Statistical analyses
To describe the data collected, mean values (for age) and median values for the first (Q1) and third (Q3) quartiles of the duration of disease were used. The cancer survivor group was compared with the group of respondents with no self-reported cancer history (non-cancer population) using the chi-squared test of independence, and the frequency of given conditions was presented. The relationship between self-reported cancer status and the presence of some limitations in functional activity, depression, dementia, need for support in everyday life, and falls, as well as poor self-rated health, was analyzed using multivariate logistic regression models. The set of possible covariates, including age (5-year groups, reference age 65–69), sex (ref. male), marital status (ref. not married), university education (ref. elementary education), and number of chronic diseases reported (none, 1, 2, or 3+), was used in all models. The results were presented as odds ratios (ORs) and 95% confidence intervals (95% CI). The next step of the analysis was to assess if there was any relation between the time from cancer diagnosis and the studied outcomes among respondents with a self-reported cancer history using multivariate logistic regression with the following covariates: age, sex, education, marital status, and number of comorbidities. For the purpose of this analysis, as a consequence of a number of missing data concerning the time of diagnosis and present status of the disease, analyses were performed with a smaller sample size. All analyses were performed in STATA 13 (StataCorp LLC, Texas, USA), and the statistical significance for all tests was 0.05.
Discussion
We found that cancer survivors over 65 years of age more frequently experienced falls and presented poor health status in comparison with those without cancer history. Moreover, the octogenarians who survived cancer presented much higher need for assistance due to limitations in functional status in comparison to with younger subjects. Elderly cancer survivors were also more likely to present multi-morbidity than patients without a history of cancer.
Around one third of cancer survivors and less than a quarter of respondents without a cancer history in our study reported having experienced a fall within the previous year. The rate of falls in community-dwelling older adults with cancer varies across studies and is estimated to be 20 to 33% [
6]. After controlling for age, sex, education, and comorbidities, a history of cancer was associated with the occurrence of falls in our study. Falls were twice as frequent in elderly men with a history of prostate cancer compared to men without a history of cancer. Exposure to androgen deprivation therapy (ADT) in treating prostate cancer is associated with a higher risk of falls than no such treatment [
7]. A study conducted by Mohile et al. also showed that elderly patients with cancer experience a higher prevalence of falls than those without a history of cancer (26.4 vs. 21.9%) [
8]. The side effects of cancer therapy, e.g., peripheral neuropathy, fatigue, and pain, are among the factors that could lead to functional impairments and increase the incidence of falls in cancer survivors [
9]. Apart from cancer and the long-term effects of treatment, the higher prevalence of falls may also be a consequence of concomitant diseases, other medications, and aging.
Although elderly patients who fall frequently tend to have deficits in functional ability, we did not observe an increased incidence of deficits in performing the IADL tasks in the group of cancer survivors studied. Other studies have also reported maintained independence in cancer survivors. Seventy-one percent of older patients with breast cancer with a median follow-up of 5 years were found to be fully independent in IADL tasks [
10]. In contrast, Mohile et al. showed that patients with a history of cancer had a significantly higher prevalence of limitations in IADL tasks than the non-cancer population (49.5 vs. 42.3%) [
11]. Furthermore, individuals with cancer reported more limitations in IADLs, with the most common limitations being heavy housework (34%) and shopping (17%) [
12]. It is of note that around half of the population in the PolSenior study had some impairment in performing IADL tasks [
13]. What is also noteworthy is that in elderly cancer survivors, functional status was found to be associated with the presence of comorbidities and level of education rather than with the presence of cancer or time from diagnosis [
14]. In our study, functional decline in cancer survivors was associated with age (5.6-fold higher odds of deficits in performing IADL tasks among patients 80 years and older compared with patients aged 65 to 79 years), but not with the presence of comorbidities or time from diagnosis. Cancer survivors who had university education were less likely to present impairments in IADLs and cognitive dysfunction, to report need for assistance or poor health status than the individuals with elementary education that might suggest that higher education could be protective against detrimental influences of cancer and its treatment or it might contribute to the better capacity to adapt to decreasing functional ability [
15]. There were no significant differences in the prevalence of impairments according to the status of cancer treatment. Previous studies have shown an association of cancer history with frailty, and falls are among the common features of frailty [
11]. The higher occurrence of falls in our population might be a clinical manifestation of frailty in elderly cancer survivors.
Numerous studies have shown that cancer increases the risk of developing depression in the elderly [
16]. Data from the Medicare Current Beneficiary Survey suggested that cancer patients were more likely to experience depression compared with non-cancer controls (OR = 1.15; 95% CI: 1.02–1.30) [
8]. In contrast, the occurrence of depression in the PolSenior population of cancer survivors was associated with female sex, age, and number of comorbidities, but not with the cancer diagnosis.
