ORIGINAL ARTICLE
Cancer Survivorship in Rural and Urban Adults: A Descriptive and Mixed Methods Study Julie A. McNulty, RN, PhD1 & Lillian Nail, PhD, RN, FAAN2 1 Arizona State University, College of Nursing and Health Innovation, Phoenix, Arizona 2 Oregon Health and Science University, School of Nursing, Portland, Oregon
Abstract Purpose: Rural-dwelling cancer survivors (CSs) are at risk for decrements in Conflict of Interest: The authors have no conflicts of interest to disclose. Funding: This work was supported by the Oregon Health and Science University Dean’s Scholarship fund. Acknowledgments: Special thanks to the cancer survivors who shared their time and experiences. The first author would like to thank Lillian Nail, PhD, RN, FAAN, who served as the chair of her dissertation committee and primary mentor, as well as committee members Rebecca Block, MSW, PhD; Brandon Hayes-Lattin, MD; and Frances Lee-Lin, PhD, CNS. For further information, contact: Julie A. McNulty, RN, PhD, 500 N 3rd St., Phoenix, AZ 85020; e-mail: [email protected]. doi: 10.1111/jrh.12106
health and well-being due to decreased access to health care and support resources. This study compares the impact of cancer in rural- and urban-dwelling adult CSs living in 2 regions of the Pacific Northwest. Methods: A convenience sample of posttreatment adult CSs (N = 132) completed the Impact of Cancer version 2 (IOCv2) and the Memorial Symptom Assessment Scale-short form. High and low scorers on the IOCv2 participated in an in-depth interview (n = 19). Findings: The sample was predominantly middle-aged (mean age 58) and female (84%). Mean time since treatment completion was 6.7 years. Cancer diagnoses represented included breast (56%), gynecologic (9%), lymphoma (8%), head and neck (6%), and colorectal (5%). Comparisons across geographic regions show statistically significant differences in body concerns, worry, negative impact, and employment concerns. Rural-urban differences from interview data include access to health care, care coordination, connecting/community, thinking about death and dying, public/private journey, and advocacy. Conclusion: The insights into the differences and similarities between rural and urban CSs challenge the prevalent assumptions about rural-dwelling CSs and their risk for negative outcomes. A common theme across the study findings was community. Access to health care may not be the driver of the survivorship experience. Findings can influence health care providers and survivorship program development, building on the strengths of both rural and urban living and the engagement of the survivorship community.
Key words cancer survivorship, impact of cancer, late effects, rural health, rural-urban cancer survivorship.
Cancer survivors (CSs) are living longer, growing in number, and are now being recognized as being at risk for decrements in health and well-being due to cancer and cancer treatment. Increases are due to an aging population, improvements in screening and treatment, and decreased mortality from other causes.1 Effects of treatment may persist many years after diagnosis, and new treatment-related problems can appear years after
282
treatment is completed.2 Challenges that confront CSs are not simply a continuation of experiences that occur during the cancer treatment phase; they are unique problems in the cancer illness trajectory, as the person treated for cancer transitions from the acute to the chronic phases of the illness.3 Persistent and late effects include physical and psychological symptoms, and social and spiritual impact. CSs are likely to face comorbid
c 2015 National Rural Health Association The Journal of Rural Health 31 (2015) 282–291
Cancer Survivorship, Rural and Urban
McNulty and Nail
illness, functional limitations, difficulties with returning to work and other role transitions, uncertainty, fear of recurrence, and barriers to quality health care, all of which complicate their lives.1,4,5
Rural Survivorship Out of the estimated 13.7 million CSs in the United States currently, 2.8 million are living in rural settings. Rural cancer survivors (RCSs) are viewed as having higher levels of risk for decrements in health and well-being due to decreased access to health care, specialty services, and support resources.6 RCSs may experience lack of symptom management, lack of access to psychosocial interventions, isolation, and decreased coping with regard to their cancer experience.7 Mental health disparities have been identified with RCSs showing poorer mental health outcomes than urban cancer survivors (UCSs).8 Studies are lacking that compare quality of life between the 2 groups. Recent studies are yielding inconsistent results or mixed findings,9 are revealing inferior outcomes in RCSs,9-11 or no differences.8 This study focuses on posttreatment CSs living in Alaska and Oregon, 2 states with high levels of rurality. Alaska has a higher percentage of rural population than the rest of the United States,12 with 52.3% of the population living in frontier areas.13 Alaska is over twice the size of Texas14 and encompasses 20% of the land mass of the United States.15 Furthermore, Alaskans face geographical challenges not present in any other state, such as limited road access into remote areas and harsh weather patterns. Travel by air is required to reach the remote areas, which make up the greatest portion of the state. Alaskans experience unique needs regarding cancer survivorship. Many health care services that would be considered standard care in other states are unavailable in rural and remote Alaska.16 Access to oncologists is limited to the urban areas of Anchorage and Fairbanks. Cancer treatment is apt to be provided far from home, often in another state, such as Washington, Oregon, or Minnesota. Similarly, the rural population of Oregon comprises 29.1% of the state population, and 86.3% of the land is defined as rural.17 People living in the eastern part of the state may be over 100 miles from cancer treatment facilities, and access to these facilities may require traveling over mountain passes. There are limited cancer treatment programs and survivorship resources in this sparsely populated area,18 and there are no practicing oncologists.19 In contrast to the cities of Anchorage and Fairbanks in Alaska, the city of Portland offers highly specialized cancer treatment centers, and the state offers more sophisticated cancer survivorship programs. In both states, travel and distance challenges create a significant disruption in
c 2015 National Rural Health Association The Journal of Rural Health 31 (2015) 282–291
the lives of rural cancer patients and their families, and their out-of-pocket costs are likely to be significant.20
Gaps in Knowledge Most health-related cancer research has focused on the acute and treatment stages of the illness, and there is a strong suggestion in the literature that more research is needed during the posttreatment stages of survivorship. Until recently this phase of the cancer trajectory has been relatively neglected in advocacy, education, clinical practice, and research.7 CSs are considered an understudied population, who may be lost in transition once cancer treatment is completed. Issues encountered by RCSs, and the long-term psychosocial effects of cancer in rural settings, are areas of needed research.21 It is currently unclear which types of psychosocial and educational interventions are most effective for RCSs. Although there is strong evidence that traditional cancer support groups are effective for patients during the acute phase of cancer treatment,22 less is known about how to effectively meet the needs of long-term RCSs. The present study is the first of its kind in these 2 states. The purpose of this descriptive and mixed methods design study is to explore and compare the impact of cancer in rural- and urban-dwelling adult CSs in 2 regions of the Pacific Northwest: Alaska and Oregon. The impact of cancer was measured by administering the Impact of Cancer version 2 (IOCv2) instrument,23,24 as well as a symptom measure.25 The impact of cancer and the survivorship experience was also explored qualitatively through in-depth interviews. In this paper, the term region will be used when referring to the state where the participant lives, and the term location will refer to whether the participant was residing in a rural or urban setting. Aim #1: To explore and compare the impact of the cancer survivorship and symptom experience in RCSs and UCSs in 2 regions of the Pacific Northwest (quantitative). We proposed that UCSs would have more positive scores on study instruments than would RCSs, that Oregonians would have more positive scores than Alaskans, and that there would be an interaction between region and location such that urban-dwelling Oregonians would have more positive scores than urban Alaskans, and rural Oregonians would have more positive scores than rural Alaskans. The exploratory hypotheses were based on the differences in access to health care between the regions and locations. Aim #2: To explore and compare the physical (including symptoms), psychological, social, and spiritual aspects of cancer survivorship in RCSs and UCSs within and between 2 regions of the Pacific Northwest (qualitative).
283
Cancer Survivorship, Rural and Urban
Methods Procedures Ethical approval for this study was obtained from the Knight Cancer Institute and the Oregon Health and Science University and Alaska Area investigational review boards. Participants were recruited by distributing e-mails and flyers to cancer survivorship programs, support group leaders, oncology and primary care providers, and through networking. Once the participants contacted the researcher, they were screened for eligibility and informed consent was obtained. Participants were given a choice of method for survey administration. If their preference was online, a link to the study questionnaires was sent via e-mail. If the choice was on paper, the instruments were mailed to the participants. In the final sample, 88% completed the surveys via the web and 12% on paper. Because of the variety of recruitment strategies utilized, it is unknown how many people were actually recruited into the study and declined to participate. Therefore, it is not feasible to calculate an overall response rate. Once potential participants contacted the researcher about the study, 70% of this group completed the study questionnaires. The convenience sample (N = 132) included adult survivors of all cancer types (excluding nonmelanotic skin cancer) who had completed active cancer treatment and were self-reported as cancer-free. Participants were able to read, speak, and write English.
