Prevalence and factors associated with chronic use of levothyroxine: A cohort study.

IMPORTANCE: Levothyroxine prescriptions are rising worldwide. However, there are few data on factors associated with chronic use.
OBJECTIVE: To assess the prevalence of chronic levothyroxine use, its rank among other chronic drugs and factors associated with chronic use. To assess the proportion of users outside the therapeutic range of thyroid-stimulating hormone (TSH).
DESIGN: Cohort study (CoLaus|PsyCoLaus) with recruitment from 2003 to 2006. Follow-ups occurred 5 and 10 years after baseline. PARTICIPANTS: A random sample of Lausanne (Switzerland) inhabitants aged 35-75 years. MAIN OUTCOMES: We evaluated the prevalence of chronic levothyroxine use and we then ranked it among the other most used chronic drugs. The ranking was compared to data from health insurance across the country. We assessed the association between each factor and chronic levothyroxine use in multivariable logistic regression models. The proportion of chronic levothyroxine users outside the usual TSH therapeutic range was assessed.
RESULTS: 4,334 participants were included in the analysis (mean±SD age 62.8±10.4 years, 54.9% women). 166 (3.8%) participants were chronic levothyroxine users. Levothyroxine was the second most prescribed chronic drug after aspirin in the cohort (8.2%) and the third most prescribed when using Swiss-wide insurance data. In multivariable analysis, chronic levothyroxine use was associated with increasing age [odds ratio 1.03, 95% confidence interval 1.01-1.05 per 1-year increase]; female sex [11.87 (5.24-26.89)]; BMI [1.06 (1.02-1.09) per 1-kg/m2 increase]; number of concomitant drugs [1.22 (1.16-1.29) per 1-drug increase]; and family history of thyroid pathologies [2.18 (1.37-3.48)]. Among chronic levothyroxine users with thyroid hormones assessment (n = 157), 42 (27%) were outside the TSH therapeutic range (17% overtreated and 10% undertreated).
CONCLUSIONS: In this population-based study, levothyroxine ranked second among chronic drugs. Age, female sex, BMI, number of drugs and family history of thyroid pathologies were associated with chronic levothyroxine use. More than one in four chronic users were over- or undertreated.

Introduction

Hypothyroidism is a common condition with unspecific symptoms, characterized by low activity of the thyroid gland [1]. About 0.1–2% of the population has overt hypothyroidism, defined as a high concentration of serum thyroid-stimulating hormone (TSH) and a low concentration of serum free thyroxine (fT4) [2, 3]. Subclinical hypothyroidism is a common feature, defined as elevated TSH concentration and normal fT4. In population surveys its prevalence ranges from 4 to 20%, is higher among women and increases with age [4, 5].

Overt hypothyroidism is usually treated with thyroid hormone replacement [6-10], but indication for such therapy is more controversial for subclinical hypothyroidism [9, 11–13]. Patients with subclinical hypothyroidism outnumber by far patients with overt hypothyroidism, and account for the majority of thyroid hormone prescriptions, which have been steadily increasing even though hypothyroidism incidence has remained steady [14, 15]. However, data about levothyroxine prescription mostly rely on health insurance claims and prescription costs, with little information on users’ characteristics [16, 17]. The available population-based studies on the topic focus mostly on the prevalence of thyroid disease and the information about users’ characteristics is scarce [15, 18].

Overtreatment or unnecessary levothyroxine use can have serious consequences for patients, as it can cause iatrogenic hyperthyroidism and has been associated with risk of atrial fibrillation, bone loss and fractures [19, 20].

The aim of this study was thus to assess the prevalence of chronic levothyroxine use, analyze which factors were associated with chronic levothyroxine use, and assess the proportion of users outside the therapeutic range of thyroid-stimulating hormone (TSH), in order to evaluate over- and undertreatment.

Methods

Study design, settings and participants

This observational study is based on data from the CoLaus|PsyCoLaus cohort (www.colaus-psycolaus.ch) [21], a large population-based cohort that includes a random sample of inhabitants of Lausanne (Switzerland) recruited between June 2003 and May 2006. The aim of the primary cohort study was to examine cardiovascular risk factors in the general population. Inclusion criteria were European origin (who are the vast majority of the Swiss population) and age (35 to 75 years at baseline). Detailed characteristics of the cohort and the recruitment process are described elsewhere [21]. The cohort was followed up at 5 and 10 years.

The Institutional Ethics Committee of the University of Lausanne approved the CoLaus|PsyCoLaus cohort study. All participants signed a written consent after having received detailed information about the aims of the study.

Participants

In the CoLaus|PsyCoLaus cohort, 6733 participants were included at baseline. By the 10-year follow-up, 459 (6.8%) had died and 1393 (20.7%) were lost to follow-up; 547 (8.1%) missed the 5-year follow-up and were excluded because our definition of chronic drug use could not be applied to these participants. Thus, we analyzed the 4334 participants with complete 5- and 10-year follow-up data (S1 Fig).

Definition of chronic levothyroxine use

Participants answered this question at each time point: “Which drugs have you been taking in the last 6 months?” and drugs were then coded using the WHO ATC (Anatomical Therapeutic Chemical classification) system [22]. Both brand-name and generic levothyroxine formulation were coded with the same ATC code. Information on the type of levothyroxine formulation (tablets, liquid, softgel) was not collected. In Switzerland, only levothyroxine is used for hormone replacement therapy [23]; over-the counter and self-prescribed levothyroxine does not exist. Only very few drugs are over-the-counter or self-prescribed and there is no electronic registry of drugs with individual patient names.

Levothyroxine use was considered as chronic if the drug was listed in the medication questionnaire at both the 2nd visit (at 5-follow-up) and 3rd visit (at 10-year follow-up). At baseline, ATC code reporting was unfortunately not precise enough to identify single molecules and baseline data were therefore not used. We analysed only participants whose medication data was available for both follow-ups, so we would not misclassify transient levothyroxine users as chronic users. However, in a sensitivity analysis, we assumed drugs from participants who only answered the 5- or the 10-year medication questionnaire (n = 1122) to be chronic used.

To rank the most common drugs in chronic use in the CoLaus cohort, we screened the database for the most frequently recorded ATC codes. Over-the counter and self-prescribed drugs were not considered.

