Disparities in the Financial Burden of Cancer Care and Access to - PowerPoint PPT Presentation

Disparities in the Financial Burden of Cancer Care and Access to Cancer Treatment during the Early Affordable Care Act Implementation Period Jennifer Tsui, PhD, MPH AcademyHealth Annual Research Meeting June 25, 2017

Costs of Cancer Care • $173 billion in 2020 • 18.1 million cancer survivors by 2020, 30% more than 2010 • Increasing: • out-of-pocket costs • oral/prescription therapies • long-term management Mariotto AB, Yabroff KR, Shao Y, Feuer EJ, Brown ML. Projections of the Cost of Cancer Care in the U.S.: 2010-2020. J Natl Cancer Inst . 2011

Financial Hardship and Cancer Care • Patients facing high financial hardship (FH) for cancer care are more likely to delay/forgo treatment & report lower QoL. 1,2 • FH more likely among younger (<65 yrs), minority (Hispanic/African American), lower-income cancer patients. 3,4 • Unclear whether ACA provisions are ameliorating disparities in FH related to cancer care. 5 • Few studies have examined financial hardship and cancer care among newly diagnosed cancer patients during the initial ACA implementation period. 1 Kent et al. Cancer 2013 2 Weaver et a;. Cancer 2010 3 Jagsi et al. J Clin Onc. 2014 4 Yabroff et al. J Clin Onc. 2016 5 McNulty et al. Curr Hem Malig Rep. 2015

Study Overview • Population-based study examining cancer care access, treatment, and outcomes in New Jersey - 6 th highest state in overall cancer incidence • Investigate disparities in cancer care by insurance type, co-morbidities, demographic factors during early ACA implementation period (2012-2015) - Medicaid expansion January 1, 2014 Overall Cancer Incidence 2010-2014 https://statecancerprofiles.cancer.gov/index.html

Study Population • The New Jersey State Cancer Registry (NJSCR) - one of 20 Surveillance, Epidemiology, and End Result (SEER) Program regions of the NCI • Inclusion criteria: First primary breast or colorectal cancer Stages I, II, III Diagnosed in Jan 2012 – Dec 2014 21-79 yrs Known insurance status at diagnosis NJ resident at time of diagnosis Alive at time of contact Not currently enrolled in another NJSCR study

Sampling and Recruitment Sampling: • Systematic random sampling by cancer site, diagnosis year and gender (for colorectal cancer only) • Stratified by age (21-64 years vs. > 65 years) • Oversampled for Medicaid-insured and uninsured Recruitment: • Mailed self-administered survey (~75 items) • September 2015 - August 2016 • $15 incentive • Response rate of 24%

Study Aims 1. Examine reports of financial hardship (FH) from cancer care in a population-based diverse sample of breast and colorectal cancer patients in New Jersey • Diagnosed and initiated treatment during the early ACA implementation period: 2012-2014 2. Assess the impact of financial hardship on access to cancer care, treatment decisions, and quality of life 3. Explore the extent of racial/ethnic and insurance-based disparities in FH

Demographic and Clinical Characteristics Breast Colorectal Breast Colorectal n=310 n=200 n=310 n=200 Age at DX AJCC Stage <50 years 32.6 14.5 52.9 Stage 0/I 24.0 50-64 years 43.9 53.5 Stage II 33.2 37.5 23.5 65+ years 32.0 13.9 Stage III 38.5 Race/Ethnicity Treatment(s) received* Hispanic 17.4 10.5 Surgery 93.9 94.5 6.1 NH-Black 9.5 Chemotherapy 55.2 57.5 NH-API/Other 11.9 13.0 Radiation therapy 71.0 23.0 64.5 NH-White 67.0 Co-morbidities Insurance at dx none 47.1 36.5 Uninsured 17.8 19.4 1 20.0 20.0 28.4 Private 32.4 19.4 Medicaid 16.7 2-3 21.9 28.0 11.0 15.5 Medicare 22.9 22.3 4+ 10.6 Other Public 10.2 * Not mutually exclusive

Financial Hardship by Race/Ethnicity 100 p<0.001 p=0.003 80 74.6 73.7 63.2 61.3 60 53.7 p=0.51 48.5 45.1 44.7 p=0.21 42.1 40.7 40 34.2 33.3 32.0 28.8 26.9 26.3 26.3 23.7 22.3 20.7 20 0 Borrowed money or go Made financial sacrifices Worried about paying Any financial hardship into debt large medical bills Total Hispanic NH White NH Black NH Asian/PI

Financial Hardship by Insurance at Diagnosis 100 p<0.001 80 P=0.001 68.1 62.6 60 p<0.001 55.3 53.7 53.3 46.1 p=0.001 45.1 42.9 40.0 39.3 40 33.3 32.9 29.7 28.8 28.2 23.0 22.4 20.0 20 11.1 10.3 0 Borrowed money or go Made financial sacrifices Worried about paying Any financial hardship into debt large medical bills Total No insurance Private insurance Medicaid Medicare

