Jane Alcyne Severson

Michigan Oncology Practices Showed Varying Adherence Rates To Practice Guidelines, But Quality Interventions Improved Care

Despite improvements in care for patients with cancer, and in their survival rates, it is not clear that best practices are uniformly delivered to patients. We measured the quality of outpatient cancer care, using validated quality measures, in a consortium of thirty-six outpatient oncology practices in Michigan. We discovered that throughout the measurement period, for breast and colorectal cancer care, there was a more than 85 percent rate of adherence to quality care processes. For end-of-life care processes, the adherence rate was 73 percent, and for symptom and toxicity management care processes, adherence was 56 percent. In particular, we found variations in care around the fundamental oncologic task of management of cancer pain. To address quality gaps, we developed interventions to improve adherence to treatment guidelines, improve pain management, and incorporate palliative care into oncology practice. We concluded that statewide consortia that assume much of the cost burden of quality improvement activities can bring together oncology providers and payers to measure quality and design interventions to improve care.

Administration of Oral Chemotherapy: Results From Three Rounds of the Quality Oncology Practice Initiative

PURPOSE Although use of oral chemotherapy is becoming more prevalent, little is known about the quality of care that patients receive when these agents are prescribed. Moreover, few practice-level systems are in place to ensure safe management of oral chemotherapy in the vulnerable population of patients with cancer. METHODS We analyzed results from 155 practices that were voluntarily participating in the American Society of Clinical Oncology Quality Oncology Practice Initiative (QOPI) program on 17 test measures of oral chemotherapy administration and management in at least one of three collection periods: spring or fall of 2012, or spring of 2013. The 17 test measures cover three domains: treatment plan documentation, patient education, and adherence/toxicity monitoring. We defined composite scores for each of the three domains. We analyzed the composite scores by secular trend and tested the difference in composite scores for the three domains. Additionally, we tested change in scores over time among practices that participated at least twice. RESULTS The majority of data was provided by QOPI-certified practices. Overall, mean practice composite scores ranged from 66% to 68% for treatment plan documentation, 51% to 57% for patient education, and 75% to 81% for adherence/toxicity monitoring. Composite scores for practices that participated more than once did not improve significantly. CONCLUSION The collection of oral chemotherapy test measure data is feasible. Composite scores for treatment plan documentation and patient education were not only lower, but had greater variability compared with adherence/toxicity monitoring. Improvement opportunities exist for patients who are prescribed oral chemotherapy.

Effect of intervention on quality measures of symptom management in the Michigan Oncology Quality Consortium (MOQC).

70 Background: Michigan oncology practice groups that participated in MOQC [JOP 5(6):281, 2009] used the Quality Oncology Practice Initiative (QOPI) tool. Adherence to processes of disease specific care was high, but poor in domains associated with palliative care. These measures did not change over time [Health Affairs. 31(4):718, 2012]. These findings prompted us to test interventions to improve quality in palliative care domains. METHODS MOQC created a process, based on the IHI Framework for Spread, to assist oncology practice groups in establishing their own primary Palliative Care services, including the implementation of Edmonton Symptom Management Scale. 8 practice groups formed teams of local change agents to participate in the Palliative Care Demonstration (PC Demo) project. The teams participated in 3 in-person and 4 online learning sessions over 8 months, led by palliative care and quality experts. Teams were provided tools, training materials, and necessary support to implement the improvements and measure their success. The learning network facilitated the sharing of best practices and lessons learned throughout the process. The teams presented their results broadly to other MOQC participants at project conclusion. RESULTS Success was measured using palliative care-focused ASCO QOPI results. PC Demo sites consecutively improved their scores in many of the QOPI measures, and their rate of improvement from Fall 2011 to Spring 2012 was greater than that of their peers. CONCLUSIONS We observed that collecting and distributing data in our consortium was insufficient to improve palliative oncology care. Providing practice groups with the appropriate infrastructure improved their capacity and capability to make the necessary changes to improve performance. [Table: see text].

Association of high symptom burden with oral oncolytic agents.

