Almost one-third of the elderly Medicare fee-for-service beneficiaries who died in 2008 underwent surgery within one year of their death, according to a retrospective study reported Oct. 6 in The Lancet. Many procedures performed on the elderly may be unnecessary or unwanted, and physicians should consider patients’ goals when making decisions about surgical intervention, researchers wrote.
End of life surgical care among those over age 65 varies based on age and U.S. region, according to the study.
Increasing age and likelihood for surgery in the final 12 months of life were inversely related. Older seniors, particularly those over 80 years, were less likely to undergo surgery compared to younger Medicare recipients, Dr. Alvin C. Kwok and his colleagues discovered (Lancet 2011 Oct. 6 [doi:10.1016/S0140-6736(11)61268-3]).
The elderly with Medicare coverage also were more likely to have surgery in the final year of life in regions of the country with a higher number of hospital beds and greater Medicare expenditures. In contrast, the numbers of surgeons in a region did not have a significant effect.
The substantial variations by age and region might suggest that health care provider discretion plays a role in allocation of end of life surgical care, according to Dr. Kwok and his colleagues. Dr. Kwok is a researcher in the Department of Health Policy and Management, Harvard School of Public Health in Boston.
Previous "studies have shown a strong association between Medicare costs and regional characteristics, including number of hospital beds or specialist physicians per head" (Ann. Intern. Med. 2003;138:273-87; Ann. Intern. Med. 2003;138:288-98).
These data suggest that "some hospital admissions could be discretionary or avoidable and some high-intensity treatments could be unnecessary or discordant with patients’ preferences," Dr. Amy S. Kelley, a member of the faculty at Brookdale Department of Geriatrics and Palliative Medicine, Mount Sinai School of Medicine, New York City, wrote in an accompanying commentary (Lancet 2011 Oct. 6 [doi:10.1016/S0140-6736(11)61420-7]).
Although multiple studies have assessed end of life medical care, only one previous study specifically examined surgery (Health Serv. Res. 2004;39:363-75), Dr. Kwok and his colleagues noted. So they assessed ICD-9-CM codes for the 1.8 million Medicare beneficiaries ages 65 and older who died in 2008. They found 32% had an inpatient surgery at an acute care hospital or skilled nursing facility during their last year of life, compared with 14% of Medicare recipients who survived that year. They also discovered 25% had a surgical procedure during their final three months of life; 18% during their final month; and 8% during their final week.
Those who had surgery within a year of death were significantly younger, an average 79.8 years of age, compared with an average 82.5 years among decedents who did not undergo surgery. So, despite increasing comorbidities with increasing age, the surgery rate actually dropped as ages climbed.
"These data should prompt careful consideration of patients' goals when assessing the need for surgical intervention at the end of life."
"Health care providers might have a high threshold to undertake surgery for elderly patients or a perception that these patients are more likely to have complications or less likely to benefit from such interventions than might younger patients," the authors wrote. "Some of these age-related differences could also be due to the preferences of patients and their families."
At age 65 years, 38% of patients were likely to undergo surgery in their final year of life. This dropped slightly to 35% by age 80 years, after which it decreased more dramatically to 24% for patients between 80 and 90 years of age.
Regional variations did not significantly differ by patient age, even for those older than 80 years. Because this was one of the first studies of its kind, the investigators devised a measure of end of life surgical intensity (EOLSI) to compare hospital referral regions. EOLSI reflects the rate of receipt of at least one surgical procedure during the last year of life for all those who died, after adjustments for age, gender, race, and income.
The adjusted EOLSI score varied by as much as three times between the lowest intensity region, Honolulu, and the highest, Munster, Ind., again indicating a wide variation in end of life surgical care by region. Dr. Kwok and his colleagues also found high-intensity areas had almost 40% more hospital beds and significantly higher Medicare spending, compared with low intensity areas.
"Our finding that surgical intensity is related to the number of hospital beds (although not the number of surgeons) suggests that wide-ranging practice patterns could exist in intensive regions that affect other aspects of care, such as when and how aggressively to intervene at the end of a patient’s life," the authors wrote. "For clinicians, these data should prompt careful consideration of patients’ goals when assessing the need for surgical intervention at the end of life."