Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2021 Apr 1.
Published in final edited form as: Support Care Cancer. 2019 Jul 24;28(4):1839–1848. doi: 10.1007/s00520-019-05005-7

Co-morbidities, treatment-related consequences and health-related quality of life among rural cancer survivors

Shaila M Strayhorn 1, Leslie R Carnahan 2, Kristine Zimmermann 2, Theresa A Hastert 3, Karriem S Watson 4, Carol Estwing Ferrans 5, Yamilé Molina 2,3,6
PMCID: PMC6980904  NIHMSID: NIHMS1535643  PMID: 31342166

Abstract

Purpose

We explored how the number of lifetime comorbidities and treatment-related cancer symptoms were associated with quality of life (QOL) in rural cancer survivors.

Methods

Survivors (n = 125) who were rural Illinois residents aged 18+ years old were recruited from January 2017-September 2018. We conducted 4 multivariable regressions with QOL domains as outcomes (social well-being, functional well-being, mental health-MHQOL, physical health - PHQOL); the number of physical and psychological comorbidities (e.g. arthritis, high blood pressure, stroke) and treatment-related cancer symptoms (e.g. worrying, feeling sad, lack of appetite, lack of energy) as predictors; and, cancer-related and demographic factors related to these variables as covariates.

Results

The number of comorbidities and treatment-related symptoms were inversely associated with functional well-being (B=−0.36, p <0.0001 and −0.18, p =0.03), and MHQOL (B = −0.30, p = 0.001 and B = −0.25, p = 0.004). Comorbidities were associated inversely with social well-being (B=−0.27, p = .003). Neither comorbidities nor treatment-related symptoms were associated with PHQOL (p=0.20–0.24). Sensitivity analyses suggested that psychological comorbidities, treatment-related psychological symptoms, and physical comorbidities were associated with social well-being, functional well-being, and MHQOL.

Conclusions

Our study highlights the utility of risk-based survivorship care plans to address the negative, additive impact of comorbidities and the treatment-related symptoms to improve the health-related QOL among rural survivors. Future research should assess how contextual factors (e.g., geographic distance to oncologists and other providers) should be incorporated in survivorship care planning and implementation for rural survivors.

Keywords: rural, co-morbidities, treatment-related consequences

INTRODUCTION

As of 2016, approximately 15.5 million cancer survivors reside in the United States [1], and an estimated 2.8 million live in rural settings [2]. Compared to non-rural cancer survivors, they experience worse quality of life (QOL) and higher mortality rates [2, 3]. These QOL and mortality disparities may reflect rural survivors’ higher number of some currently experienced co-morbidities [4] and worse treatment-related symptoms [5, 6] relative to urban survivors and rural residents without a cancer history. Rural US residents experience a higher lifetime prevalence of multiple conditions simultaneously compared to those in non-rural settings [7] including heart disease [8], diabetes [8], and mood disorders, such as depression and anxiety [3, 4]. Rural cancer survivors in particular experience worse physical [5] and psychological treatment-related symptoms [6] compared to non-rural cancer survivors. These disparities likely reflect a number of shared factors, including limited healthcare access due to traveling greater distance for oncology services [9]; financial constraints such as out-of-pocket medical costs [10]; limited insurance [2]; and transportation issues [11]. The current pilot study seeks to: 1) add to a growing body of literature examining the cumulative effects of different health issues on QOL among rural survivors; 2) compare the magnitude of association that comorbidities and treatment-related symptoms have with QOL; and, 3) compare these associations across multiple QOL domains.

The burden of multiple health issues may have a cumulative cost on QOL [12], as framed by the revised Wilson and Cleary model of health-related QOL by Ferrans, Zerwic, Wilbur, and Larson [13]. This conceptual model characterizes how individuals may have greater biological risk, which will influence the presence of multiple morbidities and symptoms, leading to worse functioning, perceptions of poor health, and subsequent worse overall QOL. This model is particularly useful to consider in rural survivorship, as it highlights how the relationship between biological risk, symptoms, functioning status, health perceptions, and QOL may be affected by individuals’ personal and environmental characteristics. There are studies that have aligned with this framework, linking comorbidities and treatment-related symptoms to worse QOL among rural cancer survivors [4, 14, 15]. These additive effects may be present due to shared biological mechanisms, leading to psychological and physical impacts that are difficult to disentangle [16, 17]. Moreover, the simultaneous occurrence of multiple health issues may inhibit survivors from focusing on any one particular health issue regardless of its intensity [18]. We are aware of only two studies that have assessed these associations among rural survivors [5, 14].

This pilot study replicates previous research to further document if rural survivors’ quality of life is impacted by a greater number of comorbidities and treatment-related symptoms, which may help provide insight to explain urban-rural disparities in survivorship outcomes (e.g., unmet needs) [19]. For this study, we specifically focus on lifetime comorbidities, which include all conditions for which patients ever been diagnosed, as well as treatment-related symptoms experienced during the past week.

A second goal this pilot study begins is to address these associations across different QOL domains. Studies of rural populations have largely focused on mental health-related QOL (MHQOL) and physical health-related QOL (PHQOL) [20, 21]. Less research has examined social well-being and functional well-being, despite their importance with regard to health outcomes and mortality [22]. Our study addresses this gap by comparing how lifetime comorbidities and treatment-related symptoms are associated with social well-being, functional well-being, MHQOL, and PHQOL. Such work has clinical implications regarding the type of QOL intervention that is needed. A third goal this pilot study is to examine the relative contributions of lifetime comorbidities (defined as a current or past comorbid condition reported by the study participant) and treatment-related symptoms. Previous research in rural populations has focused either on lifetime comorbidities [23] or treatment-related symptoms on QOL, but not both together [24]. Another gap in the literature regards examining if differences exist in terms of physical versus psychological comorbidities and treatment-related symptoms. Beginning to compare the magnitude of associations may be important particularly for clinical intervention. Lifetime comorbidities may be more strongly associated with QOL if comorbidities include multiple, severe, diverse physiological effects [25]. The impact of these factors on QOL may depend on the degree to which comorbidities and treatment-related symptoms are perceived to be stressors by survivors, which may vary depending on if they are physical versus psychological. Treatment-related symptoms may however be more strongly associated with QOL, if they are particularly acute and chronic stressors (e.g., depending on severity) [26].

