Health care access and utilization among adult cancer survivors: Results from the National Institutes of Health “All of Us” Research Program

Abstract Background Many cancer survivors face financial difficulties that prevent them from receiving appropriate health care. Racial/ethnic disparities in receipt of health care have been reported among cancer survivors, but recent data for important racial/ethnic subgroups of the US population are lacking. Methods To learn more about barriers to healthcare access faced by cancer survivors, we analyzed data from the NIH “All of Us” Research Program. Data were analyzed about demographic factors and other personal characteristics, personal medical history of cancer, healthcare utilization, and access to care. Results As of November 2020, a total of 5426 participants had a history of cancer (excluding skin cancer). About 88.2% were non‐Hispanic White; 3.9% were Black, African American, or African; 1.3% were Asian; 4.1% were Hispanic, Latino, or Spanish; and 1.2% reported more than one race. Just over one‐half had an annual income of $75,000 or greater. The majority of the participants (71.7%) were college graduates or had an advanced degree. About 47.0%% had private health insurance, 41.0% had Medicare, 6.0% had Medicaid, and the remainder had military, Veterans Affairs, other insurance, or no health insurance. Frequently cited reasons for delayed care in the past 12 months were “had to pay out of pocket for some or all of the procedures,” “deductible was too high/or could not afford the deductible,” “couldn't afford the copay,” “couldn't get time off work,” and “were nervous about seeing a health care provider.” Discussion A minority of cancer survivors who participated in the NIH “All of Us” Program had difficulty paying for health care in the past 12 months. Of particular concern are minorities such as African American and Hispanic cancer survivors along with those who are low income.


| INTRODUCTION
Cancer survival is improving over time as a result of screening and improved treatment. 1 An estimated 16.9 million cancer survivors were alive on January 1, 2019, in the US Cancer survivors often have life-long health care needs that include surveillance for recurrent or secondary primary malignancies, preventive care, and management of chronic conditions and cancer treatment-related morbidity. 2,3 Healthcare utilization among cancer survivors varies by age, race, income, insurance status, tumor type, and stage. 4 Many cancer survivors face financial difficulties that prevent them from receiving appropriate health care. 5,6 The cost of cancer care is an important barrier to health care access, particularly among those who are racial minorities, have lower income, or are uninsured. 5,7 Studies have shown that uninsured cancer survivors are more likely to avoid care and prescription medications due to cost. [8][9][10] Lack of insurance in cancer survivors has been associated with forgoing care and poorer clinical outcomes. 11 Racial/ethnic disparities in receipt of health care have been documented among cancer survivors. 3,9,10,12,13 In an analysis of data from the 2006-2010 National Health Interview Survey, Palmer et al. 3 found that non-Hispanic Whites reported receiving more health care than did Hispanics and African Americans. In a study of adolescent and young adult cancer survivors ages 15-39 years, Keegan et al. 9 found that among those who had no cancer-related medical visits in the previous year, a greater frequency were of non-Hispanic Black or American Indian/Alaska Native and Hispanic race/ ethnicity compared to those who received care. In a similar study by Kirchoff et al. 12 Hispanic cancer survivors had the lowest frequency of insurance among the survivors, and two thirds of this racial/ethnic group had no routine medical care in the previous year. American Indian/Alaska Native cancer survivors also experience challenges in accessing health care. 14,15 There is also a correlation between age and receiving health care among cancer survivors. 12,16 Health care utilization among older cancer survivors is influenced by the high frequency of comorbid diseases and treatment complications. 16 On the other hand, younger cancer survivors may be especially likely to forgo health care because of cost barriers. Kirchhoff et al. 12 found that more than one-fifth of adolescent and young adult survivors did not have a healthcare provider. More than two-thirds of uninsured survivors received no routine health care. 12 Compared to older cancer survivors, younger patients who are in their 20s are often less affluent and more likely to be without health insurance. Younger cancer patients, particularly those who are ethnic minority and low income, often have challenges such as delays in care. 17,18 To learn more about barriers to healthcare access among cancer survivors, we analyzed data from the National Institutes of Health (NIH) "All of Us" Research Program. Survey data were analyzed about demographic factors and other personal characteristics, personal medical history of cancer, healthcare utilization, and access to care. Of particular interest were racial and ethnic disparities in health care. We hypothesized that individuals who are Hispanic or African American would be less likely to have received health care and experience greater barriers to care compared to non-Hispanic Whites. The current study adds to the published literature by providing greater information about reasons for delayed care among cancer survivors. The study also adds additional information about racial disparities in access to care among cancer survivors.

