Machine learning techniques analyse effects of weight loss intervention

Machine learning techniques have revealed that while a healthy weight loss intervention is beneficial for most people, it can lead to worse outcomes in others.
The findings indicate that some patients will benefit from certain lifestyle interventions more than others, highlighting the importance of precision medicine in health care.
Scientists at the Arnhold Institute for Global Health of the Icahn School of Medicine used new, advanced machine learning techniques to review data of more than 5,000 overweight and obese people with diabetes.
The objective was to determine over a 13-year follow-up period whether modest weight loss through lifestyle intervention reduced rates of mortality and diabetes-related complications such as heart attack and stroke.
While the overall findings were null, 85 per cent of the study sample experienced significant reductions in cardiovascular mortality following weight loss intervention. The other 15 per cent, however, had a substantially increased risk of heart attack and stroke.
This group were less compliant with exercising in the interventio, while they also struggled to improve health markers such as blood sugar, blood pressure and mental health.
The researchers believe this the first suggestive evidence of adverse reactions to a healthy lifestyle interventio, and have called for greater data science and precision medication to be used in diabetes care.
“Our analysis demonstrates that recent advances in machine learning for causal inference can increase the quantity of clinically relevant findings generated from large randomized trials,” said lead author Aaron Baum, PhD.
“Being able to identify individuals that could benefit from an intervention is fundamental to patient care.”
Co-author Dr Ronald Tamler, said: “This analysis restores my faith in basic common sense. For the vast majority of people with diabetes, a healthy lifestyle with weight loss carries significant benefits; however, it’s not for everyone. Thanks to this work, clinicians can infer which patients will benefit the most from such a lifestyle intervention.”
The study was published in the Lancet Diabetes &Endocrinology.