"Clinical data" on FOX Insight participants?

Data Sharing and Publications
, ,
1

Hi guys, my doctoral student finally got back reviews after 10 months (Sci. Reports) on a GWAS of pain trajectories in early stage PD (<=3 years from diagnosis) from Fox Insight data. We are using the pain item from a PRO. One reviewer is being difficult and states:

“I cannot recommend the publication of this paper without a clear description of the clinical information of the samples used in the analysis. (e.g. Hoehn and Yahr stages, history of orthopedic diseases)”

Only ~25 of 4200 subjects have overlapping Hoehn/Yahr stages in FOX DEN. Any thoughts on my student’s rebuttal - or is there something else we can pull?

I will push back - acknowledging the 50+ publications that have leveraged Fox Insight

Here is my tentative rebuttal:

As detailed in the methods, the subjects in this study are participants in Fox Insight led/established by the Michael J Fox Foundation. It is a convenience sample of volunteers with and without PD. In partnership with 23andMe, approximately 10,000 participants with Parkinson’s disease in the US were enrolled in the Fox Insight Genetic Sub-Study (FIGS) from 2017 to 2021 - and it is from this substudy that the current genetic data originates. Details on the recruitment process and study goals have been described in detail by others: see Fox Insight collects online, longitudinal patient-reported outcomes and genetic data on Parkinson’s disease | Scientific Data and Fox Insight at 5 years - a cohort of 54,000 participants contributing longitudinal patient-reported outcome, genetic, and microbiome data relating to Parkinson’s disease | Scientific Data. Fox Insight genetic and non-genetic data have been used in >50 publications (including a recent polygenic score on age at onset in Scientific Reports The combined effect of lifestyle factors and polygenic scores on age at onset in Parkinson’s disease | Scientific Reports). Subsets of Fox Insight participants were a part of other studies that collected detailed clinical information, but only a few of them are in current study population.

It’s just annoying :face_exhaling:

1 Like
2

I may not have an answer to your question, but I truly sympathize with your situation. I must admit that lately, the attitude of many reviewers has been disappointing and lacking in understanding. Even after revisions and clarifications, they continue to insist on their comments.

2 Likes
3

I think referencing the data descriptors you mention makes good sense – I hope that we described data provenance in an acceptable manner there (at least our reviewers seemed to think so :wink: ).

Another thought: Hoehn and Yahr is only available in a small subset of Fox Insight participants through our telemedicine sub-study, as it must be administered by a healthcare provider rather than digitally as a survey (at least as I understand it). Here is a link to a paper about that work [link].

That study itself only had ~200 participants + administered instruments one time, hence the limited number of overlapping results. The linked paper describes that cohort in greater detail.

Other measures, like the UPDRS which are obviously often used in conjunction with H&Y have plentiful data in Fox DEN since it is offered to the full PD cohort, longitudinally (H&Y was only one-time).

Lastly, would note that all of the questionnaire forms themselves are made available publicly [link], as is the schedule of activities [link].

Happy to chat further, hope this is moderately helpful!

2 Likes
4

This is most useful - I’ll modify our response!
The pilot study is perfect: 95.1% (157/165) had a clinician-confirmed diagnosis

1 Like