For generalizability, we used a different source to rank drugs across over the whole country. SASIS AG, a subsidiary from SantéSuisse (an association of Swiss insurance companies), holds aggregated data on invoiced medication of approximately 8.2 million anonymized insured people (over 95% of the total Swiss population). To analyze the invoice data from SASIS, we selected the 30 most used drugs in the CoLaus|PsyCoLaus cohort and ranked them based on the SASIS invoice data (number of pills invoiced by each chronic drug user per year). We could not use exactly the same criteria to define chronic drug use because the insurance data do not allow to follow up individual drug use. Data from the years 2014–2018 were included, the same period of the 10-year follow-up of the CoLaus|PsyCoLaus cohort.

Definition of explanatory variables

To identify factors associated with chronic levothyroxine use, we considered all demographic characteristics and clinical variables available at 10-year follow-up: sex; age; body weight and height; body mass index (BMI, defined as weight/heigth2); handgrip strength (measured with a hand dynamometer); family history; and cardiovascular risk factors (hypertension, diabetes, smoking status, use of a lipid-lowering drug). All variables were assessed during the in-person interview conducted by study personnel.

TSH was assessed with an electrochemiluminescence assay at 10-year follow-up, in a blood sample taken specifically for the purpose of the study. By measuring TSH in the full cohort, instead of only by participants with pre-existent thyroid nodules of disease, we aimed to limit selection bias. For the participants with available blood samples at 10-year follow-up and concomitant chronic levothyroxine use, TSH values were evaluated to determine if they fell within the usual therapeutic range (defined as 0.4 to 4.6 mIU/l) according to most contemporary guidelines [7], although there are current discussions of using of an age-specific reference age in the elderly [24, 25].

Statistical methods

We show baseline characteristics with absolute frequencies and percentages for binary variables (e.g., sex, hypertension, diabetes, etc.), and means and standard deviations for continuous variables (e.g., age, number of drugs, etc.). Frequency and percentage of chronic levothyroxine use were calculated. We used univariable and multivariable logistic regression models to assess the association between each factor at a time and chronic levothyroxine use. In the multivariable model, each factor was adjusted for every other factor. In one model, age, BMI, and number of drugs were included as continuous variables. For assessing the association between chronic levothyroxine use and age, BMI and number of drugs as categorical variables, we included in other multivariable models the categorical variable instead of the continuous one.

STATA® Software Version 16 (Stata corp., College Station, TX, USA) was used for all our calculations and statistics models.

Results

Participants’ characteristics

At the 10-year follow-up, participants’ mean age was 62.8 years (SD 10.4 years), 54.9% were women, and mean body mass index was 26.4 kg/m2 (Table 1). The median number of concomitant drugs (other than levothyroxine) was 2; hypertension was the most frequently reported risk factor. Mean TSH was 2.5 mIU/l (SD 2.6 mIU/l); 7.9% of participants reported a family history of thyroid pathologies. Mean handgrip strength of 33.9 kg (SD 6.5 kg) corresponded to average age-dependent values for a Swiss population [26].

Table 1

Demographic characteristics at 10-years follow up for the participants who completed both 5-years and 10-years follow up (n = 4334).

Age (years)
Mean (SD)	62.8 (10.4)
Range	45.3–87.1
Female sex—% (n)	54.9 (2382)
BMI (kg/m2)—mean (SD)	26.4 (4.7)
N. of drugs a - median (p25-p75)	2 (1–4)
Hypertension—% (n)	45.8 (1984)
Diabetes—% (n)	10.1 (438)
Current smoking—% (n)	17.4 (756)
Lipid lowering drug—% (n)	23.6 (1023)
Family history of thyroid pathologies—% (n)	7.9 (343)
TSH (mIU/l)—mean (SD)	2.5 (2.8)
Handgrip (kg)—mean (SD)	33.9 (12.1)
Had TSH measurement—%(n)	94.4 (4091)

aother than levothyroxine.

Abbreviations: SD: standard deviation; n: number of participants; p25-p75: 25th– 75 percentile; TSH: Thyroid Stimulating Hormone.

Missing data: BMI 5.2%, hypertension 2.5%, diabetes 4.3%, current smoking 6.8%, TSH 5.6%, handgrip 7.3%. For the other variables there were no missing data.

The demographic characteristics of participants who completed follow-up were compared to those who dropped out (S1 Table). Not surprisingly, the dropped-out or deceased participants were significant older and had more cardiovascular risk factors.

Prevalence of chronic levothyroxine use

Of the 4334 participants, 166 (3.8%) were chronic levothyroxine users (Fig 1); 16 (0.4%) took levothyroxine at the 5-year follow-up but not at the 10-year follow-up; conversely, 70 (1.6%) who had not been on levothyroxine at the 5-year follow-up were taking it at the 10-year follow-up. The remaining 4082 (94.2%) were never-users.

Fig 1

Proportion of chronic levothyroxine users.

Among the most used chronic drugs, levothyroxine ranked second after aspirin (Table 2). Other frequently used medications were statins, anti-hypertensive drugs of many different classes, and proton pump inhibitors. Levothyroxine also ranked second in a sensitivity analysis where we assumed that participants taking drugs chronically if they had answered the drug questionnaire only for the 5- or the 10-year follow-up (S2 Table).

Table 2

Ranking of the most used chronic drugs from the CoLaus|PsyCoLaus cohort.

RANKING	DRUG
1	Aspirin
2	Levothyroxine
3	Simvastatin
4	Atorvastatin
5	Calcium+Vit Da
6	Metformin
7	Candesartan
8	Metoprolol
9	Pravastatin
10	Zolpidem
11	Lisinopril
12	Amlodipin
13	Omeprazole
14	Estradiol
15	Atenolol
16	Esomeprazole
17	Acenocoumarol
18	Allopurinol
19	Irbesartan and hydrochlorothiazidea
20	Bisoprolol
21	Paracetamol
22	Chondroitin sulfate
23	Rosuvastatin
24	Fluoxetine
25	Losartan
26	Perindopril
27	Citalopram
28	Enalapril
29	Torasemide
30	Lorazepam

acombination drug.

When the 30 most used chronic drugs from the CoLaus|PsyCoLaus cohort were ranked based on 2018 invoice data from SASIS, levothyroxine was third (after aspirin and calcium-vitamin D; S3 Table). The rankings of the same drugs between 2014 and 2017 differed slightly, with levothyroxine remaining in the third or fourth position.

Factors associated with chronic levothyroxine use

Patients’ characteristics with and without chronic levothyroxine use are reported in Table 3. Increasing age was associated with chronic levothyroxine use (Table 4). When we considered age as a categorical variable (S4 Table), odds of chronic levothyroxine use were 2.5 times higher among participants ≥75 years than among the youngest group (<55 years). Being a woman was strongly associated with chronic levothyroxine use. BMI and number of drugs (other than levothyroxine) were associated with chronic levothyroxine use both as continuous and categorical variables. Finally, a reported family history of thyroid pathologies was positively associated with chronic levothyroxine use.

