A recent study by PLOS One unveils that it is possible to identify the medical conditions of the people by analyzing their social media posts. A research team from the University of Pennsylvania has conducted a study on 999 consenting patients by compiling their electronic medical data and social media data. The study used the content which consists of around 20 million words written by them. The main objective of the study was to examine whether these contents can predict medical conditions.
The research paper reveals that the findings are positive. “We identified that patients’ Facebook status updates can predict many health conditions, suggesting opportunities to use social media data to determine disease onset or exacerbation and to conduct social media-based health interventions,” says the study. The research team also claims that it is the first report that integrates the electronic medical data of the patient with social media data.
Using the algorithm, it has found that the language pattern observed in social media posts could imply the medical conditions. As part of the research, the team checked 949,530 Facebook status updates across 999 participants. There were words like ‘drink’, ‘bottle’ which frequently appeared in the social media posts that clearly indicated Alcoholism. Depression was often implicit from the words ‘pain’, ‘tears’, ‘hurt’, etc.
The study also confirms that it is very effective in identifying medical conditions such as diabetes and mental illness. “It was particularly effective at predicting diabetes and mental health conditions including anxiety, depression and psychosis,” the researchers said in the paper. The medical conditions categorized under 21 categories in the research article include diabetes, pregnancy, anxiety, psychoses, chronic pulmonary disease, STDs, drug abuse, and alcohol abuse.
Albeit the research point out the possibility of social media in medicine, the research team also stated their concern over the issues of privacy and the consent of the users. “The power of social media language to predict diagnoses raises parallel questions about privacy, informed consent and data ownership,” the study added. The study was published online in the peer-reviewed the scientific journal PLOS One.
Matt holds a bachelor degree in Journalism. Matt is the Editor-in-Chief of Live News Herald and covers Business and Health news.