Rural and Urban Definitions Due to the stark differences in geography, travel distances and types of transportation used, a rural definition unique to each state was utilized. In Alaska, participants from the major cities were considered urban. Rural communities may be on the road system, or may be reached by boat or plane only, and are greater than 1 hour travel time to a major regional hospital.26 In Oregon, rural was defined as a geographic area that is at least 30 miles by road from an urban community.27
McNulty and Nail
concerns, life interferences, and worry. Additional subscales include employment concerns and relationship concerns. Individual items are worded to indicate level of agreement (strongly agree, agree, neutral, disagree, and strongly disagree). Strong evidence for validity has been found in CS populations.23,24 The Memorial Symptom Assessment Scale-Short Form (MSAS-SF) is multidimensional and measures 32 symptoms and their frequency and associated distress. The MSAS-SF has demonstrated validity and reliability in the cancer population.25 Participants rate each symptom from 0 to 5, with 0 indicating symptom not present and 5 indicating symptom caused very much distress. In this study, the physical symptom distress subscale was used. High and low scorers on the IOCv2 positive and negative impact scales were invited to participate in an in-depth interview, to capture a broad range of survivor experiences and impacts. A qualitative descriptive approach was utilized to explore the experiences of CSs and the impact of cancer from diagnosis to present. The interviews were conducted by the first author, either in person (3) or over the phone (16). The interview approach began with general questioning about survivorship experiences, and then it moved into more specific probing questions concerning the impact. Examples of interview questions included: (1) Tell me about your experience in getting cancer and becoming a survivor, (2) What is it like to be a cancer survivor? (3) How has the experience of having survived cancer affected you? (Probed about impact on work, leisure activities, social and psychological impact, spirituality, and sexuality), (4) What has helped you in your experiences as a cancer survivor? (5) What has not been helpful? (6) What has surprised you? and (7) How do you think living in a rural (or urban) area has impacted your experience as a survivor? Interviews lasted 40 to 90 minutes, and they were recorded and transcribed verbatim. The researcher conducted all of the transcription, which allowed for immersion in the data and assisted in generating emergent insights.
Data Analysis Measures and Data Collection Impact of Cancer Version 2 (IOCv2) The IOCv2 is designed to capture the unique aspects of cancer survivorship, including long-term aspects. It consists of 44–47 items, with 2 higher-order scales, a positive impact scale and a negative impact scale, each consisting of 4 subscales. Subscales of the positive impact scale include altruism/empathy, health awareness, meaning of cancer, and positive self-evaluation. Subscales of the negative impact scale include appearance concerns, body change
284
Quantitative SPSS software (IBM Corporation, Armonk, New York) was utilized to analyze the survey data. Descriptive statistics were obtained from the demographic data and instruments, and they were depicted according to the 4 geographical groups (Alaska urban, Alaska rural, Oregon urban, and Oregon rural) as well as the total sample. Means and standard deviations were calculated for each group. Reliability statistics on the total sample were conducted for the IOCv2. Univariate analysis of variance
c 2015 National Rural Health Association The Journal of Rural Health 31 (2015) 282–291
Cancer Survivorship, Rural and Urban
McNulty and Nail
(ANOVA) was conducted to compare the study outcome variables between the UCSs and RCSs and between the Alaska and Oregon CSs, controlling for age.
Qualitative Atlas.ti software (Scientific Software Development GmbH, Berlin, Germany) was used to assist in organizing and analyzing the interview data. Steps included line-by-line coding and content analysis, creating a codebook, sorting and synthesis of codes into categories and major themes, and identifying exemplar quotes to support themes and categories identified. The reflective process was facilitated through memoing. Methods to ensure rigor included additional researcher data analysis/inter-coder agreement, and member checking.28 A decision journal was maintained throughout the process. Thematic analysis techniques were utilized, and subcategories were combined into categories and overarching themes that depicted the survivorship experience. As the analysis progressed, particular attention was paid to the differences between region and locality, dichotomies and contrasts, and dualities emerging from the data. Exemplar quotes were selected to illustrate the rural and urban cancer survivorship experience and impact.
Findings Study Participants Total Sample The mean age of the total sample was 58, with an age range of 27 to 79, and included 84% women. The greatest proportion of the sample was employed full-time (42.7%), had private health insurance (68.2%), was married (71.2%), and had a bachelor’s degree or higher (46.2%). Ethnic representation of the sample included 5.4% Alaska Native/American Indian, 5.3% more than one race, and 88.6% white. The mean time since cancer treatment completion was 6.7 years (Table 1). Twelve cancer diagnoses were represented in the sample. The most common types of cancer were breast (56.1%), gynecologic (9.0%), lymphoma (8.3%), head and neck (6.1%), and colorectal (5.3%) (Table 1). Almost one-fifth (17%) of the sample had been diagnosed with cancer more than once. Cancer treatment modalities were surgery (80.3%), followed by chemotherapy (62.5%) and radiation (53.8%). Just under half of the participants reported having 1 to 2 comorbidities. The most common comorbidities were high blood pressure, depression, and arthritis. Over 40% reported that cancer
c 2015 National Rural Health Association The Journal of Rural Health 31 (2015) 282–291
Table 1 Sample Characteristics by Region/Location
Gender Female Male Age (mean) Time since rx completed (mean) Employment Full-time Part-time Not employed Missing Education High school Some college Associate’s Bachelor’s Graduate degree Missing Marital status Married Not marrieda Never marriedb Race/Ethnicity Alaska Native American Indian >One race White Missing Health insurance Medicaid Medicare IHS/Tribalc Private No coverage Cancer diagnosis Breast Gynecologic Lymphoma Head & neck Colorectal Melanoma Prostate Lung Hematologic Testicular Multiple myeloma Bladder
AK Urban n = 42%
AK Rural n = 26%
OR Urban n = 34%
OR Rural n = 30%
Total N = 132%
85.7 14.3 56.31 5.72
96.2 3.8 54.27 5.89
70.6 29.4 58.71 6.29
86.7 13.3 64.57 9.20
84.1 15.9 58.40 6.70
57.1 21.4 21.4
46.2 30.8 23.1
44.1 17.6 38.2
16.7 16.7 63.3 3.3
42.7 21.4 35.8 0.025
24.2 12.1
20 33.3
6.1 36.4 21.3
10.0 23.3 13.3
20 12.5 7.1 17.5 40.5
7.7 19.2 15.4 26.9 7.7
2.4
23.1
66.7 19.0 14.3
80.8 11.5 7.7
11.9 2.4 2.4 83.3
3.8
18.6 18.2 9.1 25.0 21.2 7.9
79.4 17.6 2.9
60.0 30.0 10.0
15.4 79.9 0.9
5.9 94.1
100.0
9.5 14.3 76.2
11.5
20.6
3.3 60
84.6 3.8
73.5 5.9
36.7
69 4.8 4.8 7.1 4.8 4.8 4.8
61.5 11.5 11.5 3.8 3.8 3.8
38.2 11.8 11.8 11.8 11.8 5.9 5.9
53.3 10.0 6.7 3.3
3.8
2.9
3.3 3.3 13.3 3.3
3.8 3.3
71.2 19.7 9.1 4.6 0.8 5.3 88.6 0.7 0.8 24.2 4.5 68.2 2.3 56.1 9.0 8.3 6.1 5.3 4.5 3.8 3 1.5 0.8 0.8 0.8
Rx, treatment; a Not married or partnered; b Never married or partnered; c Indian Health Service/Tribal health care beneficiary.
285
Cancer Survivorship, Rural and Urban
had had an impact on their activities of daily living. The largest proportion of the sample rated their health as “very good” (45.7%), their access to health care as “excellent” (47.3%), and their satisfaction with health care as “excellent” (46.9%). Means and standard deviations for the major variables from the total sample were IOC positive impact scale (M = 4.09, SD = 0.65), IOC negative impact scale (M = 2.67, SD = 0.82), employment concerns (M = 2.60, SD = 1.11), relationship concerns partnered (M = 2.71, SD = 0.74), relationship concerns nonpartnered (M = 2.72, SD = 1.3), and MSAS-SF physical symptom scale (M = 1.21, SD = 0.61). The Cronbach alphas on each variable for the total sample ranged from 0.62 to 0.92. Qualitative Subsample A total of 19 interview participants were evenly distributed across the 4 geographic groups. The demographics mirrored the total sample with a few slight differences being that they were slightly older (mean age 60), and mean time since treatment was less (4.2 years). The most represented cancer diagnoses included breast (53%), prostate (10%), colorectal (10%), and lymphoma (10%). Ethnicity of the subsample included 1 Alaska Native, 1 reporting more than one race, and 17 whites.
Comparisons Between Location and Region Overall comparisons between location (rural/urban) and region (Alaska/Oregon) univariate ANOVAs showed no statistically significant or meaningful differences between groups. A univariate 2 × 2 ANOVA was conducted to compare the study outcomes across the 4 geographic groups, controlling for age. No significant main effects were found. Significant interaction between region and location was found for body concerns [(F (1, 132) = 4.08, P < .05, partial n2 = 0.031], worry [F (1, 132) = 4.98, P < .05, partial n2 = 0.038], negative impact [F (1, 132) = 4.38, P < .05, partial n2 = 0.033], and employment concerns [F (1, 132) = 18.48, P < .05, partial n2 = 0.194]. Figure 1 shows the differences and patterns between the outcome variable means for each group. Follow-up post hoc tests were conducted to determine which differences were statistically significant and whether or not the hypotheses were supported (Table 2).