177 Background: Increasing numbers of cancer patients are being treated with oral oncolytics. This change represents a shift from frequent direct observation during intravenous therapy to periodic observation and increased necessity for improved monitoring and self-care management. Despite this shift, patients are at risk to develop many of the same chemotherapy-associated symptoms and toxicities. We sought to understand the symptom burden associated with use of oral oncolytics. METHODS Michigan Oncology Quality Consortium (MOQC) sponsored a quality improvement initiative focused on improving oral oncolytic care. Eight oncology practices participated. Patients were assessed with a modified Edmonton Symptom Assessment System (ESAS) prior to each outpatient visit. A total of 537 surveys were analyzed. 13 measures were categorized into mild (0 to 3), moderate (4 to 6), and severe (7 to 10) symptom burden. RESULTS Overall, the average ESAS symptom score was mild in 81% of patients, moderate in 13% of patients, and severe in 6% of patients. These average scores, however, obscure the significant burden in select ESAS domains. For example, 39% of patients categorized their overall well-being as being moderate/severely affected; 38% of patients felt they were moderate/severely fatigued, and 21% indicated moderate/severe neuropathy symptoms. Notably, 118/537 (24%) of the assessments had 4 or more symptoms rated moderate to severe. CONCLUSIONS Patients taking oral oncolytics experience significant symptom burden that impacts quality of life. Intolerance of oral oncolytics may lead to adherence issues, potentially affecting expected outcomes. Given the prevalence of symptoms and potential for toxicity, self-care strategies to improve early recognition and treatment of symptoms by patients taking oral oncolytics are necessary. [Table: see text].

Implementing a Method for Evaluating Patient-Reported Outcomes Associated With Oral Oncolytic Therapy

INTRODUCTION The paradigm shift in health care toward value-based reimbursement has brought emphasis to providing better quality of care to patients with chronic diseases, including patients with cancer. In accordance with providing better quality of care to patients, there has been a growing interest in evaluating quality of life through patient-reported outcomes (PROs). The revised Edmonton Symptom Assessment Scale (ESAS-r) is a tool that can be used to assess PROs and has been validated for use in patients with cancer. This initiative sought to use this standard assessment tool to acquire PROs concerning symptom burden from patients prescribed oral oncolytics. PATIENTS AND METHODS Eight oncology practices in the state of Michigan used a modified ESAS-r to evaluate symptom burden of patients prescribed oral oncolytics before each outpatient visit. Thirteen symptoms were categorized as mild (0 to 3), moderate (4 to 6), or severe (7 to 10). RESULTS A total of 1,235 modified ESAS-r surveys were collected and analyzed; 82.5% of symptoms were categorized as mild, 11.9% of symptoms were categorized as moderate, and 5.6% of symptoms were categorized as severe. CONCLUSION PROs can be evaluated through the use of a standardized tool, such as the ESAS-r, in oncology patients receiving oral oncolytic therapy. Implementing such a tool in both community and academic practices is feasible and may facilitate improvements in the quality of care.

Engagement in a statewide oral oncolytic collaborative and practice impact.

89 Background: The rapid shift to oral oncolytic therapy presents challenges to oncology practitioners. The purpose of this study is to describe how participation in a statewide oral oncolytic improvement collaborative where best practices and resources were shared can readily impact quality of care as measured by national standards. METHODS The Michigan Oncology Quality Consortium (MOQC) hosted a series of learning collaborative sessions focused on topics and deployment of resources specific to oral oncolytic management and quality improvement. Participating practices performed pre/post self-assessments in Oct. 2013 and Apr. 2015 (n = 3). Concordance with national ASCO QOPI and ONS standards was compared, including documentation (5 measures), patient education (7 measures), and follow-up/monitoring (4 measures). A response scale of always, sometimes, and never was used. Specifically, practices were surveyed to assess which evidence-based MOQC resources were implemented: patient intake template, drug-specific self-management guides, start date mailer, medication calendar, primary care physician communication template, Edmonton Symptom Assessment Scale (ESAS), and patient adherence questionnaires. RESULTS Practice A showed improvement in documentation, patient education, and monitoring. This practice used the initial oral chemotherapy template, ESAS, patient education templates, and patient calendar (Table). Practice B implemented 6 resources and demonstrated improvements in 15 metrics. Practice C implemented 4 resources, namely patient-focused resources, to improve all patient education and monitoring metrics. CONCLUSIONS Use of the collaborative model and supplying oncology teams with scientific evidence, standard workflows, and resources improves concordance with national standards of care. Large-scale deployment of this model program may provide a clinically efficient and effective mechanism to enhance widespread change. [Table: see text].

