Practitioners voice objections to study on malpractice suit risks

Richard L. Abbott

A study on the correlation between high-volume refractive surgery practices and lawsuits has sparked vigorous debate in the eye care profession. The study, “Medical malpractice predictors and risk factors for ophthalmologists,” further asserts that high-volume surgical cases comanaged with optometrists are associated with an increased risk of lawsuits.

Authored by Richard L. Abbott, MD, of the University of California, San Francisco, and colleagues, the study was published in the November 2003 issue of Ophthalmology. Dr. Abbott also presented information from the study at the recent American Academy of Ophthalmology meeting in Anaheim, Calif.

While practitioners do not necessarily question the data presented in the study, many consider these results to have been misinterpreted and misrepresented by the study’s authors.

“The study itself was simply a study, and it was a very well-done study,” said J. James Thimons, OD, a Primary Care Optometry News Editorial Board member who comanages refractive surgery patients in Fairfield, Conn. “But I think it was inappropriate to present a distorted view of comanagement and to imply that a comanaged relationship is in some way inferior. The statistics in the article clearly show that high volume has a lower incidence rate, but the text of the article implies otherwise, and is disingenuous in that regard.”

Dr. Abbott acknowledged that the presentation and wording of his findings engendered a good deal of confusion and controversy among optometrists. “The whole purpose of this study was to better understand the medicolegal predictors and risk factors,” he told Primary Care Optometry News. “It was written strictly from an insurance perspective and is not a reflection of patient care. Obviously, there was some misinterpretation, and it is unfortunate that this came about.”

Details of the study

The stated purpose of the retrospective, longitudinal cohort study was “to identify physician predictors in LASIK and photorefractive keratectomy (PRK) that correlate with a higher risk for malpractice liability claims and lawsuits.”

“We defined a ‘predictor’ in the study as being a statistically significant (p<0.05) characteristic in a physician’s profile that indicates the likelihood that an event (claim or lawsuit) will occur,” Dr. Abbott said. “We defined a ‘risk factor’ as an important — but not necessarily statistically significant — characteristic in a physician, patient or case profile that indicates the likelihood that an event will occur.”

According to the abstract, the study made a comparison of physician demographic and practice pattern data of 100 consecutive Ophthalmic Mutual Insurance Company (OMIC) LASIK and PRK claims and lawsuits with corresponding data for all active refractive surgeons insured by OMIC between 1996 and 2002.

Background information and data were acquired from OMIC underwriting applications, a physician survey and claims file records. Using an outcome of whether the physician had a history of a claim or lawsuit, logistic regression analysis was conducted for each predictor, and refractive surgery outcome was controlled.

According to the results, logistic regression analysis determined that the most important predictor of filing a claim was surgical volume, with those doctors performing more surgery having a greater risk of incurring a claim (odds ratio [OR] = 31.4 for >1,000 surgeries/year vs. 0 to 20 surgeries/year, 95% confidence interval [CI] = 7.9-125, p = 0.0001). Having one or more prior claims was the only other predictor examined that remained statistically significant after controlling for patient volume (OR = 6.4, 95% CI = 2.5-16.4, p = 0.0001).

The results went on to state that physician gender, advertising use, preoperative time spent with the patient and comanagement seemed to be strong risk factors in mulitvariate analyses when surgical volume was greater than 100 cases per year.

The researchers concluded that the chances for incurring a malpractice claim or lawsuit for PRK or LASIK correlate significantly with higher surgical volume and a history of claim or lawsuit.

One of the concluding statements of the study in particular was strongly refuted by optometrists. “These findings may be used in the future to help improve the quality of care for patients undergoing refractive surgery,” the study stated, “and to provide data for underwriting criteria and risk management protocols to manage proactively and perhaps reduce the risk of claims and lawsuits against refractive surgeons.”

The volume factor

Dr. Thimons discussed with Primary Care Optometry News what he considered to be the flaws in the statistical analysis of Dr. Abbott’s data. He said that, according to these data, the highest-volume surgeons performed more than 19,000 procedures and had only six complications that led to litigation.

