A new study suggests that the deadliest time for many hospital patients is not when they are in the operating room. Instead, it is the period when patients are recovering in the hospital and after they go home.

　　Researchers examined medical records of more than 40,000 patients, aged 45 years and older. The records came from 28 hospitals in 14 countries. Doctors had performed surgery on each patient, but none of the operations was related to the heart.

　　The researchers looked for patients with health problems or death within 30 days of surgery.

　　A total of five people died during the operation. That represents less than 1% of all patients. Another 500 patients, or 70%, died in the hospital. And 210 deaths, or 29%, did not happen until after patients were sent home.

　　Nearly half of all the deaths were linked to one of three problems: major bleeding, heart damage and blood infections.

　　P.J. Devereaux is with McMaster University in Canada and was the lead writer of a report on the study. The report was published in the Canadian Medical Association Journal.

　　Devereaux wrote that the research shows very few deaths happen in the operating room. He added that "there is a need to focus on post-operative care and transitional care into the home setting to improve outcomes." His comments were part of an email sent to the Reuters news agency.

　　Worldwide, 100 million patients age 45 and older have surgeries unrelated to heart-related issues every year.

　　The researchers noted that technological and medical improvements have made such surgeries safer and less invasive in recent years. However, patients are also coming to the hospital sicker and being sent home with complex care needs.

　　In the study, about half of the patients had high blood pressure. One in five had the disease diabetes, while 13% had heart disease.

　　More than one-third of them came in only for low-risk treatments that were not emergency operations.

　　Patients who experienced major bleeding after surgery were more than two times as likely to die within 30 days as people who did not have this complication. In addition, patients who developed heart injuries were also more than twice as likely to die.

　　The study was not a controlled experiment designed to identify which if any complications actually caused any deaths.

　　Barnaby Charles Reeves is with the University of Bristol in Britain. He told Reuters in an email that inflammation might be a common link to the health problems that were most responsible for deaths. He noted that, "Surgery causes a body-wide inflammatory reaction. This can lead to single or multi-organ failure which leads to death."

　　P.J. Devereaux said that patients might also not recognize that something is wrong when they begin to stop taking pain medication after surgery.

　　The researchers called for more research into improved methods for observing the condition of hospital patients after surgery. They said these methods can help identify ways to lower the risk of death following an operation.

　　I'm Jonathan Evans.

　　Lisa Rapaport reported this story for the Reuters news service. Jonathan Evans adapted it for Learning English. George Grow was the editor.