Health-Related Quality of Life After Diagnosis and Treatment of Differentiated Thyroid Cancer and Association With Type of Surgical Treatment

Key Points

Questions  What health-related quality of life outcomes are associated with diagnosis and treatment of differentiated thyroid cancer and do they vary by type of surgery received?

Findings  In this content analysis of survey responses of 1005 patients with differentiated thyroid cancer, 775 (77.1%) reported issues in their health-related quality of life after diagnosis and current treatment regimens for differentiated thyroid cancer. These issues are more prevalent in those who received a total thyroidectomy (without neck dissection) compared with those who received a hemithyroidectomy.

Meaning  Given the recent concerns regarding overdiagnosis and overtreatment in thyroid cancer, hemithyroidectomies offer fewer adverse effects of treatment and better health-related quality of life outcomes than total thyroidectomies for patients with differentiated thyroid cancer.

Importance  Concerns around possible overdiagnosis and overtreatment of differentiated thyroid cancer (DTC) have been raised. Issues concerning health-related quality of life (HRQOL) after diagnosis and treatment of DTC are understudied in this patient group.

Objective  To better understand the range of HRQOL outcomes, including possible adverse effects of treatment, associated with diagnosis and treatment of DTC and whether these outcomes vary by type of surgery received.

Design, Setting, and Participants  This content analysis assessed responses to an open-ended question about outcomes and concerns after DTC diagnosis and treatment among patients ascertained from the major postsurgical thyroid cancer treatment center and the population-based Cancer Registry in Queensland, Australia. Participants were aged 18 to 79 years and recently diagnosed with throid cancer. Responses underwent analysis to identify and code emergent themes to describe HRQOL issues and adverse effects of treatment experienced. Quantitative analysis was used to explore whether surgery type was associated with HRQOL issues and/or adverse effects of treatment. Of 1416 eligible patients, 1005 (71.0%) participated. Data were collected from July 1, 2013, through August 31, 2016, and analyzed from January 11 through April 9, 2018.

Main Outcomes and Measures  Issues concerning HRQOL.

Results  The analysis included 1005 patients (72.2% female [n = 726]; mean [SD] age, 52 [14.0 years) with DTC. Most patients were diagnosed with papillary thyroid cancer (889 of 1003 [88.6%]), had tumors smaller than 2 cm in size (564 of 1000 [56.4%]), and received a total thyroidectomy (791 of 1005 [78.7%]). Overall, 775 patients (77.1%) reported HRQOL issues after diagnosis and treatment of DTC. The following 4 main themes emerged from content analysis of patient responses: physical (663 [66.0%]), psychological (187 [18.6%]), lifestyle (82 [8.2%]), and no issue or adverse effect (246 [24.5%]). Patients who had a total thyroidectomy (without neck dissection) were 1.5 times (odds ratio, 1.49; 95% CI, 1.04-2.12) more likely to report an HRQOL issue or an adverse effect of treatment compared with patients who underwent a hemithyroidectomy.

Conclusions and Relevance  According to results of this study, patients diagnosed with DTC report wide-ranging HRQOL issues; these are more prevalent among patients who have total thyroidectomies rather than hemithyroidectomies. For patients with small, localized DTCs, hemithyroidectomy may offer fewer adverse effects of treatment and better HRQOL outcomes than total thyroidectomy. It appears that issues with HRQOL should be considered by patients and physicians when deciding on the best treatment approach after a diagnosis of DTC.

In recent years, a worldwide increase in the incidence of thyroid cancer has occurred.¹ This increase has been driven by an increase in differentiated thyroid cancer (DTC), which includes papillary thyroid (PTC) and follicular thyroid cancers. The increase has been suggested to be at least partly owing to overdiagnosis,² resulting from the widespread use of advanced imaging and increased pathologic scrutiny of thyroid specimens leading to the incidental detection of small, indolent PTCs.³ The identification of these small PTCs has resulted in an increased number of patients diagnosed with thyroid cancer, many of whom may have remained untroubled by the cancer had it not been detected.^4,5

