“High-quality health data research and the development of ML [machine learning] models requires meaningful data at a sufficient scale. Such data undoubtedly exist. Most health institutions hold clinical imaging data at a scale ranging from tens of thousands to tens of millions of scans. However, these data are often inaccessible to researchers, even where there is an intention to make them available for research, because of barriers of access and usability. Barriers of access can include: governance barriers (difficulties in understanding and working through governance frameworks regulating data usage); cost barriers (there can be considerable overhead costs to datasets and … Read More
All posts in Measurement
“Genome-wide association studies have shown that common diseases are polygenic, ie, thousands of DNA variants contribute to risk, and most of these have very small effect. In spite of this complexity, it is now possible to estimate the degree to which an individual is at risk of common illnesses owing to their genetic makeup. The so-called polygenic risk scores (PRS) are generated from DNA taken from a saliva or blood sample with DNA variants measured using genotyping technologies that are inexpensive (< US $100 per person). From these data, PRS can be calculated for a wide range of diseases (by … Read More
“In the Strategies to Reduce Injuries and Develop Confidence in Elders (STRIDE) study (July 9 issue), Bhasin et al. evaluated the effectiveness of individualized recommendations to prevent fall injuries. We find the lack of significant success with this strategy to be wholly unsurprising. Referrals or recommendations alone are insufficient to reduce falls. The STRIDE intervention successfully expanded the role of primary care providers to include fall-risk screening, assessment, and individualized care plans. It did not include any direct procedural interventions (e.g., exercise, medication management, home safety modifications, and vision care) or the necessary extensive support to ensure their receipt. [..] … Read More
“First described in 1975 by economist Charles Goodhart, Goodhart’s Law states that “when a measure becomes the target, it ceases to be a good measure.” The famous, and possibly apocryphal, example of Goodhart’s Law is the story of nail factories in the Soviet Union. To boost output, the factory performance was pegged to the number of nails produced, leading to the manufacturing of millions of tiny and useless nails. In response, the performance metric was changed to the nails’ weight. The factories, in turn, adjusted their approach and produced a small number of giant, and equally useless, nails. Hospitals, too, … Read More