We will address the three goals goals discussed above via a secondary analysis of the Illinois Rural Cancer Assessment study (IRCA). Our multivariable models will quantify the relationships between lifetime comorbidities and treatment-related symptoms with four QOL domains, allowing us to: 1) add to previous research addressing the relationships of lifetime comorbidities, treatment-related symptoms, and QOL among rural populations; 2) compare the relative magnitude of lifetime comorbidities and treatment-related symptoms; and, 3) compare associations across different QOL domains.

METHODS

Procedures

The IRCA study was a statewide cross-sectional assessment that examined mental and physical health among rural cancer survivors and caregivers. The study was approved by the University of Illinois at Chicago Institutional Review Board and the University of Illinois Cancer Center Protocol Review Committee.

Recruitment occurred across two waves (Wave 1: Jan. 2017 - Feb. 2018; Wave 2: Mar. 2018 - Sept. 2018). We targeted all Illinois counties considered non-metropolitan (Rural Urban Continuum Code (RUCC) ≥4) [27], as well as one metropolitan county with <250,000 residents (RUCC = 3) with a high proportion of African American adults and proximity to neighboring nonmetropolitan counties in an effort to increase racial diversity in our sample. However, only a small number of participants identified as being from RUCC=3. For Wave 1, we used multiple non-probability-based recruitment methods, including electronic flyers distributed through websites, listservs, and social media; physical flyers placed in clinical and community organizations that serve the rural catchment areas as described above, including cancer centers, public health departments, clinics, and hospitals, churches, hair salons, support groups; and at cancer-related events (e.g., Relay for Life); and word-of-mouth. For Wave 2, we purchased a commercial list of landline and cellular phone numbers from Marketing Systems Group. Using specific demographic targets and algorithms, the list targeted African American adults from the aforementioned targeted counties in Illinois. Research personnel sent text messages and an image of the recruitment flyer to cellular phone numbers approximately 1–2 days prior to calling potential participants. The landline numbers only received a phone call from research personnel. A total number of 3,209 phone numbers were contacted as part of Wave 2.

Interested individuals were able to visit the survey website to complete an online survey; call personnel and complete the survey by phone; or, call personnel and request a mailed survey. After undergoing screening related to eligibility criteria (self-reported as 18 years or older, a cancer survivor or caregivers of cancer patients, a rural resident of Illinois State) and providing informed consent, participants completed the survey either by phone with a member of the research team or self-administered online or by mail. The survey took 60–90 minutes to complete. Participants received a $15 or $25 incentive (the increase in incentive occurred during the recruitment period). Of the final sample (n = 227), 191 individuals participated in Wave 1 and 36 individuals participated in Wave 2. For the current study, we focused on survivor participants only (n= 139). All data were entered into a Qualtrics survey database by the participant or study staff, depending on mode of administration. Data were exported into SPSS for analysis.

Measures

Cancer-related and demographic covariates were measured using standard items from the Behavioral Risk Factor Surveillance System and the Medical Expenditure Panel Survey Cancer Survivor Supplement to obtain demographic (age, sex, race and ethnicity, marital status, employment status, annual household income) and cancer-related information (cancer site, year of diagnosis, time since last treatment) [28, 29]. Given the number of sparse cells for cancer site, we collapsed categories into breast; gynecological (i.e., cervical, endometrial, ovarian); gastro-intestinal (i.e., esophageal, pancreatic, colorectal); skin (i.e., melanoma, other skin cancer); lymphoma (i.e., Hodgkin’s, non-Hodgkin’s); and, other (i.e., head/neck, oral, thyroid, leukemia, prostate, lung, bladder, renal, brain, other cancers). Rurality was ascertained by participants’ self-reported zipcode and use of RUCC 2013 codes [27].

Lifetime number of conditions in addition to participants’ initial cancer diagnosis were measured using an adapted version of the Self-Administered Comorbidity Questionnaire [30]. Participants endorsed whether they had ever received the following 12 diagnoses (yes/no): arthritis/rheumatism; back/neck problems; fracture/bone/joint injury; heart problems; stroke; high blood pressure; diabetes; lung/breath problems; depression/anxiety/emotional problems; weight problems; musculoskeletal problems; and, thyroid problems. Participants were also able to note if they had up to five other comorbidities using open-ended questions. A clinician subsequently reviewed answers, identified redundancies (e.g., heart problems noted in the closed-ended and specified in open-ended questions) and new comorbidities (e.g., eye problems, secondary cancers - Table 1). For primary analyses, we calculated the total number of lifetime comorbidities. For sensitivity analyses, we calculated if participants had reported psychological comorbidity (yes/no; Table 1) and the number of lifetime physical comorbidities (Table 1).

Table 1.