With thyroid cancer being the fifth most common cancer diagnosed in women in the United States in 2017⁶ and the seventh most common cancer diagnosed in Australian women in 2017 (and the most commonly diagnosed cancer in women aged 25-29 years in Australia),⁷ the burden of the diagnosis and treatment on individuals as well as the health care system is significant. Standard treatment for DTC is surgery, which can include a total thyroidectomy (with or without neck dissection as required) or hemithyroidectomy. Surgery may then be followed by radioactive iodine treatment and, for some, treatment with supraphysiologic doses of thyroid hormone to suppress thyrotropin levels. All patients undergoing thyroidectomy and a small proportion undergoing hemithyroidectomy patients^8,9 require thyroid hormone replacement. All these treatments have the potential for adverse consequences.¹⁰ Thyroidectomy carries the risk of vocal cord paralysis and hypoparathyroidism, and these risks are higher for total thyroidectomy vs hemithyroidectomy.¹¹ Recent guidelines now suggest more conservative treatment options for patients diagnosed with DTC, particularly those diagnosed with small PTC, including recommending hemithyroidectomy over thyroidectomy when appropriate and the consideration of an active surveillance management approach in patients who have very low-risk tumors.¹²

Despite the excellent prognosis of DTC, several comparative studies^13-15 have demonstrated that people treated for DTC report poorer health-related quality of life (HRQOL) compared with the general population, as measured through validated questionnaires (ie, THYCA-QOL [disease-specific HRQOL questionnaire for survivors of thyroid cancer], 36-Item Short Form Health Survey, and Quality of Life Questionnaire Core 30). Furthermore, thyroid-related symptoms can negatively affect HRQOL for as long as 14 years after diagnosis and treatment, with fatigue and sleep issues being the most commonly reported complaints.¹⁶ However, previous studies on HRQOL^13-16 using standard questionnaires may not elicit the full spectrum of experiences and concerns of patients with DTC.

In light of issues surrounding the possibility of overdiagnosis and overtreatment of DTC, in particular small PTCs, this study aimed to better understand and assess the range of HRQOL outcomes, including the possible treatment-related adverse effects associated with DTC and whether these outcomes vary by the type of surgery the patient received.

Data were collected as part of the Queensland Thyroid Cancer Study, a population–based case-control study conducted in Queensland, Australia, from July 1, 2013, through August 31, 2016. Only the case participants were included in the current analysis. As part of the study, participants with recently diagnosed thyroid cancer were asked an open-ended question over the telephone about any outcomes of their thyroid cancer diagnosis or treatment that affected their HRQOL and whether they experienced or are experiencing any associated adverse effects of treatment. Their responses were recorded and summarized by a research nurse. Ethical approval for the the study was provided by the Human Research Ethics Committees of the QIMR Berghofer Medical Research Institute, Brisbane, Australia, and the Royal Brisbane and Women’s Hospital and Health Service (RBWH), Brisbane. Written informed consent for the study was obtained from all participants.

A content analysis¹⁷ of the summarized responses was conducted to identify and code emergent themes and thereby capture the overall picture of the HRQOL outcomes and adverse effects of treatment experienced by patients with thyroid cancer. Content analysis is a research method used for analyzing text data that combines qualitative and quantitative methods, allowing for the frequency of categories and the content to be reported.

Eligible participants were residents of the Australian state of Queensland aged 18 to 79 years who were diagnosed with thyroid cancer (excluding medullary cancers and lymphoma) from July 1, 2013, through August 31, 2016. Participants who were unable to complete questionnaires and interviews in English, whose physicians did not give permission for us to contact them, or who were deemed too sick to participate were excluded. Two patients with anaplastic thyroid cancer were excluded from the current analysis to limit the focus to the HRQOL outcomes of DTC only.

Participants were recruited through the RBWH thyroid cancer clinic and the Queensland Cancer Registry (QCR). The RBWH clinic is the only center in Queensland that provides postsurgical radioactive iodine treatment for patients with thyroid cancer, and thus most people diagnosed with thyroid cancer in Queensland undergo assessment at this clinic. At the clinic, potential participants were approached at their first appointment to recruit them as soon as possible after diagnosis. Those who agreed to participate were posted the study documents, including an information brochure, a health and lifestyle questionnaire to self-complete, and consent forms. Potential participants who were not approached at the RBWH clinic were invited to participate via the QCR; notification of all cancer diagnoses to state-based cancer registries is a legal requirement in Australia. A staff member at the QCR sought permission from patient’s physicians to contact the patient. Once permission was obtained, patients were sent a study information sheet and consent form in a reply-paid envelope. Those who consented to contact were then sent the study documents. After participants returned their study questionnaire, they were telephoned by the study research nurse to conduct a structured telephone interview. During this interview the nurse asked the participant to describe the process by which they came to be diagnosed with thyroid cancer. They were then asked an open-ended question about whether and how the diagnosis and treatment had affected their HRQOL or whether they had experienced any adverse effects of treatment.

The self-completed study questionnaire included questions about participants’ sociodemographic characteristics and the Kessler Psychological Distress Scale.^18,19 Detailed information about the participants’ tumors and their surgical treatment was extracted from pathology records.

