Study design and setting

We performed a qualitative study to obtain detailed personal accounts of women’s referral experiences, which are not captured by quantitative data. Participants were recruited from Princess Margaret Hospital and St. Michael’s Hospital, tertiary centres in Toronto, Canada. Unlike the majority of hospitals in the province, both centres have da Vinci surgical robots (Intuitive Surgery, Sunnyvale, California, USA), which have been shown to facilitate minimally invasive hysterectomy in patients with morbid obesity.7 As a result, these centres receive numerous surgical referrals for this population from family doctors, gynaecologists and oncologists spanning academic, community, rural and urban practices. This ensured recruitment of patients from diverse backgrounds.

Sampling and recruitment

Consecutive patients who met the following inclusion criteria were approached by a research assistant in person at the time of a clinic appointment: (1) diagnosis of low-grade endometrioid adenocarcinoma of the endometrium; (2) BMI >40 kg/m²; (3) plan to be treated by primary hysterectomy or receipt of primary hysterectomy in preceding 6 months; (4) ability to communicate in English and (5) ability to provide informed consent. Patients who were not surgical candidates due to severe medical comorbidities were excluded, as we aimed to capture barriers to surgery in patients with morbid obesity who had no contraindications.

Data collection

Literature on weight stigma and labelling theory were used to create a semi-structured interview guide (online supplementary appendix 1). Labelling theory originated in the sociology of deviance, and includes (1) primary labelling, which reflects how society reacts to a group of individuals who diverge from standard cultural norms; and (2) secondary labelling, which reflects how those individuals alter their own self-regard and social roles in response to societal stigmatisation.17 18 Questions were based on this theoretical framework and were thus designed to understand participants’ experiences in and responses to the healthcare system. The interview guide was pilot-tested on two individuals without endometrial cancer to ensure clarity, and then modified iteratively based on emerging interview findings.

Three researchers (MCC, ANS and AH) carried out one-on-one telephone interviews using the interview guide. We chose telephone interviews due to the sensitive nature of the subject matter and concerns that the body habitus of the interviewers may influence patients’ responses. All interviews were audiorecorded. After each interview, initial impressions and the tone of the conversation were documented to provide context for the analysis.19

With respect to researcher characteristics, MCC and ANS were graduate students and clinical trainees in Obstetrics and Gynaecology, and AH was a senior medical student. Instruction and oversight were provided by RH, an expert in qualitative methodology. While MCC, ANS and AH had previously interacted with endometrial cancer patients, they had no involvement in the care of any study participants. All researchers bracketed their assumptions based on previous experiences with this patient population, and the contextual perspective of the researchers was discussed during data collection and analysis.

Data analysis

Interpretive description, a qualitative research approach grounded in applied disciplines, was used to construct an account of participants’ experiences.20 Interviews were de-identified, transcribed verbatim and audited for accuracy. Throughout the auditing process, a rudimentary analysis was performed where initial ideas were documented in the transcript margins and key phrases were highlighted.19 Using NVivo (V.11), two researchers (MCC and ANS) then coded transcripts line-by-line for significant quotations, which were developed into themes (clusters of meaning) inductively and deductively drawing on labelling theory. The first three transcripts were coded by MCC and ANS concurrently so that a common codebook could be constructed, and remaining transcripts were coded independently. Researchers updated the codebook as themes emerged, and early transcripts were secondarily analysed for these new themes. Interviews were conducted until thematic sufficiency was achieved. Themes were ultimately grouped to construct a rich description of the meaning of participants’ experiences.20–22 To illustrate key themes, MCC and ANS reviewed each transcript for representative quotations and included those that were independently selected by both researchers.

Patient and public involvement

The research question and qualitative approach were selected in order to directly elicit patient perceptions of their experience, and to thus understand their concerns, preferences and priorities. However, patients or the public were otherwise not formally involved in the design, recruitment or conduct of the study.

Patients and providers lack knowledge on endometrial cancer

Prior to their diagnosis, many participants simply had no idea there was ‘such a thing as endometrial cancer’ (ID 3). One patient explained, “It’s just not a popular cancer. People don’t understand it. If you say breast cancer, people know it. If you have prostate cancer, they know what it is. I had never even heard of (endometrial cancer) (ID 11).”

Participants also lacked knowledge on the significance of postmenopausal or abnormal uterine bleeding before their diagnosis. Except for one individual who had been concerned about cervical cancer, patients had not known that these symptoms could indicate malignancy. While most notified their family physician regardless, several did not or waited years before doing so. In these cases, participants either did not think their bleeding symptoms ‘were a big deal’ (ID 4), believed they were simply a ‘change-in-life-thing’ (ID 12) or had skewed perceptions of what constituted normal cycles. One participant stated, “I’ve had menstrual problems, absent periods, infertility, ovarian cysts, for my entire life. So having pain, having odd menstruation, having periods absent for months on end, none of that was new to me (ID 3).”

Participants also described situations in which their primary care providers failed to recognise postmenopausal and abnormal uterine bleeding as symptoms concerning for cancer. While many physicians did refer to a gynaecologist, some attributed the symptoms to structural causes or perimenopause without investigating further, and others ordered lab tests or imaging rather than an endometrial biopsy. Many participants experienced worsening of symptoms before a diagnosis was made. One participant stated, “I couldn’t get the bleeding to stop on its own, and I was afraid of passing out and bleeding more. So I started calling 911 (ID 15).” For others, the experiences of family or friends led them to carry out their own research and persist in having their symptoms evaluated: “I went back to my doctor after my friend passed (away from cancer) and said, ‘This bleeding is going on unabated and it is bizarre. No woman I know puts up with this.’ (My doctor) again brought up about the fibroids. She said—‘I can send you for another ultrasound.’ And, I said—‘Yes, please do, but I also want a biopsy’ (ID 3).”

Similarly, no participant had known that obesity is a risk factor for endometrial and other types of cancers prior to their diagnosis. One participant summarised, “Obesity is a terrible thing because it affects so many things… but I did not know that it would affect cancer (ID 13).” Most agreed that messaging around endometrial cancer as an entity and obesity as a risk factor was inadequate: “It’s definitely an educational thing. My sisters are overweight and past menopause. They should be getting checked too (ID 8).”

Endometrial cancer patients with obesity are subject to stigma and poor provider communication

Participants highlighted concerns with how their healthcare providers had communicated with them about obesity. First, most providers had simply not discussed it: almost all participants learnt about the link between obesity and endometrial cancer through their own online searches and independent reading. Second, if physicians did address obesity, it was usually at the time of diagnosis, and patients questioned this approach. While some appreciated the ‘wake-up call’ (ID 14), others felt that discussing obesity at diagnosis was futile or placed blame on the patient: “I don’t know whether it was really necessary to tell me because I was so obese that that’s why I got cancer. It’s sort of after the fact. It was done in a very sensitive way, but I don’t know what the point was (ID 2).”

Most recently diagnosed participants expressed a desire to have been counselled specifically on the association between obesity and endometrial cancer earlier in life. One participant stated, “If there really is such a huge risk for women that are overweight, then it really should be more pronounced. Doctors should be bringing that to women’s attention all the time. Just like now you smoke, you get lung cancer, period. If you’re overweight, or this much obese, you’re going to get cancer. That should be told (ID 5).”

In line with primary labelling theory, some patients felt they were treated differently than normal weight individuals. For example, some were offered non-curative alternatives to surgery because of their body weight: “She didn’t like to deal with me because of my body weight. She wasn’t willing to do anything, other than radiation. That was the attitude I got. Basically it’s—Oh you’ve got cancer. Have a good life, what’s left of it—you know (ID 15). ”

In line with secondary labelling theory, several participants discussed prior stigmatising experiences that led to avoidance of healthcare. One participant stated, “I had a doctor say to me when I went for a Pap smear once—‘This is impossible, there’s so much fat down here.’ Well, I didn’t want to go for a Pap test ever again (ID 2).” Another participant who had previously been referred for knee surgery described how her doctor had implied that obesity was merely a choice: “I had the surgeon say to me—‘Why would anyone want to be fat like that?’ (ID 7).” Some participants avoided the healthcare system altogether because their previous health complaints had always been attributed to their weight: “I figured if I have like, joint pain or you know, any little issue, it would be because of my weight. If I go see the doctor, he’s going to say the obvious, you know what I mean? Lose weight (ID 14).”

However, not all participants felt that their weight adversely affected their interactions with the healthcare system. Three had never felt stigmatised, saying, “I don’t believe I’ve been treated differently at all (ID 10).” Others acknowledged a potential issue but were unwilling to let others influence their own self-worth because of their weight: “It’s not my body fat. I’m what is inside of me. I don’t let that dictate how I feel, or how people treat me. Sometimes people can be crude and make remarks and what not and you ignore that sort of thing. That’s your opinion, fine. I know who I am. I’m a good person. I’ve got a lot of self-confidence. I don’t need your opinion. I know that I’m overweight, but that’s for me to deal with, not you (ID 13).”

Potentially modifiable barriers to care exist for endometrial cancer patients with obesity

Barriers to care emerged as a common theme. Most barriers identified by participants were universal, in that they were likely experienced by both obese and non-obese patients. These included administrative, geographic and financial challenges (table 3). However, other barriers were likely obesity-specific. First, many participants described diagnostic delay due to struggles in obtaining an endometrial biopsy and/or imaging; specifically, physicians frequently needed to perform a pelvic examination and biopsy under general anaesthesia, or imaging tests were not possible or indeterminate due to body habitus. A participant whose diagnosis was made months after presentation stated, “ They tried two D&Cs eventually. The first one was inconclusive. The second one they had more material that I guess gave them some results. And then they did a CT scan. Well, they weren’t sure, because… it was a size issue. They weren’t sure their CT scan was big enough for me. Eventually they decided I’d fit their parameters. I guess I got small enough to fit in the CT scanner.”

Second, a small number of participants experienced treatment delay due to late cancellation or even intraoperative abandonment of surgery. One patient noted, “I got to the point where I had a surgery date and everything. But the anesthetist classed me as being too high risk for that hospital and said no, it could not be done there. So that was cancelled and I was referred to the other hospital (ID 15).” Finally, all participants experienced treatment delay due to requiring multiple referrals before finding an accepting surgical team. One participant, who saw four gynaecologists before surgery, described how discouraging this process was for her: “By this time I felt like I was auditioning for a show. And that if I didn’t audition well, my cancer… or my surgery would be delayed again (ID 3).”

Participants also discussed how the healthcare system is simply not equipped to accommodate individuals with morbid obesity. Deficiencies in the size or structure of waiting rooms, CT scanners, operating room tables and medical equipment were described. One patient summarised these issues when she said, “It can be embarrassing when the nurses say, oh, we don’t have a blood pressure cuff your size, so we’ll have to put it on your forearm. Or that their chairs are so darn narrow, with arms that you can’t sit on them in the waiting room (ID 6).”

Importantly, participants discussed a lack of effective weight loss interventions. Available weight loss programmes were seen as inadequate for women with morbid obesity. One participant stated, “Weight loss strategies need to be meaningful ones. Maybe there’s some people who need to lose 10 or 15 pounds for bikini season. But people who get to be my size, you wonder how can you ever get that big? If you can get that big, there’s something going on metabolically. This is not a normal process and a normal diet is not going to fix that (ID 3).”

Others discussed their repeated struggles with affording and maintaining a healthy lifestyle: “Well, what can you do? I’ve tried my whole life (ID 4).” The few patients who did take initiative to lose weight before surgery were disappointed when their efforts were not acknowledged by their healthcare providers.

Quality care for endometrial cancer patients with obesity is an achievable goal

Despite these barriers, almost all participants felt that they had been well cared for overall, in that their endometrial cancer had been ultimately treated appropriately by an expert surgeon. Indeed, participants appreciated when community physicians recognised the limitations of their centres and referred them to a more appropriate surgical team. When asked how she felt after her gynaecologist disclosed he was unable to perform her procedure due to the medical and surgical risks of morbid obesity, one participant recalled, “Actually, at the time I was, you know, kind of iffy about it but I almost immediately thought, thank you so much because at least you are saying you can’t do it and you are sending me to someone that can. We’re not beating around the bush. You’re not going to try and then tell me, oh sorry, can’t do it, so that takes more time. So I had no problem with it (ID 4).”

Many participants were ‘in awe’ (ID 13) of how quickly and smoothly their care progressed once they arrived at a centre that was comfortable offering minimally invasive surgery to patients with morbid obesity. One participant stated, “Once you got there, everything was all put in perspective. All in order. That part of the system is working (ID 7).” Another participant who was seen at five centres prior to surgery understood that the referrals were in her best interest but would have appreciated a more direct route.