| METHODS
The NIH "All of Us" Research Program is open to all eligible adults who live in the United States. Participants can sign up directly through the NIH "All of Us" Research Program website or through participating health care provider organizations. Although the data are not population-based, they reflect the diversity of cancer survivors in the US population in terms of race, ethnicity, age, gender, health status, and other factors. In the current study, cancer survivors were defined as research participants who had been diagnosed with cancer; the sample included both individuals who were undergoing cancer treatment and those who had already completed primary therapy for the disease. Individuals with skin cancer were excluded from the analysis because their cancer survivorship care needs are often relatively minor. The "All of Us" Research Program is approved by the NIH Institutional Review Board (IRB). Participants sign an informed consent document according to the Declaration of Helsinki authorizing the collection of their data. De-identified data from individual participants are available to approved researchers.
Survey data were analyzed about demographic factors and other personal characteristics (age, sex, race, Hispanic ethnicity, educational attainment, marital status, health care insurance, employment status, annual household income, and number of people in household); personal medical history of cancer; general health; and health care utilization. Under personal medical history, the participants were asked, "Has a doctor or health care provider ever told you that you have or had any of the following cancers?" and a list of 21 cancer types followed including "Other cancer." Under health care utilization, the participants were asked a number of questions as summarized in Table 1.
In summarizing the results, any group with less than 20 participants was combined with a relevant group to satisfy the safe data sharing policy of "All of Us". After frequencies and cross-tabulations of the data were completed, multivariate logistic regression methods were used to examine self-reported delayed receipt of health care according to age, sex, race, Hispanic ethnicity, education, annual household income from all sources, number of people in household, marital status, and health insurance. Ninety-five percent confidence intervals were obtained for adjusted odds ratios. Levels of statistical significance were determined using Wald chi-square tests and log-likelihood ratio tests. The goodness-of-fit of each model was examined using the log-likelihood ratio test.

| RESULTS
As of November 2020, a total of 5426 participants in the NIH "All of Us" Research Program had a history of cancer, excluding skin cancer ( Table 2). Their mean age was 67.6 years (standard deviation = 12.0). Overall, 64.3% were women and 35.7% were men. About 88.2% were non-Hispanic White; 3.9% were Black, African American, or African; 1.3% were Asian; 4.1% were Hispanic, Latino, or Spanish; and 1.2% reported more than one race. There was wide variation in self-reported annual income. Just over one-half had an annual income of $75,000 or greater. Over two-thirds were married or living with a partner. The majority of the participants (71.7%) were college graduates or had an advanced degree. About 47.0% had private health insurance, 41.0% had Medicare, 6.0% had Medicaid, and the remainder had military, Veterans Affairs, other insurance, or no health insurance. Table 3 shows the type(s) of cancer reported by the participants. The most frequent cancer types included breast, prostate, and colorectal cancer, which represented 27.4%, 14.3%, and 5.2% of the cancers, respectively. Only 2.6% of the cancer survivors had lung cancer, which has a lower survival rate. Table 4 summarizes health care access and utilization among the cancer survivors. About 9.7% were told by a health care provider or doctor's office that they did not accept their health care coverage during the past 12 months. About 8.9% reported that their health insurance or health care coverage was worse than a year ago, and 8.0% reported that it was better. Only 2.2% reported no place that they usually go to when they are sick or need advice. The majority of the participants (93.5%) had seen or talked to a general doctor or a specialist (83.4%) in the past 12 months. The most common reasons for delayed care in the past 12 months were "had to pay out of pocket for some or all of the procedures," "deductible was too high/or could not afford the deductible," "couldn't afford the copay," "couldn't get time off work," and "were nervous about seeing a health care provider." For a minority of the participants (3.4% to 10.0%), there was a time during the past 12 months when they needed prescription medications, a regular doctor or general health provider, a specialist, or follow-up care, but did not get it because they could not afford it.
We then analyzed the factors that most influenced the delays in receiving medical care in the previous 12 months among | 3649 COUGHLIN et aL. these cancer survivors. Table 5 shows the results of a multivariate logistic regression for such predictors. After adjusting for other variables in the model, cancer survivors who were younger, women, or had lower income were more likely to report delays in receiving medical care in the previous 12 months, along with those who had only completed some college.