The coding framework was developed in an iterative process involving 4 members of the study team. Two researchers (B.N. and T.T.) independently reviewed the responses from a random sample of 100 participants to generate an initial list of themes and consolidate salient overarching themes. This process allowed themes to be derived inductively from the data.²⁰ The 2 sets of independently identified major themes were documented and discussed with 2 coauthors (K.M. and S.J.) to arrive at the initial coding framework. Both coauthors further reviewed a sample of 50 responses in conjunction with the initial coding framework to finalize the coding framework. The first 2 authors (B.N. and T.T.) independently coded the same 50 responses, and the level of agreement was tested using a Cohen κ statistic (κ = 0.8, indiciating a strong level of agreement). Any discrepancies were discussed, and minor changes to the framework were made until consensus was obtained.

Data were analyzed from January 11 through April 9, 2018. The final coding was conducted by a single author (T.T.), with 500 responses (approximately 50% of the total data set) double coded by a second author (B.N.). Based on the breadth of responses, a single response could be coded to more than 1 theme. The data were then checked for interrater reliability using the Cohen κ to ensure rigor of the analysis. The pooled Cohen κ was again 0.8, indicating a strong level of agreement between the coders.

Quantitative data analysis was also conducted to explore whether the type of surgical procedure (hemithyroidectomy vs total thyroidectomy vs total thyroidectomy plus neck dissection) was associated with the reporting of HRQOL issues and/or adverse effects of treatment. Logistic regression analysis was conducted comparing those who reported any concern with those who did not report a concern, with adjustment for age, sex, educational level (postsecondary vs secondary school only), time since diagnosis (<6 vs ≥6 months), and radioactive iodine treatment (received, planned, or not received or planned), with odds ratios and 95% CIs estimated for the association between surgery type and HRQOL outcome. All quantitative analyses were conducted in SPSS software (version 24; IBM Corporation).

Overall, 1005 of the 1416 patients eligible for the study (71.0%) participated. Participants were slightly older (mean [SD] age, 52 [14] vs 48 [14] years) and had larger tumors (mean size,16 [interquartile range, 8-30] vs 10 [interquartile range, 5-20] mm) than eligible nonparticipants. Participant characteristics are summarized in Table 1. Most participants were women (726 [72.2%]) compared with men (279 [27.8%]). Most participants were diagnosed with papillary thyroid cancer (889 of 1003 [88.6%]), had tumors smaller than 2 cm (564 of 1000 [56.4%]), and had received a total thyroidectomy (695 of 1005 [69.2%]). The median time between diagnosis and interview was 23.1 weeks (range, 1.9-90.9 weeks) after their diagnosis. Most patients (636 of 982 with staging data available) had stage I disease.²¹ When findings were standardized for age and sex, a higher prevalence of high or very high psychological distress scores (Kessler Psychological Distress Scale) was found compared with the national Australian figures (20.6% vs 11.7%).²²

In relation to HRQOL and adverse effects of treatment, the following 4 main themes emerged from content analysis of the 1005 participant responses: physical, psychological, lifestyle, and no issues or adverse effects of treatment. Within each main theme were various subthemes. Table 2 outlines each theme and provides an example response. Overall 775 patients (77.1%) reported an HRQOL issue after diagnosis and treatment of DTC. Table 3 summarizes the overall frequency of responses across each theme and subtheme and by surgery type (hemithyroidectomy vs total thyroidectomy alone vs total thyroidectomy with neck dissection).

Overall, 663 participants (66.0%) reported their diagnosis had resulted in a physical adverse effect or HRQOL outcomes. This outcome was demonstrated by comments such as “constantly fatigued and struggling with daily activities” and “very frustrated by limited voice” (coded under subthemes of fatigue and voice issues, respectively). Lower energy levels, tiredness, sluggishness, or exhaustion were mentioned in 340 patient responses (33.8%). Difficulties in titrating medication levels was mentioned by 227 responses (22.6%), for example, “still having difficulty in reaching optimal levels of thyroxine.” A number of responses mentioned miscellaneous physical problems (155 [15.4%]), including “brain fog,” “hair loss,” and “having difficulties with poor appetite.”

The second major theme encompassed psychological adverse effects or HRQOL outcomes, with 187 participants (18.6%) reporting concerns relevant to this theme. Overall, 131 patients (13.0%) reported feelings of shock, sadness, fear, frustration, or stress relating to the diagnosis, treatment, or prognosis (eg, “mental anguish and anxiety and shock initially about the diagnosis”), and 39 (3.9%) reported mood issues they had not encountered before (eg, “mood swings more than prior to surgery”). Other responses included more severe changes in their mental health (eg, “had a period of depression with suicidal thoughts at the time, attributed to the trauma of cancer diagnosis and surgery” and “had an anxiety attack”).