Themes Themes from the interview data that reflected rural/urban differences included travel/access, care coordination/navigation, connecting and community, thinking
286
McNulty and Nail
about death and dying, public/private journey, and advocating for self and health care services. Travel/Access Travel was an important aspect of the cancer story for the rural CSs. Participants described challenges such as driving long distances for treatment at considerable expense: “Something has to be fairly wrong to get a referral to cross the desert, the distance and the expense is a problem.” It was important for RCSs to be in their homes and communities as much as possible, and they were willing to travel long distances for treatment. UCSs did not have to worry as much about travel logistics and access to treatment, and they had more access to public transportation. One male UCS described access to health care as being “like cutting butter.” Care Coordination/Navigation Care coordination was a challenge, especially for the RCSs. One RCS described coordinating her chemotherapy treatment: “I arranged to get my chemo closer to home, so I could travel 1½ hours instead of 6 hours, and then I could be miserable in my own home.” Another CS learned about the navigator role through her support group: “I went to a navigator because I learned about that in the group. There is one person to guide you through it.” One Alaskan male UCS had a multitude of problems with care coordination, which included going to the “lower 48” for a second opinion, having delays with scheduling surgery (over a period of several weeks and having a very aggressive cancer type), having a doctor move away, and needing to make several changes in treatment regimens. In contrast, CSs from Oregon did not express issues with care coordination. Connecting and Community Survivors connected through friends, church, neighbors, and exercise programs. RCSs described a stronger sense of community and more informal connections. One RCS described his/her town as being like “a giant support group,” and another described her neighbors as “just showing up at her doorstep.” Both groups felt the need to connect with other survivors that they could relate to, who had been through similar experiences. UCSs did not describe their sense of community as strongly and they tended to utilize more formal methods of connecting, such as programs that connect CSs with similar diagnoses for support. According to this male CS: “When I found out I had it I got to talk to others who had been through the same thing, and now I share what I went through. It makes
c 2015 National Rural Health Association The Journal of Rural Health 31 (2015) 282–291
Cancer Survivorship, Rural and Urban
McNulty and Nail
Figure 1 Comparison of Means by Region and Location of Selected IOCv2 Outcome Variables.
Comparison of IOC Means by Region & LocaƟon AK Urban
2.97
2.95
3.12
AK Rural
OR Urban
OR Rural
3.24
3.23 2.89 2.83
3.01
2.82 2.61
2.98
2.78
2.4
2.3
2.15 1.93
Body Concerns P = .045
Worry P = .027
NegaƟve Impact P = .038
Employment Concerns P = .000
Note: Significant interaction effects when controlling for age; higher number indicates higher impact or more concerns. Table 2 Follow-Up Tests, Post Hoc Comparisons Comparison Body Concerns AK Urban higher than AK rural OR Urban higher than AK rural OR rural higher than AK rural Worry AK urban higher than OR urban Negative Impact AK urban higher than AK rural Employment concerns AK urban higher than AK rural AK urban higher than OR urban OR rural higher than AK rural OR rural higher than OR urban
Mean Difference
Significance
Hypothesis Supported?
0.658 0.630 0.894
0.025 0.041 0.008
No No No
0.461
0.048
Yes
0.413
0.037
No
1.150 0.940 1.055 0.850
0.000 0.002 0.005 0.021
No Yes No Yes
Abbreviations: AK, Alaska; OR, Oregon.
them feel good and it makes me feel good.” More experienced CSs described mentoring those that were newly diagnosed. Thinking About Death and Dying Thoughts of death and dying were common. One Alaskan RCS felt that dying in her rural town would feel more normal and comfortable: “I also felt like if I died, that it was just normal, and I would have people around me and I would feel good, even if I was gonna die. I didn’t feel that here.” One UCS felt that religion made death more scary and that she should be more focused on quality of life than death. Several CSs expressed that it would be more difficult to watch their spouse die than to die
c 2015 National Rural Health Association The Journal of Rural Health 31 (2015) 282–291
themselves. Another male CS joked, “None of us are getting out of here alive.” Public/Private Journey RCSs described a very public journey of survivorship due to the close-knit nature of their communities. They struggled with how much to share with others about their situation. A rural teacher described a very public journey, but also that she gained strength and inspiration from it, through her students and their parents: I was in school, and I had a very public journey because I was teaching, and so my third graders and
287
Cancer Survivorship, Rural and Urban
their parents were aware of what was going on, and of course it was very reinforcing to present a strong front because I had an audience. And so they were very inspiring for me to get through stuff. She described it as a double-edged sword, that the public journey could be both positive and negative, and the dichotomy of public versus private journey emerged. Even though she appreciated the support of her community, she longed for privacy at times: Sometimes I was exposed more than I wanted to be as far as being out in the community. I mean everyone knows I’m a cancer survivor . . . everyone knows what you’re going through, it’s a positive and they can lift you, but it’s a negative because you can’t escape it. One RCS felt that she was less of a survivor when she traveled to an urban setting where no one knew her versus being in her home community where “everyone knows.” Yet, one UCS also experienced a small community within her urban work setting: I’ve already had my name in the Daily News and my picture, when they did a thing on cancer survivors several years ago. I gave an interview, so it’s not anything that people don’t know . . . and it’s a small town . . . when you know each other from working here and there, it’s no secret. Advocating for Self and Health Care Services CSs advocated for themselves and others. “I’ve heard so many stories from so many people all over, that it just seems like the medical profession in general that you have to be your own advocate, no matter where you are.” The RCSs were especially committed to improving health care in their communities, such as advocating for lymphedema therapists in their town or providing encouragement for the local clinic to become more patientcentered: If you’re going to be a health care facility, then you think about your patients before you think about yourself. And I went to the town and I had a little meeting with people, and I said, “You think about your patients first and not yourselves, you have to either be in the business of caring for people medically, or get some jobs where you can be in politics.” This rural survivor of lung cancer attended a presentation delivered by a health care provider. She described
288
McNulty and Nail
confronting the presenter with what she viewed as a misunderstanding about her diagnosis: For breast cancer and for colon cancer, there’s so much more awareness for those diseases. And he said, “Unlike lung cancer where you just go ‘oh well.’” Afterwards, I thanked him for all the information he gave us, and I told him, I didn’t write my death sentence yet so you probably shouldn’t either, ‘cause I’m a lung cancer survivor, and he apologized.
Discussion Across the quantitative and qualitative findings there are more similarities than differences between the RCSs and UCSs (Table 3). The majority of the comparisons from the quantitative findings did not support the hypothesis that urban dwellers would fare better (Table 2). The Alaska rural group reveals the most positive outcomes on the IOCv2 negative impact variables (Figure 1). At least half of the differences that emerged from the qualitative findings were more positive or neutral for the RCSs. These findings contradict conventional wisdom about rural health and health care, such as increased isolation and decreased access. The convenience sample is highly educated, mostly employed and highly insured, and it is very engaged with survivorship. The participants had access to the Internet or a phone; therefore, the most isolated rural dwellers lacking this type of access were absent from the sample, and overall the sample lacked diversity. Findings could have been very different in a sample that was more representative of the general survivorship population, with the potential to show a higher negative impact. In this study, most of the RCSs reported that they were very happy with their rural lifestyles. There were differences in the definitions of what was considered rural in each region, which may have influenced the findings. The definitions were similar as they were based on distances to urban areas and major medical centers. However, the travel times may have been different. While the definitions were appropriate for each region, the differences make the comparisons challenging. Alaska is often excluded from these types of studies, as the typical rural/urban classifications do not apply due to the unique geography of the state.29 The mixed findings in this study, and the results of previous studies, suggest that factors other than place of residence and access to health care may be at play in accounting for the rural/urban differences. Such factors could include ethnicity, education, socioeconomic status, support system, type of cancer, comorbidities, and personality characteristics. Future studies could consider
c 2015 National Rural Health Association The Journal of Rural Health 31 (2015) 282–291
Cancer Survivorship, Rural and Urban
McNulty and Nail
Table 3 Comparison of Quantitative and Qualitative Findings Quantitative Findings Urban/rural differences found on study outcome variables Body concerns AK urban higher than AK rural OR urban higher than AK rural OR rural higher than AK rural Worry AK urban higher than OR urban Negative Impact AK urban higher than AK rural Employment AK urban higher than AK rural AK urban higher than OR rural OR rural higher than AK rural No statistically significant differences in self-reported health, satisfaction with health care, and access to health care Elements in the positive impact scale that revealed no differences Not measured, items related to CS identity Not measured Elements in the positive impact scale that revealed no differences
Qualitative Findings Urban/rural differences emerging from interview data No differences identified between rural/urban CSs
Worry manifested itself in the themes of recurrence and uncertainty, but no differences were identified between rural/urban CSs Interwoven throughout the CSs experiences but not predominant Work impact discussed, but mostly positively
Health care Access and care coordination (urban more positive, OR more positive) Community/connecting Rural more positive Public/private journey Neutral Thinking About Death and Dying Neutral Advocacy Rural more positive
AK, Alaska; OR, Oregon.
