Medicare beneficiaries who have cancer are a growing population with unique care needs. Population-based research examining relationships between cancer patient experiences, health care utilization, and subsequent patient health outcomes is lacking.
A recently updated data resource called SEER-CAHPS links cancer registry data with Medicare information and patient surveys. It provides a comprehensive, nationally representative source of information for addressing important questions about care experiences among Medicare beneficiaries with cancer. Understanding patterns of associations between different aspects of care experiences and respondent characteristics can inform quality improvement efforts and future intervention development.
The Impact of Cancer on the Medicare Population
Most individuals aged 65 and over in the US have Medicare coverage, and cancer is common, expensive, and a leading cause of morbidity and mortality in this population. In 2016, 233,703 individuals aged 65 or older (123,606 men and 110,097 women) in the US were diagnosed with cancer, making up more than half (54%) of all individuals diagnosed with cancer in the US. Cancer is the cause of more than one in five deaths among older adults.
Cancer-related expenses are a substantial burden to individuals and also the Medicare system. Medicare payments for cancer-related health care in 2014 were approximately $29 billion [pdf] – almost four times [pdf] the expenditures for those without cancer. More than one-quarter (28%) of Medicare beneficiaries with cancer have a high out-of-pocket burden (defined as out-of-pocket costs being 20% of income or more), compared with 16% of beneficiaries without cancer.
Cancer can cause long-term, chronic pain, and disability, and the processes of diagnosis and treatment can also negatively affect patients’ quality of life. Surgery, radiation, and chemotherapy – along with hormonal and other treatments – take their own toll on health and well-being. Caring for a person with cancer places additional burdens on friends and family.
About SEER-CAHPS
SEER-CAHPS represents a collaborative effort between the National Cancer Institute’s Surveillance, Epidemiology and End Results (SEER) cancer registry and the Centers for Medicare & Medicaid Services’ (CMS) Medicare Consumer Assessment of Healthcare Providers and Systems (CAHPS®) patient experience surveys and Medicare claims and enrollment data.
SEER-CAHPS is a unique data resource that includes three different types of data:
- SEER data include cancer site, stage, initial treatment, vital status, and demographics.
- Medicare CAHPS survey data provide information on multiple aspects of care experiences, including 5 global measures (Overall Care, Personal Doctor, Specialist Doctor, Health Plan, Prescription Drug Plan) and 7 composite measures (Doctor Communication, Getting Needed Care, Getting Care Quickly, Care Coordination, Health Plan Customer Service, Prescription Drug Plan Customer Service, Getting Needed Prescriptions). The Medicare CAHPS survey also includes self-rated health status items and other indicators of illness.
- Longitudinal Medicare health care claims include data on health care utilization, such as services received, medical visits, and pharmacy claims, as well as the cost of care for FFS beneficiaries. Recently, SEER-CAHPS added Medicare Part D data for cancer cases.
The linkage includes over 250,000 people surveyed before or after their cancer diagnosis and over 800,000 Medicare beneficiaries without cancer in SEER areas. Thus, the SEER-CAHPS data resource offers a rich opportunity for analyses of Medicare beneficiaries’ experiences with their care across the cancer care continuum and in comparison to those without cancer.
SEER-CAHPS Research Published to Date
Research published to date includes studies of Medicare beneficiaries in the last year of life, those with colorectal cancer (CRC), those who survived at least a year after a cancer diagnosis, rural vs. urban survivors, and those dually enrolled in Medicare and Medicaid. Below we summarize a few interesting findings from each study.
- Individuals in the last year of life (n=5,102) who rated their general or mental health status as excellent or very good were significantly more likely to indicate excellent experience with 8 out of 9 measures examined (overall care, personal doctor, specialist, health plan, doctor communication, getting care quickly, getting needed care, and getting needed drugs), compared to those reporting poor self-rated health. This finding suggests that if individuals receive comprehensive medical and psychiatric care, even cancer patients at the end of life may rate their care experiences as excellent.
- More than half (55%) of people with CRC (n=314) rated their overall care highly (9 or 10 on a scale from 0-10). Similarly, many/most patients gave excellent ratings of their personal doctor (59%), specialist (47%), and health plan (60%) – in this case, traditional fee-for-service Medicare. Although only a small group was studied, higher ratings for personal doctors and specialist doctors were positively associated with adherence to office visits, and ratings of personal doctors were associated with adherence overall. In other words, people who perceive their physicians as high-quality providers are more likely to adhere to their treatment plans.
- Another study, using a large sample (n=19,455) of seniors with breast, colorectal, lung, or prostate cancer who survived at least a year, found that more than half provided high scores on each CAHPS measure. Again, excellent or very good general health was strongly associated with higher ratings. Among lung cancer survivors, mental health was strongly associated with better care experiences. Surprisingly, college-educated breast cancer survivors were less likely than those with less education to provide high ratings on doctor communication – a finding that deserves further exploration.
- People living in urban areas at the time of their cancer diagnosis rated Getting Care Quickly significantly lower than rural residents. This is a counter-intuitive finding, as rural residents have historically had challenges with access to care. Prior studies have also shown that rural residents experience more delayed cancer diagnoses and report more unmet needs. It may be that rural residents felt they were better able to access timely care because there was less competition from other patients. Observed differences could also be due to differences in expectations between rural and urban residents based on past experiences of their ability to get care quickly.
- The most recent SEER-CAHPS publication is a study that looks at the effects of dual (Medicare and Medicaid) enrollment on care experiences. Results showed that dual enrollees were more likely than Medicare-only enrollees to report better experiences with their health plan and prescription drug plan. On other measures, they were no more or less likely to report worse care experiences.
Future Directions
SEER-CAHPS offers a unique opportunity to focus on research across the cancer care continuum, from diagnosis and treatment through survivorship and end of life. Using this unique resource, investigators can answer important questions about how patients are affected by cancer treatments and patterns of care. Researchers interested in obtaining the data should refer to the SEER-CAHPS website. Valuable tips on working with the data were provided in the 2017 and 2019 webinars.
Interest in the SEER-CAHPS data resource is growing, and more research is in the pipeline. Studies currently in process include a study using machine-learning to estimate illness burden that incorporates self-report measures and studies looking at patterns of care and care experiences among older cancer survivors. Summaries of selected studies in the works can be found at the SEER-CAHPS website.
Conclusions
The increasing complexity of cancer care in the US makes it necessary to provide care that is respectful and responsive to cancer patients’ needs. A first step to delivering patient-centered care is to understand patterns of associations among different aspects of care experiences, respondent characteristics, and health care utilization. The SEER-CAHPS resource has already given us much food for thought. What other unexpected findings await researchers in this rich resource? Stay tuned!
Lisa M. Lines
Latest posts by Lisa M. Lines (see all)
Justin Kirschner
Latest posts by Justin Kirschner (see all)
Michael Halpern