Older Adults with Advanced Cancer are Selective in Sharing and Seeking Information with Social Networks

Abstract

Background:

Little is known about how older adults (OA) with advanced cancer interact with social network members (NM), and the nature of communication. This qualitative study aimed to characterize the processes by which OAs with cancer engage with NMs regarding their illness.

Methods:

OAs 65+ with advanced cancer and considering treatment (n=29) and NMs (n=18) underwent semi-structured interviews asking 1) about their illness understanding; 2) to identify NMs with whom OAs discuss health-related matters; and 3) to describe the content, process, and impact of those illness-related conversations. Three coders analyzed transcribed interviews. Codes were categorized and emergent themes were identified to generate hypotheses.

Results:

OAs seek NMs with medical backgrounds for cancer-related information and NMs with personal experience of a serious illness for emotional support. Patients characterize geographical location, frequency of communication, and length of NM relationship as factors that influence the nature of support the NM provides. Additionally, differences emerged between OA and NM perspectives on the depth of conversations and decision-making.

Conclusions:

A better characterization of how OAs’ seek and share information and support may improve medical communication, disease understanding, and support goals-concordant care.

Practical Implications:

Findings support more strategic involvement of key NMs in cancer-related conversations.

1. Introduction

Older adults 65 years of age and older comprise 60% of those with cancer and account for 70% of all cancer deaths [1–3]. Older adults with advanced cancer rely on their social networks for information about the disease and help making complex decisions. A social network is defined as a web of diverse social relationships that provides varying types and degrees of social support [4], for example, family members, friends, colleagues, peer groups, and healthcare providers. Yet, little is known about how older adults with cancer interact with social network members (NMs) and the nature of their communication.

Social networks can improve outcomes for individuals with cancer. Specifically, perceived high-quality social support, larger social networks, and being married are social network factors associated with decreased mortality in patients with cancer [5–7]. Support from network members is also associated with improved physical health and emotional health [8,9]. However, characteristics of older adults’ perceptions of support from social networks and social network size have not been examined in older adults with advanced cancer, and it is unknown if these findings from younger populations can be applied to older populations [10].

Older adults with cancer are least likely, of any age group, to seek information about cancer from any source [6]. However, evidence also suggests when older adults with cancer seek information, they are more likely to get information and support from social networks than younger counterparts who utilize internet resources more regularly [12]. There remains a gap in understanding the processes by which older adults with cancer seek and share information with social networks. Understanding the social network factors that influence how older adults seek and share information and the processes older adults use to engage social networks could inform medical communication. This knowledge could help to inform interventions that support the older adult’s disease understanding and decision-making.

This qualitative study aimed to characterize the processes by which older adults with advanced cancer engage with NMs regarding their illness. We elicited the contents and dynamics of networks from the perspectives of the older adults and their network members. Few previous studies have enrolled both the older adult and network member dyad to gather their respective experiences of their cancer [13–15]. Specifically, this study explored with whom older adults with cancer seek and share illness-related information, the content of that information, and the influence of information exchange with network members on disease understanding and decision-making.

2. Methods

2.1. Study Design

We performed a qualitative study of older adults with incurable cancer and members of their social networks. Using in-depth semi-structured in-person interviews, we elicited their prognostic understanding, with whom older adults talked about their cancer, the content of conversations between older adults and network members, and their perceptions of the effects of those conversations on illness understanding, management, and decision-making. Interviews were audio-recorded, transcribed, and coded to address pre-defined aims. The study was approved by the Institutional Review Board at the University of Rochester Medical Center.

2.2. Study Participants

Older adults were recruited by oncology physicians during clinic visits and screened for eligibility. Eligible older adults were ≥65 years old, within six months of diagnosis of advanced cancer, aware of their cancer diagnosis, and determined by their physician to have decision-making capacity. Eligible social network members were ≥18 years old and were recruited by three mechanisms: 1) Asking the person that accompanied the older adult to the visit, 2) having the older adult explain the study to the network member, or 3) having the older adult give a study brochure to the network member. The network member was broadly defined as someone in the older adult’s social network, they did not have to be specifically a caregiver or a specific source of support with the aim to capture diverse network member perspectives. For options 2 and 3, network members were provided with information to contact the study team if they were interested in participating. Older adults did not need to enroll with a network member to be eligible. Older adults and network members underwent informed consent.

