Is Bedtime Better?

Written By

Zachariah M. Deyo, PharmD, BCPS, CPP

Reviewed By

Michele Faulkner, Pharm.D.
Oralia Bazaldua, Pharm.D., BCPS


Bolk N, Visser TJ, Nijman J, et al. Effects of Evening vs Morning Levothyroxine Intake. Archives of Internal Medicine. 2010;170:1996-2003

Do you tell your patients to take levothyroxine in the morning 30 to 60 minutes before food and at least 4 hours apart from other medications?  It is well known that many medications and foods influence the absorption and distribution of thyroxine (T4).  Morning administration of levothyroxine, before meals and other medications, is recommended by expert consensus and the package insert.1,2  While pharmacokinetic studies highlight the importance of taking levothyroxine on an empty stomach, there is a paucity of data in regard to the optimal time of day for levothyroxine intake. 3

In 2007, Bolk and colleagues observed improvement in thyroid hormone levels among 11 patients after scheduling levothyroxine intake at bedtime.4  To confirm these findings the authors conducted a randomized, double–blind crossover trial.5  Their primary objective was to examine the difference in thyrotropin and thyroid hormone levels when levothyroxine was taken at bedtime vs in the morning. They also assessed serum creatinine, lipid profiles, body mass index, heart rate and quality of life.

Patients were eligible if they were over the age of 18 and on a stable dose of levothyroxine for at least 6 months.  Those with a gastrointestinal disorder, thyroid carcinoma, who were pregnant, or taking a medication known to interfere with levothyroxine absorption were excluded from the trial.  Patients were randomized to 1 levothyroxine capsule in the morning 30 minutes before breakfast and 1 identical placebo capsule just before going to bed or vice versa.  After 3 months, patients crossed over to the opposite schedule and were followed for another 3 months.  Study medications were compounded and provided by the hospital pharmacy at the same dose each subject was taking prior to randomization. Patients were seen in clinic at baseline and every 6 weeks by a research nurse who collected blood samples, vitals and assessed compliance through pill counts.  A quality of life questionnaire that included the 36-Item Short Form Health Survey, Hospital Anxiety and Depression Scale, 20-Item Multidimensional Fatigue Inventory and symptoms of hyper or hypothyroidism was administered at baseline, 3 and 6 months.

Data was available for 90 patients enrolled in the trial at the end of 6 months.  No patient required levothyroxine dosage adjustment during the trial.  There was not a statistically significant difference in the mean number of doses missed between the morning and evening administration periods.

Primary Outcome:

For the primary outcome, thyrotropin levels decreased when patients switched from morning to bedtime intake and increased in those that switched from bedtime to morning.  In addition, bedtime levothyroxine intake was associated with higher free Tand total triiodothyronine (TT3) concentrations.


Morning Intake 1st

Bedtime Intake 1st

Direct Treatment Effect

P value

Mean (SD)

Mean (SD)

(95% CI)

Thyrotropin Level (mcg/L)

-0.92 (2.08)

1.57 (3.87)

1.25 (0.6 to 1.89)



0.11 (0.27)

-0.04 (0.25)

-0.07 (-0.13 to -0.02)



5.80 (26.20)

-7.13 (26.30)

-6.46 (-12.10 to -0.90)


FT= Free thyroxine level, TT= Total triiodothyronine level

Secondary Outcomes:

There were no differences between groups for any of the secondary endpoints of serum creatinine, lipid profile, blood pressure, body mass index or heart rate.  In addition anxiety, depression, fatigue and symptoms of hyper or hypothyroidism did not differ between groups.

Patient Preference:

At the end of the trial, 34 of 90 (38%) patients said they felt better while taking levothyroxine in the morning, 31 (34%) preferred bedtime intake and 25 (28%) indicated no preference.  One year after completion of the trial, more than half of the patients preferred bedtime intake.


Thyrotropin was decreased and free thyroxine and total triiodothyronine were increased in patients who switched from morning to bedtime levothyroxine administration.  The opposite occurred when patients switched from bedtime to morning administration.  Why?  The authors propose several physiologic hypotheses.  Perhaps the interval of 30 minutes between levothyroxine and morning food intake is too short.  Beverages consumed (including coffee) and concomitant medications were not assessed in this trial.  Also, patients may eat dinner more than 30 minutes before bedtime allowing for a greater interval between food and levothyroxine intake.  In addition, circadian differences in gastrointestinal function including slower bowel motility and higher basal gastric acid secretion in the late evening may enhance the bioavailability of levothyroxine. 

