Patient Experience of Thyroid Cancer Active Surveillance in Japan

Key Points

Question

What is the burden of concern for patients with thyroid cancer who are under active surveillance?

Findings

In this study involving 243 patients with thyroid cancer on surveillance, more than a third of survey respondents rated their worry about cancer as occurring sometimes or more; by 3 years on surveillance, 25% said they did not worry at all, up from 14% at diagnosis. Key sources of worry were cancer growth and spread, the potential need for surgical intervention at a later time, and challenges with interpreting bodily experiences near the cancer location.

Meaning

These results are similar to those for patients whose thyroid cancer is managed with surgical intervention, suggesting that cancer worry is not necessarily prohibitive to successful active surveillance in thyroid cancer.

Abstract

Importance

The burden of concern for patients with thyroid cancer who undergo surgical intervention with or without radioactive iodine is known to be substantial. For patients under active surveillance, this aspect of the patient experience has not been described to date and could be a potential barrier to broader acceptance of surveillance as a cancer management strategy.

Objective

To describe the experiences of patients in the longest-standing and largest thyroid cancer active surveillance program.

Design, Setting, and Participants

This study used a mixed method of survey, semistructured interviews, and field observation and was conducted at Kuma Hospital in Kobe, Japan. The survey was administered from September 4, 2017, through October 18, 2017, and the field observation was conducted from August 28, 2017, to October 20, 2017. Survey participants were a consecutive sample of 249 patients under active surveillance who were attending the hospital for a surveillance visit, and the semistructured interviewees were a subset of 21 patients. The English-language survey instrument was translated by native Japanese speakers, back-translated into English, and then further refined by a panel of Japanese speakers with expertise in health research.

Main Outcomes and Measures

Survey and interview responses and field observations.

Results

In total, 249 surveys were distributed to patients with thyroid cancer on active surveillance. Two hundred forty-three patients (97.6%) completed the survey. Among the respondents, 195 (80.2%) were female and 20 (8.2%) were male (28 [11.5%] responses were missing). Among the subset of 21 patients who participated in the semistructured interview, 3 were male (14.3%), and the mean (range) age was 64 (32-85) years. Thirty-seven percent rated the frequency of cancer worry as occurring sometimes or more. Thirty-two percent said their worry affected their mood somewhat or a lot. Fourteen percent reported that their worry affected their ability to carry out daily activities somewhat or a lot. Cancer spread, later need for surgical intervention, and difficulty with interpreting bodily experiences in the general location of the cancer were among the main sources of worry. Most respondents (60.0%) said their worry was less than it was when they first found out about their cancer. By 3 years after diagnosis, the proportion of participants who reported they were not at all worried increased from 14% (95% CI, 12%-16%) to 25% (95% CI, 23%-26%). Eighty percent (95% CI, 79%-81%) of respondents agreed or strongly agreed that their decision to do active surveillance matched their personal values, and 83% (95% CI, 82%-84%) agreed or strongly agreed that choosing active surveillance was the best decision for them personally. Most patients (77%) had not heard of active surveillance before they were offered the option.

Conclusions and Relevance

Cancer concern was common among patients with thyroid cancer under active surveillance, which is comparable to the worry among actively treated patients. Levels of cancer worry reported by patients under active surveillance decreased over time, and patients expressed satisfaction with their disease management decision. These findings suggest that the possibility of cancer worry should not be viewed as prohibitive to successful active surveillance in thyroid cancer.

This study evaluates the concerns about cancer, treatment decisions, and overall experience of Japanese patients under active surveillance for thyroid cancer.

Introduction

Across the world, the incidence of small papillary thyroid cancers is dramatically increasing. In locations with access to sensitive medical testing as well as cultural mores and incentives to use it, small thyroid cancers now account for up to 50% (or more) of detected cases.^1,2,3 Papillary thyroid cancer is a disease known to have a subclinical reservoir; that is, although commonly present, it often remains asymptomatic throughout the lifespan of people who go on to die of other causes.⁴ Active surveillance, a strategy for managing small asymptomatic cancers that are unlikely to progress, involves periodic imaging or testing instead of immediate surgical intervention. First explicitly described as an approach for prostate cancer, active surveillance for cancers that are unlikely to cause harm can decrease patients’ exposure to a treatment’s adverse effects and the potential complications of cancer treatments.^5,6,7 Given the dramatic increase in small papillary thyroid cancer incidence worldwide, active surveillance as a management strategy is gaining attention.