2.3. Study Procedures

Two interviewers (GD, JB) were trained to administer the interview questions and probe for further elaboration. Older adults participated in a brief demographic survey and a 45–60-minute audio-recorded interview conducted by a research coordinator using an interview guide (Supplemental Table 1). The semi-structured interview first asked older adults about their prognostic understanding, including the purpose of treatment, curability, and estimated survival. The interview then shifted to ask questions regarding 1) to whom the older adult talks about health concerns and how they chose those individuals; 2) from where and from whom the older adult get illness-related information; 3) who provides advice and what kind of advice; and 4) who helps the older adult make decisions and in what way. We also asked the older adult to estimate the frequency of contact (daily/weekly/monthly) and length of relationship (years they have known the NM) with identified network members.

Enrolled network members completed a similar demographic survey and participated in a similar 45–60-minute audio-recorded interview regarding how they provide information, instrumental and emotional support, and advice to the older adult about the illness. Network members were asked about frequency and means of communication with the older adult with cancer, the NM’s understanding of the older adult’s illness, what they generally talk about with the older adult, and the sources of information the NM relies on for information about the older adult’s illness (Supplemental Table 2). Specifically, regarding cancer, network members were asked how often they discuss cancer with the older adult, their level of understanding regarding the older adult’s cancer, and advice, if any, they provide to the older adult about the cancer.

2.4. Data Analysis

A thematic analysis [16] was used with an initial team of eight coders. Twelve transcripts were randomly selected for initial open coding amongst four pairs of coders. Prognostic understanding was briefly asked in the interviews, but for the purpose of these aims, was not included in the analysis. The coding team met weekly to discuss and categorize codes, and create a preliminary codebook. A team of three coders (LK, CS, GD) then coded the remaining 35 transcripts according to the defined codebook using the MAXQDA 2020 qualitative data software analytic program (Berlin, Germany). Each transcript was independently coded by two randomly assigned coders. The coding team met twice a week to reach consensus on codes. Coders incorporated dyadic analysis [17] through horizontalization to examine overlaps and contrasts between older adult and network member interviews. Coders identified contrasting themes through an iterative analysis of older adult with cancer interviews and their corresponding network member interviews to generate contrasting themes between the older adult and network member perspectives. We generated codes that quantified the frequency of responses to the question regarding how often the patient communicated with the NM (daily/weekly/monthly) as well as length of relationship (patient reported number of years knowing the NM) and created coded categories for these responses. Discrepancies were handled by reconciling definitions of codes, adjusting code thresholds, and collapsing or parsing into two or more codes until reaching thematic saturation. An audit trail documented discussions regarding resolution of coding discrepancies and notes on discussion for determining final codes, categories, and emergent themes.

3. Results

3.1. Older Adult with Cancer Demographics

Twenty-nine older adults with advanced cancer were enrolled. Older adults had a mean age of 75 years (range 66–91); they were predominantly white (86%), male (72%), unmarried (59%), had a greater than high school education (76%), and were retired (79%) (Table 1).

Table 1:

Older Adult with Cancer Demographics

Variables	N=29 (100%)
Age (Standard Deviation)	75 (6.03)
Sex: N (%)
Female	8 (28)
Male	21 (72)
Race: N (%)
White	25 (86)
Black or African American	2 (7)
White and American Indian	1 (3.5)
Asian	1 (3.5)
*Ethnicity: N (%)
Non-Hispanic	27 (93)
*Marital Status: N (%)
Married	12 (41)
Separated, Widowed, Divorced	11 (38)
Never Married	3 (10)
Domestic Partner	1 (3.5)
*Education: N (%)
<High School graduate	2 (7)
>High School graduate	22 (76)
Employment: N (%)
Retired	23 (79)
Employed	4 (14)
Part-time Student, Other	2 (7)
Specific Cancer Type
Gastrointestinal	15 (51.7)
AML	3 (10)
MDS	2 (6.9)
Breast	1 (3.4)
Head and neck	1 (3.4)
Multiple Myeloma	4 (13.8)
Lung	1 (3.4)
Prostate	2 (6.9)

^*Numbers do not add to total due to missing data

3.2. Network Member Demographics

Eighteen network members, identified by 18 of the included older adults with cancer, were enrolled. The study included only one network member per older adult. Mean network member age was 58 years (range 33–88); they were predominantly white (83%), female (89%), married (55%), had a greater than high school education (83%), and half were currently employed. The majority of network members were relatives (47%), followed by spouses or partners (29%), and friends (18%) (Table 2).

