Treatment Preferences in Patients With Hypothyroidism

May combine two claims.

Three previous meta-analyses evaluated patient preference with conflicting results.

“Three previous metanalyses evaluated patient preference as reported in those RCTs comparing L-T4 vs L-T4 + L-T3, with conflicting results.”

• cites [7] Lan, 2022, Combined T4 + T3 therapy versus T4 monotherapy effect on psychological health in hypothyroidism: a systematic review and meta-analysis, Clin Endocrinol (Oxf) sourceDisputed

  The passage following Figure 4 says patients preferred T4 + T3 therapy and describes that finding as consistent with previous meta-analyses [6–8]. It therefore contradicts the claim that those meta-analyses had conflicting results on patient preference. No page number is available in our copy. (LLM, conf=0.86)

  Checked by a reasoning model — may be wrong.

  Experimental — accuracy not yet measured
• cites [8] Millan-Alanis, 2021, Benefits and harms of levothyroxine/L-triiodothyronine versus levothyroxine monotherapy for adult patients with hypothyroidism: systematic review and meta-analysis, Thyroid sourceCould not verify

  The Discussion (page unavailable) mentions a previous meta-analysis on the topic, and the Results report this paper’s own analysis of patient preference. Neither passage establishes that three previous meta-analyses evaluated preference or that their results conflicted; searches for those points found no supporting passage in our copy. (LLM, conf=0.72)

  Experimental — accuracy not yet measured
• cites [32] Akirov, 2019, A systematic review and meta-analysis of patient preferences for combination thyroid hormone treatment for hypothyroidism, Front Endocrinol (Lausanne) sourcePartly supported

  The Discussion (page unavailable) refers to prior guideline narrative summaries of combination-therapy preference rates and says their summaries differ from this paper’s findings. It does not establish that three previous meta-analyses evaluated preference or that those three had conflicting results. (LLM, conf=0.78)

  Experimental — accuracy not yet measured

Probably not a substantial claim: may not pass S3 (load-bearing).

A power analysis was calculated using the R metapower package.

“A power analysis was calculated using the R”

• cites [17] Viechtbauer, 2010, Conducting meta-analyses in R with the metafor package, J Stat Softw sourceDisputed

  The Introduction (p. 2) identifies the package described in this paper as metafor, not metapower. The paper therefore cannot substantiate the attribution of a power analysis using the metapower package. (LLM, conf=0.88)

  Checked by a reasoning model — may be wrong.

  Experimental — accuracy not yet measured

May combine two claims.

The study considered 5 additional RCTs not considered by Akirov et al, 5 additional RCTs not considered by the Millan-Alanis meta-analysis, and 4 additional RCTs not considered by Lan.

“We considered 5 additional RCTs not considered by Akirov et al (32), 5 additional RCTs not considered by the Millan-Alanis meta-analysis (8), and 4 additional RCTs not considered by Lan (7).”

• cites [32] Akirov, 2019, A systematic review and meta-analysis of patient preferences for combination thyroid hormone treatment for hypothyroidism, Front Endocrinol (Lausanne) sourceSupported
• cites [8] Millan-Alanis, 2021, Benefits and harms of levothyroxine/L-triiodothyronine versus levothyroxine monotherapy for adult patients with hypothyroidism: systematic review and meta-analysis, Thyroid sourceSupported
• cites [7] Lan, 2022, Combined T4 + T3 therapy versus T4 monotherapy effect on psychological health in hypothyroidism: a systematic review and meta-analysis, Clin Endocrinol (Oxf) sourceSupported

Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

The study was performed following PRISMA and the protocol was registered in PROSPERO (CRD42024540978).

“This study was performed following the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) (13) and the protocol was registered in PROSPERO (CRD42024540978).”

• cites [13] Page, 2021, The PRISMA 2020 statement: an updated guideline for reporting systematic reviews, BMJ sourceSupported

May combine two claims.

The present analyses performed a sensitivity analysis with the funnel plot to identify and eliminate RCTs contributing to elevated heterogeneity, an approach not followed in previous meta-analyses.