Table 3

Characteristics of participants with and without chronic levothyroxine use.

	Chronic levothyroxine use	No chronic levothyroxine use
	(n = 166)	(n = 4168)
Age (years)
median (iqr)	69.6 (16.4)	61.5 (17.1)
range	48.5–84.7	45.3–87.1
Age categories—% (n)
< 55 years	14.5 (24)	28.6 (1193)
≥55 and <65 years	21.7 (36)	31.1 (1297)
≥65 and <75 years	36.7 (61)	26.2 (1094)
≥75 years	27.1 (45)	14.0 (584)
Female sex—% (n)	92.8 (154)	53.5 (2228)
BMI (kg/m2)—mean (SD)	27.7 (5.2)	26.4 (4.7)
BMI categories—% (n)
underweight (BMI<18.5)	0.6 (1)	1.6 (66)
normal (BMI 18.5–24.9)	30.7 (51)	38.3 (1597)
overweight (BMI 25–29.9)	36.1 (60)	37.7 (1571)
obese (BMI≥30)	28.3 (47)	17.2 (716)
N. of drugsa - median (p25-p75)	5 (3–7)	2 (1–4)
N. drug categories—% (n)
<5	72.3 (120)	84.2 (3510)
5–9	15.1 (25)	12.4 (515)
≥10	12.7 (21)	3.4 (143)
Hypertension—% (n)	57.2 (95)	45.3 (1889)
Diabetes—% (n)	12.7 (21)	10.0 (417)
Current smoking—% (n)	11.4 (19)	17.7 (737)
Lipid lowering drug—% (n)	29.5 (49)	23.4 (974)
Family history of thyroid pathologies—% (n)	18.7 (31)	7.5 (312)
TSH—mean (SD)	2.4 (2.6)	2.1 (1.4)
Handgrip (kg)—mean+/-SD	25.3 (6.5)	31.8 (18.1)

aother than levothyroxine.

Abbreviations: iqr: interquartile range, SD: standard deviation, n: number of participants; p25-p75: 25th– 75 percentile, BMI: Body Mass Index; TSH: Thyroid Stimulating Hormone.

Missing data (chronic use; no chronic use): BMI (4.2%; 6.8%), hypertension (1.2%; 2.6%), current smoking (7.2%; 6.8%), TSH (5.4%; 5.6%), handgrip (7.2%; 7.3%); for the other variables there were no missing data.

Table 4

Association between chronic levothyroxine use and each factor at a time (logistic regression model).

	Univariable	p-value	Multivariable	p-value
Age (per 1-year increase)	1.05 (1.03–1.06)	< 0.01	1.03 (1.01–1.05)	<0.01
Female sex (vs male)	11.17 (6.19–20.17)	< 0.01	11.87 (5.24–26.89)	<0.01
BMI (per 1-unit increase)	1.06 (1.02–1.09)	< 0.01	1.06 (1.02–1.09)	<0.01
N. drugsa (per 1-unit increase)	1.15 (1.10–1.20)	< 0.01	1.22 (1.16–1.29)	<0.01
Hypertension (yes vs no)	1.58 (1.15–2.17)	<0.01	0.91 (0.61–1.35)	0.63
Diabetes (yes vs no)	1.28 (0.80–2.05)	0.30	0.64 (0.35–1.19)	0.16
Current smoking (yes vs no)	0.63 (0.38–1.06)	0.04	0.71 (0.41–1.22)	0.21
Lipid lowering drug (yes vs no)	1.37 (0.98–1.93)	0.07	0.80 (0.53–1.22)	0.30
Family history of thyroid pathologies (yes vs no)	2.84 (1.89–4.26)	< 0.01	2.18 (1.37–3.48)	<0.01
TSH (per 1-unit increase)	0.97 (0.88–1.06)	0.09	1.01 (0.95–1.08)	0.79
Handgrip (per 1-kg increase)	0.92 (0.91–0.94)	<0.01	1.01 (0.98–1.04)	0.72

Results are expressed as odds ratio and (95% confidence interval). In the univariable model, the association between each variable at a time and chronic levothyroxine use was assessed. In the multivariable model each variable was adjusted for the others (e.g. BMI is adjusted for age, sex, N. drugs, hypertension, diabetes, current smoking, lipid lowering drugs, family history, TSH, handgrip).

Abbreviations: iqr: interquartile range, SD: standard deviation, n: number of participants; p25-p75: 25th– 75 percentile, BMI: Body Mass Index; TSH: Thyroid Stimulating Hormone.

aother than levothyroxine.

Proportion of chronic users outside the therapeutic range

Among the 166 chronic users, 9 (5.4%) had missing TSH values, because no blood sample was taken (either because they were visited at home, contacted by phone, or because they refused the blood sample). 17% of them were overtreated and 10% undertreated (Fig 2). A minority of them had extreme TSH values (3.8%, i.e. 6 participants with TSH<0.1 mIU/l; 2.4%, i.e. 4 participants with TSH>10 mIU/l).

Fig 2

TSH levels under chronic levothyroxine therapy.

In contrast, TSH was in the normal range for 92.9% of the non-chronic user population. Among the non-chronic users, most had mild TSH elevation (only 0.6% of non-chronic users had TSH<0.4 mIU/l, 0.7% with TSH>10 mIU/l).

Discussion

In this large population-based Swiss cohort, we found that 3.8% of the participants were chronic levothyroxine users, with levothyroxine being the second most prescribed chronic drug after aspirin and similar results using insurance data covering over 95% of the Swiss population. Female sex, age, family history of thyroid pathologies and number of drugs were associated with chronic levothyroxine use. Among chronic levothyroxine users, 27% were outside the TSH therapeutic range (17% overtreated and 10% undertreated).