It has been suggested that elderly patients with cancer may have a reduced risk of dementia [
17]. These data were supported in a 2015 meta-analysis of three studies analyzing the risk of Alzheimer’s disease in patients with cancer [
18]. Also, the results of the Framingham Heart Study of 1278 participants aged 65 years or older without dementia at baseline who experienced cancer indicated that they had a lower risk of Alzheimer’s disease (HR = 0.67; 95% CI: 0.47–0.97) [
19]. Only a small number of studies have reported that cancer survivors may experience long-term cognitive deficits. In a study conducted by Heflin et al., 14.5% of cancer survivors had cognitive dysfunction compared with 8.7% of their cancer-free twins [
20]. In our study, cancer survivors presented lower prevalence of cognitive impairment in univariate analysis, but it was not confirmed in multivariate logistic regression analysis, as having cognitive impairment was associated with male sex, older age, elementary education, and not being married, but not with cancer survivorship.
The reported incidence of comorbid conditions in cancer survivors varies across studies. Results from a nationwide study among all people living in Denmark indicated a higher prevalence (40%) of a score ≥ 1 on the Charlson Comorbidity Index (CCI) in older cancer survivors than in the non-cancer population (16%). Cancer survivors had 59% higher odds of having a CCI score ≥ 1 (95% CI: 1.57–1.60) after adjustment for age and sex [
21]. In a study conducted by Holmes et al. [
4], more elderly survivors reported having two or more chronic conditions compared to controls (68 vs. 65%) and the age-adjusted prevalence of cardiovascular diseases (excluding hypertension) was higher in survivors (25 vs. 23%). Three out of four elderly cancer survivors in the PolSenior study reported having cardiovascular disorders. Cancer survivors treated with cardiotoxic therapies are at increased risk of atherosclerosis secondary to inflammation and endothelial dysfunction [
22]. We have previously observed endothelial activation in young adult survivors of childhood acute lymphoblastic leukemia early after completing treatment [
23]. On the other hand, other studies found a similar prevalence of chronic conditions in elderly cancer survivors and controls, except for a significantly increased frequency of coronary artery disease and emphysema in survivors [
24]. In this study, the average number of comorbid conditions in cancer survivors was 3, which was lower than in participants without a history of malignancy. In our study, we found an increased incidence of osteoporosis and endocrine disorders among prostate cancer survivors. The high prevalence (53%) of osteoporosis has been previously documented in men with prostate cancer on ADT [
25]. Furthermore, ADT increases the risk of diabetes in older men with prostate cancer, particularly when other comorbidities are present [
26]. In addition, ADT may be associated with an increased risk of dementia [
27]; nonetheless, we found no increased prevalence of cognitive impairment in men with a self-reported prostate cancer history and according to previously published data, the use of ADT in the studied population was low [
28]. Nevertheless, it should be emphasized that due to the cross-sectional nature of the data analyzed in the study, we cannot exclude the hypothesis that the higher number of comorbidities was related to superior health care and a more thorough diagnosis of concomitant diseases in cancer survivors.
Cancer survivorship among the older population has implications for their assessment of their general health status. Our findings indicate that elderly cancer survivors more often report poor health status than individuals without a history of cancer. As mentioned above, the occurrence of falls may contribute to a decline in perceived health status. In a cross-sectional study in older cancer survivors who had a history of falls in the previous year, falls were associated with lower scores for health-related quality of life (HRQOL) and with a prospective decline in HRQOL [
29]. In the US National Health Interview Survey in 2010, almost twice as many cancer survivors (47% aged 65 years or older) as adults without a cancer history reported poor mental health-related quality of life, and over a quarter of cancer survivors reported poor physical health-related quality of life. Having more than one comorbidity was found to be associated with poor physical health-related quality of life in both groups [
30]. In the PolSenior study, having three or more comorbidities was more strongly associated with poor self-reported health status than cancer survivorship itself.
This study has some limitations. Data were self-reported, which may have resulted in under- or over-reporting of information and patients currently under treatment for cancer might have been underrepresented in the study. Unfortunately, we did not have access to all records of the type and status of cancer treatment. The data on type of cancer, age at diagnosis, and status of cancer treatment could be verified with hospital discharge reports when provided by the patient. Furthermore, cognitive impairment was assessed using screening test (MMSE), which may be influenced by age, education, motor, and visual impairments.
In conclusion, cancer survivors over the age of 65 years experience a higher prevalence of falls, are more likely to report poor health status, and have a higher number of chronic conditions than the non-cancer elderly population, but they maintain independence in IADLs. Advanced age and lower education, but not time from cancer diagnosis, are associated with the occurrence of impairments in older cancer survivors.