Kuma Hospital in Kobe, Japan, is the site of the longest-standing and largest cohort of patients on active surveillance for papillary thyroid microcarcinoma. Since 1993, selected patients with papillary thyroid cancers measuring 1 cm or smaller with appropriate clinical characteristics have been offered the opportunity to have their cancers monitored through serial ultrasonography examination, rather than to undergo immediate thyroidectomy. Beginning with publications in 2003, researchers at Kuma Hospital and at the Cancer Institute Hospital in Tokyo have shown that this approach is safe. Among the patients on surveillance for up to 10 years, 8% of patients at Kuma Hospital and 7% of patients at the Cancer Institute Hospital showed tumor size enlargement, whereas 3.8% of patients at Kuma Hospital and 1% of patients at the Cancer Institute showed novel lymph node metastases.^8,9,10,11 All patients with increased tumor size or novel lymph node metastases were successfully treated, and none showed substantial recurrence or died of their thyroid cancer.

Active surveillance is now acknowledged as a potential management strategy in the American Thyroid Association Management Guidelines for Adult Patients with Thyroid Nodules and Differentiated Thyroid Cancer,¹² but this is the exception rather than the rule. Two other international guidelines (British Thyroid Association Guidelines for the Management of Thyroid Cancer and American Association of Clinical Endocrinologists, American College of Endocrinology, and Associazione Medici Endocrinologi Medical Guidelines for Clinical Practice for the Diagnosis and Management of Thyroid Nodules) either recommend against it unless the patient is not a surgical candidate or do not address the approach.^13,14,15 A key concern that limits uptake is patient acceptability of living with their cancer rather than undergoing a surgical intervention. Thyroid cancer prognosis is generally excellent, but patients with thyroid cancer have survivorship burdens of concern (for example, worry) that are equal to or greater than the burdens of survivors of other cancers. Many may wonder if the burden of surveillance may be even greater than that experienced after treatment. Using a mixed-methods approach of survey, semistructured interview, field observation, and audio recording of physician practices offering active surveillance, we conducted a study to understand this burden of concern. We report the patient experience of active surveillance in Japan, along with key contextual information necessary to interpret the findings outside of the Japanese setting.

Methods

This study was approved by the Ethics Committee at Kuma Hospital and the Dartmouth College Committee for the Protection of Human Subjects. For the survey, completion of the items after reading the patient information sheet served as the informed consent. For the interviews and audio recordings, patients completed a written informed consent administered by the native Japanese-speaking members of our research team (M.F., Y.I., or A.M.).

The research was conducted at Kuma Hospital. Located in the Kansai area, the hospital is historically known for its excellence in the care of women’s health problems, particularly breast and thyroid disorders. Opened in 1932, Kuma Hospital is now nationally recognized for excellence in thyroid care. Today, the patient distribution of Kuma Hospital is primarily people presenting with thyroid disorders. Currently, 8 surgeons perform approximately 40 to 50 thyroid operations weekly.

Survey

Sample Selection

The goal of sampling was to obtain a representative sample of current patients on active surveillance at Kuma Hospital. For a 6-week continuous period, we surveyed consecutive patients attending the hospital for a surveillance visit. The survey was administered from September 4, 2017, through October 18, 2017, and the field observation was conducted from August 28, 2017, to October 20, 2017. The goal was a minimum of 100 completed surveys to obtain a confidence level of 9% for dichotomous survey-item outcomes. As of 2013, the hospital had 1200 patients on surveillance, which would lead to an estimated 150 patient visits in 6 weeks, and this number was expected to have grown by the time our survey instrument was administered in 2017.

Survey Instrument Development and Administration

We began with an English-language version of a survey instrument being used for a related study in the United States. This instrument predominantly used validated items from established scales. It included 9 items related to perceived cancer risk and risk of death from cancer adapted from similar validated items,^{16,17,18,19,20} 3 items on cancer-related worry,²¹ 6 items on influencing factors in the choice of treatment,²² 6 items on satisfaction with decision,²³ 1 item on control preferences,²⁴ 4 items on decisional conflict,²⁵ 5 items on decisional regret,²⁶ and 5 items on uncertainty in illness.²² We asked the local Japanese senior investigator (A.M.) to assess the acceptability of the items in the local setting. Most questions related to perceived cancer risk and risk of death from cancer as well as those related to decisional conflict and regret were identified as problematic. These items were removed, except for 1 item on perceived risk of death from thyroid cancer, which was modified to change the word death to dire.