Table 2:

Network Member Demographics

Variables	N=18 (100%)
Age (Standard Deviation)	58 (15.56)
*Sex: N (%)
Female	16 (89)
Male	1 (11)
Race: N (%)
White	15 (83)
Black or African American	3 (17)
*Ethnicity: N (%)
Non-Hispanic	17 (94)
Marital Status: N (%)
Married	10 (55)
Separated, Widowed, Divorced	6 (33)
Never Married	1 (6)
Domestic Partner	1 (6)
Education: N (%)
<High School graduate	3 (17)
>High School graduate	15 (83)
Employment: N (%)
Retired	7 (39)
Employed	9 (50)
Part-time Student, Other	2 (11)
*Relationship to Older Adult with Cancer
Spouse/partner	5 (29)
Relative	8 (47)
Friend	3 (18)
Other	1 (6)

^*Numbers do not add to total due to missing data

3.3. Nature of Support and Network Member Characteristics

We analyzed with whom older adults wished to discuss health-related matters, what older adults discussed, and the effects of those conversations. We found that older adults engaged in “selective sharing,” a process of selecting with whom to share information, what information to share, from whom to seek information, and what factors influenced that selection process (quotes 1a, 1b). We then generated themes centered on factors influencing communication processes and content regarding disease information, support, and illness management. The nature of support needed by older adults with cancer – cancer-related information and advice, emotional or instrumental – influenced older adults’ selective sharing. Key characteristics older adults considered were the NM’s medical background, experience with cancer, geographical location, as well as frequency of communication, and length of the relationship in which respondents quantified the amount of communication and years of knowing the NM. The coding team selected the exemplar quotes (table 3) and agreed upon them as representative of each identified theme.

Table 3:

Thematic Analysis Quotes

Theme	Older Adult with Cancer	Network Member (NM)
Selective sharing	Quote 1a: If I want to talk about it. So, ah, you know, you’re limited. And by choice, I donť want, care to share too much with too many people, at this point. I’ve let people know and they took off. I have people all over the country, and you know, for weeks now, and different people I’ve have, and old girlfriends are calling me.	Quote 1b: No, um, I usually take a cue from her. If she brings it up we’ll talk about it. And she brings it up more often than, than she used to about her other ailments to gout or the diabetes. And these, these other ailments, we only discuss if she was in significant pain or if something came up about it. But, um, with the, with the cancer she has now, um, she talks about it more. Not all the time, but she, she will mention it two or three times a week, and then we'll talk about it.
Length of relationship	Quote 2a: [M] is probably my closest. I have a women’s group. We’ve been together for 40 years. The same group. So we’ve been to weddings and our kids’ weddings and funerals and divorces and remarriages.	Quote 2b: My mom and [S] are good friends. They’ve known each other probably 25 years. So I have probably known Sharon, oh, gosh, I donť know, maybe 15 of those 25. And she – you know, in some respects she’s become part of our family. She’s here for holidays and stuff like that.
Geographically closer	Quote 3a: Well, we tell her everything. But you can’t remember everything during the phone call. And you’ve got to see people face-to-face. No, actually she wants to be as involved as possible. And we would like it. But distance precludes some of that.	Quote 3b: I’m his healthcare proxy. I live around the corner from him. I’m his main caregiver. We’re close. He relies on me for transportation. I help him with meal prep and things like that. I help him mostly a couple days during the week and then on Sundays to give him some independence, but we have a close relationship
Frequency of Communication	Quote 4a: Patient (P): Well, we talk every, talk every day. Interviewer (I): Okay. And how has she been of help to you? P: Just being around.	Quote 4b: I actually talk to her in the morning on my way home from work, which is about a half an hour drive, I talk to her on my way to work, which is half an hour drive, and sometimes we talk in the middle if we need to. I: Fabulous. R: Yeah, we have a very close relationship.
Role of NM with medical background	Quote 5a: Well, my husband [T], my daughter [M], and my son-in-law [B] have been very involved with every step. They’ve been at every doctor’s – well my daughter and son-in-law are both physicians; he is a hospitalist at [hospital] and she is no longer practicing, she has a young family, but she is a pediatrician. She does a lot of research, he does a lot of research, and they communicate with [T] and I about their research and every test, everything they go over with me.	Quote 5b: Hmm, well, since he’s been sick he reaches out to me since I’m a doctor – which is new. Before that, it was more family gatherings.
Role of NM with person cancer experience or medical issue	Quote 6a: One thing I didn’t say is that his companion, she had also breast cancer. So I was trying to be helpful for his experience at that time, and then I got cancer and he was trying to be helpful, so we have these events in our life that we were closer together because of that. I think we are just understanding each other a little more.	Quote 6b: I've worked with a number of people who are cancer survivors, different types of cancer. My mother-in-law. Um, I forget what cancer she has, but she's gotten through it and they, they’re calling her a survivor. She's 80 years old and they’re like, ‘She's a survivor.’ So, um, I, these conversations I’ve had with these friends and family is, um, iťs more preparing me for what, how my mother's condition may change for the worse down the road.
Comparing Older Adult with Cancer and Network Member themes
Depth of conversations	Quote 7a: I: Have you shared details with them? P: No, I told them I got lung cancer. I: Just the basics. P: Yeah.	Quote 7b: Um, once she was diagnosed with cancer, um, and then we found out it was terminal, I would say we, um, we have, our conversations, when we do talk about it is much more frank Quote 7c: He gets scared, is that what you mean? On his lower days when he’s not feeling good, he gets depressed. He wants to talk about getting things situated and finalized and I try to turn a positive spin on it and talk about the future and what we’re doing and how we take day by day and things like that. I try to do that when he’s getting down and kind of negative.
Decision-making	Quote 8a: Well, again with the medical background, when there were two different therapies that the doctor was talking about, I asked [B], my son-in-law, and I asked my oncologist myself, which would you recommend and why? I got my answers from why. They both seemed in agreement as well. I guess I would say I reach out in making decisions about treatment first to somebody with medical experience. Oh yeah. She’s bossy. [chuckles] She tells me what to do and what not to do.	Quote 8b: He is open to hearing it but it’s up to him if he wants to follow it. Quote 8c: And I thought I’m gonna let you have it the way you want it. You know. I donť need to argue or change her opinion... Just let her be.

3.3.1. Information and Advice about Cancer

Older adults with cancer needing more cancer- and treatment-related information, advice, or decision-making support turned to NMs with a medical background, regardless of geographical location or frequency of communication. These NMs came from various health professions, such as nursing and advanced practice providers (APP), and medical specialties outside of oncology, such as family medicine (quotes 5a, 5b).

Older adults with cancer sought more general advice from NMs they describe as talking to daily or weekly and geographically closest NMs. Older adults with cancer describe sharing disease-related information with NMs they spoke to most frequently (daily or weekly) and lived closer to geographically. While older adults discussed sharing cancer-related information with NMs geographically distant, this distance appeared to act as a barrier to more detailed processing of the information (quote 3a).

3.3.2. Emotional Support Through Cancer

Several respondents articulated a preference for seeking emotional support from NMs with personal experience with cancer or other serious chronic illness, such as diabetes. The help received often involved emotional support in adjusting to a significant life change and navigating a complex healthcare system (quotes 6a, 6b) The NMs’ cancer type, stage, and treatment trajectory were rarely mentioned.

Among the factors that respondents mentioned influencing their connections with NM members, several elaborated on length of relationship (quotes 2a, 2b)]. Older adults knew some NMs for decades and felt more emotionally connected to these network members. While length and depth of the relationship with the NM did affect older adults’ selective sharing, it mattered less whether the NM was a spouse, sibling, or friend.

