Gene linked to length of hospital stay for people with psychosis

Researchers from the NIHR Maudsley Biomedical Research Centre (BRC) have discovered that variants of a gene called ZNF804A could be useful for predicting the amount of time that people suffering with psychosis will spend in hospital. 

ZNF804A has previously been confirmed as a genetic risk factor for schizophrenia, but this is the first time that a link has been made to longer-term patient outcomes in clinical practice.

“We need to find out whether these findings are repeated in other similar studies before drawing firm conclusions,” says the study’s lead author, Professor Robert Stewart, from the Institute of Psychiatry, Psychology and Neuroscience (IoPPN).  “But if the results can be confirmed, it would suggest that genes are not only deterministic of whether someone develops a mental disorder, but also of the severity of illness when it occurs.”

The research, published in the scientific journal Translational Psychiatry, was made possible by linking the Clinical Record Interactive Search (CRIS) application, an innovative text-mining tool, with routine clinical records.  The information was sourced from 291 people presenting to South London and Maudsley NHS Foundation Trust mental health services, at their first episode of psychosis. Medical records were linked to information about the genetic status of the patient, derived as part of the Genes and Psychosis (GAP) study of first-episode psychosis.

Of the three ZNF804A variants studied, carriers of the AA variant (just over 60% of the study population) spent an average of 96 days in hospital, compared to 68 days for carriers of the AC variant (around 30% of the study population) and 41 days for CC carriers (under 10% of the study population).

“Previous work done by NIHR BRC researchers has helped to highlight the potential molecular mechanism underlying this genetic association with the length of hospital stay. For instance, it is known that the risk variant studied here coordinates the activity of up to 150 different gene targets. Some of these are known to be critical for normal brain development”, says study co-author Dr Conrad Iyegbe, also from the Institute of Psychiatry, Psychology and Neuroscience (IoPPN).

According to Professor Stewart, “It’s well established that early experiences are very important in shaping the subsequent course of conditions such as psychosis, so this study could ultimately lead to opportunities for targeting treatments more effectively.  The next step is to gain a better understanding of the underlying mechanisms explaining why a genetic risk factor might influence why someone needs inpatient care more often.”

“This research also illustrates what is possible when we use the latest technology to link two large and disparate sets of data – connecting real-world clinical data to a complex biological genetic dataset,” he adds.  “This is the first time we’ve used this technology to show how a genetic characteristic is linked to a specific clinical outcome, but research using these tools is going to lead to more exciting advances like this in the future.”


Tags: Clinical and population informatics -

By NIHR Maudsley BRC at 15 Dec 2015, 16:31 PM


Back to Blog List