The items were translated into Japanese by the local investigators and then were back-translated by an American professor of Japanese literature who had no previous exposure to the instrument. This process revealed translation issues with the items on uncertainty in illness that could not be easily solved; thus, these items were removed. Further refinements to the remaining items were made, by 2 native Japanese speakers with excellent English-language skills and expertise in health outcomes research, to finalize the instrument for local use.

The final survey instrument contained 4 questions about cancer worry, 6 about decision satisfaction, 1 about cancer risk, 1 about control preferences, 6 about influencing factors in the choice of treatment, an open-response area, and 4 demographic questions. We report the responses to cancer worry and decision satisfaction items, comments in the open-response area, and the demographic information. All patients who come to Kuma Hospital for active surveillance appointments go to a single reception desk and gather forms to fill out as they wait for their blood test, ultrasonography, and visit with the physician. During the study period, every patient who arrived for a surveillance visit according to the day’s appointment list was provided with a paper copy of the sequentially numbered survey instrument to complete and return to reception along with their other forms.

We calculated the survey response rate by subtracting the number of completed surveys from the number distributed. Two hundred forty-nine surveys were distributed to the appropriate patients as they arrived for their appointments, and 243 were completed and returned, for a 98% response rate. Patients were asked to complete the survey in its entirety and were allowed to skip questions they did not understand or did not want to answer. Survey responses were directly entered into a spreadsheet (Excel; Microsoft Corp) for analysis.

Statistical Analysis

The study design was a consecutive sample and captured approximately 10% of patients on surveillance as of fall 2017. All analyses were performed using STATA statistical software, version 15 (StataCorp LLC). We calculated descriptive statistics to summarize the demographic findings. All survey items were categorical or ordinal, and they were analyzed by declaring the data set to be survey data and calculating proportions with 95% CIs. Because men made up approximately 10% of the study cohort, the analysis of the survey items was weighted for sex to accommodate the fact that the sample contained only approximately 9% men during the study period.

Semistructured Interviews

Sample Selection and Interviews

We sought a typical sample of patients under active surveillance.²⁷ A typical sample is one in which participants have had an “ordinary or usual experience of a phenomenon.”^28(p688) The goal was thematic saturation along the main interview schedule domains. Patients spend a few hours at the hospital on the day of their active surveillance visits, most of it waiting for testing, results, or visit with the physician. During the study period, the local Japanese investigator team approached patients under active surveillance who were at the hospital for a visit and had been under surveillance for at least 2 years. Patients who agreed to participate were collected from the waiting area and brought to a consultation room reserved for the interviews. The mean (range) interview length was 41 (21-69) minutes.

Data Collection and Analysis

One of us (L.D.) conducted 21 interviews, during which thematic saturation was reached and cognitive probing of survey items was completed. Two staff nurses at the hospital, who were Japanese and native speakers and had English-language proficiency, served as translators. All interviews were digitally recorded. The main interview schedule domains were as follows: how the patient’s cancer was identified (mode of detection), what it was like to live with a cancer under active surveillance, how patients handled their emotions, and their experience in the active surveillance program at Kuma Hospital. After each interview, the interviewer (L.D.) held a debriefing with the translator to discuss subtle nuances and findings and to identify emerging ideas that should be tested in subsequent iterations of the interview schedule until thematic saturation was reached. As results from the survey accumulated, the items that had low response rates or unexpected response patterns were examined and added to the interview schedule. Using cognitive probe interviewing technique, we queried interviewees to learn how they interpreted the items and compared that with the items’ intended meanings.²⁹

The interview audio recordings were transcribed and translated in the United States by a team of 3 persons with both American and Japanese cultural and language proficiency. The transcripts were coded and analyzed (by L.D.) using thematic analysis techniques. The transcription and translation team and first author (L.D.) reviewed the coding and thematic analysis findings in detail during research meetings to identify nuances that might be missed without explanation and to serve as a check on the coding and thematic analysis findings.³⁰

Field Observation

To understand the details of how the Kuma Hospital active surveillance program was implemented in terms of patient choice and decisional participation, we interviewed physicians to determine how they offered active surveillance to their patients. These data were supplemented with audio recordings of physician patient visits at which active surveillance was offered. Together, these data were used to triangulate and add depth to survey items related to patient participation in and satisfaction with decision making.

