Finding Best Practices in Chronic Disease Prevention Sally Honeycutt, MPH, CHES Evaluation Team Lead Emory Cancer Prevention & Control Research Network (CPCRN) These highlighted evaluation projects are supported by the Emory CPCRN, which is part of the Prevention Research Centers Program. It is supported by the Centers for Disease Control and Prevention and the National Cancer Institute (Cooperative agreement # 1U48DP0010909-01-1) Project Goal Generate practice-based evidence to address gaps in the research literature for cancer prevention and control by identifying and evaluating promising cancer prevention programs developed and conducted by organizations in southwest Georgia Project Activities Environmental Scan Evaluability Assessment Evaluations Dissemination Environmental Scan • Community Advisory Board (CAB) recommended scan & referred programs • Selection committee - Emory CPCRN - CAB - Southern GA Evaluation Association • Identified 8 potential programs - 4 invited to apply - 2 selected for next stage Evaluability Assessment (EA) • Pre-evaluation activity • Determine whether or not program is ready for outcome evaluation • Emory CPCRN EA Objectives - Describe and assess the program model - Determine the program’s capacity to produce needed evaluation data - Assess stakeholder interest in evaluation & intended use - Determine feasibility of outcome evaluation EA Components Site Visit Document Review Expert Review & Recommendations Literature Review EA Data Collection & Analysis • Site visits – Work with program staff to identify appropriate participants & format – Langdale: 8 participants from 4 organizations – Cancer Coalition: 18 participants from 7 organizations • Discussions/interviews recorded and transcribed verbatim • Used matrix organized by EA questions to abstract and summarize relevant information Evaluability Assessment Findings Descriptive Case Study The Langdale Company and TLC Benefits The Langdale Company • Started 1894 as a family owned timber company • Headquarters: Valdosta, GA (subsidiaries in rural areas) • Diversified enterprise, subsidiaries in forest products, automotive, banking, hospitality, land development, etc. • About 800 employees What is Unique about Langdale? Health Plan/Health Delivery Approach: • Not only self-insured, but self-administered • Necessitates preventive approach to care • Partner with organizations to provide: - Comprehensive Medical Management - Case Management - Disease Management - Compliance/Health Advocacy support Case Study Question & Methods How does an employer-owned and operated health benefits plan utilize the Chronic Care Model1 (CCM) to deliver quality chronic disease care to employees and their dependents? • Qualitative Data collection – Individual interviews (n=6) – Group discussions (n=2) – Semi-structured interview guide with questions modified from Assessment of Chronic Illness Care (ACIC) • Qualitative analysis to identify themes/concepts related to each CCM element 1 Wagner, E. H., Austin, B. T., & Von Korff, M. (1996). Organizing care for patients with chronic illness. [Review]. Milbank Q, 74(4), 511-544. Preliminary Findings CCM Element Degree of Fit with Langdale Approach Clinical Information Systems Fully developed clinical information system for chronic illness care and care coordination The Community Fully developed system of linkages between Langdale’s employee benefits program and the community Self-Management Support Reasonably good to full support for self-management care within Langdale’s benefit programs The Health System Reasonably good to full support for chronic illness care throughout the organization Delivery System Design In applicable areas, reasonably good to fully developed delivery system design around chronic illness care Decision Support Less applicable to the employee benefits setting Implications for Practice • The CCM (and particular constructs) may help provide a framework for a worksite or employee benefits program to organize the delivery of quality chronic disease care. • Future research should assess more broadly how worksites and employee benefits can be integrated into the CCM. Community Cancer Screening Program • Community Cancer Screening Program™ (CCSP) Goal: To reduce and ultimately eliminate cancer screening disparities among low-income, uninsured and under-insured patients of local community health centers and other primary care practices Promoting Colorectal Cancer Screening • Goal: increase appropriate use of colonoscopy – 304 colonoscopies in 2010 • Patient Navigation Model – Establish and maintain clinical systems to identify and enroll patients into CCSP – One-on-one education to encourage adherence to referrals for screening – Address health care system and patient barriers to screening CCSP: Evidence-based strategies The Community Guide to Preventive Services: Intervention Categories Recommended for Colorectal Cancer Screening2 Intervention Category Evidence for Method Provider Assessment & Feedback Sufficient: FOBT Provider Reminder & Recall Systems Strong: FOBT Sufficient: Sigmoidoscopy Client Reminders Strong: FOBT Small Media Strong: FOBT One-on-One Education Sufficient: FOBT Reducing Structural Barriers Strong: FOBT 2 www.thecommunityguide.org/cancer/ Used by CCSP Outcome Evaluation of the CCSP