| DISCUSSION
The results of this study indicate that a majority of cancer survivors who are participants in the NIH "All of Us" Research Program have seen a doctor or a specialist in the past 12 months, but that a significant number of participants delayed care due to out-of-pocket health care expenses such as deductibles or copayments. Approximately 3.4% to 10.0% of the participants reported no health care or follow-up care because they could not afford it, even though the participants were more educated and had higher annual incomes, on average, than the US population as a whole. It is likely that these percentages are much higher among cancer survivors in the general population, and particularly among minorities and other populations suffering significant health disparities. Cancer survivors who have completed primary therapy for the disease often have complex health care needs that include surveillance for recurrent disease or second primary malignancies, and the management of cancer-related morbidities. 2 Appropriate care during the post-initial treatment survivorship phase includes preventive care visits and screening for cancer recurrence. 2 In the current study, a high percentage (93.5%) of cancer survivors had seen a general doctor in the past 12 months and 83.4% had seen a specialist such as an oncologist. Previous studies of cancer survivors have shown high (92% to 97%) rates of adherence to routine primary care visits in the previous year. 19,20 In a study of survivors of non-Hodgkin lymphoma, 82% of the patients reported visits to oncologists. 21 Racial/ethnic disparities in general health care utilization are frequent. 3 Minorities are less likely to use health care services than are non-Hispanic Whites. 3 Factors that contribute to these disparities include low income and lack of health insurance. [22][23][24] In the current study, African Americans and Hispanics were not significantly more likely to report delaying getting care in the past 12 months than non-Hispanic Whites. Only a small percentage of participants reported that they lacked health insurance. This is likely an underrepresentation of this problem in the general population as individuals included in this database are a self-selected population; less educated and lower income patients are underrepresented and more likely to suffer from the factors most limiting of access to health care in general. In an analysis of data from the National Health Interview survey, Palmer et al. 3 found that older African American and Hispanic cancer survivors were about twice as likely as non-Hispanic Whites to not see a specialist. Similar racial/ethnic differences in health care use have been reported in Surveillance Epidemiology and End Results-Medicare studies. 25,26 Our study has limitations that need to be considered when interpreting the results. The source of data is not populationbased and, as mentioned, the participants were relatively educated and most had health insurance. Therefore, the results may not be generalizable to all cancer survivors in the US population. We compared findings from the current study with results obtained by Zeng et al. 27 in their analysis of data on cancer survivors from the 2013-2017 National Health Interview Survey (NHIS). In contrast to the current study, the NHIS is representative of the US population. In the study by Zeng et al., 27 52.6%-65.4% of the sample of cancer survivors were women compared to 64.3% of our sample from the NIH "All of Us" Research Program. In the study by Zeng et al., 27 57.9%-60% of the cancer survivors were married compared to 62.5% of the current sample. In the study by Zeng et al., 27 the percentages of the cancer survivors who were White, Black/African American, or Hispanic were 77.9%-83.8%, 7.3%-8.0%, and 5.0%-9.2%, respectively, compared to 88.2%, 3.9%, and 4.1% of the current sample. Thus, in the current study, there was a higher percentage of non-Hispanic Whites, and lower percentages of Blacks and Hispanics. The current study also included more cancer survivors who were married and fewer survivors who were uninsured. A further issue is that the numbers of participants who are American Indian, Alaska Native, or Pacific Islander were too small for separate estimates. Still, the results highlight the issues that cancer survivors face when dealing with long-term care and trying to manage their long-term medical needs associated with cancer survivorship.
In summary, a minority of cancer survivors who participated in the NIH "All of Us" Research Program had difficulty paying for health care in the past 12 months. However, these issues are magnified among minorities such as African American and Hispanic and those with lower income.