3.3.3. Instrumental Support

When older adults with cancer needed instrumental support, such as transportation to appointments, grocery shopping, meals, or household tasks, reliability was important, mentioning those with whom they communicated daily or weekly as more reliable. While patients would share information about instrumental tasks with geographically distant NMs, distance was a barrier.

3.4. Contrasting Perspectives on Impact of Selective Sharing Process

We found several patterns when comparing interviews with the OA and their respective NMs. They discussed several similar themes: 1) depth of conversations; and 2) decision-making. However, perceptions between older adults with cancer and network members diverged on the impact of those selective sharing processes. Coding between the older adult’s interview and their respective network member interview, some network members described their conversations with patients as in-depth, while patients described conversations as casual or surface-level.

Interestingly, some older adults describe wanting to have more in-depth discussions surrounding their disease and end of life, while their network members described their role as keeping discussions non-cancer-related and positive (Quotes 7a, 7b, and 7c). In addition, older adults with cancer describe decisions as shared between the older adult and networks members, while the NMs described the final decision was the older adult (Quotes 8a, 8b, and 8c).

4. Discussion and Conclusion

4.1. Discussion

The findings of this qualitative study describe the processes older adults with advanced cancer use to identify with whom to discuss illness-related information, the content of those discussions, and the influence of these network members in disease management and understanding. We found older adults with cancer do engage in a selective sharing process. Depending on the nature of support needed, older adults with cancer sought NMs with medical background, NMs with personal experiences with cancer or chronic illness, geographical proximity, longer relationships, and greater frequency of contact. We have clarified some of the contextual factors through which older adults with advanced cancer select with whom to seek and share information and support regarding their illness, such as XYZ. We also clarified the nature of information or support received, such as QRS. Yet other factors remain unclear, pointing to the need to have a better understanding of the influence and roles of network members in older adults’ information seeking, decision-making, and emotional and instrumental support that could improve medical communication, disease understanding, and goals-concordant care.

Older adults with cancer engage in a selective sharing process in identifying specific network members who provide clear types of information and support to the older adult. Previous studies have categorized the health information-seeking of patients with cancer [18,19]. Our findings provide nuance to the nature of support needed, and the circumstances and characteristics of the older adults’ challenges that influence the selection of certain NM to seek and share information and support. This selective sharing illustrates older adults’ more passive style of information seeking [20], compared to younger counterparts who are more technologically literate and more actively seek information from the internet and literature [12]. Our findings add more context to “passive information seeking” [20] by identifying the specific nature of support and the NM characteristics older adults seek for that support. For instance, if an older adult needed general emotional support, they seek out NMs they have known for decades. For more specific cancer-related coping support, the older adults intuitively sought out NMs with personal experiences of cancer and other chronic conditions.

The selective sharing theme reveals that older adults with cancer do utilize their social network depending on the information or support needed. Studies examining social networks in younger adults with cancer have found that they often withdraw from their network and experience social disruption [21–23]. However, young adults with cancer more actively utilize internet-based information and support than older adults do [12]. This difference in social network utilization between younger and older adults with cancer may be influenced by technological advances that influence how younger people engage with social networks during challenging circumstances. Older adults may have gained experience throughout their lifespan in turning to network members for varying types of support. This aligns with the intuitive nature of selective sharing; as if it is a skill set developed over years of life.

Network members with medical backgrounds were sought by older adults with cancer. They offered cancer-related informational support and advice. In other studies, older adults with cancer have reported social networks provide emotional and physical support, but not always help accessing cancer-related information [24]. After an older adult is diagnosed with cancer, it may be beneficial to identify NMs with a medical background the older adult might consider for informational support. If an older adult does not have such a network member, there is an opportunity for healthcare providers to better meet patient needs by filling in the gaps of cancer-related information they might have otherwise sought from an NM with a medical background.

The network members with medical backgrounds did not have to have cancer-specific backgrounds, but rather any area of healthcare prompted older adults to seek out those NMs for cancer-related informational support. Network member assessment tools could support oncologists to gather information about the influence of NMs with medical backgrounds on older adults’ understanding, and whether that information is aligning or conflicting with oncologists.

