Insomnia Part II - Using Treatment Scores

It all started with a question from a friend: "Does drinking cherry juice work for insomnia?" I never recommend treatments (consult with your own physician for that), however, I do review the medical literature to see what the science of medicine says. I also have no prejudice against Eastern medical treatments, herbal treatments, natural treatments, or any other alternative medical treatment. I just want to see the studies and see the data.

At we do medical review articles differently. Instead of leaving the treatment effect as a difficult statistic that few people understand, or leaving the treatment effect buried deep inside the article where it's difficult to find, we put the treatment effect front and center and describe it, or summarize it, in a new way as the treatment score.

After studying the medical literature for a while, I realized that the diagnosis I wanted to quantify was insomnia, elderly patients, with at least two weeks of follow-up. Trying to quantify the science of medicine behind medical treatments is actually an organizational nightmare. First you have to figure out the diagnosis. Then you have to figure out the outcome measure that is most important for the patient, and, just to complicate things, the best outcome measure may be different depending on the patient, so you may have to do several reviews using several different outcome measures. One patient may have trouble falling asleep at night. Another patient may have trouble staying asleep at night. Those are two significantly different outcome measures.

Many diseases have had symptom scores created for them to try to take into account many symptoms, and also so that there is standard that can be used to do studies. For insomnia there are at least five different symptom scores that need to be considered.
  1. Insomnia Severity Index (ISI) 
  2. Pittsburgh Insomnia Rating Scale (PIRS) 
  3. Visual Analogue Scale
  4. The Questionnaire Score
  5. Leeds sleep evaluation questionnaire
The development of symptom scores is an area in which patients should have a lot of influence, yet they sometimes have zero involvement. This is definitely an area for patient generated research in the future. These symptom scores can be biased, and can be designed to make certain treatments look better than other treatments. Do you know the people behind these symptom scores? Do you know their motivations? We can and should make all this transparent.

I decided to use the Insomnia Severity Index as my main outcome measure to see how many treatments I could quantify with it. Using the Treatment Scores Analyzer(TM) at my diagnosis of insomnia now has the details listed below.

My treatment list will begin with one treatment, tart cherry juice, and I hope to expand it to more treatments over time. I have found clinical studies about a tart cherry juice in the medical literature. Tart cherry juice reportedly raises melatonin levels and is also an anti-inflammatory. One study used a tart cherry juice blend produced by CherryPharm, Inc. in Geneva, New York (I have nothing to do with the company). The fact that a company may be involved raises some issues, and they will be discussed later. For the moment my treatment list consists of only one treatment:

The best study I have found is this one: "Effects of a tart cherry juice beverage on the sleep of older adults with insomnia: a pilot study," which I have put inside a STAR™ Block (STAR™ = Statistic and a Reference). Note the treatment effect statistic of 8.09% on the right.

The Insomnia Severity Index
How was this statistic calculated? The Insomnia Severity Index goes from 0 to 28. According to the Insomnia Severity Index a score from 0 to 7 means no clinically significant insomnia.

This creates a bit of a problem for interpreting this study. If the score improves, what is the denominator? Is it going to be 28 or 21? Because of the way the Insomnia Severity Index is designed, I'm going to use 21 as the denominator, which may make the treatment effect seem better than it really is.

The 15 patients had a baseline Insomnia Severity Index of 15.5. In the placebo treated group the insomnia Severity Index Score dropped to 14.9. And doing the math, we get 15.5 - 14.9 = 0.6.

When the same patients were given the tart cherry juice their Insomnia Severity Index (ISI) dropped to 13.2. So the tart cherry juice reduced the ISI from 14.9 to 13.2, which is a difference of 1.7 compared to placebo.

Taking that difference of 1.7 and putting it over the denominator of 21 gives an improvement of 8.09% in the Insomnia Severity Index. The 8.09 goes into our Treatment Score Calculator™ as below.

Because of the potential side effect of night-time urination, or nocturia, I subtracted the 0.07, (which is a somewhat random estimation because there is no hard data), and gave tart cherry juice a SOM® Treatment Score of 8, which is a Treatment Grade of "E." As far as the nocturia goes, the study tried to mitigate that side effect by giving the tart cherry juice to patients twice a day, with the second dose being at least 2 hours prior to bedtime. So the treatment score and the corresponding treatment grade are:

Is this the statistically correct way to look at this situation? I think the Cohen's d statistic used in the study is misleading as it says the treatment effect is medium to high. For me, the "net treatment benefit to the patient" as the SOM® Treatment Score, is the more accurate, and understandable, way to report the treatment effect.

What I need to do next is review the treatment scores for Ambien, Restoril, melatonin, and other treatments for insomnia. What do the statistics say about the treatment effect of each?

Wouldn't it be nice if the net treatment benefit to the patient was always clearly spelled out for you as the SOM® Treatment Score?

Follow Treatment Scores:

You must consult your own licensed physician, or other licensed medical professional, for diagnosis, treatment, and for the interpretation of all medical statistics including SOM® Treatment Scores. Treatment Scores are for educational purposes only. Treatment Scores may be incomplete, inaccurate, harmful, or even cause death if used for treatment instead of consulting a licensed medical professional. No medical advice is being given. We DO NOT CLAIM to cure, treat, or prevent any illness or condition. Nor do our services provide medical advice or constitute a physician patient relationship. Contact a physician or other medical professional if you suspect that you are ill. Call emergency services (call 911 if available) or go to the nearest emergency room if an emergency is suspected. We are not responsible for any delays in care from using our website, our services, or for any other reason. We are not responsible for any consequential damages of any nature whatsoever. We make no warranties of any kind in connection with our writings or the use of or Treatment Scores are about what happened in the past they do not predict the future.

Copyright © 2015 Treatment Scores, Inc.

No comments:

Post a Comment