factors such as resilience in rural individuals and communities that may contribute to rural/urban differences.30 In addition to individual differences, differences in the characteristics of the communities and community life may contribute to the findings. The recruitment and sampling techniques for this study differ from other studies that have found significant differences in the quality of life between RCSs and UCSs. Sampling strategies from previous studies included analyzing data from national surveys8,10 and recruiting from cancer centers and health departments.9 No studies focused on cancer survivorship were found that examined specific characteristics of rural communities or aspects of community life. These factors are important to consider in future studies. Given the sample characteristics, the findings with regard to health care utilization and coordination are not surprising. The CSs in this study, rural and urban alike, demonstrated that they were well informed and utilizing resources available to them. This is in contrast with other studies that have shown that RCSs are less likely to be aware of services and utilize less mental health resources.9 In this study, CSs from Oregon did not express concerns about lack of care coordination or navigation; however, it was a concern for the Alaskans. Models for navigation are beginning to emerge in
c 2015 National Rural Health Association The Journal of Rural Health 31 (2015) 282–291
survivorship care. However, there is conflicting evidence for their effectiveness,31 and there is limited focus on rural populations. Although there were minimal differences found, this study reveals themes that have not been previously described in the literature. RCSs described connections to their communities and networks very positively. They were more informal, occurred more naturally, and they relied heavily on them. RCSs talked about the importance of their neighbors, and in contrast, UCSs relied on formal social ties that were more planned, such as work networks and joining formal organizations. Both RCSs and UCSs relied heavily on their church communities, and being active in a religious community has been shown to be beneficial for CSs.32 Other studies have described a strong sense of community and dense social networks in rural settings, but not within the context of cancer survivorship.33 The notion of public/private journey is also new to this body of literature. The RCSs in this study talked about their identity as CSs more strongly than their urban counterparts. They described the dichotomy between others knowing and not knowing about their cancer journey. The notion of lack of anonymity in small towns has been well described,34 and the wish to maintain privacy may
289
Cancer Survivorship, Rural and Urban
not be a choice for RCSs. The process of integrating a new identity as a CS has been previously described,35,36 but once again not within the context of a rural setting. The strength in advocacy came through more strongly for the RCSs. They advocate for themselves, their diagnosis, survivorship, and for improved health care in their communities. The advocacy emerged as seeking a second opinion, accessing support resources, asking questions and seeking answers, and fighting for their financial and employment rights. The RCSs in particular engage in community advocacy by fundraising, volunteering with survivorship organizations, and speaking publicly about survivorship issues. Overall, there was a sense of strong empowerment from this highly engaged group of CSs. The fact that the majority of the sample was female, and breast CSs, could have heavily influenced the findings, since there are typically more resources available for breast CSs than for survivors of other malignancies. The sample size for each group was small, and the sample was very heterogeneous with regard to cancer diagnosis, type of treatment, and time since treatment. These factors could have influenced the ability to detect differences between the RCSs and the UCSs, and made it difficult to make cancer site-specific recommendations. The methodological limitations of this study include the following: (1) there were no comparisons between the CSs and the general population; therefore, it cannot be confirmed that the findings are unique to CSs; (2) quantitative data were collected by self-report and self-administered (not during an interview); therefore, the participants may not have been answering within the context of their cancer survivorship experience; (3) medical diagnosis and treatment profiles were not confirmed; (4) nonverbal cues were lost in the phone interview data collection process; and (5) the cross-sectional design did not capture changes over time. Future studies could benefit from conducting community-based participatory research (CBPR) studies as an effective strategy to engage CS populations. The CBPR approach could capitalize on the high level of engagement with survivorship and lead to successful engagement of CSs with future research, especially in their communities. This study provides rural/urban comparisons across multiple states and includes CSs in some of the most remote locations of the United States. The sample includes CSs who are under-represented in the current body of research, including rural dwellers, Alaska Natives/American Indians, men, and survivors of a variety of cancer diagnoses. It is one of the few studies to explore the potential positive aspects of rural living within the context of cancer survivorship. The study provides evidence for the effectiveness of the IOCv2 instrument in detecting differences in a diverse population, for the
290
McNulty and Nail
constructs associated with negative impact. Overall, the study has demonstrated that mixed methods is a valuable approach to studying cancer survivorship. The results of this study challenge prevalent assumptions about RCSs and their risk for negative outcomes. Findings provide insight into the differences and similarities between rural and urban CSs. A profile of rural survivorship emerges, with a common thread being community. Access to health care may not be the driver of the survivorship experience. There is an opportunity to influence health care providers and developing programs serving this growing population with complex issues, building on the strengths of both rural and urban living, and the engagement of the cancer survivor community.
References 1. American Cancer Society. Cancer treatment and survivorship facts and figures 2012–2013; 2012. Available at: http://www.cancer.org/acs/groups/content/@ epidemiologysurveilance/documents/document/acspc033876.pdf. Accessed Febuary 1, 2013. 2. Yabroff K, Lawrence W, Clauser S, et al. Burden of illness in cancer survivors: findings from a population-based national sample. J Natl Cancer Inst. 2004;96(17):13221330. 3. Feuerstein M. Defining cancer survivorship. J Cancer Surv. 2007;1:5-7. 4. Aziz N, Rowland J. Cancer survivorship research among ethnic minority and medically underserved groups. Oncol Nurs Forum. 2002;29(5):789-801. 5. Ganz P. Monitoring the physical health of cancer survivors. J Clin Oncol. 2006;24:5105-5111. 6. Angell K, Kreshka M, McCoy R, et al. Psychosocial intervention for rural women with breast cancer: the Sierra Stanford Partnership. J Gen Intern Med. 2003;18:499-507. 7. Hewitt M, Greenfield S, Stovall E. From Cancer Patient to Cancer Survivor: Lost in Transition. Washington, DC: National Academies; 2006. 8. Burris J, Andrykowski M. Disparities in mental health between rural and nonrural cancer survivors: a preliminary study. Psycho-Oncology. 2010;19(6):637-645. 9. DeSipio T, Hayes S, Newman B, et al. Does quality of life among breast cancer survivors one year after diagnosis differ depending on urban and non-urban residence? A comparative study. Health Qual Life Outcomes. 2012;8(3):1-10. 10. Lyons M, Shelton M. Psychosocial impact of cancer in low-income rural/urban women: phase I. Online J Rural Nurs Health Care. 2004;4(1):10-21. 11. Weaver K, Geiger A, Lu L, et al. Rural-urban disparities in health status among US cancer survivors. Cancer. 2013;119(5):1050-1057.
c 2015 National Rural Health Association The Journal of Rural Health 31 (2015) 282–291
McNulty and Nail
12. US Census Bureau. Current Population Reports, P60-229, Income, Poverty, and Health Insurance Coverage in the United States: 2004. 2005. Available at: www.census.gov/ prod/2005pubs/p60-229.pdf. Accessed December 10, 2014. 13. National Network of Libraries of Medicine. State information for Alaska. 2007. Available at: http://nnlm.gov/pnr/characteristics/alaska.html. Accessed December 10, 2014. 14. City-Data. State of Alaska. 2007. Available at: http://www.city-data.com/states/Alaska.html. Accessed December 10, 2014. 15. Alaska Experience Project. Health care in the last frontier: rural challenges/Alaska innovations. 2007. Available at: http://www.agnewbeck.com/pdf/statewide/AEP/AEP_ Healthcare-in-the-Last-Frontier.pdf. Accessed December 10, 2014. 16. GAP Report. Mental Health in Remote Rural Developing Areas: Concepts and Cases. Washington, DC: American Psychiatric Press; 1995. 17. Oregon State University. Defining rural Oregon: an exploration. 2007. Available at: http://extension. oregonstate.edu/catalog/pdf/em/em8937-e.pdf. Accessed December 10, 2014. 18. American College of Surgeons. Cancer programs 2012. Available at: http://datalinks.facs.org/cpm/CPMApproved Hospitals_Result.cfm. Accessed November 1, 2012. 19. Oregon Society of Medical Oncologists. Members. 2012. Available at: www.osmo.org/members. Accessed November 1, 2012. 20. Palmer N, Geiger A, Lu L, et al. Impact of rural residence in forgoing healthcare after cancer because of costs. Cancer Epidemiol Biomarkers Prev. 2013;22(10):1668-1676. 21. Aziz N, Rowland J. Trends and advances in cancer survivorship research: challenge and opportunity. Sem Radiat Oncol. 2003;13(3):248-266. 22. Zabalegui A, Sanchez S, Sanchez P, et al. Nursing and cancer support groups. J Adv Nurs. 2005;51(4):369-381. 23. Crespi C, Ganz P, Petersen L, et al. Refinement of the impact of cancer scale. J Natl Cancer Inst. 2008;100:15301541.
c 2015 National Rural Health Association The Journal of Rural Health 31 (2015) 282–291
Cancer Survivorship, Rural and Urban
24. Crespi C, Smith S, Petersen L, et al. Measuring the impact of cancer: a comparison of non-Hodgkin lymphoma and breast cancer survivors. J Cancer Surviv. 2010;4: 45-58. 25. Chang V, Hwang S, Feuerman M, et al. The memorial symptom scale short form (MSAS-SF): validity and reliability. Cancer. 2000;89(5):1162-1171. 26. Leipert B. Rural and remote women developing resilience to manage vulnerability. In: Lee H, Winter C, eds. Rural Nursing: Concepts, Theory and Practice. 2nd ed. New York, NY: Springer; 2006:79-95. 27. Crandall M, Weber B. Oregon State University, Rural Studies Program. Defining rural Oregon: An exploration. 2005. Available at: http://ruralstudies.oregonstate.edu/ sites/default/files/pub/pdf/rsp reports/rsp05-03.pdf. Accessed November 1, 2012. 28. Creswell J, Plano-Clark V. Designing and Conducting Mixed Methods Research. 2nd ed. Thousand Oaks, CA: Sage; 2010. 29. Ormond B, Zuckerman S, Lhila A. Rural/urban differences in healthcare are not uniform across states. The Urban Institute. 2000; Series B, No B-11. 30. Buikstra D, Ross H, King C, et al. The components of resilience-perceptions of an Australian rural community. J Community Psychol. 2010;38(8):975-991. 31. Wells K, Battaglia T, Dudley D, et al. Patient navigation; State of the art or is it science? Cancer. 2008;113(8):19992010. 32. Vachon M. Psychosocial distress and coping after cancer treatment. Am J Nurs. 2006;106:26S-31S. 33. Johnson I. Patient perspective on health, health needs, and health care services in a rural Irish community: a qualitative study. Rural Remote Health. 2011;11: 1659. 34. Long K, Weinert C. Developing the theory base. In: Lee H, Winter C, eds. Rural Nursing: Concepts, Theory and Practice. 2nd ed. New York, NY: Springer; 2006:3-16. 35. Zebrack B. Cancer survivor identity and quality of life. Cancer Pract. 2000;8(5):238-242. 36. Breaden K. Cancer and beyond: the question of survivorship. J Adv Nurs. 1997;26:978-984.