Study sample characteristics (n = 125)

DEMOGRAPHIC n %
Age
 18–53 years old 45 36
 54–64 years old 44 35
 65–83 years old 36 29
Sex
 Male 23 18
 Female 102 82
Race
 non-Latino White 112 90
 Other 13 10
Education
 < Bachelor’s degree 71 57
 ≥ Bachelor’s degree 54 43
Marital Status
 Married 86 69
 Not married 39 31
Employment
 Full-time employed 68 54
 Other 57 46
Household income
 <$35,000 36 29
 $35,001-$50,000 25 20
 ≥$50,001 64 51
County rurality1
 Populations of ≥20,000 (RUCC <4) 48 38
 Populations of <20,000 (RUCC 4+) 77 62
CANCER-RELATED n %
First cancer site2
 Breast cancer 49 39
 Other 31 25
 Skin 12 10
 Lymphoma 13 10
 Gastro-intestinal 11 9
 Gynecological 9 7
Time since last treatment
 ≥5 years ago 44 34
 1–4 years ago 39 31
 <1 year ago 42 35
TREATMENT-RELATED SYMPTOMS n %
Individual treatment-related symptoms
Difficulty concentrating 67 54
Feeling nervous 50 40
Difficulty sleeping 81 65
Feeling sad 62 50
Worrying 62 50
Feeling irritable 54 43
Pain 75 60
Lack of energy 85 68
Cough 35 28
Dry Mouth 49 39
Nausea 37 30
Feeling Drowsy 52 42
Numbness/tingling in the hands/feet 59 47
Feeling bloated 47 38
Problems with urination 21 17
Vomiting 26 21
Shortness of breath 42 34
Diarrhea 31 25
Sweats 42 34
Problems with sexual interest or activity 54 43
Itching 29 23
Lack of appetite 37 30
Dizziness 36 29
Difficulty swallowing 27 22
Treatment-Related Symptoms Scales3 M SD
Total MSAS 0.78 0.70
Physical Symptoms 0.66 0.68
Psychological Symptoms 1.11 0.98
LIFETIME COMORBIDITIES n %
Individual comorbidities
Arthritis/rheumatism 55 44
Back/neck problems 65 52
Fracture/bone/joint injury 52 42
Heart problems 28 22
Stroke 7 6
High blood pressure/hypertension 54 43
Diabetes 30 24
Lung/breath problem 33 26
Depression/anxiety/emotional problems 58 46
Obesity 67 54
Musculoskeletal problem 34 27
Thyroid problems 39 31
Other comorbidity 75 60
Lifetime Comorbidities Summary Scales4 M SD
All 5.48 3.23
Physical Comorbidities 5.02 3.07
QUALITY OF LIFE5 M SD
Social Well-being 18.66 5.16
Functional Well-being 19.50 6.59
MHQOL 48.82 10.48
PHQOL 44.95 11.07
Open in a new tab
1

Rural Urban Continuum Code (RUCC). RUCC 1–3 codes incorporate populations with ≥20,000 residents in metropolitan areas. RUCC 4–9 incorporate urban and completely rural areas with populations of <20,000.

2

Among the 24 survivors who had multiple cancer diagnoses.

3

Treatment-related symptoms were measured using the the first 24 items of the Memorial Symptom Assessment Scale (MSAS). [31]. Each symptom was assessed in terms of its presence (yes/no), frequency (rarely, occasionally, frequently, almost constantly), and associated severity/distress (not at all, a little bit, somewhat, quite a bit, very much) during the past week.

4

Lifetime number of conditions (in addition to participants’ initial cancer diagnosis) were measured using an adapted version of the Self-Administered Comorbidity Questionnaire [30]. Responses for comorbidities are dichotomous with participants selecting either “yes” or “no.” The ranges are respectively 0 to 14. Only 1 item pertained to psychological comorbidities, which is reported in the individual comorbidities section (depression/anxiety/emotional problems).

5

Range values social well-being and functional well-being are 0–28 with higher scores indicating better quality of life.[34, 35] The score for both mental health-related quality of life (MHQOL) and physical health-related QOL (PHQOL) range from 0 to 100 with a higher score also indicating better QOL. [36]

Treatment-related symptoms were measured using the first 24 items of the Memorial Symptom Assessment Scale (MSAS) [31]. The following symptoms were included: difficulty concentrating, feeling nervous, difficulty sleeping, feeling sad, worrying, feeling irritable, pain, lack of energy, cough, dry mouth, nausea, feeling drowsy, numbness/tingling in the hands/feet, feeling bloated, problems with urination, vomiting, shortness of breath, diarrhea, sweats, problems with sexual interest or activity, itching, lack of appetite, dizziness, and difficulty swallowing. Each symptom was assessed in terms of its presence (yes/no), frequency (rarely, occasionally, frequently, almost constantly), and associated severity/distress (not at all, a little bit, somewhat, quite a bit, very much) during the past week. Scoring for physical symptoms incorporates all dimensions (0 = symptom not present; 0.8 symptom present with no distress; 1.6 =symptom present and causes a little bit of distress; 2.4 = symptom present and causes some distress; 3.2 = symptom present and causes quite a bit of distress; 4.0 = symptom present and causes much distress). Scoring psychological symptoms incorporates the presence and frequency (0 = symptom not present; 1 = symptom present and occurs rarely; 2 = symptom present and occurs occasionally; 3 = if symptom is present and occurs frequency; 4 = symptom present and occurs almost constantly). For primary analyses, we calculated the Global Distress Index score (possible range = 0–4), which was the average of all 24 symptom scores. Cronbach’s alpha was 0.93 for our sample. For sensitivity analyses, we used an adjusted Physical Symptom Subscale, which was the average of the 8 prevalent physical symptom scores (lack of appetite, lack of energy, pain, feeling drowsy, dry mouth, nausea, vomiting, dizziness) and the Psychological Symptom Subscale, which was the average of the 6 psychological symptom scores (worrying, feeling sad, feeling nervous, difficulty sleeping, feeling irritable, difficulty concentrating). Cronbach’s alphas for both scales were 0.86.

Social and Functional Well-being were measured with the Functional Assessment of Cancer Therapy - General (FACT-G) instrument [32], a patient-reported outcome measure used to assess health and non-health dimensions of quality of life among patients undergoing cancer therapy. Due to survey response burden, only the social well-being and functional well-being subscales were assessed within this instrument.. Social well-being is assessed with 7 items, including family emotional support, friend support, family acceptance of illness, communication of illness, and closeness to partner. Functional well-being is also assessed with 7 items, including being able to work, perceived fulfillment of work, ability to enjoy life, acceptance of illness, sleep quality, enjoyment of “fun” activities, and feeling content with one’s life. Previous research suggests that a difference of 2 points is clinically meaningful for both the social and functional well-being scores [33]. The ranges for the FACT-G instrument are 0–28 for both social well-being and functional well-being, with higher scores indicating better QOL [34, 35]. Cronbach’s alphas for social and functional well-being were 0.87 and 0.89.

