SfAA 2017 Conference Feature Part 2: “Experiences and Identity in Long-term and Chronic Illnesses”

This week on the blog we are continuing our feature of a paper session from the 2017 Society for Applied Anthropology (SfAA) Annual Meeting which took place in Santa Fe, New Mexico from March 28th through April 1st. This session was entitled “Experiences and Identity in Long-term and Chronic Illnesses” and featured Beth Moretzsky, Karen Dyer, Marlaine Gray, and Ellen Rubinstein (full program from the SfAA meeting available here). Here, we present a summary of Karen Dyer and Marlaine Gray’s presentations. Part one of this feature is available here.

Karen Dyer (VCU) – Examining Health and Illness after Treatment for Colorectal Cancer: Long-term Healthcare Needs and Quality of Life

From her research, Dyer discusses the attachments formed between people with a history of colorectal cancer and their oncology team, which stems from both a fear of recurrence of illness and an especially strong emotional bond forged through mutual experiencing of a life-threatening disease. Dyer also discusses the consequences of the ambiguity of cancer follow-up care. Questions frequently include, who is the best doctor to see? If multiple care providers are seen by people with a history of colorectal cancer, will tests will be duplicated or even missed? Which emerging symptoms are serious enough to warrant further investigation? This continuing surveillance of a “survivor’s” body encompasses many medical repercussions from the treatment of cancer and reality of recurrence risk, often transforming people into life-long patients.

Dyer interviewed 30 participants in Virginia who had a history of colorectal cancer and were at least five years post-treatment. As this category of people with a history of colorectal cancer increases, there is a growing number of individuals using, and in need of, follow-up care. Yet this need is contrasted against a shortage of oncologists and primary care providers who are able to treat this group of people. Dyer asks, how do we treat and provide adequate long-term care to people for years, possibly for the rest of their lives, in a way that is not going to compromise or strain the oncology clinics? What are the physical, social, and emotional needs of longer-term colorectal cancer “survivors,” and how does their cancer experience impact these needs?

Dyer explains that most participants with a history of colorectal cancer did see an oncology team regularly for a follow-up care. For this group, as with other groups with specific types of cancer histories, the five-year mark is a critical period where individuals get discharged from monitoring and care because the risk of recurrence is statistically very low. Yet from Dyer’s research, a large number of her participants were planning to continue to see their oncology team after they had passed the five-year mark. Most had no formal “survivorship” care plan, and in general, there was not an understanding of what future care would entail.

Many individuals with a history of colorectal cancer continue to see their oncology team because of the intense bond and emotional connection they have developed. Going through a serious life-threatening experience created an attachment and deep sense of friendship. Dyer discusses that the oncology team fills a role of social friendship and support during the cancer experience when many other relationships may change. For Dyer’s participants, the oncology team has seen them in their “worst moments” and guided them through this demanding treatment. This type of connection and support is difficult to abandon. One woman said, “I need my security blanket, and yeah, I guess that’s what Dr. L [her oncologist] is.” This sense of being cared for and understood will be greatly missed. Any kind of care planning needs to take that strength of bond and trust into account.

Ambiguity surrounding cancer follow-up care is also an important dimension of Dyer’s work. Many participants report difficulty several years post-treatment when symptoms or health problems manifest in uncertain ways. Participants report difficulties distinguishing between normal aging processes, potential non-cancer related problems such as diabetes, or potential cancer-related, or cancer treatment-related, effects. Dyer uses fatigue as an example. Many participants spoke about being considerably more fatigued than they usually were. While this fatigue could be part of a normal aging process, it could also signal a variety of diseases or indicate the arrival of a cancer recurrence. Ambiguous symptoms such as fatigue lead to a high stakes, complex decision-making process.

People experiencing these types of indistinct symptoms often express an uncertainty about who they should contact with questions and when. Participants frequently did not want to “bother” their oncologist or be perceived as overreacting. Self-diagnosis and self-assessments of the level of seriousness of these symptoms were often the responses.


Marlaine Gray (GHC) – Shouldn’t We Be Listening?: Using Twitter for Recruitment, Patient Engagement, and Data Collection in a Study about How Young Adults with End Stage Cancer Make Medical Decisions

Gray begins by discussing research methodological complications when researching young people with metastatic cancer who are geographically spread all across the United States. It can be difficult to find and access this understudied population. Gray also discusses the sensitive nature of the topic is often compounded by time constraints; asking a patient for an hour of their time as part of an interview is difficult when that individual may not have a lot of time left.

Using an already active Twitter community, Gray investigated how young adults with metastatic cancer made medical decisions and whether or not their care matched their ultimate goals. The research was called The Clare Project, named after and featuring a personal story of metastatic cancer. The intention was to understand what these patients wanted for their remaining life and quality of life and translate those goals back into the medical discourse in order to match up treatments. Most participants wanted more quality time with their family, yet they were often being advised to get surgeries. Gray explains this disjunction can be problematic since metastatic cancer patients may never return home from the hospital after these types of surgeries, or they be unable to recover completely and be unable to fully engage with their families again.

Twitter became a way to contact people who were already publically speaking about their cancer experiences. The population of young people online is very active in seeking treatment, finding other patients to connect with, and finding out what the treatments are like. While there are also blogs, threads like Reddit, Facebook groups and pages, and other online message boards, Twitter emerged as the most successful way of communicating with this population. People are online constantly to discuss their cancer experiences.

The metastatic cancer Twitter community uses hashtags such as #mayacc (metastatic adolescent and young adult cancer community), #hpm (hospice and palliative medicine), or #metsmonday, where people with metastatic cancer post about their experiences on Monday. After launching their call for recruitment on Twitter, the Clare Project (Twitter page available here) achieved 200% of their recruitment goal within 24 hours. By using established hashtags and following prominent community members, Gray was able to reach an extensive participant audience.