291
Cancer Survivorship in Rural and Urban Adults: A Descriptive and Mixed Methods Study Julie A. McNulty, RN, PhD1 & Lillian Nail, PhD, RN, FAAN2 1 Arizona State University, College of Nursing and Health Innovation, Phoenix, Arizona 2 Oregon Health and Science University, School of Nursing, Portland, Oregon
Abstract Purpose: Rural-dwelling cancer survivors (CSs) are at risk for decrements in Conflict of Interest: The authors have no conflicts of interest to disclose. Funding: This work was supported by the Oregon Health and Science University Dean’s Scholarship fund. Acknowledgments: Special thanks to the cancer survivors who shared their time and experiences. The first author would like to thank Lillian Nail, PhD, RN, FAAN, who served as the chair of her dissertation committee and primary mentor, as well as committee members Rebecca Block, MSW, PhD; Brandon Hayes-Lattin, MD; and Frances Lee-Lin, PhD, CNS. For further information, contact: Julie A. McNulty, RN, PhD, 500 N 3rd St., Phoenix, AZ 85020; e-mail: [email protected]. doi: 10.1111/jrh.12106
health and well-being due to decreased access to health care and support resources. This study compares the impact of cancer in rural- and urban-dwelling adult CSs living in 2 regions of the Pacific Northwest. Methods: A convenience sample of posttreatment adult CSs (N = 132) completed the Impact of Cancer version 2 (IOCv2) and the Memorial Symptom Assessment Scale-short form. High and low scorers on the IOCv2 participated in an in-depth interview (n = 19). Findings: The sample was predominantly middle-aged (mean age 58) and female (84%). Mean time since treatment completion was 6.7 years. Cancer diagnoses represented included breast (56%), gynecologic (9%), lymphoma (8%), head and neck (6%), and colorectal (5%). Comparisons across geographic regions show statistically significant differences in body concerns, worry, negative impact, and employment concerns. Rural-urban differences from interview data include access to health care, care coordination, connecting/community, thinking about death and dying, public/private journey, and advocacy. Conclusion: The insights into the differences and similarities between rural and urban CSs challenge the prevalent assumptions about rural-dwelling CSs and their risk for negative outcomes. A common theme across the study findings was community. Access to health care may not be the driver of the survivorship experience. Findings can influence health care providers and survivorship program development, building on the strengths of both rural and urban living and the engagement of the survivorship community.
Key words cancer survivorship, impact of cancer, late effects, rural health, rural-urban cancer survivorship.
Cancer survivors (CSs) are living longer, growing in number, and are now being recognized as being at risk for decrements in health and well-being due to cancer and cancer treatment. Increases are due to an aging population, improvements in screening and treatment, and decreased mortality from other causes.1 Effects of treatment may persist many years after diagnosis, and new treatment-related problems can appear years after
282
treatment is completed.2 Challenges that confront CSs are not simply a continuation of experiences that occur during the cancer treatment phase; they are unique problems in the cancer illness trajectory, as the person treated for cancer transitions from the acute to the chronic phases of the illness.3 Persistent and late effects include physical and psychological symptoms, and social and spiritual impact. CSs are likely to face comorbid
c 2015 National Rural Health Association The Journal of Rural Health 31 (2015) 282–291
Cancer Survivorship, Rural and Urban
McNulty and Nail
illness, functional limitations, difficulties with returning to work and other role transitions, uncertainty, fear of recurrence, and barriers to quality health care, all of which complicate their lives.1,4,5
Rural Survivorship Out of the estimated 13.7 million CSs in the United States currently, 2.8 million are living in rural settings. Rural cancer survivors (RCSs) are viewed as having higher levels of risk for decrements in health and well-being due to decreased access to health care, specialty services, and support resources.6 RCSs may experience lack of symptom management, lack of access to psychosocial interventions, isolation, and decreased coping with regard to their cancer experience.7 Mental health disparities have been identified with RCSs showing poorer mental health outcomes than urban cancer survivors (UCSs).8 Studies are lacking that compare quality of life between the 2 groups. Recent studies are yielding inconsistent results or mixed findings,9 are revealing inferior outcomes in RCSs,9-11 or no differences.8 This study focuses on posttreatment CSs living in Alaska and Oregon, 2 states with high levels of rurality. Alaska has a higher percentage of rural population than the rest of the United States,12 with 52.3% of the population living in frontier areas.13 Alaska is over twice the size of Texas14 and encompasses 20% of the land mass of the United States.15 Furthermore, Alaskans face geographical challenges not present in any other state, such as limited road access into remote areas and harsh weather patterns. Travel by air is required to reach the remote areas, which make up the greatest portion of the state. Alaskans experience unique needs regarding cancer survivorship. Many health care services that would be considered standard care in other states are unavailable in rural and remote Alaska.16 Access to oncologists is limited to the urban areas of Anchorage and Fairbanks. Cancer treatment is apt to be provided far from home, often in another state, such as Washington, Oregon, or Minnesota. Similarly, the rural population of Oregon comprises 29.1% of the state population, and 86.3% of the land is defined as rural.17 People living in the eastern part of the state may be over 100 miles from cancer treatment facilities, and access to these facilities may require traveling over mountain passes. There are limited cancer treatment programs and survivorship resources in this sparsely populated area,18 and there are no practicing oncologists.19 In contrast to the cities of Anchorage and Fairbanks in Alaska, the city of Portland offers highly specialized cancer treatment centers, and the state offers more sophisticated cancer survivorship programs. In both states, travel and distance challenges create a significant disruption in
c 2015 National Rural Health Association The Journal of Rural Health 31 (2015) 282–291
the lives of rural cancer patients and their families, and their out-of-pocket costs are likely to be significant.20
Gaps in Knowledge Most health-related cancer research has focused on the acute and treatment stages of the illness, and there is a strong suggestion in the literature that more research is needed during the posttreatment stages of survivorship. Until recently this phase of the cancer trajectory has been relatively neglected in advocacy, education, clinical practice, and research.7 CSs are considered an understudied population, who may be lost in transition once cancer treatment is completed. Issues encountered by RCSs, and the long-term psychosocial effects of cancer in rural settings, are areas of needed research.21 It is currently unclear which types of psychosocial and educational interventions are most effective for RCSs. Although there is strong evidence that traditional cancer support groups are effective for patients during the acute phase of cancer treatment,22 less is known about how to effectively meet the needs of long-term RCSs. The present study is the first of its kind in these 2 states. The purpose of this descriptive and mixed methods design study is to explore and compare the impact of cancer in rural- and urban-dwelling adult CSs in 2 regions of the Pacific Northwest: Alaska and Oregon. The impact of cancer was measured by administering the Impact of Cancer version 2 (IOCv2) instrument,23,24 as well as a symptom measure.25 The impact of cancer and the survivorship experience was also explored qualitatively through in-depth interviews. In this paper, the term region will be used when referring to the state where the participant lives, and the term location will refer to whether the participant was residing in a rural or urban setting. Aim #1: To explore and compare the impact of the cancer survivorship and symptom experience in RCSs and UCSs in 2 regions of the Pacific Northwest (quantitative). We proposed that UCSs would have more positive scores on study instruments than would RCSs, that Oregonians would have more positive scores than Alaskans, and that there would be an interaction between region and location such that urban-dwelling Oregonians would have more positive scores than urban Alaskans, and rural Oregonians would have more positive scores than rural Alaskans. The exploratory hypotheses were based on the differences in access to health care between the regions and locations. Aim #2: To explore and compare the physical (including symptoms), psychological, social, and spiritual aspects of cancer survivorship in RCSs and UCSs within and between 2 regions of the Pacific Northwest (qualitative).
283
Cancer Survivorship, Rural and Urban
Methods Procedures Ethical approval for this study was obtained from the Knight Cancer Institute and the Oregon Health and Science University and Alaska Area investigational review boards. Participants were recruited by distributing e-mails and flyers to cancer survivorship programs, support group leaders, oncology and primary care providers, and through networking. Once the participants contacted the researcher, they were screened for eligibility and informed consent was obtained. Participants were given a choice of method for survey administration. If their preference was online, a link to the study questionnaires was sent via e-mail. If the choice was on paper, the instruments were mailed to the participants. In the final sample, 88% completed the surveys via the web and 12% on paper. Because of the variety of recruitment strategies utilized, it is unknown how many people were actually recruited into the study and declined to participate. Therefore, it is not feasible to calculate an overall response rate. Once potential participants contacted the researcher about the study, 70% of this group completed the study questionnaires. The convenience sample (N = 132) included adult survivors of all cancer types (excluding nonmelanotic skin cancer) who had completed active cancer treatment and were self-reported as cancer-free. Participants were able to read, speak, and write English.