Mental and Physical Health-related QOL was measured using Short Form-12 Health Survey (SF-12) composite scores for mental and physical health, ranging from 0 to 100, where 0 indicates the lowest level of health and 100 indicates the highest. The SF-12 [36] has 12 questions with subdomains of general health, physical functioning, role functioning (physical), bodily pain, vitality, role functioning (emotional), mental health, social functioning. The PHQOL and MHQOL scales are created by combining, scoring, and weighting the questions.

Statistical Analysis

All analyses were performed with SPSS, version 25.0 [37]. We first assessed missingness. Second, we characterized study sample characteristics (Table 1). Next, we used bivariate analyses to identify covariates that were related to comorbidities, treatment-related consequences or QOL domains (Table 2). We were unable to examine relationships by sex, race, and cancer site due to sparse cell sizes while conducting the bivariate analyses. As a result, the covariates utilized within this analysis were age, marital status, education, household income, rurality, and time since last treatment. For our primary analyses (Table 3), we conducted 4 multivariable linear regressions that included our 2 predictors of interest (comorbidities and treatment-related consequences) and identified covariates related to the outcome variables of interest. We conducted five sets of sensitivity analyses with each of the four outcomes, including examining physical and psychological conditions only (Table 3).

Table 2.

Bivariable linear regression models concerning demographic factors, time since treatment, lifetime comorbidities, treatment-related consequences, and QOL domains (n = 125).1

Lifetime Comorbidities Treatment-related symptoms Social well-being Functional well-being Mental Health QOL Physical Health QOL
Std B (95%CI) p Std B (95%CI) p Std B (95%CI) p Std B (95%CI) P Std B (95%CI) p Std B (95%CI) p
Age −0.18 (0.004, 0.36) 0.05 −0.10 (−0.26, 0.07) 0.27 0.09 (−0.09, 0.27) 0.30 0.02 (−0.15, 0.20) 0.81 0.21 (0.04, 0.39) 0.02 −0.09 (−0.27, 0.09) 0.32
Marital Status
 Not Married REF REF REF REF REF REF REF REF REF REF REF REF
 Married −0.10 (−0.28, 0.09) 0.25 0.001 (−0.14, 0.14) 0.99 0.18 (0.01, 0.36) 0.04 0.19 (0.02, 0.37) 0.03 0.07 (−0.11, 0.25) 0.44 0.26 (0.09, 0.43) 0.004
Education
 <Bachelor’s REF REF REF REF REF REF REF REF REF REF REF REF
 ≥Bachelor’s 0.03 (−0.15, 0.21) 0.74 −0.03 (−0.21, 0.16) 0.77 0.09 (−0.08, 0.27) 0.30 0.16 (−0.02, 0.33) 0.08 0.06 (−0.12, 0.24) 0.52 0.28 (0.10, 0.45) 0.002
Household income −0.14 (−0.32, 0.03) 0.11 −0.17 (−0.35, 0.004) 0.05 0.23 (0.06, 0.40) 0.01 0.31 (0.14, 0.48) <0.0001 −0.15 (−0.02, 0.33) 0.08 0.36 (0.19, 0.52) <0.0001
County rurality2
 ≥20,000 REF REF REF REF REF REF REF REF REF REF REF REF
 <20,000 0.10 (−0.07, 0.28) 0.26 0.15 (−0.03, 0.33) 0.10 −0.07 (−0.25, 0.11) 0.43 −0.15 (−0.32, 0.03) 0.11 −0.09 (−0.26, 0.09) 0.35 −0.14 (−0.31, 0.04) 0.13
Cancer site
 Other cancers REF REF REF REF REF REF REF REF REF REF REF REF
 Breast cancer −0.08 (−0.26, 0.09) 0.35 0.07 (−0.11, 0.24) 0.47 −0.10 (−0.27, 0.08) 0.29 0.07 (−0.25, 0.11) 0.44 −0.13 (−0.31, 0.05) 0.15 0.09 (−0.09, 0.27) 0.32
Time since last treatment
 <5 years ago REF REF REF REF REF REF REF REF REF REF REF REF
 ≥5 years ago 0.02 (−0.15, 0.20) 0.81 0.21 (0.03, 0.38) 0.02 −0.16 (−0.34, 0.02) 0.08 −0.30 (−0.47, −0.13) 0.001 −0.20 (−0.37, −0.02) 0.03 −0.33 (−0.49, 0.16) <0.0001
Open in a new tab
1

Significant associations (p ≤0.05) are in bold. Non-significant associations (p<0.10) are italicized.

2

Rural Urban Continuum Code (RUCC). RUCC 1–3 codes incorporate populations with ≥20,000 residents in metropolitan areas. RUCC 4–9 incorporate urban and completely rural areas with populations of <20,000.

Table 3.

Multivariable regression models concerning lifetime comorbidities, treatment-related consequences, and QOL domains (n = 125).1