Results

In total, 249 surveys were distributed to patients with thyroid cancer on active surveillance. Two hundred forty-three patients (97.6%) completed the survey. Among the respondents, 195 (80.2%) were female and 20 (8.2%) were male (28 [11.5%] responses were missing). The survey respondent demographics (Table 1) closely reflected the previously published characteristics of those who receive care at Kuma Hospital and who have been offered active surveillance.³¹ The sex makeup was predominantly female, which is consistent with the overall patient attendance by sex at this hospital. Most patients were older than 50 years, reflecting those selected as appropriate candidates for active surveillance. (Cancers of younger patients were found to be more likely to grow than cancers of older patients.¹¹)

Table 1. Demographic Characteristics of Survey Respondents.

Characteristic	Survey Respondents, No. (%)a
No. of surveys distributed	249
Overall response rate	243 (97.6)
Sex
Male	20 (8.2)
Female	195 (80.2)
Missing	28 (11.5)
Heard of active surveillance prior to coming to Kuma Hospital
Yes	27 (11.1)
No	187 (76.9)
Missing	29 (11.9)
Age group, y
19-30	3 (1.2)
31-40	11 (4.5)
41-50	29 (11.9)
51-60	52 (21.4)
61-70	65 (26.7)
71-80	49 (20.2)
≥81	9 (3.7)
Missing	25 (10.3)
Years on active surveillance
≤23 mo	24 (9.9)
24-35 mo (2 to <3 y)	49 (20.2)
36-47 mo (3 to <4 y)	40 (16.5)
48-59 mo (4 to <5 y)	31 (12.8)
60 mo-119 mo (5 to <10 y)	59 (24.3)
≥120 mo (≥10 y)	14 (5.8)
Missing	26 (10.7)

^aPercentages may not add up to 100 because of rounding.

Among the subset of 21 patients who agreed to participate in the semistructured interview, 3 were male (14.3%). The mean (range) age of interviewees was 64 (32-85) years, and the mean (range) number of years on surveillance was 6 (2-17) years.

At the time of the survey, 73 patients (30.0%) had been on surveillance for less than 3 years, 71 (29.2%) for 3 to 5 years, and 73 (30.0%) for more than 5 years. An initial analysis of the survey item responses by these tertiles showed little numerical difference for those who had been under surveillance for more than 3 years. Thus, the subgroup analyses were dichotomized to groups comprising patients who had been on surveillance either less than 3 years or more than 3 years.

According to R. M. Tuttle, MD, Memorial Sloan Kettering (email communication, August 2018), unlike the largest US cohort of patients under active surveillance, most patients (77%) at Kuma Hospital had never heard of active surveillance prior to coming to the hospital for their care. In semistructured interviews, some patients reported they had been offered the option of surgical intervention or active surveillance, and others reported they were simply told, “they did not need an operation, because their cancer was very small.” These descriptions were confirmed by triangulation with the audio recordings of patient visits at which active surveillance was offered and with the information obtained from physician interviews. Some physicians said they always offered the 2 options, whereas others said they suggested 1 specific option. Published data on physician uptake of active surveillance at Kuma Hospital also confirms this information.³²

To provide a cultural context of health care–seeking behavior in this Japanese setting, we analyzed interviewee descriptions of how their cancer was found to classify the mode of detection.³³ The most common mode of detection was requested screening. In Japan, virtually everyone has an annual health checkup, offered either through a person’s employer or through the local government’s public health system as part of the national health care system. However, these health checkups generally do not include thyroid ultrasonography, and people need to seek this service specifically. Retail offerings of screenings or imaging studies are common, including highly sophisticated tests such as fluorodeoxyglucose positron emission tomographic scans. Twelve of 21 interviewees (57.1%) said their cancer was found through a screening: 3 underwent a screening physical examination that prompted ultrasonography, and 9 had elective retail (not endorsed by guidelines) thyroid ultrasonography, carotid ultrasonography, or fluorodeoxyglucose positron emission tomographic scan. The rest of the patients’ cancers were identified through diagnostic cascade,³³ a medical workup for vague symptoms that uncovers a thyroid cancer that does not explain the presenting symptoms (8 interviewees), or radiologic serendipity,³³ an incidental identification of cancer on diagnostic imaging that was done for other purposes (1 interviewee).