Levothyroxine chronic use prevalence

Previous population-based studies across Europe (Finland, Norway, Italy, and Spain) found a prevalence of levothyroxine use varying between 3.6 and 10% [15, 17, 27, 28]. In Norway, prevalence of treated subclinical hypothyroidism doubled between 1996–2006, regardless of the concentrations of TSH or the presence of antibodies [15]. Levothyroxine is the most frequently prescribed drug in the US (data from 2014–2016) [29, 30]; and the second most frequently prescribed drug in the UK [31]. Reasons for the vast increase in levothyroxine use might include misinterpretation of individual variations in TSH levels, greater inclination to treat transient, mild or asymptomatic subclinical hypothyroidism [32], or misinterpretation of symptoms of hypothyroidism that overlap with other conditions unrelated to the thyroid [33]. Moreover, once levothyroxine started, the vast majority of patients continue treatment for a long term [34]. Unfortunately, most studies on the prevalence of levothyroxine treatment were not based on a measure of chronic use, as we could perform in our cohort study. In our study, 3.8% of the participants were chronic levothyroxine users. This includes both those with overt and subclinical hypothyroidism, as we could not differentiate between the indications. Given the prevalence of overt hypothyroidism of less than 1% [35], many of the 166 chronic levothyroxine users can be assumed to be taking levothyroxine for subclinical hypothyroidism. Two-thirds of chronic levothyroxine users in our cohort were over 65 years old. A recent randomized controlled trial [11] and a systematic review and meta-analysis [12] showed that levothyroxine therapy for subclinical hypothyroidism provides no benefits for elderly patients. Hence, prescribing levothyroxine to elderly to treat this condition had not been recommended in recent guidelines issued by a panel of independent experts [13], although other guidelines are more liberal regarding prescription [6, 9]. Based on this new evidence, it is possible that chronic levothyroxine use for subclinical hypothyroidism is going to decrease in the near future.

Previous studies also found an association of levothyroxine use with increasing age and female sex [15, 17, 27, 28], which is not surprising since thyroid diseases are more common as people age and among women [36]. Obesity was associated with a higher likelihood of chronic levothyroxine use. As a higher BMI could be secondary to the thyroid dysfunction in itself, the directional causality is complex to establish. It has been suggested that doctors tend to prescribe levothyroxine to patients with high cardiovascular profile [23]; still, no association between levothyroxine prescription and most cardiovascular risk factors was found in the multivariable analysis. In our study, chronic levothyroxine use was also associated with the overall number of drugs, which raises concerns about the potential risk of drug-to-drug interactions. Several drugs impair absorption of levothyroxine [37] and certain drugs can produce hypothyroidism as a side effect of other drugs (e.g., amiodarone [38]). Finally, the association with family history of thyroid disease is not surprising. Hypothyroidism has a known genetic component; and members of affected families could as well have a tendency to undergo more frequent TSH assessments.

Proportion of levothyroxine users within the therapeutic range

Our finding that 27% of chronic levothyroxine users were outside the therapeutic range is both clinically relevant and alarming: 16.6% of patients were overtreated, which caused iatrogenic hyperthyroidism. Other studies have found similar or higher proportions of overtreated patients, e.g., in the US (41% overtreatment among elderly ≥65 years) [39] and the UK (15% overtreatment) [33]. Overtreatment raises the risks of atrial fibrillation and fractures [20, 40]. Undertreatment (i.e. TSH >4.6 mIU/l) is less risky, though undertreated patients (10.2%) might be burdened by daily treatment without potential clinical benefits. If TSH is measured regularly, these conditions could be avoided by properly adjusting the dosage of levothyroxine [7]. The assessment of the therapeutic range was based on a single available TSH measurement, similar to several other large cohort studies. For instance, in a large meta-analysis on subclinical thyroid dysfunction and fracture risk, only 5 out of 13 cohort had repeated TSH measurements [20]. Adherence to treatment, weight changes or recent use of proton pump inhibitors were not evaluated in our study.

Limitations

Our definition of chronic drug use, based on participants who completed 2nd and 3rd visit at 5- and 10-year follow-up, may have introduced selection bias, as older and sicker participants were more likely to drop out. However, age and number of drugs (an indirect measure of multimorbidity) were both positively associated with levothyroxine use, so that selection bias would have led us to underestimate overall levothyroxine use. In the participants who completed both follow-ups the percentage of women were slightly higher, which could have introduced a small bias of opposite magnitude. The participants with a positive family history of thyroid disease were also less likely to drop out, but this concerned a small percentage of participants. To address the significance of those biases, we performed a sensitivity analysis where we assumed that participants taking their medication chronically if they had answered the questionnaire only in one follow-up, and the place of levothyroxine in the ranking remained the same. 8.7% of the participants did not have complete data and were thus excluded from the multivariable analysis. As participants self-reported the medications they took, reporting bias is a possibility; nevertheless, self-reporting offers advantages over an analysis based on medical prescriptions because it reflect the way patients usually take drugs instead of only capturing the drugs doctors prescribe. However, we did not have a measure of the adherence to the treatment and the dosing of levothyroxine was not reported. These factors may alter the interpretation of the proportion of the users outside the TSH therapeutic range. Information about current pregnancy is lacking, but only 3 participants taking levothyroxine were women younger than 50 years. Information about the indication for levothyroxine was not available and thus we could not identify and analyze the characteristics of participants with subclinical versus overt hypothyroidism. Our cohort was located in a single city in the French-speaking part of Switzerland and was limited to people with European descent. Therefore, it may not represent the Swiss population, but when we compared our results with those from invoice data that covered almost the whole population of Switzerland, we still found levothyroxine was one of the most commonly prescribed drugs for chronic conditions. Since the first drug assessment was in 2009, this ranking does not contain drugs that were released after this date; if the ranking was updated, it is possible that newer drugs (e.g. rivaroxaban) could take the place of older drugs (e.g. acenocoumarol). However, the prevalence of common chronic condition (hypertension, atrial fibrillation, hypercholesterolemia) remained stable over the last decade, so the position of the levothyroxine in the ranking should not be strongly influenced.

Conclusions

In a population-based study, levothyroxine ranked second among chronic drugs, with 3.8% of the participants taking levothyroxine during at least 5 years. Female sex, age, family history of thyroid pathologies and number of drugs were associated with chronic levothyroxine use. Moreover, the observation that more than one in four patients treated with levothyroxine in a population-based cohort are outside the therapeutic TSH range implies that doctors prescribing levothyroxine need to (more) closely monitor their patients in order to avoid over- and undertreatment.

Flow chart.

(DOCX)

Click here for additional data file.

Demographic characteristics at baseline comparing participants who attended 5- and 10-year follow up (n = 4334) and participants lost to follow up (n = 2399).

aother than levothyroxine. Abbreviations: SD: standard deviation; n: number of participants; p25-p75: 25th– 75 percentile; BMI: Body Mass Index; TSH: Thyroid Stimulating Hormone.

(DOCX)

Click here for additional data file.

Sensitivity Analysis including participants who only answered drug questionnaire at one follow-up–Ranking of the most used chronic drugs from the CoLaus cohort.

(DOCX)

Click here for additional data file.