Social well-being2 Functional well-being2 MHQOL3 PHQOL3
Std B (95%CI) p Std B (95%CI) P Std B (95%CI) p Std B (95%CI) p
Primary Analyses
Lifetime comorbidities −0.27 (−0.44, −0.09) 0.003 −0.36 (−0.52, −0.20) <0.0001 −0.30 (−0.46, −0.13) 0.001 −0.10 (−0.27, 0.07) 0.24
Treatment-related symptoms −0.16 (−0.34, 0.01) 0.08 −0.18 (−0.34, −0.02) 0.03 −0.25 (−0.42, −0.08) 0.004 −0.11 (−0.28, 0.06) 0.20
Psychological Conditions Only
Lifetime psychological comorbidities −0.22 (−0.40, −0.05) 0.01 −0.30 (−0.47, −0.14) <0.0001 −0.41 (−0.55, −0.27) <0.0001 0.01 (0.15, 0.18) 0.88
Treatment-related psychological symptoms −0.22 (−0.40, −0.03) 0.03 −0.22 (−0.39, −0.05) 0.01 −0.39 (−0.54, −0.24) <0.0001 −0.10 (−0.28, 0.09) 0.29
Physical Conditions Only
Lifetime physical comorbidities −0.25 (−0.43, −0.07) 0.007 −0.33 (−0.50, −0.17) <0.0001 −0.26 (−0.43, −0.08) 0.005 −0.10 (−0.27, 0.06) 0.23
Treatment-related physical symptoms −0.11 (−0.30, 0.06) 0.20 −0.14 (−0.30, 0.02) 0.09 −0.15 (−0.32, 0.03) 0.10 −0.12 (−0.28, 0.05) 0.18
Open in a new tab
1

All models were controlled for the following covariates: age, marital status (ref :not married), education, annual household income, rurality, and time since treatment. Significant associations (p ≤0.05) are in bold. Non-significant associations (p<0.10) are italicized

2

Range values social well-being and functional well-being are 0–28 with higher scores indicating better quality of life. [34, 35]

3

The score for both mental and physical health range from 0 to 100 with a higher score also indicating better QOL. [36]

RESULTS

Only 9 cases had any missing data, of which 7 did not report their household income, 2 did not report their cancer site, and 2 did not report the time since their last treatment. Five cases also noted they did not need treatment for their cancers and were excluded from this analysis, given the focus on treatment-related symptoms. Relative to individuals in the final analytic sample (n = 125), the 14 excluded participants were less likely to have completed treatment >5 years before their participation in the study (p = 0.01).

Table 1 depicts the sample’s demographic and cancer-related characteristics. With regard to demographic factors, approximately 29% were 65–83 years old, 82% were female, 90% identified as non-Latino White, 69% were married, 43% had a Bachelor’s degree or greater educational attainment, and 51% had a household income of $50,000 or greater. Approximately 62% lived in counties with <20,000 residents. With regard to cancer-related factors, the most common type was breast cancer and 35% of participants had completed treatment <1 year ago. The most prevalent treatment-related symptoms within the study sample was lack of energy (n=85, 68%) followed by difficulty sleeping (n=81, 65%). With regard to Global Distress Index scores, 19% of respondents reported they had not experienced any of the included symptoms and 32–73% reported no or little distress in relation to these specific experienced symptoms. Weight problems (n=67, 54%) and back/neck problems (n=65, 52%) were the most prevalent lifetime comorbidities. The average scores for social, functional, mental, and physical QOL were respectively 18.66 (Std Dev = 5.16), 19.50 (Std Dev = 6.59), 48.82 (Std Dev = 10.48), and 44.95 (Std Dev = 11.07).

Table 2 provides results from bivariate relationships between demographic and cancer-related variables to lifetime comorbidities, treatment-related symptoms, and QOL domains. Older individuals reported more lifetime comorbidities. Individuals with lower income and those who had completed treatment ≥5 years ago reported more treatment-related symptoms. Social well-being and functional well-being were associated with being married and having a higher income. Functional well-being was also associated with completing treatment ≥5 years ago. MHQOL was associated with older age and completing treatment 5 or more years ago. PHQOL was associated with being married, having a Bachelor’s degree, higher income, and completing treatment ≥5 years ago.

Table 3 depicts the results of multivariable regressions, after adjusting for age, marital status, education, income, rurality, and time since last treatment. Lifetime comorbidities were inversely associated with social well-being (β=−0.27, CI:−0.44,−0.09), functional well-being (β=−0.36, CI:−0.52,−0.20), and MHQOL (β=−0.30, CI:−0.46, −0.13). Treatment-related symptoms were inversely associated with only functional well-being (β=−0.18, CI:−0.34, - 0.02) and MHQOL (β=−0.25, CI:−0.42, −0.08). Based on the beta coefficients, there was a slightly stronger magnitude of association observed for lifetime comorbidities compared to treatment-related symptoms. Sensitivity analyses suggested that treatment-related psychological symptoms and psychological comorbidities were comparably related to social well-being, functional well-being, and MHQOL. In contrast, physical comorbidities, but not physical treatment-related symptoms, were associated with social well-being, functional well-being, and MHQOL.

DISCUSSION

Our work adds to a growing body of research regarding the additive impacts of multiple health costs on QOL among rural cancer survivors [2–4]. Specifically, our research highlights that the multiple disproportionate health burdens rural populations face, as approximated by the number of lifetime comorbidities and treatment-related symptoms, may lead to worse outcomes. Further, our work provides preliminary data regarding the more common lifetime comorbidities (e.g., obesity, back/neck problems) and treatment-related symptoms (e.g., lack of energy, difficulty sleeping) rural cancer survivors face. These findings will inform future work that examines if specific combinations of comorbidities and symptoms may be particularly important to address among rural survivors.

We further provided preliminary data to suggest how relationships may vary by QOL domain. Specifically, there were stronger associations with social well-being, functional well-being, and MHQOL relative to PHQOL. Lifetime comorbidities and treatment-related symptoms may be more likely to impact these domains because of cognitive [38] and interpersonal factors [39] which may not be as strongly related to PHQOL. For example, comorbidities and treatment-related symptoms may more strongly and consistently affect perceived life satisfaction, which would more strongly affect social well-being, functional well-being, and MHQOL [40]. This work partially supports the causal pathway provided by Ferrans and colleagues [13], wherein functioning/functional well-being and QOL domains that are closely linked to health perceptions may be more proximally connected to biological function and symptoms than PHQOL. In turn, PHQOL may be more impacted by factors such as type of treatment and cancer stage in the long term.