Evaluation Goals • To explore differences in CRC screening rates at 4 intervention clinics as compared to 9 comparison clinics • To explore the degree of patient navigator effectiveness towards improving colonoscopy screening rates Research Design • Quasi-experimental design – 2 conditions – No randomization to condition • 18-month study period Eligible Population Non-Randomized Assignment – Nov. 1, 2009-Apr. 30, 2011 CCSP 4 clinics Colonoscopy No Colonoscopy No CCSP 9 clinics Colonoscopy No Colonoscopy Eligibility Criteria • Seen by a clinic primary health care provider at least once during the 18-month study period • Age 50-64 • Sliding fee scale eligible Setting: All 13 FQHCs in region Blue stars = CCSP Intervention Clinics Green stars = Comparison Clinics FQHC: Federally Qualified Health Center Southwest GA Federally Qualified Health Centers # Patients seen (2010) # Clinics < 1,000 1 1,001 – 2,000 5 • % Uninsured Patients 2,001 – 3,000 2 – 0-25%: 6 clinics – 25-50%: 7 clinics 3,001 – 4,000 4 > 4,000 1 • Four FQHC Systems – 13 total clinics • Provider/Patient Ratio – Mean: 0.0039 – Range: 0.0015-0.0087 Sample Size • Intervention – – – – – • Comparison 4 clinics Serve 3,009 patients 1,267 eligible patients Take 25% 350 charts to review – – – – – 975 patients 9 clinics Serve 11,001 patients 2,506 eligible patients Take 25% 625 charts to review Data Collection Methods • Data source: Patient medical charts (EMR & paper) • Time period: Aug. 2011 – Mar. 2012 • Randomly select charts from list of eligible patients provided by clinic • Trained abstractors collect data in clinics • Rigorous quality control methods – 10% of records double-abstracted – 100% double-abstraction for primary outcomes (Colonoscopy referral & exam) Data Abstraction Form • Used to abstract data from patient charts • Provides a standard way to collect data • Captures – Demographics – CRC history – CRC screening • Colonoscopy • Sigmoidoscopy • Blood Stool Test Data Analysis • Analysis to date – Descriptive statistics – Identify potential confounders – Preliminary assessment of differences between intervention and comparison clinics • Pending analysis – Controlling for clustering within clinics – Screening rates by clinic – Relationship between navigator contacts and colonoscopy Preliminary Findings Evaluation Goal: To determine whether the colorectal cancer screening component of the Coalition’s CCSP is associated with increased rates of colonoscopy screening. • n=809 patients at normal colorectal cancer (CRC) risk • Patient Demographics – 66% female – Mean age 56 years (range 50-64) – 61% Black; 36% White Preliminary Findings Are rates of colonoscopy screening among uninsured/ underinsured patients age 50-64 at the four intervention clinics significantly higher than at the nine comparison clinics? Outcome Had colonoscopy referral during study Had a colonoscopy exam during study Compliant on any test No Yes No Yes Total ˄ No Yes Total† Intervention 108 (42.0%) 149 (58.0%) 167 (65.0%) 90 (35.0%) 257 166 (57.4%) 123 (42.6%) 289 Note. Statistics not yet adjusted for clustering within clinics ˄ Among patients due for colonoscopy during study † Among all eligible patient * p < .0001 Comparison 388 (76.1%) 122 (23.9%) 477 (93.5%) 33 (6.5%) 510 464 (89.2%) 56 (10.8%) 520 Total 496 (64.7%) 271 (35.3%) 644 (84.0%) 123 (16.0%) 767 630 (77.9%) 179 (22.1%) 809 c2 86.738* 103.439* 108.962* Preliminary Findings What is the degree of CCSP effectiveness towards improving colonoscopy screening behavior? Outcome Wald (c2) Had colonoscopy referral during study (among due) Had colonoscopy exam during study (among due) Compliant on any test 75.447* 79.669* 89.448* Note. Statistics not yet adjusted for clustering within clinics ˄ Controlling for Race (Black) † Controlling for Race (Black) and Age (50-59 and 60-64) * p < .0001 Odds Ratio 4.260˄ 7.708† 6.013† Limitations • Non-random assignment – Possibility program implemented in higher capacity clinics • Variable quality of chart data – Intervention clinics: all EMRs – Comparison clinics: mix of paper and EMR – CCSP designed to improve quality of medical info in charts • Contamination – Patients from comparison clinics referred to intervention clinics for colonoscopy Acknowledgments Emory University CPCRN Kimberly Jacob Arriola, PhD, MPH*† Lucja Bundy, MEd, MA Michelle Carvalho, MPH Cam Escoffery, PhD, MPH† April Hermstad, MPH Sally Honeycutt, MPH Michelle Kegler, DrPH, MPH† Joseph Lipscomb, PhD† Natasha Ludwig-Barron, MPH Gillian Schauer, MPH* Iris Smith, PhD, MPH† Deanne Swan, PhD*† Amanda Wyatt, MPH Vera (Jingqi) Yang, MPH * Evaluability assessment or evaluation project leader † Expert Review Committee Member CCSP Staff, Stakeholders & Local Team Cancer Coalition of South GA: Denise Ballard, MEd Diane Fletcher, RN Rhonda Green James Hotz, MD Medical College of GA: Alex Brueder Shavonda Thomas Jennifer Yam SW GA Family Medical Residency: Teri Stapleton, MD TSTC Health IT program: Aisha Viquez Langdale Staff, Stakeholders & Local Team The Langdale Company: Barbara Barrett Mark Wilson Lowndes County Partnership for Health: Alan Powell John Sparks TLC Benefits Solutions, Inc.: Kate Waagner Doctor’s Direct Health Care: Tina Wise, RN Questions? • Ask now, or… • Look for our posters at the National Cancer Conference in August!