Caregiving associated with selected cancer risk behaviors and screening utilization among women: cross-sectional results of the 2009 BRFSS

There is recent concern that caregivers are less likely than non-caregivers to practice health-promoting behaviors [1–3], which may include seeking cancer screening tests. Caregivers are generally defined as persons who provide unpaid assistance or supervision with personal or instrumental activities of daily living (i.e., bathing, eating, dressing, medication management, handling finances) to a relative or friend who cannot perform these tasks due to cognitive, physical, or psychological impairments [4]. The increasing number of older caregivers in the United States [5], and their chronically high levels of self-reported stress and potential for adverse health consequences [6], underscores the importance of health maintenance and disease prevention in this population. Receipt of appropriate cancer screening tests is an important component of health maintenance, yet little is known about the cancer screening practices of caregivers as compared to non-caregivers.

Of the estimated 43.5 million U.S. caregivers in 2009, 54% were aged 50 years or older, and 26% spent ≥20 hours/week performing caregiving activities [5]. As the majority (67%) of caregivers are female [5], the health effects of caregiving are an increasingly important women’s health issue. Caregivers consistently report more psychological distress, such as depressive symptoms and anxiety, as well as self-reported stress than non-caregivers [6]. According to the Caregiver Stress Process model, stress may result from characteristics of the caregiving situation, including the care recipient’s condition, level of care required, the necessity to balance caregiving responsibilities with other employment and family responsibilities, and the availability of social support and other factors that may reduce caregiving-related stress, leading to adverse effects on psychological or physical health [7]. As a result of this stress, or a way to cope with caregiving demands, caregivers may engage in risky health behaviors, such as cigarette smoking and alcohol consumption. Likewise, they may neglect health promotion activities, such as regular cancer screening, due to feeling stressed or the time constraints of caregiving [1, 2, 8]. However, there is conflicting evidence that caregivers have an increased risk of physical health decline or mortality [4, 9, 10]. In fact, older women caregivers had fewer medical conditions and functional limitations than their non-caregiver peers [4], suggesting that older caregivers may attempt to maintain good health in order to continue providing care to their care recipient.

Scant research has examined the association between caregiving and modifiable health behaviors related to cancer risk or utilization of cancer screening. The results of existing studies are inconsistent [2, 3, 8, 11], with caregivers having some better health habits [2, 8, 11] and pro-active cancer screening practices [1] as well as some poorer health habits [8]. Likewise, of studies comparing breast and cervical cancer screening behaviors, one reported increased utilization among caregivers [3], while the other reported no association [1]. Most studies to date have been limited by focusing on spousal caregivers [1, 3, 8], restriction to persons older than age 50 [1, 11] or 65 [2, 8], not stratifying by gender [1–3, 8, 11], and lack of adjustment for covariables [2]. Thus, the results of these studies may not relate to cancer-screening guidelines that are generally gender- and age-specific.

Effective screening methods are widely available for breast and cervical cancers, two of the most common cancers among middle-aged and older women. Among U.S. women, an estimated 209,060 incident cases of breast cancer and 12,200 incident cases of cervical cancer were expected in 2010 [12], with the highest incidence rates among women age ≥40 [7]. Though cervical cancer guidelines have recently been revised [13], the American Cancer Society (ACS) previously recommended Pap tests either yearly or every two to three years depending on personal history of Pap test results for women age ≥21 to screen for cervical cancer and annual clinical breast exams (CBE) and mammograms for women age ≥40 to screen for breast cancer [12].

Given the increases in the aging population and resultant need for informal caregiving, it is important to understand whether caregiving is associated with cancer risk and screening behaviors, especially among women. Therefore, we used data from female respondents to the 2009 Behavioral Risk Factor Surveillance System (BRFSS) to examine the association between caregiving and modifiable health behaviors known to affect overall cancer risk, such as smoking, alcohol use, obesity, physical activity, and fruit/vegetable consumption, and on breast and cervical cancer screening. We hypothesized that caregivers would engage in more negative health behaviors and be less likely to receive recommended cancer screenings as compared to non-caregivers. We additionally sought to examine if associations between caregiving and health behaviors and cancer screening differed by age or race. If caregiving is associated with poorer health behaviors or lack of screening utilization, this might identify an opportunity for intervention, as caregivers are likely to have frequent contacts with the health care system due to their caregiving responsibilities.

In this cross-sectional analysis of the 2009 BRFSS, we found that caregiving was associated with behaviors that increase cancer risk (i.e. obesity and smoking) as well as those that reduce cancer risk (i.e. physical activity). Further, caregiving was positively associated with ever having had a mammogram or CBE, but were generally similar in their usage of other cancer screening tests. The associations with obesity, smoking, and physical activity varied by age and race, although tests of interaction terms were not always statistically significant. The association between caregiving and ever having had a mammogram or CBE was observed among whites but not non-whites, though no statistically significant interaction was found.