Finally, our sensitivity analyses suggest relationships vary when examining physical and psychological health factors separately. We found that physical treatment-related symptoms were less associated with QOL domains than lifetime comorbidities (physical, psychological) and psychological treatment-related symptoms. Physical lifetime comorbidities may be more strongly associated with QOL due to multiple, severe, diverse physiological effects [25]. It is possible that the impact of physical treatment-related symptoms have resolved over time, so that they no longer impact QOL substantially for many of our participants. Thus, the persistence of psychological treatment-related symptoms found in our study is even more striking, demonstrating that these symptoms have ongoing and even long-term effects among rural cancer survivors [41].

LIMITATIONS

There are several limitations within this study. The first limitation concerns convenience-based sampling, which resulted in a lack of ethnic and gender diversity. Our findings may not be generalizable. Studies that leverage probability-based sampling are needed. Second, our findings and small sample size should be interpreted with caution. Larger studies are needed. Third, our pilot study treated lifetime comorbidities and treatment-related symptoms as independent factors. Yet, as described above, treatment-related symptoms can result in post-treatment comorbidities [42] and lifetime comorbidities can result in more treatment-related symptoms [43]. There is a need to disentangle the relationship of these factors in order to understand their likely nuanced associations with QOL. Relatedly, our study assessed lifetime comorbidities and was not able to measure current comorbidities specifically. There is a need for future research that can examine the relative contributions of lifetime versus current comorbidities as well as disparities between rural and non-rural populations of survivors and non-cancer survivors. Toward that goal, more detailed examination of length of time since cancer diagnosis and the duration of psychological or physical health conditions are warranted in the future via longitudinal designs. Lastly, the entire FACT-G instrument was not included within this study to measure QOL to minimize survey burden. Future researchers may be able to obtain a more comprehensive understanding of the QOL of rural cancer survivors by measuring each subscale within the FACT-G instrument along with the SF-12 instrument.

CONCLUSION AND IMPLICATIONS

The findings of this study suggest that the social well-being, functional well-being, and MHQOL among rural cancer survivors are associated with lifetime comorbidities and treatment-related psychological conditions. The implementation of effective psychosocial interventions to improve rural cancer survivors’ QOL may need to assess these characteristics and consider their simultaneous presence/additive impact prior to attempt to mitigate their effects. Our findings also reinforce the importance of incorporating comorbidities and persistent treatment-related symptoms into survivorship care plans for rural cancer survivors. Currently survivorship care plans are providing guidance for cancer survivors with preexisting health issues post-treatment [44]. The findings of this study specifically highlight the utility of risk-based survivorship care plans, which have grown in popularity [45–47]. Our work reinforces the need to consider the types of treatment-related symptoms and lifetime comorbidities as well as their cumulative and potentially interdependent effects, in line with recent recommendations for shifting from “point of care to point of need” [47]. Yet, a growing concern regards how to provide such comprehensive care in light of provider shortages and growing cost [47], especially for minority health and health disparity populations [48]. For example, rural cancer survivors are less likely to receive such survivorship care plans compared to urban cancer survivors [49]. Our work adds to a growing collective call to incorporate contextual and health factors in personalized, risk-based survivorship care planning for all survivors, including rural populations. For example, when developing personalized survivorship care plans for rural populations, it may be important to consider patients’ geographic access to their provider team (e.g. community nurses; primary care physicians; oncologists) in complement with existing research regarding who may be optimal for providing risk-based care [50]. In rural settings, substantial distance from any provider, compounded by distance from the most expert providers, poses unique challenges to for delivery of timely and effective care, particularly preventative care, creating increased survivorship risks. Relatedly, it will be worthwhile to consider both individual- and area-level socioeconomic disadvantage [4], which may underlie comorbidities and exacerbate comorbidities’ impact on rural survivors’ QOL. Accordingly, person-centered care and patient engagement when developing personalized plans will be crucial [51].

Acknowledgment

We thank Dr. Rodolfo Rodriguez and Emily Hallgren for their technical assistance. We would like to thank the rural cancer survivors who participated in this study and without whom this research would not be possible.

Conflict of Interest Statement: The authors declare that there are no conflicts of interest and no financial disclosures to report. This work was supported by the Center for Research on Women and Gender, University of Illinois at Chicago and the University of Illinois Cancer Center. Dr. Yamile Molina’s time was supported by the National Cancer Institute under the grant number K01CACA193918. Dr. Shaila Strayhorn’s time was supported by the National Cancer Institute training grant: Cancer Education and Career Development Program (T32CA057699). Dr. Molina has full control of all primary data and would agree for the Journal of Supportive Care in Cancer offices to review data, if requested.

Footnotes

Publisher's Disclaimer: This Author Accepted Manuscript is a PDF file of a an unedited peer-reviewed manuscript that has been accepted for publication but has not been copyedited or corrected. The official version of record that is published in the journal is kept up to date and so may therefore differ from this version.