Cancer Worry

Survey respondents (N = 243) endorsed worrying about their cancer (Table 2). Thirty-seven percent of survey respondents rated their worry about cancer as occurring sometimes or more, 32% said it affected their mood somewhat or a lot, but only 14% said it affected their ability to carry out daily activities somewhat or a lot. Representative quotations suggest that patients with other health issues in addition to their cancer had more worry than patients who did not (Box). For example, some people with other illnesses, such as depression, or who used medications that might affect immune function described more worry, whereas those who had few illnesses and no worrisome bodily experiences described less worry. Most respondents (60.0%) said their worry was less than it was when they first found out about their cancer. Consistent with this finding, the result from the subgroup analysis regarding length of time under surveillance showed improvements in the proportion of respondents reporting that they were not at all worried 3 years after diagnosis, increasing from 14% (95% CI, 12%-16%) to 25% (95% CI, 23%-26%).

Table 2. Worry About Thyroid Cancer Among 243 Patients on Active Surveillance.

Question and Response	Overall, %a	Weighted Sample, % (95% CI)
		<3 y on Active Surveillance	≥3 y on Active Surveillance
Looking back on your feelings from the past month: How much have you worried about your thyroid cancer?
Not at all	21	14 (12-16)	25 (23-26)
Rarely	42	42 (39-45)	42 (40-44)
Sometimes (or more)	37	44 (41-47)	33 (31-35)
How much has your worry affected your mood?
Not at all	24	18 (16-20)	27 (25-29)
A little	44	44 (42-47)	44 (42-46)
Somewhat or a lot	32	37 (35-40)	29 (27-30)
How much has your worry affected your ability to perform your daily activities?
Not at all	38	35 (32-38)	40 (38-42)
A little	48	44 (41-47)	50 (47-52)
Somewhat or a lot	14	21 (19-23)	10 (9-12)
Compared to when you first found out about your cancer: What is your level of worry about your cancer now?
Greater than it was	10	10 (8-12)	9 (8-11)
Less than it was	60	55 (52-58)	63 (61-65)
About the same as it was	30	35 (33-38)	27 (25-29)

^aSurvey result percentages are weighted estimates to account for the distribution of sex in the sample compared with the population under surveillance at Kuma Hospital. The sample size was 249, and respondents were 91% female and the population under surveillance consisted of 90% females. Percentages may not add up to 100 because of rounding.

Box. Representative Qualitative Survey and Interview Responses on the Experience of Worry.

Specific Response Ordered Broadly From High Levels of Worry to Low

• I worry about it metastasizing to another place. (Survey respondent #11)

• I check my neck frequently and if something is different, I go to see the doctor soon. (Interviewee #3)

• Breaking up the calcification in order to confirm the diagnosis means that the fragments then scatter into the thyroid.… I worry that the fragmented pieces from the cancerous area might lead to spread in the future. (Survey respondent #9)

• Ten years ago I was diagnosed with depression and it didn’t seem to be getting any better; I looked up the symptoms on the internet and myself, thought that maybe it was related to my thyroid, so I came to the Kuma Hospital to have blood work done… the 2 ~ 3 y after the diagnosis was a time in which I was psychologically depressed and full of anxiety. When I had checkups the doctors’ exam and their explanations allowed me to go home feeling secure. Though it was not where the cancerous cells were, at times I felt a throbbing swelling in the area around the thyroid, and regardless of when my next appointment was scheduled, I would just go on into the hospital. Then, too, I would be nervous until I was examined, but the doctor’s examination helped me relax. (Survey respondent #86)

• I have rheumatoid arthritis.… I get an injection of a biological preparation every two weeks. The biological preparation medication I take makes me worried about the thyroid cancer. (Survey respondent #110)

• When my cancer was discovered last year in January it was 9 mm but it is now 11 mm…. My doctor encourages me to have surgery if it grows to over 12 mm. Since it has grown 2 mm in the course of one year and eight months, I worry that about this eventuality.… An acquaintance of mine had thyroid cancer that metastasized; they died from that. I hope to spend my days with the peace of mind coming from surveillance. (Survey respondent #114)

