Individualized Graves Disease Remission Rates and Chronic Use of Antithyroid Drugs: A Videographic Analysis of Medical Encounters

Introduction

Graves’ disease (GD) is the most common cause of hyperthyroidism, with a lifetime probability of 2.9% for women and 0.5% for men [1]. Because of the many metabolic effects of abnormally elevated thyroid hormone levels, GD is associated with many debilitating symptoms; and over time, GD decreases the quality of life with an impact on physical and social function, mental health, and overall wellbeing. There are three treatments for GD: thyroidectomy, radioactive iodine, and antithyroid drugs (ATDs). These options differ in their efficacy (i.e., remission), safety, convenience, and costs; none is clearly superior to the others for all patients [2]. Yet, remission rate has been identified as the most important attribute that influences treatment decision-making [3].

In a recent discrete choice experiment of 286 patients and 61 clinicians, participants were offered hypothetical treatment options (ATD, thyroidectomy, or radioactive iodine) which differed in rates of remission, severe side effects, permanent voice changes, and hypocalcemia. It was noted that all attributes included in the study significantly influenced treatment choice, yet remission rate was the dominant factor explaining more than one-third of treatment choices in both patients and clinicians [3]. The importance of remission as a decision-making factor is particularly relevant with ATD. Remission rates for ATD are highly dependent on patients’ characteristics, and they vary significantly as opposed to remission rates for surgery and RAI, which are 100% and between 90–100%, respectively [4]. The extent to which clinicians and patients with GD talk about ATD remission rate is unclear.

Decision-making between patients with GD and their clinicians should also include conversations that include all available options. Evidence about the safety of long-term ATD has resulted in many patients using ATD beyond the 18 months recommended length of therapy; in fact, approximately 25% of patients in the representative cohort in the United States are currently on chronic ATD therapy [5]. Therefore, it is important that patients with GD discuss whether chronic ATD use is an option for them.

This study aims to determine how often ATD remission rates are discussed between clinicians and patients, whether these remission rates are individualized by patient characteristics (e.g., degree of hyperthyroidism), and if conversations about treatment options also include chronic use of ATDs.

Methods

We conducted a secondary analysis of 55 video recordings of clinical encounters obtained during a controlled before and after study aiming to assess the impact of a clinical decision aid (GD choice) on the decision-making process. This study had two cohorts: for the first nine months (first cohort), all patients with GD received usual care, and for the following 15 months(second cohort), all patients with GD experienced care supplemented by GD Choice.

GD choice is a paper-based encounter shared decision-making tool that includes information about the three treatment modalities focused on: (i) effectiveness of therapy (when am I going to start feeling better?), (ii) side effects, (iii) the need for thyroid replacement therapy, (iv) cost of therapy, (v) frequency of follow-up during therapy [6]. GD choice does not include the option of chronic use of ATDs and it does not provide an individualized estimate of GD remission, but it includes an estimate of remission for the general population of patients with GD. The original study took place between 2012–2014 at the Thyroid Clinic (with 9 endocrinologists) within the Division of Endocrinology at the Mayo Clinic (Rochester, MN), and included adults with GD who, as judged by their clinicians, needed treatment for this condition.

For this video analysis, we developed an extraction code scheme that included questions to assess: i) if the probability of remission was discussed (remission defined as being euthyroid after stopping ATDs and not needing additional treatment), ii) whether this probability was provided as an average estimate or individualized by patients’ characteristics, and iii) whether clinicians described the option of taking ATD chronically (more than 18 months). Two study members were trained to use the coding scheme and asked to review an initial set of 5 videos to calibrate. Given that there were no disagreements between extractors, only one extractor completed the video-graphic review. We used descriptive statistics to analyze results: categorical variables were reported as frequencies and percentages, whereas continuous variables were reported as mean and standard deviation (SD). Due to the potential effect of GD choice on decision making, we tested for differences between the group that used the tool in the encounter (SDM group) vs. the group that did not (usual care) using χ² tests for categorical and t-tests for continuous variables.

Results

Patients in the 55 medical encounters were mostly female (78%), white (91%), with a mean age (SD) of 42 (14.7), with higher than high school education (71%), thyrotropin level (SD) of 0.1 mIU/L (0.4), thyroid receptor antibody level of 9.8 IU/L (11.6), free thyroxine level of (SD) of 2.8 ng/dl (1.5), and income less than $40,000 (25%). Patients in the SDM group had similar characteristics to patients in the control group.

Of the 55 videos, the probability of remission of GD with ATDs was discussed in 69%. Among these conversations, 3% included estimates of remission that reflected individual patients’ characteristics; the remaining 97% were given an average estimate for the population. Clinicians who used the GD choice had a higher frequency of conversations regarding remission of GD with ATDs (83% vs. 54%, p=0.008). In all but one encounter, the clinician initially brought up the probability of remission and not patients. Conversations about the option of taking ATD chronically, as a treatment option, occurred in 7% of clinical encounters, which did not differ among the SDM or control group.

Discussion and Conclusion

Remission rate is the most important factor in GD treatment decision-making. We found that conversations about GD remission with ATDs happened in 7 out of 10 medical encounters. These conversations mainly occurred when the decision aid (GD choice) was used and included estimates of remission, but these were not individualized by the patient’s characteristics. Furthermore, conversations about treatment decisions rarely included the option of chronic use of ATD. Limitations of this study include potential selection bias since we only examined encounters in which clinicians agreed to be video-recorded. Another limitation is the analysis of a single encounter. Additional conversations about remission or chronic use of ATD could have occurred in subsequent clinician-patient interactions. Finally, the videos were collected between 2012–2014, and it is possible that they may not reflect more contemporary patient-clinician conversations.

This is the first analysis of clinician-patient interactions to estimate whether and how remission rates are discussed when GD treatment options are considered. The gaps in the care for patients with GD shown here demand additional research to support treatment conversations. For instance, clinicians could use available validated scores to individualize conversations about relapse risk and support shared decision-making conversations.