Identification of Primary Medication Concerns Regarding Thyroid Hormone Replacement Therapy From Online Patient Medication Reviews

[helvella]

We have all read of the negative side of data mining (e.g. to manipulate elections, reviews, advertising, etc.). This is an interesting and, hopefully, somewhat more positive direction.

My personal experience of WebMD has not been that good, but the message boards/forums might have much the same issues as anywhere else.

(Be aware, if you are not in the USA, you might be diverted to a local WebMD - such as Boots in the UK.)

This sentence is concerning:

Furthermore, treatment satisfaction on levothyroxine depended on what primary medication concerns the patient had.

Isn't it likely that people post about primary medication concerns precisely because their treatment satisfaction isn't that good?

J Med Internet Res. 2018 Oct; 20(10): e11085.

Published online 2018 Oct 24. doi: 10.2196/11085

PMCID: PMC6231751

PMID: 30355555

Identification of Primary Medication Concerns Regarding Thyroid Hormone Replacement Therapy From Online Patient Medication Reviews: Text Mining of Social Network Data

Monitoring Editor: Gunther Eysenbach

Reviewed by Judith Taylor and Tinh-Hai Collet

So Hyun Park, PharmD#1,2 and Song Hee Hong, PhD

1 Research Institute of Pharmaceutical Science, College of Pharmacy, Seoul National University, Seoul, Republic Of Korea

2 Korea Institute of Drug Safety and Risk Management, Drug Safety Information Office of Adverse Drug Reaction Relief, Anyang, Republic Of Korea

Abstract

Background

Patients with hypothyroidism report poor health-related quality of life despite having undergone thyroid hormone replacement therapy (THRT). Understanding patient concerns regarding levothyroxine can help improve the treatment outcomes of THRT.

Objective

This study aimed to (1) identify the distinctive themes in patient concerns regarding THRT, (2) determine whether patients have unique primary medication concerns specific to their demographics, and (3) determine the predictability of primary medication concerns on patient treatment satisfaction.

Methods

We collected patient reviews from WebMD in the United States (1037 reviews about generic levothyroxine and 1075 reviews about the brand version) posted between September 1, 2007, and January 30, 2017. We used natural language processing to identify the themes of medication concerns. Multiple regression analyses were conducted in order to examine the predictability of the primary medication concerns on patient treatment satisfaction.

Results

Natural language processing of the patient reviews of levothyroxine posted on a social networking site produced 6 distinctive themes of patient medication concerns related to levothyroxine treatment: how to take the drug, treatment initiation, dose adjustment, symptoms of pain, generic substitutability, and appearance. Patients had different primary medication concerns unique to their gender, age, and treatment duration. Furthermore, treatment satisfaction on levothyroxine depended on what primary medication concerns the patient had.

Conclusions

Natural language processing of text content available on social media could identify different themes of patient medication concerns that can be validated in future studies to inform the design of tailored medication counseling for improved patient treatment satisfaction.

Keywords: medication counseling, social network data, primary medication concerns, satisfaction with levothyroxine treatment

Full paper freely available here:

ncbi.nlm.nih.gov/pmc/articl...

[m7-cola]

I agree with your point: people are likely to be critical of their treatment

if their symptoms had not improved.

[diogenes Remembering]

This is all very well. It is acknowledged that patients on T4 therapy can be dissatisfied with their treatment outcome. And that more care should be taken in understanding and correcting their medication if necessary. BUT the overt step to openly suggesting multi-therapy (T3 combination or NDT) has been ducked in my opinion. If all that is suggested is tweaking of mono therapy to get better satisfaction then all data mining systems based solely on this will be compromised.

[helvella Administrator in reply to diogenes]

Couldn't agree more, diogenes.

I find it depressing that so many people join this forum (and others) with the absolute most basic questions. The sorts that are likely to be picked up by this data mining. It is so obvious that a doctor handing out a prescription for levothyroxine often leaves the patient adrift. Little to no explanation of what is expected, why the prescription was issued, that further tests and titration of dose is usual, etc.

If the medical establishment can't even manage to do that, the chances of patients who need combination or T3 monotherapy being identified are slim.

[Adam10 in reply to helvella]

What are the typical ways for a patient to identify if he/she needs multi-therapy (T3 combo or NDT)?

I have Hashimotos and have been on mono T4 therapy for four years (recently TSH rose 0.5 to 3.4, i increased my T4 from 100 mcg to 125 mcg 6 weeks ago) but am feeling completely exhausted and needing to sleep every 3-4 hours.

How does someone in my position find out if they need multi-therapy?

[helvella Administrator in reply to Adam10]

I think you need to start with blood tests - FT4, FT3 and TSH - and consider if it looks like your FT3 is inadequate despite sufficiently high FT4.

I do not have Hashimoto's so do not have the personal experience. I suggest that you ask as a new post so that more other members see your question and can help to answer for you.