Most prior studies have reported no differences in health behaviors such as smoking or alcohol use between caregivers and non-caregivers [3, 8, 11], though these studies generally included only spousal caregivers while our analysis was not restricted to spouse caregivers but did not specify the caregivers’ relationship to the care recipient. Caregiving was associated with increased physical activity in some [4], but not all previous [3, 8, 11] studies. Another study of BRFSS respondents [2] reported that caregivers were more likely to meet guidelines for weekly physical activity, but did not differ from non-caregivers on fruit and vegetable consumption, smoking and alcohol use, or BMI. This previous study only included individuals age ≥65 and did not examine men and women separately [2], though evidence suggests that caregiving may have different effects for men and women [5]. The suggestion of age and race differences we noted is novel and requires confirmation by additional studies.

Previous studies of caregiving and cancer screening tended to rely on convenience samples and lacked comparison groups of non-caregivers [15–17]. Of studies that compared caregivers to non-caregivers, one reported increased breast and cervical cancer screening in unadjusted analyses of spousal caregivers of cancer patients in South Korea [3], though these results may have reflected having a spouse with cancer more than the experience of caregiving. By contrast, the other study observed no association between caregiving and these screening tests in a nationally representative U.S. sample [1]. We were unable to examine the reasons for our finding that caregivers were more likely to have had mammography or CBE in their lifetime but did not differ on screening within ACS guidelines. One might hypothesize that caregivers were formerly more vigilant about screening exams, yet their caregiving responsibilities diminished their ability to be screened at the recommended frequency.

It is important to note that we relied on cancer screening guidelines published by the American Cancer Society in 2009 [18] to determine whether or not women had received appropriate cancer screening. Other organizations, such as the U.S. Preventive Services Task Force, also make screening recommendations to the public. We chose to use a single set of guidelines for simplicity and consistency, though screening recommendations may differ across organizations. Further, breast and cervical cancer screening guidelines have changed recently [13, 19], so our findings may not be generalizable to current cancer screening guidelines.

Our study had several limitations. First, the BRFSS included a single question on caregiving status, and lacked details on the caregiving situation and duration. Thus the caregivers in our study likely represent a wide range of caregiving intensity, reason for caregiving, and familial relationship to the care recipient. Cancer risk and screening behaviors of caregivers may vary by amount of strain associated with caregiving or by the health and the relationship between the care recipient and caregiver. In fact, one study found that caregivers’ health promotion and screening behaviors declined 2–5 years after a spouse’s diagnosis, and then improved after 5 years post-diagnosis [3]. Our results may have revealed significant differences between caregivers and non-caregivers if we were able to use a stricter definition of caregiving, such as requiring caregivers to be assisting the care recipient with one or more ADL/IADL tasks. Further, women who were caregivers in the past but not currently, and who may have changed their health and screening behaviors as a result of caregiving, were counted as non-caregivers in our analysis. This classification might minimize the differences between groups, thus attenuating true associations. Additionally, we were unable to exclude individuals who were care recipients, and these individuals would be classified as non-caregivers in our analysis. This misclassification would likely serve to attenuate any true differences between caregivers and non-caregivers. Further, other modifiable breast and/or cervical cancer risk factors, such as hormone therapy and condom use, were not assessed, nor was history of Pap test results, which is necessary for ACS guidelines. Our definition of meeting guidelines as having a Pap test within the previous three years is likely to be overly conservative as many women may have histories for which more frequent screening is recommended. Finally, the Women’s Health Module was administered in only four states, which may not be generalizable to the population of U.S. women.

However, our study is strengthened by the high quality of the BRFSS, its sampling procedures, and large racially/ethnically diverse sample. The few previous studies on the association between caregiving and cancer risk and screening behaviors have focused on older caregivers. Our age- and race-stratified results make an important contribution to this literature.

In summary, our results demonstrate that middle-aged and older women caregivers are more likely than non-caregivers to engage in health behaviors that increase, as well as some that reduce, cancer risk. Although receipt of screening tests for breast and cervical cancer was similar between caregivers and non-caregivers, approximately one-third of caregivers had not been screened for breast cancer following ACS guidelines. Thus interventions to reduce cancer risk and mortality among caregivers should focus on reducing obesity and smoking and increasing frequency of mammography and CBE, depending on both age and race. Given their frequent contacts with the health care system due to their roles as caregivers, this is a population subgroup that may be especially amenable to such interventions [6].