Rural and Urban Definitions Due to the stark differences in geography, travel distances and types of transportation used, a rural definition unique to each state was utilized. In Alaska, participants from the major cities were considered urban. Rural communities may be on the road system, or may be reached by boat or plane only, and are greater than 1 hour travel time to a major regional hospital.26 In Oregon, rural was defined as a geographic area that is at least 30 miles by road from an urban community.27
McNulty and Nail
concerns, life interferences, and worry. Additional subscales include employment concerns and relationship concerns. Individual items are worded to indicate level of agreement (strongly agree, agree, neutral, disagree, and strongly disagree). Strong evidence for validity has been found in CS populations.23,24 The Memorial Symptom Assessment Scale-Short Form (MSAS-SF) is multidimensional and measures 32 symptoms and their frequency and associated distress. The MSAS-SF has demonstrated validity and reliability in the cancer population.25 Participants rate each symptom from 0 to 5, with 0 indicating symptom not present and 5 indicating symptom caused very much distress. In this study, the physical symptom distress subscale was used. High and low scorers on the IOCv2 positive and negative impact scales were invited to participate in an in-depth interview, to capture a broad range of survivor experiences and impacts. A qualitative descriptive approach was utilized to explore the experiences of CSs and the impact of cancer from diagnosis to present. The interviews were conducted by the first author, either in person (3) or over the phone (16). The interview approach began with general questioning about survivorship experiences, and then it moved into more specific probing questions concerning the impact. Examples of interview questions included: (1) Tell me about your experience in getting cancer and becoming a survivor, (2) What is it like to be a cancer survivor? (3) How has the experience of having survived cancer affected you? (Probed about impact on work, leisure activities, social and psychological impact, spirituality, and sexuality), (4) What has helped you in your experiences as a cancer survivor? (5) What has not been helpful? (6) What has surprised you? and (7) How do you think living in a rural (or urban) area has impacted your experience as a survivor? Interviews lasted 40 to 90 minutes, and they were recorded and transcribed verbatim. The researcher conducted all of the transcription, which allowed for immersion in the data and assisted in generating emergent insights.
Data Analysis Measures and Data Collection Impact of Cancer Version 2 (IOCv2) The IOCv2 is designed to capture the unique aspects of cancer survivorship, including long-term aspects. It consists of 44–47 items, with 2 higher-order scales, a positive impact scale and a negative impact scale, each consisting of 4 subscales. Subscales of the positive impact scale include altruism/empathy, health awareness, meaning of cancer, and positive self-evaluation. Subscales of the negative impact scale include appearance concerns, body change
284
Quantitative SPSS software (IBM Corporation, Armonk, New York) was utilized to analyze the survey data. Descriptive statistics were obtained from the demographic data and instruments, and they were depicted according to the 4 geographical groups (Alaska urban, Alaska rural, Oregon urban, and Oregon rural) as well as the total sample. Means and standard deviations were calculated for each group. Reliability statistics on the total sample were conducted for the IOCv2. Univariate analysis of variance
c 2015 National Rural Health Association The Journal of Rural Health 31 (2015) 282–291
Cancer Survivorship, Rural and Urban
McNulty and Nail
(ANOVA) was conducted to compare the study outcome variables between the UCSs and RCSs and between the Alaska and Oregon CSs, controlling for age.
Qualitative Atlas.ti software (Scientific Software Development GmbH, Berlin, Germany) was used to assist in organizing and analyzing the interview data. Steps included line-by-line coding and content analysis, creating a codebook, sorting and synthesis of codes into categories and major themes, and identifying exemplar quotes to support themes and categories identified. The reflective process was facilitated through memoing. Methods to ensure rigor included additional researcher data analysis/inter-coder agreement, and member checking.28 A decision journal was maintained throughout the process. Thematic analysis techniques were utilized, and subcategories were combined into categories and overarching themes that depicted the survivorship experience. As the analysis progressed, particular attention was paid to the differences between region and locality, dichotomies and contrasts, and dualities emerging from the data. Exemplar quotes were selected to illustrate the rural and urban cancer survivorship experience and impact.
Findings Study Participants Total Sample The mean age of the total sample was 58, with an age range of 27 to 79, and included 84% women. The greatest proportion of the sample was employed full-time (42.7%), had private health insurance (68.2%), was married (71.2%), and had a bachelor’s degree or higher (46.2%). Ethnic representation of the sample included 5.4% Alaska Native/American Indian, 5.3% more than one race, and 88.6% white. The mean time since cancer treatment completion was 6.7 years (Table 1). Twelve cancer diagnoses were represented in the sample. The most common types of cancer were breast (56.1%), gynecologic (9.0%), lymphoma (8.3%), head and neck (6.1%), and colorectal (5.3%) (Table 1). Almost one-fifth (17%) of the sample had been diagnosed with cancer more than once. Cancer treatment modalities were surgery (80.3%), followed by chemotherapy (62.5%) and radiation (53.8%). Just under half of the participants reported having 1 to 2 comorbidities. The most common comorbidities were high blood pressure, depression, and arthritis. Over 40% reported that cancer
c 2015 National Rural Health Association The Journal of Rural Health 31 (2015) 282–291
Table 1 Sample Characteristics by Region/Location
Gender Female Male Age (mean) Time since rx completed (mean) Employment Full-time Part-time Not employed Missing Education High school Some college Associate’s Bachelor’s Graduate degree Missing Marital status Married Not marrieda Never marriedb Race/Ethnicity Alaska Native American Indian >One race White Missing Health insurance Medicaid Medicare IHS/Tribalc Private No coverage Cancer diagnosis Breast Gynecologic Lymphoma Head & neck Colorectal Melanoma Prostate Lung Hematologic Testicular Multiple myeloma Bladder
AK Urban n = 42%
AK Rural n = 26%
OR Urban n = 34%
OR Rural n = 30%
Total N = 132%
85.7 14.3 56.31 5.72
96.2 3.8 54.27 5.89
70.6 29.4 58.71 6.29
86.7 13.3 64.57 9.20
84.1 15.9 58.40 6.70
57.1 21.4 21.4
46.2 30.8 23.1
44.1 17.6 38.2
16.7 16.7 63.3 3.3
42.7 21.4 35.8 0.025
24.2 12.1
20 33.3
6.1 36.4 21.3
10.0 23.3 13.3
20 12.5 7.1 17.5 40.5
7.7 19.2 15.4 26.9 7.7
2.4
23.1
66.7 19.0 14.3
80.8 11.5 7.7
11.9 2.4 2.4 83.3
3.8
18.6 18.2 9.1 25.0 21.2 7.9
79.4 17.6 2.9
60.0 30.0 10.0
15.4 79.9 0.9
5.9 94.1
100.0
9.5 14.3 76.2
11.5
20.6
3.3 60
84.6 3.8
73.5 5.9
36.7
69 4.8 4.8 7.1 4.8 4.8 4.8
61.5 11.5 11.5 3.8 3.8 3.8
38.2 11.8 11.8 11.8 11.8 5.9 5.9
53.3 10.0 6.7 3.3
3.8
2.9
3.3 3.3 13.3 3.3
3.8 3.3
71.2 19.7 9.1 4.6 0.8 5.3 88.6 0.7 0.8 24.2 4.5 68.2 2.3 56.1 9.0 8.3 6.1 5.3 4.5 3.8 3 1.5 0.8 0.8 0.8
Rx, treatment; a Not married or partnered; b Never married or partnered; c Indian Health Service/Tribal health care beneficiary.
285
Cancer Survivorship, Rural and Urban
had had an impact on their activities of daily living. The largest proportion of the sample rated their health as “very good” (45.7%), their access to health care as “excellent” (47.3%), and their satisfaction with health care as “excellent” (46.9%). Means and standard deviations for the major variables from the total sample were IOC positive impact scale (M = 4.09, SD = 0.65), IOC negative impact scale (M = 2.67, SD = 0.82), employment concerns (M = 2.60, SD = 1.11), relationship concerns partnered (M = 2.71, SD = 0.74), relationship concerns nonpartnered (M = 2.72, SD = 1.3), and MSAS-SF physical symptom scale (M = 1.21, SD = 0.61). The Cronbach alphas on each variable for the total sample ranged from 0.62 to 0.92. Qualitative Subsample A total of 19 interview participants were evenly distributed across the 4 geographic groups. The demographics mirrored the total sample with a few slight differences being that they were slightly older (mean age 60), and mean time since treatment was less (4.2 years). The most represented cancer diagnoses included breast (53%), prostate (10%), colorectal (10%), and lymphoma (10%). Ethnicity of the subsample included 1 Alaska Native, 1 reporting more than one race, and 17 whites.
Comparisons Between Location and Region Overall comparisons between location (rural/urban) and region (Alaska/Oregon) univariate ANOVAs showed no statistically significant or meaningful differences between groups. A univariate 2 × 2 ANOVA was conducted to compare the study outcomes across the 4 geographic groups, controlling for age. No significant main effects were found. Significant interaction between region and location was found for body concerns [(F (1, 132) = 4.08, P < .05, partial n2 = 0.031], worry [F (1, 132) = 4.98, P < .05, partial n2 = 0.038], negative impact [F (1, 132) = 4.38, P < .05, partial n2 = 0.033], and employment concerns [F (1, 132) = 18.48, P < .05, partial n2 = 0.194]. Figure 1 shows the differences and patterns between the outcome variable means for each group. Follow-up post hoc tests were conducted to determine which differences were statistically significant and whether or not the hypotheses were supported (Table 2).