Adolescent and young adult cancer patients are already actively using social media, many joining Twitter after their diagnosis. Twitter becomes a means of social connection. Gray articulates people are using Twitter to discuss decisions they have to make surrounding their metastatic cancer treatments. Even though patients talk with their doctors and family members, they are using the Twitter support groups to find out what the treatment experiences are. It is these treatment narratives from fellow metastatic cancer sufferers which holds more decision-making weight. Some of these decisions are very high stakes and are based on their peer, rather than medical, advice.

Accessing the first-hand expertise of other patients is labeled as a different kind of expertise than they can get from the medical community. Additionally, for side effects, participants express that doctors can tell them what the treatment is, but their fellow patients will express what the treatment is like and how to manage it. This social support is crucial when participants often do not know anyone else with these types of cancer.

Gray also discusses a kind of “legacy activism,” where people would know they were terminal with few options in their own treatment, but they wanted to advocate for more research funding and attention to metastatic cancer. Social media became a way to engage in social activism. Even though people could not physically go to advocacy events, they could virtually participate from their bedrooms and still spread their message. Through Twitter, people can participate in research and campaigning who would otherwise be unable to do so.

Advertisements

Issue Highlight: Vol 39 Issue 4, Incarceration & Medical Anthropology

This blog post is the second installment of our three-part issue highlight on the new December 2015 issue of Culture, Medicine & Psychiatry (the full issue is accessible here.) In this week’s blog, we examine Carolyn Sufrin’s article on the shared roles of clinicians and anthropologists working with incarcerated women in the United States.


cropped-cards.jpg

Clinician-anthropologists are in a unique position to lend voice to their participants and to care for their medical needs. For Carolyn Sufrin, who served both as a physician and an ethnographic researcher for incarcerated women in the United States, the clinical and anthropological roles intersected in deeply meaningful ways. Sufrin notes that in these dual roles, she observed with female prisoners and analyzed their experiences, and provided reproductive health care and delivering the same women’s babies in a county hospital, where imprisoned women gave birth. The women were alternately enrolled in Sufrin’s study upon ceasing to officially be her patients, or were individuals that Sufrin cared for during her tenure as their OB/GYN.

Sufrin uses her case as an example of how to interpret the ethical consequences of working as a clinician-anthropologist. Part of the challenge she faced as both a physician and anthropologist to the women was that IRB and HIPAA regulations divided the types of data she could use in her anthropological research. Though some details of her participants’ lives, such as past trauma or childhood abuse, were essential to their experiences as mothers and as incarcerated women, she could not use this data inasmuch as it had been shared with her in the context of a patient-practitioner interaction: in other words, it was considered classified personal health information that could not be subsequently utilized in her ethnographic research. These methodological hurdles, Sufrin explains, shed light on the complexities of what constitutes “sacrosanct” data. It also suggests how the safeguarding of medical data does not necessarily translate to the “protective” collection of ethnographic data which is not isolated to the case of individual patients, and which relies on knowledge of participants and their connections to other people (in this instance, to individuals in the participant’s life who had caused past psychological harm.)

Likewise, as stated earlier, relationships between the participant/patient and the clinician/anthropologist are another form of interpersonal connection which must be reconfigured depending on the nature of the exchange taking place. In Sufrin’s case, this meant being clear with her participants that– upon entry into her anthropological study– she was no longer their physician, and that the nature of their exchanges and their professional relationship would take a new form. She could share their information (albeit de-identified), unlike information drawn solely from their medical records or from an examination.

Yet here, Sufrin notes that the anthropologist– like the clinician– is still engaged in an ethic of care. For example, one of Sufrin’s former patients in the prison was charged with child endangerment after giving birth in an alley and handing her child to a stranger, unable to afford treatment at a hospital following her release from prison. The story made news headlines, and as public knowledge, was within the realm of information that Sufrin employed in her ethnographic analysis of reproductive health in the American prison system. However, Sufrin knew an added piece of information that was not already publicized from her interaction with the woman in prison: a clinical detail that she understood as a physician, but was shared during her time as an ethnographic researcher. This detail would have enhanced her analysis of the situation, but she chose to omit it out of concern that to publicize the detail would be to betray the woman’s trust in her as a researcher and as someone who served in the prison as a physician.

In other instances that Sufrin discusses, her role as both an anthropologist and clinician led the women she worked with in each capacity to feel as if she was especially attentive and caring. Rather than always relying on her medical authority, Sufrin underscores her decision to remain non-judgmental and supportive even when the women in her study had made potentially harmful choices; for example, she chose not to openly admonish a woman who was continuing to use crack upon release from jail, despite being thirty-two weeks pregnant. Sufrin notes that her silence hid her own frustration with this choice, but it strengthened the woman’s trust in her as an anthropologist and “as a doctor.”

In sum, these dual roles ultimately bolstered her relationships to the incarcerated and formerly incarcerated women that she served. Thus, the author argues that both anthropologists and clinicians share an “ethic of care” in their relationships with research subjects or patients. This ethic involves a careful fostering of the professional, interpersonal connections that ethnographers make with research participants, or that physicians make with their patients. This care entails the protection of private information and the researcher/clinician’s conscious attendance to interpersonal exchanges that bolster the patient/participant’s trust in the clinician/anthropologist. Likewise, an analysis of the clinician role alongside the anthropological one demonstrates both the delineations between multiple forms of care and social connection, as well as the shared commitment in medicine and anthropology to the sensitive attendance to patient and participant experiences.