Ranking of the most used drugs in Switzerland.

a Paracetamol would rank first according to the invoice data. However, in the analysis of the CoLaus data we chose to omit on-demand and over-the-counter drugs. This exclusion could not be done with the SantéSuisse data, we therefore chose not to report paracetamol in the first position. b combination drug; HCT = hydrochlorothiazide Ranking of most invoiced chronic drugs using insurance data of 2014–2018 from SASIS, analysed by SantéSuisse, based on number of pills invoiced per insured person per year in Switzerland.

(DOCX)

Click here for additional data file.

Association between chronic levothyroxine use and each factor at a time from the multivariable models including categorical variables.

aother than levothyroxine. Results are expressed as odds ratio and (95% confidence interval) To assess the association with the categorical variables age, BMI and number of drugs respectively, we included the categorial variable in the multivariable model instead of the continuous related one (e.g. the model with BMI as categorical has the following variables: BMI categories, age (continuous), sex, N. drugs (continuous), hypertension, diabetes, current smoking, lipid lowering drugs, family history, TSH, handgrip). Abbreviations: iqr: interquartile range, SD: standard deviation, n: number of participants; p25-p75: 25th– 75th percentile, BMI: Body Mass Index. TSH: Thyroid Stimulating Hormone. Definition of BMI categories: underweight (<18.5), normal (18.5–24.9), overweight (25–29.9), obese (≥30).

(DOCX)

Click here for additional data file.

30 Mar 2021

PONE-D-20-35307

Prevalence and factors associated with chronic use of levothyroxine: a cohort study

PLOS ONE

Dear Dr. Rodondi,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

Please submit your revised manuscript by May 14 2021 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Angela Lupattelli, PhD

Academic Editor

PLOS ONE

Journal requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. We note that the grant information you provided in the ‘Funding Information’ and ‘Financial Disclosure’ sections do not match.

When you resubmit, please ensure that you provide the correct grant numbers for the awards you received for your study in the ‘Funding Information’ section.

3.  Thank you for stating the following in the Competing Interests section:

"The authors have declared that no competing interests exist. "

We note that one or more of the authors are employed by a commercial company: SASIS AG Solothurn, santésuisse Solothurn.

3.1. Please provide an amended Funding Statement declaring this commercial affiliation, as well as a statement regarding the Role of Funders in your study. If the funding organization did not play a role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript and only provided financial support in the form of authors' salaries and/or research materials, please review your statements relating to the author contributions, and ensure you have specifically and accurately indicated the role(s) that these authors had in your study. You can update author roles in the Author Contributions section of the online submission form.

Please also include the following statement within your amended Funding Statement.

“The funder provided support in the form of salaries for authors [insert relevant initials], but did not have any additional role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript. The specific roles of these authors are articulated in the ‘author contributions’ section.”

If your commercial affiliation did play a role in your study, please state and explain this role within your updated Funding Statement.

3.2. Please also provide an updated Competing Interests Statement declaring this commercial affiliation along with any other relevant declarations relating to employment, consultancy, patents, products in development, or marketed products, etc.

Within your Competing Interests Statement, please confirm that this commercial affiliation does not alter your adherence to all PLOS ONE policies on sharing data and materials by including the following statement: "This does not alter our adherence to  PLOS ONE policies on sharing data and materials.” (as detailed online in our guide for authors http://journals.plos.org/plosone/s/competing-interests) . If this adherence statement is not accurate and  there are restrictions on sharing of data and/or materials, please state these. Please note that we cannot proceed with consideration of your article until this information has been declared.

Please include both an updated Funding Statement and Competing Interests Statement in your cover letter. We will change the online submission form on your behalf.

Please know it is PLOS ONE policy for corresponding authors to declare, on behalf of all authors, all potential competing interests for the purposes of transparency. PLOS defines a competing interest as anything that interferes with, or could reasonably be perceived as interfering with, the full and objective presentation, peer review, editorial decision-making, or publication of research or non-research articles submitted to one of the journals. Competing interests can be financial or non-financial, professional, or personal. Competing interests can arise in relationship to an organization or another person. Please follow this link to our website for more details on competing interests: http://journals.plos.org/plosone/s/competing-interests

4. We note that you have indicated that data from this study are available upon request. PLOS only allows data to be available upon request if there are legal or ethical restrictions on sharing data publicly. For information on unacceptable data access restrictions, please see http://journals.plos.org/plosone/s/data-availability#loc-unacceptable-data-access-restrictions.

In your revised cover letter, please address the following prompts:

a) If there are ethical or legal restrictions on sharing a de-identified data set, please explain them in detail (e.g., data contain potentially identifying or sensitive patient information) and who has imposed them (e.g., an ethics committee). Please also provide contact information for a data access committee, ethics committee, or other institutional body to which data requests may be sent.

b) If there are no restrictions, please upload the minimal anonymized data set necessary to replicate your study findings as either Supporting Information files or to a stable, public repository and provide us with the relevant URLs, DOIs, or accession numbers. Please see http://www.bmj.com/content/340/bmj.c181.long for guidelines on how to de-identify and prepare clinical data for publication. For a list of acceptable repositories, please see http://journals.plos.org/plosone/s/data-availability#loc-recommended-repositories.

We will update your Data Availability statement on your behalf to reflect the information you provide.

5. Please include captions for your Supporting Information files at the end of your manuscript, and update any in-text citations to match accordingly. Please see our Supporting Information guidelines for more information: http://journals.plos.org/plosone/s/supporting-information.

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Partly

Reviewer #2: Yes

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: No

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: No

Reviewer #2: No

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: The study by Janett-Pellegri et al. showed the results of a population-based cohort study investigating the prevalence of chronic use of levothyroxine and its associated factors in Lausanne (Switzerland).

Below some comments.

- Levothyroxine was the second most prescribed drug. However, there are no data on the underlying cause. What was the reason for levothyroxine prescription? What was the underlying disease? What is the screening program for thyroid diseases in Switzerland? Finally, what is the screening policy for thyroid nodules in Switzerland? The finding of a thyroid nodule lead to measurement of thyroid hormones and TSH, which increases the probability of detecting a subclinical hypothyroidism. This information is important to help interpreting the data.

- The authors found that in 27% of the cases of patients under chronic levothyroxine treatment the TSH was outside reference ranges. The authors should discuss about the relevance of such a result, which was based on a single TSH measurement without any assessment of the potential underlying cause. For instance, was the adherence to the treatment evaluated? Was recent weight loss or gain considered? Or recent treatment with proton pump inhibitors? Were patients with TSH outside reference ranges managed in primary care setting or referral Endocrinological Centers?

- Which levothyroxine formulations were used? Tablets? Liquid? Both?

