Scripps researchers: Hypothesis testing framework can help with mHealth data analysis

Data pulled from digital devices and apps can give providers insight into a patient's health and biology, according to researchers from the Scripps Translational Science Institute.

The researchers present a hypothesis testing framework and investigate data related to blood pressure readings of 38 participants in a six-month health monitoring intervention trial.

"The approach we present can be adapted to other similarly structured data," the article's authors write. "We find that by leveraging all data across individuals, we were able to detect an approximately 2 mmHg decrease in blood pressure over, despite considerable intra- and inter-individual variation."

The hypothesis testing framework used unstructured time series data and features a mixed model approach.

Data analysis is a growing focus in the mHealth. However, issues with such information remain. One recent study noted how a dozen wearables aren't producing reliable or accurate data when it comes to energy expenditure tracking, and a data deluge, created from increasing use of digital and mHealth tools, is prompting researchers to double check validity and accuracy of the information attained.

However, a data testing framework can lead to informed health decisions and allows researchers a pathway to addressing what it terms as innovative health and biology questions, the researchers say.

"The hypothesis testing framework we present can be applied in future studies utilizing digital medicine technology or implemented in the technology itself to support the quantified self," they say.

For more information:
- read the article