“Throughout the present analyses, we performed a sensitivity analysis with the funnel plot to identify and eliminate RCTs that contributed to an elevated heterogeneity (without affecting the robust preference for L-T4 + L-T3 and DTE), an approach that was not followed in the previous meta-analyses (7,8,32).”

• cites [7] Lan, 2022, Combined T4 + T3 therapy versus T4 monotherapy effect on psychological health in hypothyroidism: a systematic review and meta-analysis, Clin Endocrinol (Oxf) sourceSupported
• cites [8] Millan-Alanis, 2021, Benefits and harms of levothyroxine/L-triiodothyronine versus levothyroxine monotherapy for adult patients with hypothyroidism: systematic review and meta-analysis, Thyroid sourceSupported
• cites [32] Akirov, 2019, A systematic review and meta-analysis of patient preferences for combination thyroid hormone treatment for hypothyroidism, Front Endocrinol (Lausanne) sourceSupported

Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

Akirov et al considered 7 RCTs and asked whether combination therapy is the preferred treatment for hypothyroidism.

“Akirov et al (32) considered 7 RCTs and asked whether combination therapy is the preferred treatment for hypothyroidism.”

• cites [32] Akirov, 2019, A systematic review and meta-analysis of patient preferences for combination thyroid hormone treatment for hypothyroidism, Front Endocrinol (Lausanne) sourceSupported

Probably not a substantial claim: may not pass S3 (load-bearing).

Bias was assessed by 2 authors independently using the Cochrane Risk of Bias tool (RoB V.2.0).

“Bias was assessed by 2 authors (F.E.L.B. and G.C.) independently using the Cochrane Risk of Bias tool (RoB V.2.0) (14).”

• cites [14] Sterne, 2019, Rob 2: a revised tool for assessing risk of bias in randomised trials, BMJ sourceSupported

Probably not a substantial claim: may not pass S3 (load-bearing).

Funnel plot asymmetry was visually assessed for 5 or more studies.

“Funnel plot asymmetry, involving 5 or more studies, was visually assessed due to biased results in statistical tests for prevalence outcomes in single-group studies (15,16).”

• cites [15] Lin, 2018, Empirical comparison of publication bias tests in meta-analysis, J Gen Intern Med sourcePartly supported

  The Analysis Approach section says the paper considered meta-analyses containing at least five studies. The Introduction discusses inflated false-positive rates for some publication-bias tests with binary outcomes. It does not say that funnel plots were visually assessed for prevalence outcomes in single-group studies; its analysis concerns statistical tests in Cochrane meta-analyses. (LLM, conf=0.86)

  Experimental — accuracy not yet measured
• cites [16] Deeks, 2008 sourceNot checked — no full text of the source

Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

Neither Millan-Alanis et al nor Lan et al used funnel plots to reduce heterogeneity when treatment preference was evaluated.

“Neither used funnel plots to reduce heterogeneity when treatment preference was evaluated.”

• cites [8] Millan-Alanis, 2021, Benefits and harms of levothyroxine/L-triiodothyronine versus levothyroxine monotherapy for adult patients with hypothyroidism: systematic review and meta-analysis, Thyroid sourceSupported
• cites [7] Lan, 2022, Combined T4 + T3 therapy versus T4 monotherapy effect on psychological health in hypothyroidism: a systematic review and meta-analysis, Clin Endocrinol (Oxf) sourceSupported

May combine two claims.

The other 2 meta-analyses by Millan-Alanis et al and Lan et al did not consider patient preference as a primary objective, and neither focused on randomized RCTs.

“The other 2 meta-analyses by Millan-Alanis et al (8) and Lan et al (7) did not consider patient preference as a primary objective, and neither focused on randomized RCTs.”

• cites [8] Millan-Alanis, 2021, Benefits and harms of levothyroxine/L-triiodothyronine versus levothyroxine monotherapy for adult patients with hypothyroidism: systematic review and meta-analysis, Thyroid sourcePartly supported

  The Introduction (page unavailable) names comparison of treatment benefits and harms—not patient preference—as the review’s primary aim, although Methods says preference was evaluated. The claim that the review did not focus on randomized trials is not supported: Methods explicitly includes RCTs, and Results reports that 17 of the 18 included studies were RCTs. This verdict addresses Millan-Alanis et al.; this source cannot establish Lan et al.’s objectives. (LLM, conf=0.88)

  Experimental — accuracy not yet measured
• cites [7] Lan, 2022, Combined T4 + T3 therapy versus T4 monotherapy effect on psychological health in hypothyroidism: a systematic review and meta-analysis, Clin Endocrinol (Oxf) sourceSupported

May combine two claims.

