Association of digital measures and self-reported fatigue: a observational study in healthy . . . and chronic inflamm rheumatic disease, 2023, Chaitra

Full title:
Association of digital measures and self-reported fatigue: a remote observational study in healthy participants and participants with chronic inflammatory rheumatic disease


Background: Fatigue is a subjective, complex and multi-faceted phenomenon, commonly experienced as tiredness. However, pathological fatigue is a major debilitating symptom associated with overwhelming feelings of physical and mental exhaustion. It is a well-recognized manifestation in chronic inflammatory rheumatic diseases, such as Sjögren’s Syndrome and Systemic Lupus Erythematosus and an important predictor of patient’s health-related quality of life (HRQoL). Patient reported outcome questions are the key instruments to assess fatigue. To date, there is no consensus about reliable quantitative assessments of fatigue.

Method: Observational data for a period of one month were collected from 296 participants in the United States. Data comprised continuous multimodal digital data from Fitbit, including heart rate, physical activity and sleep features, and app-based daily and weekly questions covering various HRQoL factors including pain, mood, general physical activity and fatigue. Descriptive statistics and hierarchical clustering of digital data were used to describe behavioural phenotypes. Gradient boosting classifiers were trained to classify participant-reported weekly fatigue and daily tiredness from multi-sensor and other participant-reported data, and extract a set of key predictive features.

Results: Cluster analysis of Fitbit parameters highlighted multiple digital phenotypes, including sleep-affected, fatigued and healthy phenotypes. Features from participant-reported data and Fitbit data both contributed as key predictive features of weekly physical and mental fatigue and daily tiredness. Participant answers to pain and depressed mood-related daily questions contributed the most as top features for predicting physical and mental fatigue, respectively. To classify daily tiredness, participant answers to questions on pain, mood and ability to perform daily activities contributed the most. Features related to daily resting heart rate and step counts and bouts were overall the most important Fitbit features for the classification models.

Conclusion: These results demonstrate that multimodal digital data can be used to quantitatively and more frequently augment pathological and non-pathological participant-reported fatigue.

https://www.frontiersin.org/journals/digital-health/articles/10.3389/fdgth.2023.1099456/full

 

Make that "multimodal digital data can be <mis>used to quantitatively make garbage data look more impressive".

Maybe filling out daily depressing questionnaires makes them tired?

 

Very liberal use of "predictive" all over the place here. May as well say that negative transactions in a bank account predicts purchases, or that people throwing away the boxes of recently purchased items predicts those purchases, when really they are just facets of the same thing.

Nothing predicts anything here. Evidence-based medicine is a total failure at predicting anything at all, never does, and systematically confuses correlation with causation. This is observational data in real-time and it's basically just as useful as simply asking people to rate their level of functioning, since people can properly assess their level of functioning, and their assessment of their level of functioning does not predict their daily function, rather it's simply that their ratings reflect their ability to function. Just as someone who completed a marathon in under 3h would rate their performance as good, and someone who failed at the 5 km would rate it as poor.

All they really have to do is devise unbiased assessment methods, but they can't even do that, because they never know what questions to ask, and don't trust patients to lead the way. They have to lead the way, even if they always end up getting us all lost. All of this amounts to the most bizarre convoluted way of trying to run a restaurant that refuses to take orders, and instead tries to sift through a bunch of random data for ways to predict what people actually want, when you can never do that for any individual case. And they do this because simply trusting that more often than not people will rate their ability to function correctly makes their entire ideology useless. And it sure is entirely useless.

Ironically, the biggest reason why PROMs are unreliable isn't even that people are unreliable at using them, it's that they are systematically misused by clinicians and researchers, always biased in the choice of questions and their interpretation. This is what makes most uses of PROMs in research useless. It's not that it uses patients' subjective evaluation, it's that they are systematically misused and biased towards the researchers or clinicians preferences, goals and agendas.

This actually ends up similar to how retail theft is always emphasized when even for the retail industry alone it only amounts to a fraction of wage theft. Yes, subjective ratings can be a problem in patients' hands, but it doesn't hold a candle to how much of a giant problem it is in the hands of biased researchers who aren't ever subjected to any basic level of accountability, oversight and are never expected to actually deliver anything of actual value. Or worse yet, are expected to always deliver the same pseudoscientific garbage conclusions no matter how utterly pointless it all is.