NSRR staff
Boston, MA
0000-0002-0506-8368
Top Topics
Recent Topics
Hi,
Sorry, something unexpected happened with your STAGES request. Can you please fill out and submit another one? We'll also try sending your agreement from a couple weeks ago back to you. Thanks!
Hey - we didn't receive any documentation about this dataset, other than that >=3% desaturations were used. Presumably "sleep_time" is in seconds, e.g. 25800/60 = 430 minutes.
The other columns seem to pertain to event counts (e.g. n_obs -> number of obstructive apneas), average event lengths (e.g. hyp_dur -> average length of hypopneas), and an overall AHI.
I see you are inquiring about the CHAT dataset. The only wheezing variable I could find that was asked at baseline and follow-up are these:
The Sleep and Health Questionnaire was (evidently) changed sometime in the middle of data collection, hence the slightly offset variable names. I could only find asthma-related questions on the medical and family history forms, which were only collected at baseline.
Thanks for using the site!
Hey Pranavan,
I attempted a fresh run of nsrr download ccshs using NSRR gem 8.0.0 and it completed without any errors both times. I had success with both Ruby 2.7 and 3.02p107.
The tool is designed with catching incomplete/interrupted downloads in mind. Did you get these failures on your first run? Perhaps the files downloaded incompletely, the tool checked the checksums, and found the mismatch. Ideally you could then re-run the same command again and it would fix/re-download any files where the checksums mismatch. All other files will get skipped quickly and be reported as "identical" if the checksums match.
Hi Yuval,
We are only sharing the 'mesa-sleep-dataset-0.4.0.csv' file: https://sleepdata.org/datasets/mesa/files/datasets
This dataset contains 2,237 rows, corresponding to the number of subjects who participated in the MESA Sleep ancillary study as part of MESA Exam 5. The dataset is keyed on 'mesaid'.
Note that NSRR, at this point, does not share any data from other MESA exams. I will send you a followup email.
I suggest using arm (https://sleepdata.org/datasets/chat/variables/arm) to determine the randomized treatment arm (n=226 in surgery arm).
I'm guessing had_surgery also includes subjects who crossed over from the watchful waiting arm. Some subjects randomized to surgery did not ultimately have surgery, too.
Hey Wolfgang,
Good question - the "Unsure" events in SHHS are still a bit of a mystery to us as well. Our current recommendation is to ignore these events in SHHS. The "Unsure" tag was not used in SHHS scoring, but may have been inadvertently introduced (on a very limited number of overall events) during various reprocessing steps over the years. From what I recall, a handful of studies have a high number of Unsure events, in which case those individual studies may need to be entirely excluded.
In other datasets, e.g. MrOS, MESA, the "Unsure" tag corresponds to Hypopneas as described in the snippet you copy/pasted.
To your other question, the NSRR XML files should be consistent in their presentation of events/staging, though the underlying scoring rules may differ from dataset to dataset.
Grace,
Unfortunately not. As best I can tell, the actigraphy devices were very limited. The minute by minute output contains columns for "activeness", "mode", and "steps".
Hey Shuo,
The actigraphy .zip file opens/unzips for me in Windows 10/11, using both the built-in archive handling and 7-Zip.
The file is in this folder: https://www.sleepdata.org/datasets/stages/files/m/browser/original/De-identified%20Data/
The file contains individual folders per subject with a .JSON minute by minute file. Some subjects have empty data files because they did not wear the device.
Joohyun,
Thank you for being diligent in reading the data request language and for caring about this requirement. In short, yes, all the numbered things you listed would be sufficient to meet the data request requirement.
We look forward to reviewing your data request!