- Table 3 is not clear. Please, provide a Table with descriptive statistics (chronic LT4 treatment vs non chronic prescription), with appropriate statistics, and a separate table with the results of the logistic regression. Additionally, provide all P values of the logistic regression. The authors state that the model is adjusted for age and sex. What do they mean? Age and sex were forced into the regression model? Please, clarify which kind of model was used (stepwise? other?).

- Footnote of Table 3: “Missing data did not substantially differ between the two subgroups”. What missing data are the authors referring to? Moreover, “substantially” does not mean anything when dealing with statistics. The differences between groups may be statistically or not statistically significant.

- Lines 217-219. The association may also be explained by the fact that patients with cardiovascular risk factors undergo more frequent TSH assessments, increasing the probability of finding subclinical hypothyroidism.

- Figure 1. Please, provide numbers together with percentages.

- Please, provide more explanation on what the current manuscript may add to the knowledge on the field. What is the added value of this research and what are the implications for the treatment of the patients, if any?

Reviewer #2: This paper describes the number of chronic levothyroxine users in a random sample of Caucasian inhabitants of Lausanne. The study population consisted of 4,334 participants of which 3.8% were self-reported chronic levothyroxine users. Not surprisingly, the results indicate that known risk factors for hypothyroidism, e.g. age, female sex and positive family history for thyroid disease were associated with chronic use of levothyroxine.

Comments

It appears that the regression model only adjusted for age and sex and there is overlap between the factors. Obesity and hypertension are included in the factor ‘number of CVRF’, while BMI and hypertension are also included as separate factors. In addition, BMI and hypertension are not independent factors. The authors should consider leaving out the ‘number of CVRF’ and adjust the model not only for age and sex, but also for BMI.

TSH was in the normal range is 92.9% of the non-chronic user population. Consequently 299 subjects in the non-chronic user population had a TSH outside the reference range. What is the explanation for this relatively high number and why were these subjects not receiving levothyroxine in case of increased TSH? How does this impact the validity of the non-users control group?

In the current study is not possible to distinguish between the effect of the underlying thyroid condition and the effect of levothyroxine use in the chronic user group. This point needs more discussion with specific focus on the relevance and potential implications (or absence thereof) of the observed association between chronic levothyroxine use and cardiovascular risk factors.

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

15 Jun 2021

Dear Dr. Lupatelli,

Thank you for informing us that our manuscript has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands and for giving us the opportunity to revise our manuscript. Please find below our detailed point-by-point reply.

Comments Reviewer #1

1. Levothyroxine was the second most prescribed drug. However, there are no data on the underlying cause. What was the reason for levothyroxine prescription? What was the underlying disease?

Thanks for these relevant questions. As you underline, we did not have information about the diagnosis motivating the start of the levothyroxine therapy. However, subclinical hypothyroidism is by far more prevalent (up to 10:1) than overt hypothyroidism [1]. Therefore, we speculate that the majority of patients under levothyroxine were treated for subclinical hypothyroidism. Nevertheless, we added this point to the limitations (p.16, lines 265-267): “Information about the indication for levothyroxine was not available and thus we could not identify and analyze the characteristics of participants with subclinical versus overt hypothyroidism.”

What is the screening program for thyroid diseases in Switzerland? Finally, what is the screening policy for thyroid nodules in Switzerland? The finding of a thyroid nodule leads to measurement of thyroid hormones and TSH, which increases the probability of detecting a subclinical hypothyroidism. This information is important to help interpreting the data.

There is no screening program for thyroid diseases or thyroid nodules in Switzerland and recommendations vary widely across the different medical societies and experts groups [2,3]. For this reason, screening in Switzerland is individualized to persons with potential symptoms or who are at high risk (e.g. autoimmune disorders, family history). No systematic ultrasound screening for nodules is offered. Nodules are mostly incidental findings.

In this cohort, the measurement of the thyroid hormones was tested in a blood sample obtained specifically for the purpose of the study. Selection bias would occur if only participants with potential thyroid nodules or diseases were selected. As all participants underwent thyroid hormone assessment, selection bias was avoided. We clarified the circumstances in which the blood samples were taken in the “Methods” section (p. 7, lines 108-111): “TSH was assessed with an electrochemiluminescence assay at 10-year follow-up, in a blood sample taken specifically for the purpose of the study. By measuring TSH in the full cohort, instead of only by participants with pre-existent thyroid nodules or disease, we aimed to limit selection bias.”

2. The authors found that in 27% of the cases of patients under chronic levothyroxine treatment the TSH was outside reference ranges. The authors should discuss about the relevance of such a result, which was based on a single TSH measurement without any assessment of the potential underlying cause. For instance, was the adherence to the treatment evaluated? Was recent weight loss or gain considered? Or recent treatment with proton pump inhibitors?

The available information from the cohort participants, originally planned to evaluate cardiovascular risk factors in the general population, includes the self-reported medication list and various demographic variables (see “Definition of explanatory variables”, line 101). TSH and thyroid hormones were assessed in blood at a single follow up, as frequently performed in other large cohorts. For instance, in a large meta-analysis on subclinical thyroid dysfunction and fracture risk, only 5 out of 13 cohorts had repeated TSH measurements [4].

Information on the adherence to treatment, the weight changes or the recent introduction with proton pump inhibitors was not available. We adapted the section limitations in the “Discussion” as follows (p. 15, lines 239-243): “The assessment of the therapeutic range was based on a single available TSH measurement, similar to several other large cohort studies; for instance, in a large individual participant data analysis on subclinical thyroid dysfunction and fracture risk, only 5 out of 13 cohorts had repeated TSH measurements. Adherence to treatment, weight changes or recent use of proton pump inhibitors were not evaluated in our study.“

Were patients with TSH outside reference ranges managed in primary care setting or referral Endocrinological Centers?

We do not know if the management of the thyroid substitution was followed in primary or specialized care. Based on our experience, most patients are managed in primary care setting for this condition in Switzerland.

3. Which levothyroxine formulations were used? Tablets? Liquid? Both?

In Switzerland, many levothyroxine formulations (tablets, softgel capsules and oral liquid formulations) exist. Liquid formulations were approved in Switzerland in 2018 only and are thus not used in this study (the last follow-up took place in 2013-2016). Conversely, softgel capsules are approved since 2006.

We do not know which formulations were used by the participants of this study. To our knowledge, excluding pediatric patients, patients who are enterally-fed or unable to swallow, and despite promising data on better absorption under liquid/softgel formulation, there is still no evidence that one formulation is superior to the other. The American Thyroid Association considers the liquid/softgel formulation as an alternative in case of allergies to tablet excipients and other guidelines do not prefer one over the other [5,6,7]. The pharmacokinetic of both formulations is very similar.