• I have complete confidence in my doctor here. It was all quite a shock at first but now it does not worry me too much as I go about my everyday life. I look forward to continued good care. (Survey respondent #129)

• I eat fine and my body feels fine. I don’t really worry about it. Even now, the head doctor says I’m ok, I’m just carrying on with my daily life. (Interviewee #11)

Among interviewees, the most common source of worry was that the cancer would grow or spread, with some patients also expressing concern that the cancer was near their vocal cords. They expressed worry that they someday might need surgery, which could entail risks: “I worry I will need to have surgery, and if the cancer is bigger, I will not be able to talk; I will lose my voice.” Interviewees also described the challenge of interpreting bodily experiences, knowing they had cancer: “My tumor is small, so I’m not too worried about it. But, as I get older and older, eating is hard. I can’t eat big things because it’s hard to swallow, and if the cancer is getting bigger, that would also make it harder to swallow.” Most interviewees said they were glad they knew about their cancer.

Decisional Comfort With Active Surveillance

The survey contained items to assess patient comfort with and participation in decision making. In interviews, a few respondents described this group of items as difficult because they did not see what their personal values or ideas about what was best for them had to do with their decisions about cancer. They saw cancer as a medical problem that fell under the purview of their doctors. This perception might be hypothesized as part of the reason for the 6% to 9% nonresponse rate for this group of survey items. Among the 91% to 94% of patients who did respond to these questions, most believed they were informed (87%; 95% CI, 85%-88%), they had made the best decision for them personally (83%; 95% CI, 82%-84%), their decision to do active surveillance was consistent with their personal values (80%; 95% CI, 79%-81%), and the decision was theirs to make (79%; 95% CI, 78%-80%) (Table 3).

Table 3. Decision Satisfaction Among 243 Patients on Active Surveillance.

Survey Item	Response Rate, %	Weighted Sample, % (95% CI)a
		Strongly Agree or Agree	Uncertain	Disagree or Strongly Disagree
I am satisfied that I am adequately informed about the issues important to my decision.	94	87 (85-88)	12 (11-13)	1 (1-2)
The decision I made was the best decision possible for me personally.	93	83 (82-84)	17 (16-18)	0
I am satisfied that my decision was consistent with my personal values.	93	80 (79-81)	19 (18-20)	1 (0.6-1.4)
I am satisfied that this was my decision to make.	91	79 (78-80)	18 (17-20)	2 (1.5-2.5)

^aSurvey result percentages are weighted estimates to account for the distribution of sex in the sample compared with the population under surveillance at Kuma Hospital. The sample size was 249, and respondents were 91% female and the population under surveillance consisted of 90% females. Percentages may not add up to 100 because of rounding.

Discussion

Cancer worry was common among patients with thyroid cancer on active surveillance at Kuma Hospital. However, by both self-report and cross-sectional analysis, their levels of worry improved over time. Overall, 37.0% percent of survey respondents said they worried at least sometimes, and 31.7% said it affected their mood somewhat or a lot. Most respondents (60.0%) said their worry was less than it had been when they first found out about their cancer. Three years after the cancer diagnosis, the proportion of patients who said they did not worry at all rose from 14% to 25%. While worry was common, respondents largely expressed satisfaction with their decision. Most interviewees said they were glad they knew about their cancer. This comment follows logically from the finding that 12 of the 21 interviewees’ cancers were found on elective retail screenings, which patients sought and paid for, and echoes the US studies on cancer screening enthusiasm.³⁴

These levels of worry mirror those of treated patients. In a study of thyroid cancer survivors in Canada, drawn from the broad membership of the Thyroid Cancer Canada patient support group, participants rated their worry about health as 2.77 of 4 (in which 1 was the least worry and 4 was the most worry). Worry about health decreased slightly (to 2.66) among long-term survivors. Compared with a reference of mixed-population cancer survivors, Bresner et al³⁵ found thyroid cancer survivors had slightly higher levels of health worry (health worry in the mixed population was 1.73 overall and 1.86 among long-term survivors). Although useful as a general comparator, this study included patients with any type of thyroid cancer, and 25% reported they were currently experiencing possible or definite evidence of recurrence. In addition, Bresner et al³⁵ published separately about the experience of patients with low-risk thyroid cancer, and more than one-third of respondents said they were somewhat or very concerned about possible long-term adverse effects of cancer treatment and disease recurrence. Almost half of respondents worried about the long-term effects of cancer on their health (28% of this cohort had received radioactive iodine treatment).³⁶