The third theme of lifestyle included 82 patient responses (8.2%). Responses expressed, for example, work-related concerns “about ability to cope with full-time work again” and “stress of not knowing when can return to work—uncertainty is hard,” and lifestyle changes such as “to assist in absorption of thyroxine,” “on gluten-free diet,” and “increased probiotics intake.”

Although most participants reported an adverse effect or some HRQOL issue, almost one-quarter (246 [24.5%]) indicated that little had changed overall in their HRQOL and that they had had few or no adverse effects of treatment. Responses in this theme included “feels no different” and “hasn’t had any problems.”

The multivariable logistic regression model (adjusted for age, sex, educational level, time since diagnosis, and use of radioactive iodine treatment) provided strong evidence of a difference in reporting of a change in or concern about HRQOL and adverse effects of treatment (any vs none) between surgery types. Patients who received a total thyroidectomy were 1.5 times more likely to report a change in their HRQOL or an adverse effect of treatment compared with patients who underwent a hemithyroidectomy (adjusted odds ratio, 1.49; 95% CI, 1.04-2.12), whereas patients who received a total thyroidectomy plus neck dissection were 2.3 times (adjusted odds ratio, 2.29; 95% CI, 1.10-4.80) more likely to report an HRQOL issue or an adverse effect of treatment compared with patients who underwent a hemithyroidectomy (Table 4). The estimate of the association of outcomes with surgery type did not change when we restricted the analysis to those patients with tumors smaller than 1 cm or tumors 1 to 4 cm or when we restricted the analysis to patients without a previous cancer diagnosis (other than thyroid cancer).

Our content analysis of the qualitative responses of 1005 patients with DTC among a population-based study of patients newly diagnosed with thyroid cancer found that more than three-quarters of participants reported that their thyroid cancer diagnosis and/or treatment was associated with HRQOL issues, including some treatment-related adverse effects. This finding is consistent with a number of other HRQOL studies^13-15 citing a reduced HRQOL compared with the general population. The most common theme that emerged from patient responses was physical. Among participants reporting physical issues, the most common concern was fatigue, a previously well-documented adverse effect of thyroid cancer treatment.²³ The second most commonly reported physical concern was medication, which has not been effectively captured in previous HRQOL studies. Medication issues encompassed the inconvenience of lifelong medication therapy and comments on adjusting to the timing and storage of levothyroxine sodium.

Emotional distress ranked as the most frequent psychological subtheme reported, including shock, fears, and mental anguish related to diagnosis. Many patients within this subtheme reported a fear of cancer recurrence, which emphasizes the importance of appropriate communication regarding the low risk of progression or recurrence of these cancers. In addition, we found that overall, participants had higher psychological distress scores than current population norms. The association with lifestyle also emerged as a major theme, with patients voicing concerns related to the inability to resume work, similar to findings from a recent study on unemployment risk and decreased income 2 to 4 years after a thyroid cancer diagnosis.²⁴

Guidelines now recommend hemithyroidectomy rather than thyroidectomy for small intrathyroidal cancers, and more limited surgery may also be considered for tumors as large as 4 cm.¹² Furthermore, the same guidelines also support consideration of active surveillance for patients diagnosed with low-risk PTC smaller than 1 cm, and recently published evidence lends additional support to this approach.^25,26 Although treatment-related adverse effects may be reduced in patients undergoing management with active surveillance, our participant group did not include anyone undergoing active surveillance, so we are unable to assess whether HRQOL concerns would be lower in this group. The practice of active surveillance for a thyroid cancer diagnosis is very uncommon in Australia at this time, so no one in the sample of 1005 patients fell into this category. We also did not have information from other potential control groups, including people with thyroid conditions who were unaffected by cancer, so we cannot definitively ascribe participants’ concerns to their treatment pathways or their condition. However, our analysis demonstrated that patients who received a total thyroidectomy alone were 1.5 times more likely to report an HRQOL issue or an adverse effect of treatment than patients who underwent a hemithyroidectomy, and this association did not vary by tumor size. Most participants in our sample were treated with a total thyroidectomy, half of whom had DTC tumors smaller than 2 cm and may have been candidates for hemithyroidectomy or even active surveillance (if their tumors were low risk and <1 cm). Understandably, we also found that those who received lateral neck dissection, a necessary form of treatment for those who had more advanced disease, were much more likely to report HRQOL concerns and adverse effects of treatment than those who received a total thyroidectomy or hemithyroidectomy.