Dopamine turnover rates are reduced in the NAc of TRβ -/- mice and in mice with perinatal hypothyroidism.

“Whereas dopamine turnover rates are reduced in the NAc of TRβ -/- mice (35) and in mice with perinatal hypothyroidism (36)”

• cites [35] Ookubo, 2015, Aberrant monoaminergic system in thyroid hormone receptor-β deficient mice as a model of attention-deficit/hyperactivity disorder, Int J Neuropsychopharmacol sourceSupported
• cites [36] Umezu, 2019, Hyperactive behavioral phenotypes and an altered brain monoaminergic state in male offspring mice with perinatal hypothyroidism, Toxicol Rep sourceSupported

May combine two claims.

Approximately 47% of patients had at least 1 serum FT4 level above the control reference range (>1.52 ng/dL) and 15.6% had at least 1 serum FT3 below the reference range (<0.23 ng/dL).

“This is so dramatic that approximately 47% of patients had at least 1 serum FT4 levels above the control reference range (>1.52 ng/dL) and 15.6% had at least 1 serum FT3 below the reference range (<0.23 ng/dL) (41).”

• cites [41] Penna, 2024, Thyroid hormone homeostasis in levothyroxine-treated patients: findings from ELSA-brasil, J Clin Endocrinol Metab sourceSupported

Probably not a substantial claim: may not pass S3 (load-bearing).

Neurons take up and package T3 into endosomes, which are subsequently retrogradely transported to the cell nucleus, where it regulates gene transcription.

“Neurons take up and package T3 into endosomes, which are subsequently retrogradely transported to the cell nucleus, where it regulates gene transcription (43).”

• cites [43] Salas-Lucia, 2023, Axonal T3 uptake and transport can trigger thyroid hormone signaling in the brain, Elife sourceSupported

Probably not a substantial claim: may not pass S3 (load-bearing).

Akirov et al found a positive association between the dose of L-T3 (0-20 μg/day) and treatment preference (regression slope 0.043; 95% CI, 0.007-0.078; P = .020).

“They found a positive association between the dose of L-T3 (0-20 μg/day) and treatment preference (regression slope 0.043; 95% CI, 0.007-0.078;P = .020)”

• cites [32] Akirov, 2019, A systematic review and meta-analysis of patient preferences for combination thyroid hormone treatment for hypothyroidism, Front Endocrinol (Lausanne) sourceSupported

May combine two claims.

Treatment with L-T4 that normalizes TSH levels also results in an upshift in the serum free T4 (FT4) levels and a downshift in the serum T3 levels.

“Notably, treatment with L-T4 that normalizes TSH levels also results in an upshift in the serum free T4 (FT4) levels and a downshift in the serum T3 levels.”

• cites [41] Penna, 2024, Thyroid hormone homeostasis in levothyroxine-treated patients: findings from ELSA-brasil, J Clin Endocrinol Metab sourceSupported

May combine two claims.

These areas of concern include a focus on the diagnosis of hypothyroidism, the types and doses of medications, and the patient-doctor relationship.

“These include a focus on the diagnosis of hypothyroidism, the types and doses of medications, and the patient-doctor relationship (33).”

• cites [33] Ettleson, 2022, Brain fog in hypothyroidism: understanding the patient's perspective, Endocr Pract sourcePartly supported

  The Results section (Textual Data Analysis of Open-Ended Comments; page unavailable) identifies disease/diagnosis, patient/doctor, and medication as categories of patients’ concerns. The Discussion also identifies the patient-doctor relationship as an important concern. The body text does not establish that medication doses specifically were a concern; its mention of dose changes in the Introduction concerns a historical treatment shift, not the patients’ responses. (LLM, conf=0.82)

  Experimental — accuracy not yet measured

May combine two claims.