We added this information in the Methods section (p. 6, lines 76-77): “Information on the type of levothyroxine formulation (tablets, liquid, softgel) was not collected.”

4. Table 3 is not clear. Please, provide a Table with descriptive statistics (chronic LT4 treatment versus non-chronic prescription), with appropriate statistics, and a separate table with the results of the logistic regression. Additionally, provide all P values of the logistic regression. The authors state that the model is adjusted for age and sex. What do they mean? Age and sex were forced into the regression model? Please, clarify which kind of model was used (stepwise? other?).

Following your suggestion, we split the Table 3 into two different tables. The new Table 3 (p. 10-11) contains the descriptive statistics for the participants with and without chronic levothyroxine use. The new Table 4 (p. 11-12) shows the univariable and multivariable association between the assessed risk factors and chronic levothyroxine use (logistic regression model). We have also added the p-values from the logistic model as suggested.

In the original manuscript, sex was adjusted for age, age was adjusted for sex and each of the other factors were adjusted for both age and sex. After your comment and the remarks of Reviewer #2, we decided to replace the model with a multivariable analysis in which each factor was adjusted for every other factor. The Statistical methods in the abstract (p.1, lines 13 and 18) and in the revised manuscript (p. 7, lines 120-127) were adapted accordingly.

Abstract:

“We assessed the association between each factor and chronic levothyroxine use in multivariable logistic regression models.”

Manuscript:

“We used univariable and multivariable logistic regression models to assess the association between each factor at a time and chronic levothyroxine use. In the multivariable model, each factor was adjusted for every other factor. In one model, age, BMI, and number of drugs were included as continuous variables. For assessing the association between chronic levothyroxine use and age, BMI and number of drugs as categorical variables, we included in other multivariable models the categorical variable instead of the continuous one.”

The results are shown in Table 4, including the p-values as requested by the Reviewer, and explained in the Discussion (p. 14, lines 216-229). For more clarity, we decided to report the results from the multivariable models including age, BMI and number of drugs as categorical variables in a separate table (S1 Table, supplementary material).

5. Footnote of Table 3: “Missing data did not substantially differ between the two subgroups”. What missing data are the authors referring to? Moreover, “substantially” does not mean anything when dealing with statistics. The differences between groups may be statistically or not statistically significant.

Thank you for the remark. We agree that the statement was not clear and we deleted it. The missing data were explicitly added to a footnote of Table 3 (p. 12), similar to Table 1 (p. 8): “Missing data (chronic use; no chronic use): BMI (4.2%; 6.8%), HTN (1.2%; 2.6%), current smoking (7.2%; 6.8%), TSH (5.4%; 5.6%), handgrip (7.2%; 7.3%); for the other variables there were no missing data.” We also added following statement to the Limitations (p. 16, lines 256-257): “8.7% of the participants did not have complete data and were thus excluded from the multivariable analysis”.

6. Lines 217-219. The association may also be explained by the fact that patients with cardiovascular risk factors undergo more frequent TSH assessments, increasing the probability of finding subclinical hypothyroidism.

When we analyzed every cardiovascular risk factor separately in the new multivariable analysis (see response to Comment 4), we did find a significant association between chronic levothyroxine use with BMI only (see Table 4, p. 11). As you highlight, the association between hypothyroidism, levothyroxine and any cardiovascular risk factor is complex and we added this point to the Discussion: “Obesity was associated with a higher likelihood of chronic levothyroxine use. As a higher BMI could be secondary to the thyroid dysfunction in itself, directional causality is complex to establish. It has been suggested that doctors tend to prescribe levothyroxine to patients with high cardiovascular profile [3]; still, no association between levothyroxine prescription and most cardiovascular risk factors was found in the multivariable analysis.” (p.14, lines 218-223).

7. Figure 1. Please, provide numbers together with percentages.

We adapted the Figure accordingly.

8. Please, provide more explanation on what the current manuscript may add to the knowledge on the field. What is the added value of this research and what are the implications for the treatment of the patients, if any?

The observation that more than one in four patients treated with levothyroxine in a population-based cohort are outside the therapeutic TSH range implies that doctors prescribing levothyroxine need to (more) closely monitor their patients in order to avoid over- and undertreatment, as mentioned in the Conclusion (p. 16, lines 279-285). It was already known that levothyroxine is among the most widely used drugs in western countries (see Ref. 14 and 15 in the Bibliography). We confirm this by adding data from Switzerland in a large population-based study.

Comments Reviewer #2:

1. It appears that the regression model only adjusted for age and sex and there is overlap between the factors. Obesity and hypertension are included in the factor ‘number of CVRF’, while BMI and hypertension are also included as separate factors. In addition, BMI and hypertension are not independent factors. The authors should consider leaving out the ‘number of CVRF’ and adjust the model not only for age and sex, but also for BMI.

Thank you for your comment. As you and Reviewer #1 suggested (Comment 4), we decided to include further covariates in the multivariable logistic regression model. In our original manuscript we adjusted for age and sex. In our revised manuscript we performed a multivariable analysis in which each variable was adjusted for every other variable. Therefore, our multivariable model includes the following variables: age, sex, BMI, number of drugs, hypertension, diabetes, current smoking, lipid lowering drugs, family history, TSH, and handgrip. Age, BMI and number of drugs were included as continuous variables (see Table 4, p.11 and answer to point 4 of Reviewer #1). Since every risk factor is evaluated separately, we deleted the variable “number of CVRF”; the variable “obesity” was also deleted due to its overlap with BMI.

2. TSH was in the normal range is 92.9% of the non-chronic user population. Consequently, 299 subjects in the non-chronic user population had a TSH outside the reference range. What is the explanation for this relatively high number and why were these subjects not receiving levothyroxine in case of increased TSH? How does this impact the validity of the non-users control group?

The mean age of our population were 62.8 years and prevalence of thyroid disorders increase with age. Most participants had increased TSH (6.6% of the whole population without chronic levothyroxine use). These participants could have a transient TSH elevation or an untreated overt or subclinical hypothyroidism. If we add to those participants the participants treated with levothyroxine, the resulting rate (9.8%) correspond to the expected proportion of hypothyroidism in the general population. Our primary hypothesis is that the TSH elevation in the non-treated population corresponds to an incidental finding, as the blood sample was taken for the purpose of the study and not because of clinical suspicion of thyroid disease (or in the context of any acute disease). An alternative explanation could be that, in those participants, the TSH elevation was previously known but not treated because of missing indication; in fact, only 26 participants had a TSH>10 mIU/l, while the others had only mildly elevated TSH values. Since it is controversial to treat subclinical hypothyroidism with mildly elevated TSH (<10 mIU/l) [8], this could correspond to an increasing awareness among physicians about the delicate risk-benefit balance of thyroid substitution. Therefore, this element should not substantially affect the validity of the non-users control group, as incidental finding of elevated TSH is expected in a random sample of general population.