Themes Themes from the interview data that reflected rural/urban differences included travel/access, care coordination/navigation, connecting and community, thinking
286
McNulty and Nail
about death and dying, public/private journey, and advocating for self and health care services. Travel/Access Travel was an important aspect of the cancer story for the rural CSs. Participants described challenges such as driving long distances for treatment at considerable expense: “Something has to be fairly wrong to get a referral to cross the desert, the distance and the expense is a problem.” It was important for RCSs to be in their homes and communities as much as possible, and they were willing to travel long distances for treatment. UCSs did not have to worry as much about travel logistics and access to treatment, and they had more access to public transportation. One male UCS described access to health care as being “like cutting butter.” Care Coordination/Navigation Care coordination was a challenge, especially for the RCSs. One RCS described coordinating her chemotherapy treatment: “I arranged to get my chemo closer to home, so I could travel 1½ hours instead of 6 hours, and then I could be miserable in my own home.” Another CS learned about the navigator role through her support group: “I went to a navigator because I learned about that in the group. There is one person to guide you through it.” One Alaskan male UCS had a multitude of problems with care coordination, which included going to the “lower 48” for a second opinion, having delays with scheduling surgery (over a period of several weeks and having a very aggressive cancer type), having a doctor move away, and needing to make several changes in treatment regimens. In contrast, CSs from Oregon did not express issues with care coordination. Connecting and Community Survivors connected through friends, church, neighbors, and exercise programs. RCSs described a stronger sense of community and more informal connections. One RCS described his/her town as being like “a giant support group,” and another described her neighbors as “just showing up at her doorstep.” Both groups felt the need to connect with other survivors that they could relate to, who had been through similar experiences. UCSs did not describe their sense of community as strongly and they tended to utilize more formal methods of connecting, such as programs that connect CSs with similar diagnoses for support. According to this male CS: “When I found out I had it I got to talk to others who had been through the same thing, and now I share what I went through. It makes
c 2015 National Rural Health Association The Journal of Rural Health 31 (2015) 282–291
Cancer Survivorship, Rural and Urban
McNulty and Nail
Figure 1 Comparison of Means by Region and Location of Selected IOCv2 Outcome Variables.
Comparison of IOC Means by Region & LocaƟon AK Urban
2.97
2.95
3.12
AK Rural
OR Urban
OR Rural
3.24
3.23 2.89 2.83
3.01
2.82 2.61
2.98
2.78
2.4
2.3
2.15 1.93
Body Concerns P = .045
Worry P = .027
NegaƟve Impact P = .038
Employment Concerns P = .000
Note: Significant interaction effects when controlling for age; higher number indicates higher impact or more concerns. Table 2 Follow-Up Tests, Post Hoc Comparisons Comparison Body Concerns AK Urban higher than AK rural OR Urban higher than AK rural OR rural higher than AK rural Worry AK urban higher than OR urban Negative Impact AK urban higher than AK rural Employment concerns AK urban higher than AK rural AK urban higher than OR urban OR rural higher than AK rural OR rural higher than OR urban
Mean Difference
Significance
Hypothesis Supported?
0.658 0.630 0.894
0.025 0.041 0.008
No No No
0.461
0.048
Yes
0.413
0.037
No
1.150 0.940 1.055 0.850
0.000 0.002 0.005 0.021
No Yes No Yes
Abbreviations: AK, Alaska; OR, Oregon.
them feel good and it makes me feel good.” More experienced CSs described mentoring those that were newly diagnosed. Thinking About Death and Dying Thoughts of death and dying were common. One Alaskan RCS felt that dying in her rural town would feel more normal and comfortable: “I also felt like if I died, that it was just normal, and I would have people around me and I would feel good, even if I was gonna die. I didn’t feel that here.” One UCS felt that religion made death more scary and that she should be more focused on quality of life than death. Several CSs expressed that it would be more difficult to watch their spouse die than to die
c 2015 National Rural Health Association The Journal of Rural Health 31 (2015) 282–291
themselves. Another male CS joked, “None of us are getting out of here alive.” Public/Private Journey RCSs described a very public journey of survivorship due to the close-knit nature of their communities. They struggled with how much to share with others about their situation. A rural teacher described a very public journey, but also that she gained strength and inspiration from it, through her students and their parents: I was in school, and I had a very public journey because I was teaching, and so my third graders and
287
Cancer Survivorship, Rural and Urban
their parents were aware of what was going on, and of course it was very reinforcing to present a strong front because I had an audience. And so they were very inspiring for me to get through stuff. She described it as a double-edged sword, that the public journey could be both positive and negative, and the dichotomy of public versus private journey emerged. Even though she appreciated the support of her community, she longed for privacy at times: Sometimes I was exposed more than I wanted to be as far as being out in the community. I mean everyone knows I’m a cancer survivor . . . everyone knows what you’re going through, it’s a positive and they can lift you, but it’s a negative because you can’t escape it. One RCS felt that she was less of a survivor when she traveled to an urban setting where no one knew her versus being in her home community where “everyone knows.” Yet, one UCS also experienced a small community within her urban work setting: I’ve already had my name in the Daily News and my picture, when they did a thing on cancer survivors several years ago. I gave an interview, so it’s not anything that people don’t know . . . and it’s a small town . . . when you know each other from working here and there, it’s no secret. Advocating for Self and Health Care Services CSs advocated for themselves and others. “I’ve heard so many stories from so many people all over, that it just seems like the medical profession in general that you have to be your own advocate, no matter where you are.” The RCSs were especially committed to improving health care in their communities, such as advocating for lymphedema therapists in their town or providing encouragement for the local clinic to become more patientcentered: If you’re going to be a health care facility, then you think about your patients before you think about yourself. And I went to the town and I had a little meeting with people, and I said, “You think about your patients first and not yourselves, you have to either be in the business of caring for people medically, or get some jobs where you can be in politics.” This rural survivor of lung cancer attended a presentation delivered by a health care provider. She described
288
McNulty and Nail
confronting the presenter with what she viewed as a misunderstanding about her diagnosis: For breast cancer and for colon cancer, there’s so much more awareness for those diseases. And he said, “Unlike lung cancer where you just go ‘oh well.’” Afterwards, I thanked him for all the information he gave us, and I told him, I didn’t write my death sentence yet so you probably shouldn’t either, ‘cause I’m a lung cancer survivor, and he apologized.
Discussion Across the quantitative and qualitative findings there are more similarities than differences between the RCSs and UCSs (Table 3). The majority of the comparisons from the quantitative findings did not support the hypothesis that urban dwellers would fare better (Table 2). The Alaska rural group reveals the most positive outcomes on the IOCv2 negative impact variables (Figure 1). At least half of the differences that emerged from the qualitative findings were more positive or neutral for the RCSs. These findings contradict conventional wisdom about rural health and health care, such as increased isolation and decreased access. The convenience sample is highly educated, mostly employed and highly insured, and it is very engaged with survivorship. The participants had access to the Internet or a phone; therefore, the most isolated rural dwellers lacking this type of access were absent from the sample, and overall the sample lacked diversity. Findings could have been very different in a sample that was more representative of the general survivorship population, with the potential to show a higher negative impact. In this study, most of the RCSs reported that they were very happy with their rural lifestyles. There were differences in the definitions of what was considered rural in each region, which may have influenced the findings. The definitions were similar as they were based on distances to urban areas and major medical centers. However, the travel times may have been different. While the definitions were appropriate for each region, the differences make the comparisons challenging. Alaska is often excluded from these types of studies, as the typical rural/urban classifications do not apply due to the unique geography of the state.29 The mixed findings in this study, and the results of previous studies, suggest that factors other than place of residence and access to health care may be at play in accounting for the rural/urban differences. Such factors could include ethnicity, education, socioeconomic status, support system, type of cancer, comorbidities, and personality characteristics. Future studies could consider
c 2015 National Rural Health Association The Journal of Rural Health 31 (2015) 282–291
Cancer Survivorship, Rural and Urban
McNulty and Nail
Table 3 Comparison of Quantitative and Qualitative Findings Quantitative Findings Urban/rural differences found on study outcome variables Body concerns AK urban higher than AK rural OR urban higher than AK rural OR rural higher than AK rural Worry AK urban higher than OR urban Negative Impact AK urban higher than AK rural Employment AK urban higher than AK rural AK urban higher than OR rural OR rural higher than AK rural No statistically significant differences in self-reported health, satisfaction with health care, and access to health care Elements in the positive impact scale that revealed no differences Not measured, items related to CS identity Not measured Elements in the positive impact scale that revealed no differences
Qualitative Findings Urban/rural differences emerging from interview data No differences identified between rural/urban CSs
Worry manifested itself in the themes of recurrence and uncertainty, but no differences were identified between rural/urban CSs Interwoven throughout the CSs experiences but not predominant Work impact discussed, but mostly positively
Health care Access and care coordination (urban more positive, OR more positive) Community/connecting Rural more positive Public/private journey Neutral Thinking About Death and Dying Neutral Advocacy Rural more positive
AK, Alaska; OR, Oregon.