In addition to NMs with medical backgrounds, older adults sought NMs with personal experiences of cancer or chronic illness. These NMs offered coping and emotional support. Because of their shared experience of having had a serious illness, NMs can provide emotional support, promote healthier coping behaviors, physical activity and social interaction [25], - and ultimately improve quality of life [26][27].

The older adults with cancer and their identified network members had described their mutual conversations differently. In several OA-NM pairings, we noted that an NM described avoiding end-of-life conversations in favor of more “positive” topics, believe that they have a lesser role in decision-making (and feel that older adults make their own decisions), and describe their conversations as more “in-depth” than the older adults reported. This observation is consistent with prior studies which revealed a need for normalization and a desire to talk to others about normal life, hobbies, and family [28] and avoid talking in-depth about their prognosis [29]. While avoidance and positive focus may have a temporary benefit of promoting emotional coping for older adults to counter the intensity of their illness, older adults also articulated unmet needs for deeper discussions. Communication tools that support older adults and network members to examine, “What did you hear [older adult/network member] say?” compared to examining the intended message being communicated might reduce these contrasting perspectives on roles, depth of conversation, and decision-making, and overall support more functional communication between older adults and network members.

Studies suggest older adults with cancer are less likely to want active participation in decision-making [30,31]. Additionally, findings indicate older adults have varying preferences about control over decision-making. Current scholars found mixed perspectives on control over decision-making. One interview study found older adults with cancer did not want to fully relinquish control for decisions to their families [32], while older adults with early-stage breast cancer preferred a shared decision-making style [33]. In relation to our findings, the older adult’s perception of shared decision-making may be influenced by factors such as cancer type, stage, or present health status. Future studies should consider the differing perspectives on decision-making between older adults and network members and the influence this has on selective sharing.

This study has several limitations. Firstly, the data consisted of transcribed interviews, in which participants may have not been fully forthcoming about their perspectives and may have misinterpreted questions. Respondent disclosure may also have been affected by similarities and differences between interviewer and respondent demographic characteristics. Our findings reflect the sample in one geographic area and cannot be generalized to other cancer populations and their network members. Additionally, our sample of older adults with cancer was mostly male and the network member sample was mostly female, which may bias generated themes. Our sample also included NMs identified by the patient, which may indicate the patient felt closer to the selected NMs. This may limit the breadth of perspectives on information seeking and sharing. Future studies should consider larger and more diverse samples in regard to demographics, but also selection of network members.

4.2. Conclusions

To our knowledge, no studies have described the influences of distinct categories of members of social networks of older adults with cancer and specific areas of support those NMs provide. We present novel identification of specific contextual factors, such as NMs with medical backgrounds acting as a source of information and advice about treatment decisions. Additionally, the differing perspectives between older adults and NMs in regard to their depths of conversations, roles in decision-making, and end-of-life conversations deepen our insights into target points to improve communication between older adults and NMs. Our findings demonstrate a need for more qualitative work to gather older adults with cancer and network member perspectives about their unmet information and medical needs and more deeply explore the processes older adults use to gather information. Additionally, future studies could examine the relationship between these selective sharing processes with patient satisfaction about oncologist communication, prognostic understanding, and decision-making processes.

4.3. Practice Implications

Clinicians’ knowledge of the specific contextual factors that drive older adults’ choices regarding from whom they seek health-related information and with whom they share it -- such as the network member’s background in health care, physical proximity and ability to provide instrumental care -- can help the care team involve key social network members in care more strategically and effectively, improve adherence and follow-through, and potentially improve patient outcomes. Additionally, the resulting improvements in communication may prevent future discordance between patients and providers.

Our study presents a challenge to clinical teams when older adults and NMs are misaligned in how they perceive their conversations. At the very least, clinicians will be disabused of the notion that even intimate partners share common perspectives on the illness, treatment, and need for and meaning of support, and may have implications for the older adult’s ability to cope with their illness, make informed decisions, and understand their disease. Our findings provide an avenue for interdisciplinary teams, including but not limited to physicians, social workers, nurses, or other advanced practice providers (APP) to support older adult and NM communication on these topics and facilitate explicit and aligned understanding about older adults and NM needs and perceptions while navigating cancer.