To our knowledge, this study is the first to analyze free-text responses among patients with thyroid cancer to an open-ended question about effects of diagnosis and treatment. The aim was to obtain unprompted and unanticipated responses of HRQOL issues and adverse effects for patients recently diagnosed with thyroid cancer. Use of the well-established method of content analysis allowed for us to make rigorous judgments and valid inferences on the range and frequency of self-reported effects on patients.¹⁷ Furthermore, using the outcome of any issue or concern vs no issue or concern in the multivariable logistic regression model was the most conservative way of analyzing the data, given that some of the reported issues or concerns could overlap the assigned categories. The size and representativeness of the sample is a major strength of the study. Moreover, we used an iterative process of framework development involving a number of reviewers and the strong level of agreement on themes in the initial independent coding of responses.

However, this study has several limitations, including that recorded participant responses were succinct, with approximately less than 50 words per response. Furthermore, the verbal responses were subject to the nurse’s accuracy in recording and summarizing comments. The relative severity of an issue raised by a patient could not be confidently quantified based on our data set, including the researcher’s interpretation of lifestyle issues, some of which may have been seen by the patient as positive. In addition, we did not have specific data on prediagnosis psychological health and thus cannot assess whether higher levels of prediagnostic anxiety or psychological distress were associated with the extent of surgery received. We also do not know whether or not patients were receiving levothyroxine; however, because we had data on surgery type, we can assume that all patients undergoing total thyroidectomy and a small proportion undergoing hemithyroidectomy were receiving levothyroxine.^8,27 Another limitation is that participation in our study was less than 100%. Those who did not take part were slightly younger and tended to have smaller tumors. Thus, those who participated may have had more HRQOL concerns, and as such, we may have overestimated the proportion with HRQOL issues. Nevertheless, our response proportion was 71.0%, and our large population-based sample suggests that HRQOL concerns affect a substantial proportion of patients with thyroid cancer.

Our data demonstrate that patients with DTC whose clinical management is based on current treatment regimens report wide-ranging concerns about HRQOL and adverse effects of treatment. In light of our findings and the recent discussions surrounding the appropriateness of more conservative treatment, hemithyroidectomies may offer less treatment adverse effects and better HRQOL outcomes than total thyroidectomies for patients with DTC, and in particular may help reduce overtreatment in small PTCs.²⁸ Patients and physicians should consider the implications of a thyroid cancer diagnosis and treatment to mitigate any negative outcomes it may have for patients’ HRQOL in the future.

Accepted for Publication: November 11, 2018.

Corresponding Author: Brooke Nickel, BSc (Hons), MIPH, Wiser Healthcare, Sydney School of Public Health, The University of Sydney, Sydney, New South Wales, Australia 2006 (brooke.nickel@sydney.edu.au).

Published Online: January 17, 2019. doi:10.1001/jamaoto.2018.3870

Author Contributions: Drs McCaffery and Jordan are joint senior authors. Ms Nickel and Dr Jordan had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

Concept and design: Tan, Youl, McCaffery, Jordan.

Study concept and design: Nickel.

Acquisition, analysis, or interpretation of data: Nickel, Tan, Cvejic, Baade, McLeod, Pandeya, McCaffery, Jordan.

Drafting of the manuscript: Nickel, Tan, McCaffery, Jordan.

Critical revision of the manuscript for important intellectual content: All authors.

Statistical analysis: Nickel, Tan, Cvejic, Pandeya, McCaffery, Jordan.

Obtained funding: McCaffery, Jordan.

Administrative, technical, or material support: Tan, McCaffery.

Supervision: Youl, Jordan.

Conflict of Interst Disclosures: Drs McCaffery, McLeod, and Jordan reported receiving grants from National Health and Medical Research Council of Australia (NHMRC) during the conduct of the study. No other disclosures were reported.

Funding/Support: The study was funded by project grant APP1047733 from the NHRMC and fellowships APP1092153 (Dr McLeod), APP1029241 (Dr McCaffery), and APP1061341 (Dr Jordan) from the NHMRC.

Role of the Funder/Sponsor: The funder/sponsor had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

Additional Contributions: Andrea McMurtrie, RN, and Emily Dickson, BA, employees of QIMR Berghofer Medical Research Institute, managed and coordinated the data collection and were not compensated outside of their usual salary. The Royal Brisbane and Women’s Hospital assisted in participant recruitment.