Volume, complications inversely related for cataract surgeons

November 1, 2007

Toronto-Researchers conducting a population-based analysis of cataract surgery in Ontario found a volume-outcomes relationship in which surgeons performing the highest number of procedures per year had the lowest complication rates. Overall, the selected adverse event rate for surgeons performing 50 or more cataract procedures was less than 1 in 200; surgeons who performed more than 1,000 procedures a year had a complication rate of 0.1%.

Toronto-Researchers conducting a population-based analysis of cataract surgery in Ontario found a volume-outcomes relationship in which surgeons performing the highest number of procedures per year had the lowest complication rates. Overall, the selected adverse event rate for surgeons performing 50 or more cataract procedures was less than 1 in 200; surgeons who performed more than 1,000 procedures a year had a complication rate of 0.1%.

“The more you do, the better you get, and it continues,” said Chaim Bell, MD, PhD, an assistant professor of medicine and health policy, management and evaluation at the University of Toronto, and St. Michael’s Hospital, and adjunct scientist at the Institute for Clinical Evaluative Sciences, Toronto. “We expected a plateau rate, but we didn’t find one, or any evidence of diminishing returns. There is something different about people who do a lot more surgeries, and it affects their complication rate.”

Dr. Bell and colleagues found a statistically significant linear relationship between volume and outcome (p <0.001) in each of 3 study years in their analysis.

As part of a study of wait times for cataract surgery at hospitals in Ontario, they tapped population-based administrative health records to analyze the number of surgeries, rates of surgery, and adverse effects. One of the objectives was to determine whether increasing the rate of surgery increased complications.

The researchers were aware that relationships between volume and outcome had been found for certain surgeries and treatments in specialties such as cardiology, with a doctor’s performance of more procedures associated with better outcomes.

“The interesting thing that we found for most of those types of surgeries in the literature was more along the lines that doing 10 surgeries was far better than doing two surgeries a year, or doing 30 surgeries is better than doing five surgeries a year. The difference between a high- and a low-volume surgeon might be 10, 20, maybe 50 procedures,” Dr. Bell continued.

Volume higher

Cataract surgery volumes, however, tend to be much higher. Whereas studies of surgeries in other specialties entailed single- or double-digit surgery volumes, cataract surgeons often have annual volumes in triple or quadruple digits.

With that large volume of cataract surgeries and an extensive, diverse sample of patients, surgeons, and hospitals available for analysis, the researchers had an excellent opportunity to study the association between surgeon volume and the risk of surrogate markers for important postoperative events within 2 weeks of surgery.

“We were wondering if there was a plateau effect in cataract surgery,” Dr. Bell said. “Cataract surgery had not been well-described, particularly in this era of new surgical techniques.”

He and his colleagues began their research by identifying patients aged at least 20 years who had undergone cataract surgery using the Ontario Health Insurance Plan physician claims database for the period between April 1, 2001, and March 31, 2004. Ophthalmologists were identified using their physician specialty codes and unique billing identifiers. Each surgeon’s annual cataract surgery volume was calculated based on the number of claims submitted for each fiscal year.

Only those who performed 50 or more cataract surgeries in 1 year were included in the analysis. Those physicians were linked with their respective patients to form the cohort. The baseline comparison group consisted of physicians who each performed 50 to 250 procedures a year; additional groups were physicians who performed 251 to 500, 501 to 1,000, or more than 1,000 procedures per year. Physicians with exceptionally low numbers of procedures annually were excluded to lessen miscalculation bias.

The number of cataract surgeries performed in Ontario during the study period ranged from 89,556 in the 2001–2002 fiscal year to 99,333 in 2003–2004. The reviewers looked at more than 284,000 procedures during the study period after excluding surgeons who performed fewer than the established minimum number per year. They reviewed annual volumes and also performed an analysis over the entire 3-year period by pooling all the data.

Between 231 and 243 surgeons each performed more than 50 cataract procedures per year at 70 hospitals or eye-surgery centers. The median surgeon volume was 353 in the first year, 391 in the second, and 392 in the third.

Complications evaluated

The evaluation of complications was determined by a composite of adverse events, including vitrectomy, dislocated lens extraction (with vitrectomy), air or fluid exchange (with vitrectomy), and vitreous aspiration or injection of medication performed between 1 and 14 days after cataract surgery. Researchers used physician billing claims to track the procedures, which were surrogate markers for retinal detachment, lost lens or lens fragment, and suspected endophthalmitis. A maximum of one adverse event was assigned to each surgery regardless of the actual number.

“The overall adverse event rate was less than 0.5% a year,” Dr. Bell said. The figure was 0.4% in the first year and 0.3% in the subsequent 2 years. That rate was inversely correlated with surgeon volume in each study year. For example, in 2003, surgeons who performed 50 to 250 procedures had an adverse event rate of 0.8%, whereas those who performed 251 to 500 procedures had a rate of 0.4% and those who performed 501 to 1,000 annually had a rate of 0.2%. Surgeons who performed more than 1,000 surgeries a year had the lowest complication rate, 0.1%.

Data pooled

The pooled data from the entire study showed the same adverse event rates at the same volume levels as the 2003 data. The volume-outcome relationship persisted after adjustment for age and gender and when adverse events were analyzed separately.

Dr. Bell, who is not an ophthalmologist, said that experts from that specialty would be best able to explain the volume-outcomes relationship for cataract surgery observed in this study. He speculated, however, that the low adverse event rate for very high-volume surgeons may be due to their ability to recognize problems during surgery and correct them at the time rather than treat them on a follow-up visit or refer the patient to another ophthalmologist. It is also possible that the high-volume surgeons have a different technique from other surgeons, he added.

Not only was the overall adverse event rate low, but the complications that occurred also were, for the most part, treatable, and patients had reasonably good visual outcomes, Dr. Bell said.

He cautioned, however, that the findings are general trends that might not be applicable to individual surgeons. “Just because an individual is a low-, medium-, or high-volume surgeon does not mean that [his or her] adverse events are low, medium, or high relative to others,” he said.

From a policy perspective, the findings of this study should be helpful to professional organizations and policy-makers who want to make recommendations on the appropriate level of cataract surgery required to maintain a high level of proficiency.

The analysis was published in the March 2007 issue of Ophthalmology.OT