Confidence in self-managing side effects from oral oncolytics among a sample of Michigan oncology practices.

68 Background: The aim of this study was to assess patient confidence to self-manage common side effects of oral oncolytic agents among cancer patients whose providers participated in the Michigan Oncology Quality Consortium (MOQC). METHODS This was a cross-sectional study conducted between August and September 2015, enrolling a convenience sample of 125 patients currently taking oral oncolytics in 10 oncology practices who completed a brief online survey. Demographics, clinical information, health literacy (3 items) and Patient Activation Measure (PAM, 13 items) were assessed. The PAM instrument assesses patient self-reported knowledge, skills and confidence for self-management. Participants were also asked to rate on a scale 0-10 how confident they felt self-managing 3 common oral oncolytic side-effects: fatigue, nausea and diarrhea. Differences by age, gender, education level, health literacy, and PAM level were examined. RESULTS Respondents were 66.2 (SD, 13.6) years old and 57.7% were female. The mean length of oral oncolytic therapy was 12.9 (SD, 13.8) months and average number of medications was 5.9 (SD, 3.2). Approximately three-quarters of patients assessed had high levels of patient activation (41.5% level 3 and 32.5% level 4), with the remaining 26.0% having low involvement in their care (17.1% level 1 and 8.9% level 2). Confidence to manage fatigue was high for 79% of patients, moderate for 12.1% and low for 8.9%. Confidence to manage nausea and diarrhea was higher with 85.5% and 88.7% patients reporting high confidence, respectively, but 6.5% and 4.8% reporting low confidence to manage these two symptoms. When combining the confidence measures to manage each of the three symptoms, patients with higher health literacy (p = 0.001), higher PAM levels (p = 0.018) and older patients (p = 0.041) exhibited higher confidence at a statistically significant degree. CONCLUSIONS Healthliteracy and PAM were associated with patient confidence to self-manage oral oncolytic side effects. Health literacy screening in clinical practice may be an efficient means to identify patients with low self-management confidence and determine needs of further education and support.

Abstract 3728: Rapid deployment of statewide tobacco cessation program for cancer patients

Purpose Tobacco use is not only associated with the development of cancer, but after a patient develops cancer, its continued use is associated with poorer cancer treatment outcomes. Despite this fact, most oncologists do not regularly provide tobacco cessation support. The purpose of this study was to implement an expedited process of patient identification and tobacco cessation for cancer patients. Methods The Michigan Oncology Quality Consortium (MOQC) collaborated with the Michigan Department of Community Health, Blue Cross and Blue Shield of Michigan, and the Michigan Cancer Consortium to enhance tobacco cessation support for cancer patients. A statewide tobacco cessation demonstration project using a lean engineering approach was developed to provide structured tobacco assessments, referral to state quitline resources, and provide tailored cessation support to cancer patients consisting of behavioral counseling and free pharmacotherapy where appropriate. These steps were rapidly deployed in over 30 locations including those in private oncology practices and cancer centers in the first year. Data were analyzed to evaluate participation, retention, and quit rates for cancer patients enrolled in the program. Results Previous reports have described high rates of patient contact and participation in the cessation program. In the 690 patients with at least 12 months since program enrollment, 200 (29%) participated in only 1 call, 313 (45%) participated in 1-2 follow-up calls, and 177 (26%) participated in 3 or more follow-up calls. Follow-up 30 day quit rate data was obtained in 198, 138, and 199 patients at 3, 6, and 12 months respectively. Self-reported quit rates remained largely consistent with 29.8% at 3 months, 27.5% at 6 months, and 30.2% at 12 months. Conclusions The MOQC demonstration project represents a model partnership between payors, clinicians, and statewide quitlines that maintained consistent and moderate quit rates in cancer patients. Large-scale deployment of this model program may provide a clinically efficient and effective mechanism to enhance widespread tobacco cessation efforts for cancer patients. Citation Format: Laura Petersen, Hilary Baca, Karen Brown, Jamie Lindsay, Sarah Mott, Jane Severson, Anna Schulze, Graham Warren. Rapid deployment of statewide tobacco cessation program for cancer patients. [abstract]. In: Proceedings of the 106th Annual Meeting of the American Association for Cancer Research; 2015 Apr 18-22; Philadelphia, PA. Philadelphia (PA): AACR; Cancer Res 2015;75(15 Suppl):Abstract nr 3728. doi:10.1158/1538-7445.AM2015-3728