Have to say, it reads as if you could need some T3 in there! But there are all the other possible issues - like iron deficiency, B12 deficiency, etc. So many things contribute to tiredness/exhaustion/sleepiness.

[TSH110 in reply to helvella]

You don’t see what you don’t look for

[JGBH in reply to helvella]

Totally agree with you helvella. So very depressing!

[diogenes Remembering in reply to helvella]

What's needed is a practical way to detect those such as on this forum who would be likely to benefit from more than T4 monotherapy, at the onset of the need for treatment and not sometime afterwards after a futile attempt at successful T4 therapy alone. This needs a technique to separate out subgroups from the main "satisfied" group. After we believed it essential first for understanding the detailed physiology of thyroid/body/pituitary interaction and the consequences for treatment, we're now concentrating on just this problem - how to discover patients that may need different treatment from the norm. We have submitted a paper to Frontiers which addresses this problem and develops the statistical methods for doing so. It shows how to identify such subjects and gives the reasons for devising properly organised clinical trials that actually will address the problem properly, not lumping everyone in together, but attempting to distinguish subgroups from the main group. I believe it is probably the first study of this kind, as it has used patient data over nearly a decade, where patients routinely are assessed for the success or otherwise of T4 monotherapy. It has wider implications for linking eg OP and AF with thyroid parameters and patient variation in risk. I only hope it is not too strong meat for the reviewers.

[helvella Administrator in reply to diogenes]

Suggestion, diogenes, would it make sense to start everyone on combination therapy - and wean them off when they stabilise to see if it worsens their wellness?

[diogenes Remembering in reply to helvella]

The trouble is with that is that just as there is a subgroup with poor conversion, there will be the opposite of very good and perhaps oversensitive conversion (those with health low-normal FT3). I wouldn't want to overdose those at the start with T3.

[PR4NOW in reply to diogenes]

Diogenes, do you think this will work with people like me, outliers, who are not even on the curve? Or is this primarily for 'normal' people. PR

[diogenes Remembering in reply to PR4NOW]

I can't see why it shouldn't, as your initial response trajectory on T4 only would show you well outside normal trajectories of response. This would invite alternative dosage.

[helvella Administrator]

I wonder how good the "natural language" processing is at handling sarcasm, facetiousness, humour, and all the other ways in which we express ourselves?

After all, I'm totally convinced that researchers in Korea will understand every nuance of what I post! The processing will, of course, "get it".

[Sybilla14 in reply to helvella]

This was the first thing that came to my mind when I read your post helvella. Interpreting statements and classifying them can be so subjective, even in native language never mind a foreign one. The statement you quoted “...treatment satisfaction...depends on what primary medication concerns the patient had” tells me that the dissatisfaction may be imaginary and a result of a personal bias of the patient. Fits quite well with the premise that our symptoms are in our heads🙂

[Hidden]

"I'm totally convinced that researchers in Korea will understand every nuance of what I post!"

I once tried a translation site with a short poem (unfair of course). It translated "Caterpillar" (the butterfly larva) as caterpillar tracks on vehicles, and the bill of a bird as in cash accounts!

[humanbean]

Furthermore, treatment satisfaction on levothyroxine depended on what primary medication concerns the patient had.

This sounds like typical patient blaming. The patient is probably assumed to have read some forums, realised that lots of people were unhappy, so had been primed to think Levo was a nasty thing to be on.

It seems to me that researchers and doctors always suggest a treatment for practically anything is perfect. For example, many doctors tell patients that statins have no side effects and they have no other patients who complain.

[helvella Administrator in reply to humanbean]

Whereas one of the most common posts here is the one in which someone has, usually after a long time, found this forum and come to realise that their ongoing suffering just might have an answer. Truly I suspect many would never have looked or found the forum if they were even reasonably OK.

[MissGrace]

‘Furthermore, treatment satisfaction on levothyroxine depended on what primary medication concerns the patient had.’

I find this such a vague and ambiguous statement - it lacks any scientific integrity.

It makes it sound like any concerns are a self-fulfilling prophecy - you think it’s going to be cr*p, so it is cr*p. It suggests we are too stupid to know what feeling well is like (though it’s a distant memory for some of us!) Many people may have doubts Levo will be the answer - but it doesn’t mean they don’t want it to be and don’t give it a good go, full of hope! And for many it could still be the answer if only they were allowed by doctors to take enough of it rather than having their dose f*cked about with for erroneous reasons!

Alternatively, this could read as, people who have low expectations of the efficacy of Levo could well be satisfied with the treatment as they never expected much from it in the first place.

I doubt a study completed in this way has any scientific validity at all, for these reasons and all the many reasons sited in other posts above. It’s a pity, as there is, without doubt, a lot that medical science could learn about thyroid treatments from the experiences and writings of those of us who know them best - the patients.

This study sounds best placed in the dustbin with a match. 🔥 🤸🏿‍♀️