REFERENCES

  • 1.Cancer Statistics. (2018) In: National Cancer Institute. https://www.cancer.gov/about-cancer/understanding/statistics. Accessed 6 Nov 2018
  • 2.Weaver KE, Palmer N, Lu L, Case LD, Geiger AM (2013) Rural-urban differences in health behaviors and implications for health status among US cancer survivors. Cancer Causes & Control 24:1481–1490 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Burris JL, Andrykowski M (2010) Disparities in mental health between rural and nonrural cancer survivors: a preliminary study. Psycho-Oncology: Journal of the Psychological, Social and Behavioral Dimensions of Cancer 19:637–645 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Gunn KM, Berry NM, Meng X, Wilson CJ, Dollman J, Woodman RJ, … & Koczwara B (2019). Differences in the health, mental health and health-promoting behaviours of rural versus urban cancer survivors in Australia. Supportive Care in Cancer, 1–11. [Google Scholar]
  • 5.DiSipio T, Hayes SC, Newman B, Aitken J, Janda M (2010) Does quality of life among breast cancer survivors one year after diagnosis differ depending on urban and non-urban residence? A comparative study. Health and quality of life outcomes 8:3. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Lashbrook M, Bernardes CM, Kirshbaum MN, Valery PC (2018) Physical functioning and psychological morbidity among regional and rural cancer survivors: A report from a regional cancer centre. Australian Journal of Rural Health [DOI] [PubMed] [Google Scholar]
  • 7.O’Connor A, Wellenius G (2012) Rural-urban disparities in the prevalence of diabetes and coronary heart disease. Public health 126:813–820 [DOI] [PubMed] [Google Scholar]
  • 8.Azuero A, Benz R, McNees P, Meneses K (2014) Co-morbidity and predictors of health status in older rural breast cancer survivors. SpringerPlus 3:102. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Blake KD, Moss JL, Gaysynsky A, Srinivasan S, Croyle RT (2017) Making the case for investment in rural cancer control: an analysis of rural cancer incidence, mortality, and funding trends. Cancer Epidemiology and Prevention Biomarkers 26:992–997 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Palmer NR, Geiger AM, Lu L, Case LD, Weaver KE (2013) Impact of rural residence on forgoing healthcare after cancer because of cost. Cancer Epidemiology and Prevention Biomarkers 22:1668–1676 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Baldwin L, Cai Y, Larson EH, Dobie SA, Wright GE, Goodman DC, Matthews B, Hart LG (2008) Access to cancer services for rural colorectal cancer patients. The Journal of Rural Health 24:390–399 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Franceschini J, dos Santos AA, El Mouallem I, Jamnik S, Uehara C, Fernandes ALG, Santoro IL (2008) Assessment of the quality of life of patients with lung cancer using the Medical Outcomes Study 36-item Short-Form Health Survey. Jornal Brasileiro de Pneumologia 34:387–393 [DOI] [PubMed] [Google Scholar]
  • 13.Ferrans CE, Zerwic JJ, Wilbur JE, Larson JL (2005) Conceptual model of health-related quality of life. Journal of nursing scholarship 37:336–342 [DOI] [PubMed] [Google Scholar]
  • 14.Beck SL, Towsley GL, Caserta MS, Lindau K, Dudley WN (2009) Symptom experiences and quality of life of rural and urban older adult cancer survivors. Cancer Nursing 32:359–369 [DOI] [PubMed] [Google Scholar]
  • 15.Chabowski M, Polanski J, Mazur G, Janczak D, Rosinczuk J (2017) Sociodemographic and clinical determinants of quality of life of patients with non-small cell lung cancer In: Clinical Management of Pulmonary Disorders and Diseases. Springer, pp 1–10 [DOI] [PubMed] [Google Scholar]
  • 16.Miller AH, Ancoli-Israel S, Bower JE, Capuron L, Irwin MR (2008) Neuroendocrine-immune mechanisms of behavioral comorbidities in patients with cancer. Journal of clinical oncology: official journal of the American Society of Clinical Oncology 26:971. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Cleeland CS, Bennett GJ, Dantzer R, Dougherty PM, Dunn AJ, Meyers CA, Miller AH, Payne R, Reuben JM, Wang XS (2003) Are the symptoms of cancer and cancer treatment due to a shared biologic mechanism? A cytokine-immunologic model of cancer symptoms. Cancer: Interdisciplinary International Journal of the American Cancer Society 97:2919–2925 [DOI] [PubMed] [Google Scholar]
  • 18.Gerend MA, Pai M (2008) Social determinants of Black-White disparities in breast cancer mortality: a review. Cancer Epidemiology and Prevention Biomarkers 17:2913–2923 [DOI] [PubMed] [Google Scholar]
  • 19.Beesley V, Eakin E, Steginga S, Aitken J, Dunn J, Battistutta D (2008) Unmet needs of gynaecological cancer survivors: implications for developing community support services. Psycho-Oncology: Journal of the Psychological, Social and Behavioral Dimensions of Cancer 17:392–400 [DOI] [PubMed] [Google Scholar]
  • 20.Pedro L (2012) Theory derivation: adaptation of a contextual model of health related quality of life to rural cancer survivors. Online Journal of Rural Nursing and Health Care 10:80–95 [Google Scholar]
  • 21.Naughton MJ, Weaver KE (2014) Physical and mental health among cancer survivors considerations for long-term care and quality of life. North Carolina medical journal 75:283–286 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Reid-Arndt SA, Cox CR (2010) Does rurality affect quality of life following treatment for breast cancer? The Journal of Rural Health 26:402–405 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Andrykowski MA (2012) Physical and mental health status of survivors of multiple cancer diagnoses: findings from the national health interview survey. Cancer 118:3645–3653 [DOI] [PubMed] [Google Scholar]
  • 24.Russell L, Gough K, Drosdowsky A, Schofield P, Aranda S, Butow PN, Westwood JA, Krishnasamy M, Young JM, Phipps-Nelson J (2015) Psychological distress, quality of life, symptoms and unmet needs of colorectal cancer survivors near the end of treatment. Journal of cancer survivorship 9:462–470 [DOI] [PubMed] [Google Scholar]
  • 25.Sehl M, Lu X, Silliman R, Ganz PA (2013) Decline in physical functioning in first 2 years after breast cancer diagnosis predicts 10-year survival in older women. Journal of Cancer Survivorship 7:20–31 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Deimling GT, Arendt JA, Kypriotakis G, Bowman KF (2009) Functioning of older, long-term cancer survivors: The role of cancer and comorbidities. Journal of the American Geriatrics Society 57:S289–S292 [DOI] [PubMed] [Google Scholar]