However, changes in thyroxine, FT4  and TTvalues did not translate into changes in quality of life or symptoms.  The authors suggest this could be due to a difference in plasma and tissue (the brain) thyroid levels.  They also hypothesize it could  be due to the fact that the primary etiology of hypothyroidism in both groups was a chronic autoimmune disease (Hashimoto’s disease) or that levothyroxine does not replace newly discovered thyroid hormone metabolites (thyronamine).   These findings may not be surprising given the fact that no dose adjustments were needed during the study.

While the strength of this trial is that each patient acted as his/her own control, there are limitations including the single center design and cultural differences in eating habits in other countries or cultures. However, a retrospective chart review of skilled nursing facility patients and a prospective study conducted at an academic medical center, both in the US, have found similar results.6,7  Exclusion of patients with gastrointestinal disorders, pregnancy, and interfering medications may limit generalizability.

It appears that the most important factor to maximize levothyroxine absorption is that it is taken on an empty stomach.  However, other plausible physiologic advantages may exist to bedtime levothyroxine intake.  The results of this trial show that levothyroxine administration at bedtime is an effective alternative – perhaps preferable in some patients.  For those patients who find evening doses more convenient or who are not euthyroid with morning intake most likely caused by interacting foods or medications, evening administration seems the way to go.  Given that adherence with medication was not adversely impacted during the bedtime administration period, we shouldn’t hesitate to recommend this alternative dosing time.

Is bedtime administration simply a good alternative for a select few or should we preferentially recommend bedtime administration for all patients?  What are your thoughts?



<p>The only reason I've suggested morning dosing for patients is that some patients have complained of insomnia with bedtime dosing. I'm not sure if this is a placebo effect or what. I found it interesting that this was not mentioned in the article (unless I missed it somewhere?...). The biggest cause of TSH fluctuations, I think, is poor adherence and taking it with food and other medications. I think some patients do not adequately wait the full 30 minutes. I do have some that take their levothyroxine with food on a daily basis that have stable levels. I think consistency, no matter what that may be, is most important. The key takeaway with this article is that it's "okay" to go against common practice sometimes if it is what works best for the patient.</p>

<p>Levothyroxine is one of the medications I routinely suggest patients keep on their nightstand with a glass/bottle of water in the absence of young children/animals/etc that could get into it.&nbsp; This can work for both morning and nighttime dosing, and I have seen it increase compliance with meds across the board, not just levothyroxine, although it is especially advantageous for those people taking levothyroxine for a couple reasons.</p><p>In my experience, most patients I've seen are awake for 30 minutes&nbsp;before&nbsp;they consume breakfast...they're usually brushing their teeth, showering, fixing breakfast, getting the kids ready for school, etc. in that first 30.&nbsp; If these people take their medication immediately upon waking, they will be starting their day while the medication has the time it needs to be absorbed.&nbsp; Likewise, patients usually have a span of at least 4 hours after dinner before they go to bed.&nbsp; By making taking their levothyroxine the first or last thing they do in their day, it takes the guesswork out of the timing for patients.&nbsp; If they are a patient who needs their cup of morning coffee first thing, it usually works out to take the medication at bedtime - and if they're a diabetic patient that needs an evening snack, morning usually works out well since some of their time is also usually taken up by testing their fasting BG first thing in the morning.</p><p>However, usually the most common reasons I choose to recommend one time over the other - and unfortunately a downfall in the external validity of the study - are ubiquitous interacting medications.&nbsp; Since most patients I see taking levothyroxine are also on multiple medications, this can be tricky.&nbsp; For example, a younger patient may take their oral contraceptive in the morning and&nbsp;multivitamin at&nbsp;night.&nbsp; An older patient may take their iron tabs three times a day, as well as warfarin.&nbsp; Afternoons, whether for eating habits or compliance issues, are usually not a viable option for most patients.&nbsp;</p><p>All that being said, I haven't seen a clear advantage in either administration time since there are so many patient variables to consider; I agree with Dr. Dixon in that the most important element seems to be consistancy of administration in each patient, although it is&nbsp;reassuring to know there is a study showing the&nbsp;efficacy of evening dosing.&nbsp; I think it would be interesting to see what the front runner would be if patients with interacting medications were included.</p>