LAS VEGAS – A comorbid mental illness may predispose surgical patients to poor outcomes, increasing the risk of postoperative complications, a prolonged length of stay, and – in some cases – even in-hospital mortality.
The link between mental illness and physical response to surgery is not well elucidated, and is likely an extremely complicated one, Elizabeth Bailey, MD, said at the Association for Academic Surgery/Society of University Surgeons Academic Surgical Congress.
Dr. Bailey said there is an extreme paucity of data on the relationship between mental illness and surgical outcomes. To investigate it, she examined 580,000 patient records contained in the.
The cases spanned 2009-2011 and represented the four most common surgical procedures in the United States: cholecystectomy, appendectomy, adhesion excision/lysis, and colorectal resection.
She compared surgical outcomes among patients without a DSM-IV diagnosis and those with one of the five most common: mood disorder, anxiety, impulse control, schizophrenia, and substance abuse disorder.
The study’s primary outcomes were length of hospital stay, in-hospital mortality, and postoperative complications. Her analysis controlled for age, gender, race, admission status, operative approach, non–mental health comorbidities, insurance, and income.
Of the 580,000 in the study group 7% (39,000) had at least one of the mental health comorbidities. Mood disorder was the most common (59%), followed by substance abuse (24%), schizophrenia (13%), anxiety disorder (12%), and impulse control disorder (5%).
There were a number of significant baseline differences between those with a mental diagnosis and those without. Those with a DSM-IV diagnosis were younger (52 vs. 54 years), more often women (61% vs. 57%), and white (78% vs. 69%). They more often had additional physical comorbidities (80% vs. 68%). They were more likely to be admitted through the emergency room (74% vs. 71%), to have nonlaparoscopic surgery (60% vs 63%), to be on public insurance (53% vs. 43%), and to be in the lowest income quartile (28% vs. 25%).
Surgical outcomes were almost universally significantly worse among these patients. They were 41% more likely to experience a prolonged length of stay and 18% more likely to experience a complication. These included wound disruption, ileus, and small bowel obstruction. They faced a 24% increased risk for needing total parenteral nutrition; a 29% increased risk of abdominal pain; an 18% increased risk of percutaneous abdominal drainage; and a 15% increased risk of needing another operation in the same admission.
Dr. Bailey also broke down overall risks by DSM-IV diagnosis.
• Patients with a mood disorder were 35% more likely to have a prolonged length of stay and 13% more likely to have a surgical complication.
• Patients with an anxiety disorder were 16% more likely to have a prolonged length of stay and 10% more likely to have a complication.
• Patients with schizophrenia were 77% more likely to have a prolonged length of stay, 3% more likely to die, and 28% more likely to have a complication.
• Patients with substance abuse were 70% more likely to have a prolonged length of stay, 6% more likely to die, and 39% more likely to have a complication.
Interestingly, Dr. Bailey said, the risk of in-hospital death was 16% lower in patients with a mood disorder, 59% lower in those with an anxiety disorder, and 77% lower in those with an impulse control disorder.
She stressed that the National Inpatient Sample provides a limited look into a patient’s hospital experience. The study can’t assess how long patients were sick before they came to the hospital, their medications or medication adherence, or how well they managed their mental and physical comorbidities.
“While we lacked the means to delve into potential clinical mediators, look at unplanned readmissions, or the use of inpatient psychiatric consults, we can clearly see the association with worse surgical outcomes,” Dr. Bailey said. “Recognizing this is the first step in learning how to optimize care for this frequently marginalized population.”
She had no financial disclosures.