We added a short description of the TSH values among non-chronic users in the Result section (p. 12, lines 181-182).

3. In the current study is not possible to distinguish between the effect of the underlying thyroid condition and the effect of levothyroxine use in the chronic user group. This point needs more discussion with specific focus on the relevance and potential implications (or absence thereof) of the observed association between chronic levothyroxine use and cardiovascular risk factors.

As you highlight (similar to point 6 of Reviewer 1), the association between thyroid condition, levothyroxine treatment and cardiovascular risk is very complex. It might be a result of a tendency among doctors to prescribe the drug to patients with high cardiovascular risk, or of the higher probability of diagnosing subclinical hypothyroidism by more frequent TSH assessment in such a population. However, in the fully adjusted multivariable model (based on Comment 4 of Reviewer 1), only an association between chronic levothyroxine use and BMI was found (see Table 4, p.11). We detailed the complex association in the Discussion (see Comment 6 of Reviewer 1).

Journal requirements:

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming.

We adapted the style of the manuscript accordingly.

2. We note that the grant information you provided in the ‘Funding Information’ and ‘Financial Disclosure’ sections do not match. When you resubmit, please ensure that you provide the correct grant numbers for the awards you received for your study in the ‘Funding Information’ section.

The Funding Information and Financial Disclosure sections were corrected.

3. Thank you for stating the following in the Competing Interests section: "The authors have declared that no competing interests exist.” We note that one or more of the authors are employed by a commercial company: SASIS AG Solothurn, santésuisse Solothurn.

The Competing Interests section was adapted (p. 18, lines 309-311).

3.1. Please provide an amended Funding Statement declaring this commercial affiliation, as well as a statement regarding the Role of Funders in your study. If the funding organization did not play a role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript and only provided financial support in the form of authors' salaries and/or research materials, please review your statements relating to the author contributions, and ensure you have specifically and accurately indicated the role(s) that these authors had in your study. You can update author roles in the Author Contributions section of the online submission form.

The Funding Statement was amended (p. 17, lines 297-307).

Please also include the following statement within your amended Funding Statement. “The funder provided support in the form of salaries for authors [insert relevant initials], but did not have any additional role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript. The specific roles of these authors are articulated in the ‘author contributions’ section.” If your commercial affiliation did play a role in your study, please state and explain this role within your updated Funding Statement.

The statement was included (p. 17, lines 303-307)

3.2. Please also provide an updated Competing Interests Statement declaring this commercial affiliation along with any other relevant declarations relating to employment, consultancy, patents, products in development, or marketed products, etc. Within your Competing Interests Statement, please confirm that this commercial affiliation does not alter your adherence to all PLOS ONE policies on sharing data and materials by including the following statement: "This does not alter our adherence to PLOS ONE policies on sharing data and materials.” If this adherence statement is not accurate and there are restrictions on sharing of data and/or materials, please state these. Please note that we cannot proceed with consideration of your article until this information has been declared.

The Competing Interest section was updated.

Please include both an updated Funding Statement and Competing Interests Statement in your cover letter. We will change the online submission form on your behalf.

The cover letter was updated as requested.

4. We note that you have indicated that data from this study are available upon request. PLOS only allows data to be available upon request if there are legal or ethical restrictions on sharing data publicly. In your revised cover letter, please address the following prompts:

a) If there are ethical or legal restrictions on sharing a de-identified data set, please explain them in detail (e.g., data contain potentially identifying or sensitive patient information) and who has imposed them (e.g., an ethics committee). Please also provide contact information for a data access committee, ethics committee, or other institutional body to which data requests may be sent.

b) If there are no restrictions, please upload the minimal anonymized data set necessary to replicate your study findings as either Supporting Information files or to a stable, public repository and provide us with the relevant URLs, DOIs, or accession numbers.

We clarified this point and updated the cover letter accordingly.

5. Please include captions for your Supporting Information files at the end of your manuscript, and update any in-text citations to match accordingly.

The captions were included and the in-text citations checked.

Sincerely yours

Camilla Janett-Pellegri (first author)

Prof. Nicolas Rodondi (corresponding author)

Bibliography

[1] Canaris GJ, Manowitz NR, Mayor G, et al. The Colorado Thyroid Disease Prevalence Study. Arch. Intern Med. 2000;160(4):526-534

[2] Biondi B, Cooper DS. The clinical significance of subclinical thyroid dysfunction. Endocrine reviews. 2008;29(1):76-131

[3] Baumgartner C, Blum MR, Rodondi N. Subclinical hypothyroidism: summary of evidence in 2014. Swiss medical weekly. 2014;144:w14058.

[4] Blum MR, Bauer DC, Collet TH, et al. Subclinical thyroid dysfunction and fracture risk: a meta-analysis. JAMA 2015;313(20):2055-65

[5] Virili C, Trimboli P, Romanelli F, et al. Liquid ans softgel levothyroxine use in clinical practice: state of the art. Endocrine 2016;54:3-14

[6] Pearce SH, Brabant G, Duntas LH, et al. 2013 ETA Guideline: Management of Subclinical Hypothyroidism. European thyroid journal. 2013;2(4):215-228.

[7] Garber JR, Cobin RH, Gharib H, et al. Clinical practice guidelines for hypothyroidism in adults: cosponsored by the American Association of Clinical Endocrinologists and the American Thyroid Association. Thyroid : official journal of the American Thyroid Association. 2012;22(12):1200-1235.

[8] Bekkering GE, Agoritsas T, Lytvyn L, et al. Thyroid hormones treatment for subclinical hypothyroidism: a clinical practice guideline. Bmj. 2019;365:l2006.

Submitted filename: Response to Reviewers.docx

Click here for additional data file.

26 Nov 2021

Prevalence and factors associated with chronic use of levothyroxine: a cohort study

PONE-D-20-35307R1

Dear Dr. Rodondi,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Angela Lupattelli, PhD

Academic Editor

PLOS ONE

9 Dec 2021

PONE-D-20-35307R1

Prevalence and factors associated with chronic use of levothyroxine: a cohort study

Dear Dr. Rodondi:

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Angela Lupattelli

Academic Editor

PLOS ONE