costs by cancer site and stage at diagnosis. Discussion: Estimates reported in this study improve our understanding of the magnitude of long-term economic resources needed to provide cancer care in the United States by describing cancer care costs among patients in integrated care settings and those age < 65 years. Our results will be directly applicable to future analyses of cancer care costs, serving as a foundation for cost comparisons across different health care systems. Background/Aims: Colorectal cancer (CRC) is the second leading cause of cancer death in the United States. CRC screening allows for early detection of CRC or its precursors, and leads to a decrease in CRC incidence and mortality. National guidelines recommend screening with either an annual fecal immunochemical test (FIT), a colonoscopy every 10 years, or a flexible sigmoidoscopy every 5 years for individuals age 50-75 years. Nationally, the overall age-adjusted combined CRC screening rates increased from 52.3% to 65.4% between 2002 and 2010; however, screening rates among Asian Americans were 11% lower compared with whites, but differences among Asian subgroups is unknown. Thus, we examined CRC screening rates among Asian subgroups in a large managed care setting. Methods: We used electronic data to characterize CRC screening among Asian subgroups (Japanese, Korean, Filipino, Asian-Indian, Chinese, Vietnamese, Other Asian) compared with white non-Hispanics (WNHs). Using descriptive and multivariable models, we evaluated 408,242 screening-eligible members of Kaiser Permanente Southern California aged 50-89 years as of January 1, 2010. Results: Compared with WNHs over the past 3 years, Japanese were least likely and Koreans were most likely to have had any CRC screening exam (odds ratio [OR] 0.89, 95% confidence interval [CI] 0.84-0.95; and OR 1.84, 95% CI 1.68-2.02, respectively). Of those screened, compared with WNHs, Japanese men were older (mean age 65.2 vs. 62.6). Screened Asian-Indians and Vietnamese had an equal distribution of men and women, while the other subgroups had more females. In the previous 3 years, Japanese were less likely to have had a FIT (41%, 1,742 of 4,301), and Vietnamese were more likely (56%, 2,248 of 4,018). In the previous 10 years, Koreans were more likely to have had a colonoscopy (56%, 1,797 of 3,189) and Filipinos were least likely (42%, 9,135 of 21,688). Discussion: CRC screening rates and choice of screening test differ among Asian subgroups in an insured population. Further research is needed to understand the reasons for the differences among Asian subgroups, particularly the Japanese population.