factors such as resilience in rural individuals and communities that may contribute to rural/urban differences.30 In addition to individual differences, differences in the characteristics of the communities and community life may contribute to the findings. The recruitment and sampling techniques for this study differ from other studies that have found significant differences in the quality of life between RCSs and UCSs. Sampling strategies from previous studies included analyzing data from national surveys8,10 and recruiting from cancer centers and health departments.9 No studies focused on cancer survivorship were found that examined specific characteristics of rural communities or aspects of community life. These factors are important to consider in future studies. Given the sample characteristics, the findings with regard to health care utilization and coordination are not surprising. The CSs in this study, rural and urban alike, demonstrated that they were well informed and utilizing resources available to them. This is in contrast with other studies that have shown that RCSs are less likely to be aware of services and utilize less mental health resources.9 In this study, CSs from Oregon did not express concerns about lack of care coordination or navigation; however, it was a concern for the Alaskans. Models for navigation are beginning to emerge in
c 2015 National Rural Health Association The Journal of Rural Health 31 (2015) 282–291
survivorship care. However, there is conflicting evidence for their effectiveness,31 and there is limited focus on rural populations. Although there were minimal differences found, this study reveals themes that have not been previously described in the literature. RCSs described connections to their communities and networks very positively. They were more informal, occurred more naturally, and they relied heavily on them. RCSs talked about the importance of their neighbors, and in contrast, UCSs relied on formal social ties that were more planned, such as work networks and joining formal organizations. Both RCSs and UCSs relied heavily on their church communities, and being active in a religious community has been shown to be beneficial for CSs.32 Other studies have described a strong sense of community and dense social networks in rural settings, but not within the context of cancer survivorship.33 The notion of public/private journey is also new to this body of literature. The RCSs in this study talked about their identity as CSs more strongly than their urban counterparts. They described the dichotomy between others knowing and not knowing about their cancer journey. The notion of lack of anonymity in small towns has been well described,34 and the wish to maintain privacy may
289
Cancer Survivorship, Rural and Urban
not be a choice for RCSs. The process of integrating a new identity as a CS has been previously described,35,36 but once again not within the context of a rural setting. The strength in advocacy came through more strongly for the RCSs. They advocate for themselves, their diagnosis, survivorship, and for improved health care in their communities. The advocacy emerged as seeking a second opinion, accessing support resources, asking questions and seeking answers, and fighting for their financial and employment rights. The RCSs in particular engage in community advocacy by fundraising, volunteering with survivorship organizations, and speaking publicly about survivorship issues. Overall, there was a sense of strong empowerment from this highly engaged group of CSs. The fact that the majority of the sample was female, and breast CSs, could have heavily influenced the findings, since there are typically more resources available for breast CSs than for survivors of other malignancies. The sample size for each group was small, and the sample was very heterogeneous with regard to cancer diagnosis, type of treatment, and time since treatment. These factors could have influenced the ability to detect differences between the RCSs and the UCSs, and made it difficult to make cancer site-specific recommendations. The methodological limitations of this study include the following: (1) there were no comparisons between the CSs and the general population; therefore, it cannot be confirmed that the findings are unique to CSs; (2) quantitative data were collected by self-report and self-administered (not during an interview); therefore, the participants may not have been answering within the context of their cancer survivorship experience; (3) medical diagnosis and treatment profiles were not confirmed; (4) nonverbal cues were lost in the phone interview data collection process; and (5) the cross-sectional design did not capture changes over time. Future studies could benefit from conducting community-based participatory research (CBPR) studies as an effective strategy to engage CS populations. The CBPR approach could capitalize on the high level of engagement with survivorship and lead to successful engagement of CSs with future research, especially in their communities. This study provides rural/urban comparisons across multiple states and includes CSs in some of the most remote locations of the United States. The sample includes CSs who are under-represented in the current body of research, including rural dwellers, Alaska Natives/American Indians, men, and survivors of a variety of cancer diagnoses. It is one of the few studies to explore the potential positive aspects of rural living within the context of cancer survivorship. The study provides evidence for the effectiveness of the IOCv2 instrument in detecting differences in a diverse population, for the
290
McNulty and Nail
constructs associated with negative impact. Overall, the study has demonstrated that mixed methods is a valuable approach to studying cancer survivorship. The results of this study challenge prevalent assumptions about RCSs and their risk for negative outcomes. Findings provide insight into the differences and similarities between rural and urban CSs. A profile of rural survivorship emerges, with a common thread being community. Access to health care may not be the driver of the survivorship experience. There is an opportunity to influence health care providers and developing programs serving this growing population with complex issues, building on the strengths of both rural and urban living, and the engagement of the cancer survivor community.
References 1. American Cancer Society. Cancer treatment and survivorship facts and figures 2012–2013; 2012. Available at: http://www.cancer.org/acs/groups/content/@ epidemiologysurveilance/documents/document/acspc033876.pdf. Accessed Febuary 1, 2013. 2. Yabroff K, Lawrence W, Clauser S, et al. Burden of illness in cancer survivors: findings from a population-based national sample. J Natl Cancer Inst. 2004;96(17):13221330. 3. Feuerstein M. Defining cancer survivorship. J Cancer Surv. 2007;1:5-7. 4. Aziz N, Rowland J. Cancer survivorship research among ethnic minority and medically underserved groups. Oncol Nurs Forum. 2002;29(5):789-801. 5. Ganz P. Monitoring the physical health of cancer survivors. J Clin Oncol. 2006;24:5105-5111. 6. Angell K, Kreshka M, McCoy R, et al. Psychosocial intervention for rural women with breast cancer: the Sierra Stanford Partnership. J Gen Intern Med. 2003;18:499-507. 7. Hewitt M, Greenfield S, Stovall E. From Cancer Patient to Cancer Survivor: Lost in Transition. Washington, DC: National Academies; 2006. 8. Burris J, Andrykowski M. Disparities in mental health between rural and nonrural cancer survivors: a preliminary study. Psycho-Oncology. 2010;19(6):637-645. 9. DeSipio T, Hayes S, Newman B, et al. Does quality of life among breast cancer survivors one year after diagnosis differ depending on urban and non-urban residence? A comparative study. Health Qual Life Outcomes. 2012;8(3):1-10. 10. Lyons M, Shelton M. Psychosocial impact of cancer in low-income rural/urban women: phase I. Online J Rural Nurs Health Care. 2004;4(1):10-21. 11. Weaver K, Geiger A, Lu L, et al. Rural-urban disparities in health status among US cancer survivors. Cancer. 2013;119(5):1050-1057.
c 2015 National Rural Health Association The Journal of Rural Health 31 (2015) 282–291
McNulty and Nail
12. US Census Bureau. Current Population Reports, P60-229, Income, Poverty, and Health Insurance Coverage in the United States: 2004. 2005. Available at: www.census.gov/ prod/2005pubs/p60-229.pdf. Accessed December 10, 2014. 13. National Network of Libraries of Medicine. State information for Alaska. 2007. Available at: http://nnlm.gov/pnr/characteristics/alaska.html. Accessed December 10, 2014. 14. City-Data. State of Alaska. 2007. Available at: http://www.city-data.com/states/Alaska.html. Accessed December 10, 2014. 15. Alaska Experience Project. Health care in the last frontier: rural challenges/Alaska innovations. 2007. Available at: http://www.agnewbeck.com/pdf/statewide/AEP/AEP_ Healthcare-in-the-Last-Frontier.pdf. Accessed December 10, 2014. 16. GAP Report. Mental Health in Remote Rural Developing Areas: Concepts and Cases. Washington, DC: American Psychiatric Press; 1995. 17. Oregon State University. Defining rural Oregon: an exploration. 2007. Available at: http://extension. oregonstate.edu/catalog/pdf/em/em8937-e.pdf. Accessed December 10, 2014. 18. American College of Surgeons. Cancer programs 2012. Available at: http://datalinks.facs.org/cpm/CPMApproved Hospitals_Result.cfm. Accessed November 1, 2012. 19. Oregon Society of Medical Oncologists. Members. 2012. Available at: www.osmo.org/members. Accessed November 1, 2012. 20. Palmer N, Geiger A, Lu L, et al. Impact of rural residence in forgoing healthcare after cancer because of costs. Cancer Epidemiol Biomarkers Prev. 2013;22(10):1668-1676. 21. Aziz N, Rowland J. Trends and advances in cancer survivorship research: challenge and opportunity. Sem Radiat Oncol. 2003;13(3):248-266. 22. Zabalegui A, Sanchez S, Sanchez P, et al. Nursing and cancer support groups. J Adv Nurs. 2005;51(4):369-381. 23. Crespi C, Ganz P, Petersen L, et al. Refinement of the impact of cancer scale. J Natl Cancer Inst. 2008;100:15301541.
c 2015 National Rural Health Association The Journal of Rural Health 31 (2015) 282–291
Cancer Survivorship, Rural and Urban
24. Crespi C, Smith S, Petersen L, et al. Measuring the impact of cancer: a comparison of non-Hodgkin lymphoma and breast cancer survivors. J Cancer Surviv. 2010;4: 45-58. 25. Chang V, Hwang S, Feuerman M, et al. The memorial symptom scale short form (MSAS-SF): validity and reliability. Cancer. 2000;89(5):1162-1171. 26. Leipert B. Rural and remote women developing resilience to manage vulnerability. In: Lee H, Winter C, eds. Rural Nursing: Concepts, Theory and Practice. 2nd ed. New York, NY: Springer; 2006:79-95. 27. Crandall M, Weber B. Oregon State University, Rural Studies Program. Defining rural Oregon: An exploration. 2005. Available at: http://ruralstudies.oregonstate.edu/ sites/default/files/pub/pdf/rsp reports/rsp05-03.pdf. Accessed November 1, 2012. 28. Creswell J, Plano-Clark V. Designing and Conducting Mixed Methods Research. 2nd ed. Thousand Oaks, CA: Sage; 2010. 29. Ormond B, Zuckerman S, Lhila A. Rural/urban differences in healthcare are not uniform across states. The Urban Institute. 2000; Series B, No B-11. 30. Buikstra D, Ross H, King C, et al. The components of resilience-perceptions of an Australian rural community. J Community Psychol. 2010;38(8):975-991. 31. Wells K, Battaglia T, Dudley D, et al. Patient navigation; State of the art or is it science? Cancer. 2008;113(8):19992010. 32. Vachon M. Psychosocial distress and coping after cancer treatment. Am J Nurs. 2006;106:26S-31S. 33. Johnson I. Patient perspective on health, health needs, and health care services in a rural Irish community: a qualitative study. Rural Remote Health. 2011;11: 1659. 34. Long K, Weinert C. Developing the theory base. In: Lee H, Winter C, eds. Rural Nursing: Concepts, Theory and Practice. 2nd ed. New York, NY: Springer; 2006:3-16. 35. Zebrack B. Cancer survivor identity and quality of life. Cancer Pract. 2000;8(5):238-242. 36. Breaden K. Cancer and beyond: the question of survivorship. J Adv Nurs. 1997;26:978-984.
291