Financial Hardship by Change in Insurance Between Diagnosis and Treatment 100 P<0.001 82.9 P<0.001 78.7 80 p=0.00 p=0.001 60 56.7 53.7 52.3 50.0 49.3 48.9 45.5 45.1 44.7 39.4 40 36.4 32.4 28.8 25.7 24.3 22.4 20.5 18.8 20 0 Borrowed money or go Made financial sacrifices Worried about paying Any financial hardship into debt large medical bills Total No insurance change Change in insurance type Gained insurance between dx to trt No insurance at dx and trt

Financial Hardship by Diagnosis Year 100 2012-2013 2014 85 85 83 82 80 69 68 64 60 59 59 60 55 54 54 53 51 51 50 50 49 48 42 41 38 39 40 20 0 Insurance at Dx Insurance at Trt Change in Insurance

Adjusted Models for Cancer Care Access Did not receive all Treatment Decisions Treatment delay needed cancer care OR 95% CI OR 95% CI OR 95% CI Any Financial 3.55 ** 2.29 ** 1.47 + (0.95-2.25) (1.53-8.23) (1.35-3.89) Hardship Insurance at Treatment Private 1.0 1.0 1.0 Uninsured 1.16 (0.35-3.89) 1.07 (0.49-2.34) 2.00 + (0.98-4.06) Medicaid 0.94 (0.33-2.70) 1.52 (0.78-2.94) 2.17 ** (1.23-3.85) Medicare 0.88 (0.30-2.61) 0.48 + (0.22-1.07) 1.10 (0.61-2.00) Year of Dx 2012-2013 1.0 1.0 1.0 2014 1.07 (0.49-2.33) 1.30 (0.75-2.27) 1.57 + (0.99-2.50) + p < 0.10, * p < 0.05, ** p < 0.01, *** p < 0.001 Models adjusted for age at dx, marital status, \stage, household income, race/ethnicity, # of co-morbidities

Financial Hardship and Quality of Life

Summary & Implications • High rates of FH among cancer cases in early ACA period. o Borrow money/go into debt (22%) vs MEPS Cancer Supp 2011(7.1%) 1 o Made financial sacrifices (28%) vs. MEPS (9.4%) o Worrying about medical bills (45%) vs MEPS (23%) • FH highest among Hispanics & APIs (75%) and uninsured (68%). • FH continues to impact access to care, care decisions, and QoL. • Similar rates of treatment delays between Medicaid and uninsured groups and increasing rates of treatment delays in 2014: o Transitional expansion year o Changing composition of Medicaid patients o System-level barriers to timely care 1 Yabroff et al. J Clin Oncol. 2016

Next Steps • Further exploration of financial hardship specific to 1.) cancer patients gaining coverage and 2.) stages of the care continuum • Concurrent study of a Medicaid-cancer registry linkage to examine patterns of care and reasons for treatment delays • Mixed methods study of Medicaid providers to identify care delivery processes that impact access/barriers, timeliness and management of cancer care (ACS MRSG 2017-2022) • Inform and develop strategies to address cancer care-related financial hardship, particularly for vulnerable groups.

Acknowledgements Rutgers Cancer Institute of New Jersey New Jersey State Cancer Registry • • David Rotter Antoinette Stroup • • Carolina Lozada Natalia Herman • Aishwarya Kulkarni Rutgers School of Public Health • Jie Li • Dirk Moore • Kitaw Demissie Sidney Kimmel Cancer Center at Jefferson • Grace Lu-Yao Rutgers Center for State Health Policy • Joel C. Cantor This project was supported by the Rutgers Biomedical Health Sciences Team Science Initiative. The New Jersey State Cancer Registry is funded by NCI SEER Program contract #HHSN261201300021I, the CDC NPCR #5U58DP003931-02, the State of New Jersey, and the Rutgers Cancer Institute of New Jersey.

Questions? jennifer.tsui@rutgers.edu

Total breast and colorectal Records excluded (n=21,446) cases • Stage IV diagnosed in 2012-2014 in • Diagnosed outside of 2012- NJSCR (n=35,107) 2014 period • Diagnosed outside the state of NJ • Age<20 years or age>79 years • Absence of invasive cancer Eligible cases (n=13,661) • Actively enrolled in other studies Records excluded (n=11,253) based on sampling frame Target sample based on sampling frame (n=2408) Records excluded (n=42) • Deceased • MD refused Eligible cases contacted by mail (n=2366) Records excluded (n=1832) • Ineligible: based on survey screener, incomplete survey, incorrect address Total breast and colorectal • Active refusal cancer participants in • Passive refusal IMPACT study • Deceased post contact (n=534)