  • 27.United States Department of Agriculture, (2013) Rural Urban Continuum Codes. https://www.ers.usda.gov/data-products/rural-urban-continuum-codes/
  • 28.Yabroff KR, Dowling E, Rodriguez J, Ekwueme DU, Meissner H, Soni A, Lerro C, Willis G, Forsythe LP, Borowski L (2012) The Medical Expenditure Panel Survey (MEPS) experiences with cancer survivorship supplement. Journal of Cancer Survivorship 6:407–419 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Centers for Disease Control and Prevention (CDC) (2013) About the Behavioral Risk Factor Surveillance System (BRFSS).(2016)
  • 30.Sangha O, Stucki G, Liang MH, Fossel AH, Katz JN (2003) The Self-Administered Comorbidity Questionnaire: a new method to assess comorbidity for clinical and health services research. Arthritis Care & Research 49:156–163 [DOI] [PubMed] [Google Scholar]
  • 31.Portenoy RK, Thaler HT, Kornblith AB, Lepore JM, Friedlander-Klar H, Kiyasu E, Sobel K, Coyle N, Kemeny N, Norton L (1994) The Memorial Symptom Assessment Scale: an instrument for the evaluation of symptom prevalence, characteristics and distress. European Journal of Cancer 30:1326–1336 [DOI] [PubMed] [Google Scholar]
  • 32.Cella DF, Tulsky DS, Gray G, Sarafian B, Linn E, Bonomi A, Silberman M, Yellen SB, Winicour P, Brannon J (1993) The Functional Assessment of Cancer Therapy scale: development and validation of the general measure. J Clin Oncol 11:570–579 [DOI] [PubMed] [Google Scholar]
  • 33.Brucker PS, Yost K, Cashy J, Webster K, Cella D (2005) General population and cancer patient norms for the Functional Assessment of Cancer Therapy-General (FACT-G). Evaluation & the health professions 28:192–211 [DOI] [PubMed] [Google Scholar]
  • 34.Yost KJ, Thompson CA, Eton DT, Allmer C, Ehlers SL, Habermann TM, Shanafelt TD, Maurer MJ, Slager SL, Link BK (2013) The Functional Assessment of Cancer Therapy-General (FACT-G) is valid for monitoring quality of life in patients with non-Hodgkin lymphoma. Leukemia & lymphoma 54:290–297 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Fairclough D, Cella D (1996) Functional Assessment of Cancer Therapy (FACT-G): Non-response to individual questions. Quality of Life Research 5:321–329 [DOI] [PubMed] [Google Scholar]
  • 36.Ware JE Jr, Kosinski M, Keller SD (1996) A 12-Item Short-Form Health Survey: construction of scales and preliminary tests of reliability and validity. Medical care 34:220–233 [DOI] [PubMed] [Google Scholar]
  • 37.SPSS I (2013) SPSS statistics for Windows. Armonk, NY: IBM, Corp [Google Scholar]
  • 38.Ciucă A, Băban A (2017) Psychological factors and psychosocial interventions for cancer related pain. Romanian Journal of Internal Medicine 55:63–68 [DOI] [PubMed] [Google Scholar]
  • 39.Slavich GM, Auerbach RP (2018) Stress and its sequelae: Depression, suicide, inflammation, and physical illness. APA Handbook of Psychopathology: Psychopathology: Understanding, Assessing, and Treating Adult Mental Disorders 1:0000064–016 [Google Scholar]
  • 40.Tate DG, Forchheimer M (2002) Quality of life, life satisfaction, and spirituality: Comparing outcomes between rehabilitation and cancer patients. American Journal of Physical Medicine & Rehabilitation 81:400–410 [DOI] [PubMed] [Google Scholar]
  • 41.Mehnert A, Koch U (2008) Psychological comorbidity and health-related quality of life and its association with awareness, utilization, and need for psychosocial support in a cancer register-based sample of long-term breast cancer survivors. Journal of psychosomatic research 64:383–391 [DOI] [PubMed] [Google Scholar]
  • 42.Hewitt M, Greenfield S, Stovall E (2005) From cancer patient to cancer survivor: lost in transition. National Academies Press [Google Scholar]
  • 43.Given CW, Given B, Azzouz F, Kozachik S, Stommel M (2001) Predictors of pain and fatigue in the year following diagnosis among elderly cancer patients. Journal of pain and symptom management 21:456–466 [DOI] [PubMed] [Google Scholar]
  • 44.Mayer DK, Birken SA, Check DK, Chen RC (2015) Summing it up: an integrative review of studies of cancer survivorship care plans (2006–2013). Cancer 121:978–996 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 45.McCabe MS, Bhatia S, Oeffinger KC, Reaman GH, Tyne C, Wollins DS, Hudson MM (2013) American Society of Clinical Oncology statement: achieving high-quality cancer survivorship care. Journal of Clinical Oncology 31:631. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 46.Mayer DK, Alfano CM (2019) Personalized Risk-Stratified Cancer Follow-Up Care: Its Potential for Healthier Survivors, Happier Clinicians, and Lower Costs. Journal of the National Cancer Institute 111:442–448 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 47.Alfano CM, Leach CR, Smith TG, Miller KD, Alcaraz KI, Cannady RS, Wender RC, Brawley OW (2019) Equitably improving outcomes for cancer survivors and supporting caregivers: A blueprint for care delivery, research, education, and policy. CA: a cancer journal for clinicians 69:35–49 [DOI] [PubMed] [Google Scholar]
  • 48.Halpern MT, McCabe MS, Burg MA (2016) The cancer survivorship journey: models of care, disparities, barriers, and future directions. American Society of Clinical Oncology Educational Book 36:231–239 [DOI] [PubMed] [Google Scholar]
  • 49.Schootman M, Homan S, Weaver KE, Jeffe DB, Yun S (2013) Peer reviewed: the health and welfare of rural and urban cancer survivors in Missouri. Preventing chronic disease 10:1–11 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 50.McCabe MS, Partridge AH, Grunfeld E, & Hudson MM (2013) Risk-based health care, the cancer survivor, the oncologist, and the primary care physician. In Seminars in oncology 40: 804–812 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 51.Loonen J, Blijlevens NM, Prins J, Dona DJ, Den Hartogh J, Senden T, van Dulmen-Den Broeder E, van der Velden K, Hermens RP (2018) Cancer survivorship care: person centered care in a multidisciplinary shared care model. International journal of integrated care 18 [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES