For Whom the Cancer Bell Tolls — Or Should It?

For Whom the Cancer Bell Tolls — Or Should It?

Matthew S. Katz, MD

Feb 19, 2020

A recently published clinical trial about patients with cancer ringing a bell at the end of radiation treatment attracted a lot of attention. The study highlights the importance of the patient experience, but people are jumping to conclusions prematurely based upon this provocative but inconclusive study.

When I started working in community practice in 2014, one of my hospitals had been using a bell for years and still does. The other hospital where I work doesn’t have a bell. The recent media coverage made me more aware that we should guide patients through treatment in a way that minimizes distress, including making us question existing practices we thought helped.

So let’s look critically at the prospective trial evaluating the association of ringing a bell with distress experienced by patients with cancer. We can then ask what it may teach us about how to study using the cancer bell further, and whether patients should still ring bells to finish treatment now.

Ringing the Bell: Trial Details

The study is a prospective clinical trial registered with the National Cancer Institute in 2016 and was conducted from June 2016 to February 2018 at the Norris Comprehensive Cancer Center’s Department of Radiation Oncology.1 210 patients enrolled in the trial to complete a survey to assess distress levels related to cancer treatment and to obtain patient perspectives on the treatment experience. About half of the patients rang a cancer bell next to the nurse’s station before enrollment to answer the survey. 163 patients (78%) then completed the same survey at “routine scheduled follow-up visits,” answering only distress-related questions on the survey. Patients ranked total distress, peak distress, unpleasantness of treatment, and current distress on a 1-10 Likert scale.

Patients had to speak English or Spanish, and the study excluded people with brain tumors or receiving palliative radiation. Anyone else interested was eligible, with about 40% of the participants having breast cancer, and the other 60% evenly split between participants with genitourinary, gastrointestinal, gynecologic, and head and neck cancers.

The study authors started without the bell in the department, then they installed it later in the study. An interim analysis found worse distress with the bell, but they moved forward to complete the study, removing the bell again to balance the numbers in both groups for analysis.

Mean distress scores were not statistically different on the final day of treatment on univariate analysis (p = 0.07), but the patients ringing the bell did have higher distress scores at follow-up (p = 0.009). Both scores appeared significant on multivariate analysis. Follow-up differed, however, occurring an average of 103 days after treatment for patients ringing the bell versus 130 days for those who didn’t.

Trial Flaws Limit Conclusions

It’s possible I have some bias because one of my hospitals has a bell, but I’ve mainly been agnostic on its use. All patients are offered the opportunity to ring it; none are required. It’s framed as a graduation, not a victory. I would like a good answer to the question of whether bell ringing is actually distressing rather than empowering. But this trial doesn’t answer it for me, for several reasons.

First, the hypothesis was that ringing the bell may end treatment on a good note, psychologically lessening the patient’s experience/memory of distress, called the peak-end rule. This is a neat concept that has been tested but may oversimplify the complexity of the cancer experience. Using the concept to measure distress at the end of radiation for everyone, when the disease experience varies widely between different cancers, doesn’t inherently make sense. For prostate cancer, radiation may be the entire curative treatment, while in breast cancer, radiation is done to improve surgical results and may not be perceived as necessary. Without involving someone with expertise in psycho-oncology, the study may not have defined the problem well at the outset.

Second, I found the survey instrument problematic. The 1-10 Likert scale had even numbers in black and odd numbers in gray. Lots of instructions were put in quotation marks for some reason, which might bias interpretation. (This style of punctuation is sometimes called scare quotes or shudder quotes, and can imply irony, skepticism, derision, or doubt.)  Although the article describes discussing “nonphysical negative consequences” as part of distress, the survey only asks about physical symptoms: "all side effects, such as pain, nausea, hair loss, trouble swallowing, diarrhea, etc.” Interestingly, the survey also asked patients to agree/disagree with whether they had a choice about treatment, whether it was “valuable to me,” and about their confidence that “I have ‘beaten/defeated’ my cancer by undergoing this treatment.” This section was done in the department, which might make it difficult to get an unbiased/honest answer returned by the patient to the treating physician.

Third, the cohort and trial setting make it hard to draw conclusions. 78% of the cohort was Hispanic with no breakdown of non-Hispanic patients, making it hard to generalize. The cancer experience differs by ethnicity, disease type, and other sociodemographic factors.2-8 In Lowell, I see many patients from Cambodia, Nigeria, Greece, Portugal, and the Dominican Republic. In Manchester, I see many patients of French Canadian or Bosnian descent. I have to vary my discussion of treatment options to account for differing perspectives. I imagine the meaning of the bell differs also.

Fourth, there were time-related issues that could influence the results. The trial couldn’t be conducted simultaneously ringing a bell in the same department, but it introduced another variable—seasonal bias, which might influence mood and remembering distress.9 Because the follow-up surveys had to be completed at a clinic visit, patients ringing the bell had a near-statistically significantly shorter time to recall distress compared to the patients in the control arm. Recall bias or actual persistent symptoms might vary with that 1-month difference. It’s possible this latter issue could be the reason that the follow-up study indicated higher mean distress with ringing the bell. Unfortunately, time from treatment to survey was not included in the multivariate analysis.

Finally, the discussion section didn’t bring up any of these issues. It was an opportunity to suggest alternate study designs, or raise other important questions that the trial generated. Instead, it was a jump to concluding the bell is detrimental. Strangely, the authors then back away from the logical conclusion of their results, which would be to remove the bell: “Given the apparent joy that patients derive from ringing the bell, we are disinclined to propose stopping the tradition.”

Legitimate Questions of Inducing Distress

The study is flawed but brings up important questions for oncologists to consider on how we try to help people through the cancer experience:

  • How do doctors, nurses, and radiation therapists frame the use of a cancer bell? Are we setting up patients for a brief but false sense of victory, especially when using militaristic metaphors?
  • Are all patients allowed to ring the bell, regardless of prognosis? It could be upsetting to patients receiving palliative care if they not included.10
  • What if patients can’t finish the planned treatment due to side effects? If they don’t get to ring the bell, does it imply they “failed”? Or, by ringing it, does the bell reinforce a reminder of treatment toxicity?
  • Can the bell increase distress in patients with metastatic cancer regardless of inclusion? Radiation treatments end; chemotherapy may not, or the end may signal having to change drugs for progression or to transition to supportive care only.
  • Where should the bell be placed, somewhere public or more private?
  • Who really benefits? Is it for the patient, their family, or to help the staff feel validated in doing a good job?

We don’t know. We need more research to learn if we’re really helping rather than worsening the cancer experience.

Seven Ways to Get Better Answers

  1. Collaborate outside your specialty. At a minimum, include psychologists or psychiatrists with a nuanced understanding of the published literature. Include all treating specialties, including non-physician professionals. Get all perspectives from the therapeutic side.
  2. Ask patients. Defining the value of the bell depends upon why people ring it, or why they don’t. You can ask informally what patients think at follow-up, but better yet would be a well-designed survey addressing some of the complexity in the cancer experience and how ringing a bell helps, or doesn’t.
  3. Focus. A clearer understanding of what the bell means may emerge by studying more homogeneous groups. Collaborate to accrue many patients for a single disease state (e.g. patients with luminal subtype breast cancer receiving chemotherapy, or post-prostatectomy salvage radiation, or with HPV+ oropharyngeal cancer receiving concurrent chemoradiation).
  4. Standardize framing. In a trial setting, there needs to be a defined way to describe use of the bell consistently. Stop the military metaphors unless there are published surveys of patients’ experiences suggesting those are actually helpful.
  5. Inclusion. Ethnic and/or cultural differences may emerge, but only if we include everyone. Studies should also include educational level and account for potential influence of financial toxicity on distress.
  6. Standardize timing. Consider sending follow-up surveys to the patient at home a set time period after treatment. If you really want to test bell versus no bell, conduct the study in separate years with the same time frames (e.g. summer/fall of 2020 with bell, summer/fall of 2021 without).
  7. Bell location. Seriously think about whether the bell should be in a public or private space within your department. In a trial setting, it should be standardized location if possible in a multicenter study.

What to Do for Now?

Based upon this study and the conversations taking place, I will continue to do what I have done discussing the bell with my patients—nothing. It has not come up in conversations between me and my patients in over a decade. But it is stimulating a discussion among our staff to review how they discuss using the bell. We’ll make sure we try to minimize the chance we’re making it harder on patients, but our team already did some of the things I listed above.

For now, I would favor telling patients receiving radiation that they define the value of ringing the bell. Not us. It’s not for victory, or as a milestone in a journey. Empower patients to choose what it means, so that the bell tolls for them if they want, the way they want.


  1. Williams PA, Hu J, Yang D, et al. The cancer bell: too much of a good thing? Int J Radiat Oncol Biol Phys. 2019;105:247-53.
  2. Leach CR, Vereen RN, Rao AV, et al. Impact of individual- and area-level race/ethnicity on illness intrusiveness among cancer survivors. Transl Behav Med. 2019;9:1208-15.
  3. Yee MK, Sereika SM, Bender CM, et al. Symptom incidence, distress, cancer-related distress, and adherence to chemotherapy among African American women with breast cancer. Cancer. 2017;123:2061-9.
  4. Rivas C, Matheson L, Nayoan J, et al. Ethnicity and the prostate cancer experience: a qualitative metasynthesis. Psychooncology. 2016;25:1147-56.
  5. Janz NK, Li Y, Beesley LJ, et al. Worry about recurrence in the multi-ethnic population of breast cancer survivors and their partners. Support Care Cancer. 2016;24:4669-78.
  6. Patel-Kerai G, Harcourt D, Rumsey N, et al. The psychosocial experiences of breast cancer amongst Black, South Asian, and White survivors: do differences exist between ethnic groups? Psychooncology. 2017;26:515-22.
  7. Kwok W, Bhuvanakrishna T. The relationship between ethnicity and the pain experience of cancer patients: a systematic review. Indian J Palliat Care. 2014;20:194-200.
  8. Saunders CL, Abel GA, Lyratzopoulos G. Inequalities in reported cancer patient experience by socio-demographic characteristic and cancer site: evidence from respondents to the English Cancer Patient Experience Survey. Eur J Cancer Care. 2015;24:85-98.
  9. Pidduck W, Wan BA, Zhang L, et al. Seasonal fluctuations in psychological distress amongst women diagnosed with early breast cancer receiving radiotherapy. Psychooncology. 2019;28:2091-7.
  10. Gale KJ. Inappropriate celebration in the presence of patients with poor prognosis – the cancer center victory bell. JAMA Oncol. 2019;5:146-7.


The ideas and opinions expressed on the ASCO Connection Blogs do not necessarily reflect those of ASCO. None of the information posted on is intended as medical, legal, or business advice, or advice about reimbursement for health care services. The mention of any product, service, company, therapy or physician practice on does not constitute an endorsement of any kind by ASCO. ASCO assumes no responsibility for any injury or damage to persons or property arising out of or related to any use of the material contained in, posted on, or linked to this site, or any errors or omissions.

Back to Top