Abstract 5042: Using a statewide collaborative approach to improve tobacco cessation referral rates for cancer patients

Proceedings: AACR Annual Meeting 2014; April 5-9, 2014; San Diego, CA Purpose: Though tobacco use by cancer patients is associated with poor therapeutic outcomes, most oncologists do not regularly provide tobacco cessation support for cancer patients. The purpose of this study was to determine if dedicated statewide services could substantially enhance delivery of evidence based cessation support to broad cancer patient populations across many oncology clinics. Methods: The Michigan Oncology Quality Consortium (MOQC) collaborated with the Michigan Department of Community Health and the Michigan Cancer Consortium to identify and address tobacco cessation support that could be improved for cancer patients. A statewide Tobacco Cessation Demonstration Project was developed to reduce patient barriers to access by increasing education, identifying all cancer patients who use tobacco, referring them to the Michigan Tobacco Quitline, and providing funding for free Quitline telephone counseling services and nicotine replacement therapy for cancer patients. Reported are results in 19 clinical practices. Results: Implementation of this project resulted in tobacco assessments in more than 98% of patients and increased tobacco cessation referral rates for participating practices from 15% in May 2013 to over 60% in July and September of 2013 (p < .001). Total statewide referrals to the Quitline increased by between 32% and 60% over the first 6 months, with an increasing proportion of patients (between 31% and 50%) being cancer patients referred by MOQC participating sites. View this table: Conclusions: Reducing patient barriers to access to tobacco cessation services and supplying clinic teams with scientific evidence, standard workflows, and lean tools improved the referral rate to a tobacco cessation program for oncology patients. Critical factors for success included strong physician leadership, adoption of an auto referral process, and front line staff accountability for identifying all patients who use tobacco and assessing their tobacco cessation status at subsequent clinic visits. Citation Format: Jane Severson, Hilary Baca, Douglas Blayney, Karen Brown, Grayce Galiyas, Mara Minasian, Joan Schmidt, Anna Schulze, Graham Warren. Using a statewide collaborative approach to improve tobacco cessation referral rates for cancer patients. [abstract]. In: Proceedings of the 105th Annual Meeting of the American Association for Cancer Research; 2014 Apr 5-9; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2014;74(19 Suppl):Abstract nr 5042. doi:10.1158/1538-7445.AM2014-5042

Using a statewide collaborative approach to improve tobacco cessation referral rates for cancer patients.

74 Background: Tobacco use by cancer patients decreases the effectiveness of cancer treatment, increases treatment toxicity, increases the risk of developing a second primary cancer, and increases mortality. Stopping tobacco use may reduce many of these adverse effects in cancer patients. METHODS The Michigan Oncology Quality Consortium (MOQC) collaborated with the Michigan Cancer Consortium (MCC) to implement the Tobacco Cessation Demonstration Project. This entailed designing a standard workflow and other lean tools to assist oncology practices in referring all cancer patients who use tobacco to the free Michigan Tobacco QuitLine or other cessation services. 19 practices participated in three learning sessions during which the following were provided: education by subject matter experts, use of data management and lean tools, and the sharing of barriers and successes. RESULTS Examination of baseline MOQC Quality Oncology Practice Initiative (QOPI) data demonstrated that < 47% of patients who use tobacco were advised to quit or referred for tobacco cessation (n = 574 charts/44 sites), which was equivalent to the national QOPI mean in Fall 2012. Root cause analysis of this lack of action in tobacco cessation identified limited knowledge to the risks of continued smoking and the lack of available resources as barriers to adopting necessary change. Implementation of the MOQC/MCC Tobacco Cessation Demonstration Project increased total statewide referrals to the QuitLine by 30.9% over the first 2 months with an increasing proportion of patients (41%) being cancer patients referred by MOQC participating sites. CONCLUSIONS Supplying scientific evidence, collaboration, and the use of lean tools improved the referral rate to a tobacco cessation program for oncology patients. [Table: see text].