Millan-Alanis et al and Lan et al included one quasi-randomized study, and Lan et al also included 3 nonrandomized studies.

“Millan-Alanis et al and Lan et al included one quasi-randomized study, and the latter also included 3 nonrandomized studies.”

• cites [8] Millan-Alanis, 2021, Benefits and harms of levothyroxine/L-triiodothyronine versus levothyroxine monotherapy for adult patients with hypothyroidism: systematic review and meta-analysis, Thyroid sourceSupported
• cites [7] Lan, 2022, Combined T4 + T3 therapy versus T4 monotherapy effect on psychological health in hypothyroidism: a systematic review and meta-analysis, Clin Endocrinol (Oxf) sourceCould not verify

  Section 2.1 describes eligibility for prospective studies, and Section 2.2 describes risk-of-bias assessment, but neither identifies the included studies by randomization status. Searches for quasi-randomized and nonrandomized studies found no passage establishing that Lan et al. included one of the former and three of the latter. The cited paper's copy also does not establish the count attributed to Millan-Alanis et al. (LLM, conf=0.72)

  Experimental — accuracy not yet measured

May combine two claims.

Excluding 4 studies reduced heterogeneity to I2 = 24% without affecting the pooled effect (RR 1.97; 95% CI 1.52-2.54; P < .00001).

“Excluding 4 studies reduced the high heterogeneity (I 2 = 81%) without affecting the results (RR: 1.97; 95% CI, 1.52-2.54;P < .00001;I 2 = 24%).”

May combine two claims.

Among the 11 RCTs, 5 identified a statistically significant preference for combination therapy, none identified a preference for L-T4 monotherapy, and 6 did not identify a difference in preference.

“Among the 11 RCTs (3 parallel) (22)(23)(24) and 8 crossover studies (21, 25-30, 47), 5 (263 participants) identified a statistically significant preference for combination therapy (25-29), none identified a preference for L-T4 monotherapy, and 6 (21-24, 28, 30) (435 participants) did not identify a difference in preference.”

May combine two claims.

Among the included studies, 592 patients received combination therapy (L-T3 + L-T4 or DTE) and 543 received L-T4 monotherapy.

“These studies involved 592 patients receiving combination therapy (L-T3 + L-T4 or DTE) and 543 receiving L-T4 monotherapy.”

May combine two claims.

Based on NMA, DTE had an 84% probability of being the preferred therapy, followed by L-T3 + L-T4 (65%) and L-T4 (1%).

“Based on the data presented in Supplemental Table 2 ( analysis indicated that DTE had an 84% probability of being the preferred therapy, followed by L-T3 + L-T4 (65%) and L-T4 (1%) (see Fig. 11A).”

May combine two claims.

The systematic review included 11 randomized controlled trials (8 crossover) with a total of 1135 patients.

“Eleven RCTs (8 cross-over studies) with a total of 1135 patients were considered.”

• Also stated in: § Literature Search · plain

  “In total, 11 studies met the inclusion criteria and were eligible for the meta-analysis (21)(22)(23)(24)(25)(26)(27)(28)(29)(30)(31).”

The quote only partly states this claim.

In network meta-analysis, DTE had RR 2.34 (95% CI 1.07-5.11) vs L-T4 across all selected studies.

“DTE (RR: 2.34; 95% CI, 1.07-5.11)”

May combine two claims.

The difference in final TSH values in crossover studies was statistically significantly associated with the results (P = .01), while daily L-T3 dose, follow-up duration, age, and body weight difference were not.

“The difference in the final TSH values in crossover studies was statistically significantly associated with the results (P = .01),but none of the other variables were (P = .527for dose of L-T3; P = .117for duration of follow-up; P = .399for age; and P = .956for body weight difference) (Table 2 and Supplemental Figs.89101112).”

May combine two claims.

Overall, 24% of patients preferred L-T4, 52% preferred L-T4 + L-T3 or DTE, and 24% had no preference.

“Overall, 24% of patients preferred L-T4 vs 52% who preferred L-T4 + L-T3 or DTE; 24% had no preference.”

Probably not a substantial claim: may not pass S4 (based).

The patient preference for DTE alone has not been considered in previous meta-analyses.

“The patient preference for DTE alone has not been considered in previous meta-analyses.”

