Comparing apples and apples

Drug trials (or trials of any new treatment for that matter) are designed by comparing a group on the new treatment against a group on something else. That may be no treatment, the best available treatment, or placebo (or dummy treatment). In order to make a fair comparison, it is essential to compare like with like.

An extreme example of this would be to say that “men are faster than women”, having compared the Jamaican mens olympic sprint team, against a selection of girls from the local secondary school. This is obviously ridiculous in so many ways: adult men vs teenage girls, trained atheletes vs untrained and so on…

In studies of Parkinson’s, it is also important that we compare like with like. For the past 50 years, researchers have used the “Hoehn and Yahr Scale” to create standardised definitions. (The scale is named after Margaret Hoehn and Melvin Yahr who pioneered research into progression and epidemiology of Parkinson’s in the 1960s.) Designed to give researchers a simple and global assessment of disease severity, this is probably the most widely used Parkinson’s staging system in use. According to the latest version of this scale there are 6 stages from 0= Asymptomatic to 5 = wheelchair bound or bedridden unless unaided.

Despite its widespread use, until now there have been no systematic attempts to prove how useful and reproducible it is. The researchers in this study assessed the acceptability, reproducibility, content and validity of the scale. They compared how different raters scored people (known as inter-rater reliability), and how that compared with other scoring systems of stiffness, tremor and other features of Parkinson’s. They also looked at whether the same assessor gave the same patient the same score at different times (known as test-retest reliability).

The good news is that inter-rater reliability and test-retest reliability are incredibly high. In other words, if one clinician judges a person to have stage II, so will almost every other (trained) clinician. It also correlated quite well with other physical assessment scores.

As ever, it is not a perfect test. The difference between stages is not the same (in other words, going from I to II is not the same as going from III-IV). Nor does it correlate at all well with quality of life. It also does not take into account non-motor features of Parkinson’s. Furthermore, only 2.9% of more than 3000 participants were at the most advanced stage V, so may not be quite as reliable for these people.

In summary, we have a pretty good measure of giving a number to the amount that an individual is affected by Parkinson’s. Until we have a blood test/scan/marker of progression, this is the best we have, and we need to make sure we have ways of addressing its weaknesses. Using it we can compare like with like, and, hopefully, get ever closer to preventing people advancing through the scale.

RNR

http://onlinelibrary.wiley.com/doi/10.1002/mds.27242/abstract;jsessionid=A596451DCF79B33BF34C1802EFD6E2CF.f04t03

Martinez-Martin P, Skorvanek M, Rojo-Abuin JM, Gregova Z, Stebbins GT, Goetz CG, et al. Validation study of the hoehn and yahr scale included in the MDS-UPDRS. Mov Disord. 2018 Jan 11;17:427.