To our knowledge, this study is the first report about worry among patients with thyroid cancer on active surveillance; thus, examining comparative data from other diseases is informative. Prostate cancer is an excellent comparator because active surveillance has been under study for this disease since the 1980s.⁵ Published reports of outcomes date back to 2002,³⁷ and patient-reported outcomes of both treated and surveilled patients with prostate cancer have been extensively examined. A 2015 systematic review showed that quality of life, determined with standard measures, was no worse in patients undergoing active surveillance than among patients who chose to undergo active management.³⁸ Although quality of life is not exactly the same as worry, it is conceptually similar. In addition, data published in 2016 on 10-year outcomes from the ProtecT study, a randomized clinical trial comparing intervention to active surveillance for early-stage prostate cancer, showed no difference in anxiety, depression, or SF-12 Health Survey mental health component scores between active surveillance and active intervention groups.³⁹

Limitations

A reasonable concern about the findings in this study is whether the results are applicable to other cultural settings. To our knowledge, no long-term active surveillance of patients with thyroid cancer exists in other countries to enable such a comparison. However, we believe these findings about how active surveillance affects overall patient experience are potentially generalizable to settings outside of Japan. Findings from this study are similar to the patient experiences described earlier, which were from studies about both thyroid cancer and prostate cancer; included data from Canada, England, Europe, and Australia; and included high-quality levels of evidence (data from systematic reviews and randomized clinical trials). For maximal confidence in the results, our findings would need to be confirmed in a randomized clinical trial.

Translating surveys from one culture and language to another is difficult, and it is a potential limitation of this study. We used robust techniques consistent with the World Health Organization recommendations.⁴⁰ We used local Japanese language speakers to translate the survey, had it back-translated by a native English-language speaker fluent in Japanese, and made further revisions with an additional panel of native Japanese speakers. Because we were not able to perform cognitive interviews and final revisions prior to field use, we used a mixed-methods approach in survey administration, combining the survey instrument with semistructured interviews to elicit detailed respondent interpretations of the meaning and personal experiences related to the survey items. This study design not only provided more robust data but also served as a check on the face validity of the items in this particular cultural context. These findings led us to remove from final analysis those items that did not function as intended.

We surveyed roughly 10% of the total population of patients with thyroid cancer under active surveillance at Kuma Hospital. Thus, the results may not be representative of the entire group. However, consecutive patients attending their follow-up visits were surveyed, and the response rate was 97.6%. Patients schedule their own surveillance visits, and we had no reason to believe that patients who came for their appointment during the 6-week period of the survey were substantially different from patients who had scheduled their visits outside of that time.

Conclusions

The role of the physician, and medical care more broadly, is to relieve suffering. It is reassuring to find that the burden of concern, such as worry, is similar between patients who are treated and those who are surveilled, but the observation that one-third of patients with thyroid cancer receiving long-term active surveillance harbor worry that affects their mood deserves attention. Research on patients with thyroid cancer who receive treatment shows that, in general, they have similar or slightly lower quality of life than normative populations. Specific aspects of their health experiences are the key contributors to lower quality, such as hormone withdrawal for remnant ablation or follow-up procedures and short- and long-term adverse effects from radioactive iodine treatment.^41,42,43 Many patients believe that care could be improved by explicit offer or availability of a patient support group, referral to psychological care, or advice on handling concerns about cancer recurrence. Currently, most patients do not receive these services.^44,45,46 Findings from this study will provide useful benchmark data for future studies of whether and how much these services improve the burden of concern from a cancer diagnosis.

Cancer worry is common among patients with thyroid cancer who are under active surveillance, but this worry improves over time. The study findings are comparable to published reports of worry among patients who underwent a surgical intervention with or without radioactive iodine treatment. Given this finding, cancer worry should not necessarily be viewed as uniformly prohibitive to successful active surveillance in thyroid cancer. Regardless of the disease management method chosen, the burden of cancer worry that patients describe is troubling. Our findings suggest that efforts to improve the survivorship experience of this patient population should include both those whose first management choice is surgical intervention and those whose first management choice is surveillance.