May combine two claims.

When only crossover studies were considered (8 studies), no differences between DTE and L-T3 + L-T4 were observed, with 1-SUCRA ranging between 0.73 and 0.75.

“However, when only crossover studies were considered (8 studies), no differences between DTE and L-T3 + L-T4 were observed, with 1-SUCRA ranging between 0.73 and 0.75 (Figs.11 and 12).”

Probably not a substantial claim: may not pass S3 (load-bearing).

Discrepancies in bias assessment were resolved by a third researcher.

“Discrepancies were resolved by a third researcher (A.B.).”

May combine two claims.

Four studies presented a low risk of bias across the 5 domains; six had some concerns; one had high risk.

“Four studies presented a low risk of bias across the 5 domains (21)(22)(23)(24).Six other studies were classified as exhibiting some concerns (25)(26)(27)(28)(29)(30), and one was deemed to have a high risk (21) (Fig. 2).”

Probably not a substantial claim: may not pass S3 (load-bearing).

The inclusion criteria were predefined.

“The inclusion criteria were predefined as follows: (1) RCTs of parallel or crossover designs comparing the efficacy of combination therapy vs L-T4 monotherapy; (2) investigation of adults (age ≥18 years) with overt primary hypothyroidism (autoimmune, postsurgical, or radioactive iodine-induced hypothyroidism).”

Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

Patient preference is assessed after the patient is asked to decide on different treatments, which requires weighing the pros and cons of each treatment.

“This is assessed through patient preference after the patient is asked to decide on different treatments, which requires weighing the pros and cons of each treatment (12).”

• cites [12] Patient Preference InformationNot matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

Classified as probably not a substantial claim.

Network consistency was evaluated through node-split models.

“Network consistency was evaluated through node-split models.”

Probably not a substantial claim: may not pass S3 (load-bearing).

The exclusion criteria were predefined.

“Exclusion criteria were as follows: (1) case reports, retrospective studies, observational studies, or post hoc analyses of RCTs; (2) studies that did not evaluate patient preference or acceptability; (3) simultaneous use of other drugs; (4) studies that had overlapping populations; and (5) studies that included patients with subclinical hypothyroidism or central hypothyroidism.”

Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

Two authors independently searched PubMed, Embase, and Cochrane, with disagreements resolved by a third independent reviewer.

“Two authors (F.E.L.B. and G.C.) performed, independently, an electronic search of the PubMed, Embase, and Cochrane databases for relevant studies.Disagreements between the reviewers were solved through discussion with a third independent reviewer (A.B.).”

Classified as probably not a substantial claim. May combine two claims.

PROs assess a patient's perceptions of their health status before, during, and after therapy but do not indicate how much the patient values a particular outcome or therapy compared to other options.

“While they assess a patient's perceptions of their health status before, during, and after therapy, they do not indicate how much the patient values a particular outcome or therapy compared to other options (12).”

• cites [12] Patient Preference InformationNot matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

May combine two claims.

The absence of endorsement from clinical guidelines resulted in many physicians declining treatment with L-T4 + L-T3 and a considerable number of patients feeling deeply dissatisfied.

“The absence of endorsement from clinical guidelines resulted in many physicians declining treatment with L-T4 + L-T3, and in a considerable number of patients feeling deeply dissatisfied (11).”

• cites [11] Peterson, 2018, An online survey of hypothyroid patients demonstrates prominent dissatisfaction, Thyroid sourceNot checked — no full text of the source

May combine two claims.

Comorbidities were considered in only 3 RCTs, but possible differences in intestinal absorption of iodothyronines were not considered.

“Fourth, comorbidities were considered in only 3 RCTs (were excluded or randomized; Supplemental Table 1 (31)), but possible differences in intestinal absorption of iodothyronines were not considered.”

• cites [31] Beltrao FEL, Giulia C, Beltrao DCA ,  et al  Treatment preferences in patients with hypothyroidism: an analysis of eleven randomized controlled trials. Zenodo. Deposited 30 June 2024. 10.5281/zenodo.13312680Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

May combine two claims.

