Socioeconomic influence on treatment with liothyronine and desiccated thyroid extract in Denmark

Introduction

Persons treated with levothyroxine (L-T4) describe a decreased quality of life more often than the general population and the explanation seems to be multifactorial. Whether to use routine treatment of hypothyroidism to increase the quality of life has been debated for many years (1), while the use of thyroid hormone has increased worldwide. In Denmark, from 2001 to 2015, an increase of 132% was seen, simultaneously, with a decrease in the thyroid-stimulating hormone (TSH) threshold for initiating L-T4 therapy and approximately 25% started treatment without biochemical hypothyroidism (2). The increase in L-T4 therapy and falling threshold may be due to increasing demand from patients (3). A recent study demonstrates that the probability of starting thyroid hormone therapy despite normal TSH was higher in persons with long education (4). Educational level is a marker of social status and may indicate the ability to access health information and increased argumentative skills (5). Thus, we speculate that these patients were more likely to request off-label treatments in a dialog with the general practitioner (GP).

During the last few decades, the interest in thyroid hormone treatments – not ‘just L-T4’ – has increased, and it has been suggested that some hypothyroid patients could improve their quality of life if treated with liothyronine (L-T3) or desiccated thyroid extract (DTE). However, L-T3 and DTE therapy is still controversial among endocrinologists and GPs, due to the lack of evidence of treatment effect in large-scale randomized clinical trials. Guidelines suggest that L-T4 is standard therapy. If symptoms persist after normalization of TSH, combination therapy with L-T3 may be considered an experimental approach according to the European guidelines (ETA) (1) but not the American guidelines (ATA) (6). Testing and prescribing on patients’ request is known to happen (3) and people with longer education more often seek the medical system (5). Consequently, we speculated that educational level impacts the usage of L-T3 and/or DTE as seen in treatment initiation for L-T4 with normal TSH.

Methods

Results

We identified 180,063 persons who redeemed thyroid hormone replacement prescriptions and ended up including 154,360 persons in the study (see Fig. 1). In short, of the total study population, 3829 (2.48%) were treated with L-T3 and 430 (0.28%) with DTE. The majority taking L-T3 and/or DTE were women (97.3%), and the most common age for this was 40–59 years (57.9%). For further description, see Table 1.

View Full Size

Consort diagram.

Citation: European Thyroid Journal 11, 6; 10.1530/ETJ-22-0149

• Download Figure
• Download figure as PowerPoint slide

Results of logistic regression models 1 and 4 are presented in Table 2. When analyzing whether educational level influenced treatment with L-T3 and DTE, we found that the odds rose with an increasing educational level for medium vs short 1.64 (95% CI: 1.50–1.8) and long vs short 1.95 (95% CI: 1.79–2.13) in the age- and sex-adjusted model. Test for trend: OR: 1.39 (95% CI: 1.33–1.44) P < 0.001) (Table 2). The results only changed marginally in the partly adjusted models (models 2+3) and in sensitivity analyses including disposable income as a covariable (results not shown). The same pattern of increasing odds of treatment with L-T3 or DTE was seen when increasing disposable income was applied as the exposure in the sensitivity analysis.

Further, we found an age-dependent probability of treatment with L-T3 and/or DTE with age 40–49 years at highest odds compared with age 30–39 (OR 2.53 (95% CI: 2.23–2.87)) decreasing towards lower odds with age ending at age 80+ (OR 0.12 (95% CI: 0.09–0.15)) (Table 2). We found a higher OR of a redeemed prescription of DTE or L-T3 after thyroid surgery OR 1.18 (95% CI: 1.06–1.31) and medication for depression OR 1.37 (95% CI: 1.28–1.45). Also, the region of residence played a role as residents in the Capital region had higher ORs of treatment with DTE or L-T3 than residents in the Middle or South regions (Table 2). As for educational level, the results of all other exposures only changed marginally in the adjusted models.

Discussion

In this Danish national register-based cross-sectional study, we found that less than 3% of persons treated with thyroid hormones were treated with either L-T3 or DTE from 2017 to 2020. Compared with short education, there is almost double the odds (OR 1.95) that people with a long education are treated with L-T3 or DTE, illustrating social inequality in treatment. Further, we find that sex, age, former thyroid surgery, region of residence, and medical therapy with antidepressants played a role in the odds of treatment but only a minor role compared to education level.