Comparisons between L-T4 vs L-T4 + L-T3, L-T4 vs DTE, and L-T4 + L-T3 vs DTE revealed no statistically significant discrepancies between direct and indirect comparisons, with P values ranging from .54 to .95.

“comparisons between L-T4 vs L-T4 + L-T3, L-T4 vs DTE, and L-T4 + L-T3 vs DTE revealed no statistically significant discrepancies between direct and indirect comparisons.The P values, ranging from .54 to .95,indicate a consistent evaluation of treatment effects across all comparisons, including both the overall and crossover studies.”

Classified as probably not a substantial claim.

A reduction in plasma T3 could in theory be compensated for by a homeostatic acceleration of the D2 pathway, increasing locally produced T3.

“A reduction in plasma T3 could in theory be compensated for by a homeostatic acceleration of the D2 pathway, increasing locally produced T3 (44).”

• cites [44] Peeters, 2001, Regional physiological adaptation of the central nervous system deiodinases to iodine deficiency, Am J Physiol Endocrinol Metab sourceNot checked — no full text of the source

May combine two claims.

These alterations in TH homeostasis associated with L-T4 treatment set the stage for an incomplete restoration of TH signaling in the brain, potentially affecting the reward and motivation circuitry (and the NAc).

“These alterations in TH homeostasis associated with L-T4 treatment set the stage for an incomplete restoration of TH signaling in the brain, potentially affecting the reward and motivation circuitry (and the NAc).”

• AssumptionRests on an assumption: The reward system in the brain is a network of structures that process reward, motivation, pleasure, and reinforcement.
• AssumptionRests on an assumption: Key structures of the reward system include the ventral tegmental area, nucleus accumbens (NAc), prefrontal cortex, and amygdala.
• AssumptionRests on an assumption: The NAc receives dopaminergic inputs from the ventral tegmental area, and the dopamine release is associated with the experience of pleasure and reward, influencing preferences toward stimuli that produce these feelings.

May combine two claims.

The influence of ethnic diversity and polymorphisms on patient preferences was not considered, as all studies were conducted in Europe, the United States, and Australia, and only one study evaluated the effect of gene polymorphisms.

“Third, the influence of ethnic diversity and polymorphisms on patient preferences was not considered, as all studies were conducted in Europe, the United States, and Australia, and only one study evaluated the effect of gene polymorphisms.”

Probably not a substantial claim: may not pass S2 (checkable), S3 (load-bearing).

The finding of preference for therapy with L-T4 + L-T3 or DTE in the RCTs could be linked to neural mechanisms such as reward processing.

“Thus, the finding of preference for therapy with L-T4 + L-T3 or DTE in the RCTs could be linked to neural mechanisms such as reward processing.”

• AssumptionRests on an assumption: The reward system in the brain is a network of structures that process reward, motivation, pleasure, and reinforcement.
• AssumptionRests on an assumption: Key structures of the reward system include the ventral tegmental area, nucleus accumbens (NAc), prefrontal cortex, and amygdala.
• AssumptionRests on an assumption: The NAc receives dopaminergic inputs from the ventral tegmental area, and the dopamine release is associated with the experience of pleasure and reward, influencing preferences toward stimuli that produce these feelings.

Probably not a substantial claim: may not pass S2 (checkable), S3 (load-bearing), S4 (based).

The preference for combination therapy over L-T4 monotherapy could reflect a suboptimal normalization of TH signaling in the brain.

“Although we do not understand fully the mechanistic basis for such preference, it could reflect a suboptimal normalization of TH signaling in the brain.”

Probably not a substantial claim: may not pass S2 (checkable), S3 (load-bearing).

Patient preference for one therapy over another remains a subjective and possibly mutable outcome.

“Finally, patient preference for one therapy over another remains a subjective and possibly mutable outcome.”

May combine two claims.

In the NMA, only 1 study evaluated all 3 treatments simultaneously (direct evidence), and indirect comparisons could introduce heterogeneity.

“Second, in the NMA, only 1 study evaluated all 3 treatments simultaneously (direct evidence), and indirect comparisons could introduce heterogeneity.”

Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

Patient preference is often measured as a binary choice, although strength of preference may vary between individuals.

“It is often measured as a binary choice, although strength of preference may vary between individuals.”