Treatment with L-T3 or DTE is still controversial. Current guidelines from ETA and ATA are different on the issue of L-T3 supplementary treatment, where ETA is open for a trial period – ATA advises against routine use due to a lack of evidence of benefit (1, 6). Both agree upon not recommending DTE. At the same time, there is a consensus statement between ETA, the British thyroid association, and ATA that more knowledge is needed on the effectiveness and side effects of combination therapy (9); thus, the debate on whether to use L-T3 is ongoing. The discussion on DTE, which most endocrinologists consider obsolete, might be refueled by the recent blinded crossover study on DTE vs L-T4 and L-T3 with 75 patients (10). In this study, the overall result was neutral, but they postulate that the patients having the lowest quality of life had an effect of L-T3/DTE from a subgroup analysis. Currently, there are more neutral/negative trials on LT4 supplementary treatment (DTE and L-T3), but still, many questions are unanswered as stated in the formerly mentioned consensus statement (9).

Initiation of therapy, not recommended as standard treatment in Guidelines, will often be based on a combination of patient requests and the willingness of the doctors to initiate a specific therapy.

Regarding the doctor’s willingness to initiate combination therapy, a recent international questionnaire (THESIS study) asked endocrinologists if they would prescribe L-T3 combination therapy in L-T4-treated hypothyroid patients with persistent symptoms. In Denmark, 71% answered yes (11), which is like Swedish data (78%) (12) but is much higher compared with other European countries, i.e. Italy (40%) and France (26%) (13). The difference in doctors’ willingness could be explained by local guidelines and the availability of the medication. The patient request is related to the national focus of patients’ societies and social media on the possible effects of this specific therapy. In Denmark, high focus and availability of the medication are present, and the Danish Endocrine Society guidelines suggest a trial of L-T3 therapy (but not DTE) in patients with persistent symptoms despite long-term euthyroidism on L-T4.

Sometimes patients persuade the doctors to prescribe other treatments than the treatments the doctors prefer according to guidelines. The known discrepancy between patients’ wish for treatment and doctors wanting to follow guidelines is supported by a qualitative study from the USA among ATA members showing that patients’ wish was reported as a barrier to therapy following guidelines (14). The same mechanism may be the case with L-T3 treatment.

If we believe in a possible effect of L-T3/DTE, the number of patients treated with these drugs is low (<3%) – compared with the estimate that 5–10% may be undertreated with L-T4 alone (15, 16). On the other hand – if the treatment is harmful, many patients are harmed unnecessarily. We lack solid randomized sufficiently powered long-term trials to know whether side effects do exist. There are, however, newer retrospective data that raises concerns with one study showing increased risk of heart failure and possible stroke and another with increased risk of anti-psychotic prescription (17, 18).

It is not a surprise that educational level and sex influence prescriptions of DTE and L-T3, as it was recently described that educational level (as an indicator of cognitive and analytical skills (5)) was a factor in initiating L-T4 treatment with normal TSH (4). The same applies to the fact that women are more prone to DTE and L-T3 therapy and starting therapy with normal TSH (19). We speculate that a similar mechanism exists here.

We found that former thyroid surgery was related to initiating L-T3/DTE. We excluded those who had a total thyroidectomy due to thyroid cancer. By this, we excluded the often-used L-T3 pretreatment before radioiodine ablation or uptake measurement as the explanation. The explanation could be that this group of people are more often seen by endocrinologists who are more prone to start L-T3 treatment than GPs.

We also demonstrated that the region of residence was related to initiating this treatment. The explanation could be that local doctors’ preferences follow key opinion leaders that are more willing to start this treatment and vice versa. Another explanation could be that in some regions more people are getting DTE and/or L-T3 via the internet and are thereby not included in the database. In a previous study done by questionnaire, we estimated that about 5% of the patients in Denmark on L-T3/DTE buy the hormones on the internet (20), meaning that we most likely have data on 95% of the patients.

Conclusion

In this large epidemiological study on more than 154,000 persons treated with thyroid hormones, we discover that the odds of being treated with L-T3 or DTE almost doubled with the longest educational level compared with the shorter and that approximately 3% of persons treated with thyroid hormones were treated with L-T3 or DTE.

Declaration of interest

The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of the research reported.

Funding

This study received funding from the non-profit Musikforlæggerne Agnes og Knut Mørks Fond.

Author contribution statement

J L C conceived the study and drafted the manuscript, L T M conceived the study and did the data analytical work and helped on the manuscript, B R M helped to conceive the study and helped on the manuscript, C Z J helped to conceive the study and helped on the manuscript, A A B helped to with the data analytical work and helped on the manuscript, B N conceived the study and helped on the manuscript.