“That is an incidence rate of less than 0.03% — that’s one in 3,167 procedures. On the other hand, the lowest-volume surgeons had an average complication rate of 0.12% — one out of every 863 procedures,” he said. “This is a 3.7 times greater risk to the patient of having a complication leading to a lawsuit by using a low-volume, inexperienced surgeon than by using a high-volume, experienced surgeon.”

Dr. Thimons cited data from Table 3 of the study (physician predictors for incurring a claim or lawsuit). “In that database group, the study shows that the ophthalmologists who performed 100 or fewer procedures per year were at a lower risk for malpractice litigations than were surgeons who had performed more than 1,000 procedures per year,” Dr. Thimons said. “That’s intuitive. If you perform fewer procedures, you should probably have less risk of action being taken against you. What is important is not the raw number of occurrences, but the percentage occurrence rate.”

Dr. Thimons added that comanaging optometrists often choose to work with higher-volume surgeons due to their level of experience. “The majority of comanaging optometrists do not choose a surgeon who is doing five cases a month,” he said. “They choose someone who is doing a volume that is reasonable to underwrite an increased skill set and better outcomes for their patients.”

Dr. Abbott agreed that his findings on volume were intuitive, but not yet proven scientifically. “I said that it is intuitive when we first began the study,” he said. “But we didn’t have any data, so we wanted to analyze these cases and look specifically. Consistant with our use of the data for insurance operations, the findings and interpretation of the data were on an absolute basis and not on a relative or per procedure basis.”

Significance of comanagement

Michael D. Twa, OD, MS, FAAO, a researcher at The Ohio State University, discussed the significance of the comanagement variable. “The type of analysis that was done — a logistical regression model — was appropriate,” he said in an interview. “Using this kind of analysis, you can consider any of the variables they measured: surgical volume, sex, age, geographic location, comanagement, etc., as a predictor for being sued, and you look at a lot of those things together. Any one of them could turn out to be significant, or not.”

He added that it is then important to control the factors that are most dominant in determining the outcome. “In this case, they found one that was more important than others, and that was the number of cases that the doctor performed,” he said. “So volume is driving who gets sued and who doesn’t.”

Dr. Twa also evaluated the p value of comanagement as a risk factor. “When you look at the p value, you are determining how probable it is that something was statistically significant or important,” Dr. Twa said. “If you look at the probability of comanagement before you adjust for volume, it appears to be a significant factor, but when you adjust for volume, it is not.”

Dr. Thimons also assessed the study’s data pertaining to comanagement. He said, according to these data, the odds ratio of a comanaged patient having a risk factor vs. a non-comanaged patient is 1.03. “If you look at that number, when you look at odds ratios, 1.0 is equal odds — that means it could happen on either side,” he said. “Statistically, 1.03 is an insignificant difference in the odds ratio between a comanaged patient and a non-comanaged patient.”

Dr. Thimons pointed out that comanagement is very often linked to high-volume practices. “Comanaged patients tend to be more typically associated with higher-volume practices. This is related to the referring doctor selecting experience, access to technology and improved patient outcomes,” he said. “So while the number of occurrences may seem to be higher, an odds ratio analysis shows that this is actually not the case.”

Dr. Abbott acknowledged that his study did not find comanagement to be a statistically significant risk factor independent of volume. “As we stated in the paper, the most important findings in the study were high surgical volume and history of a prior claim or suit,” he said. “As the volume increased, the association of a higher litigation risk with comanagement was related to volume, not to comanagement itself. We found nothing inherent in comanagement that caused a higher claim or risk.”

However, according to Louis E. Probst, MD, a medical director for TLC Laser Eye Centers, these findings were contradicted by statements in the text of the paper, which suggests a strong correlation between lawsuits and comanagement.

“The authors actually show (in Table 3) that there is no relationship between comanagement and malpractice suits, yet in the text of their paper, they state that comanagement is a ‘significant risk factor’ to lawsuits,” Dr. Probst told Primary Care Optometry News. “It is not once mentioned in the text that, once adjusted for volume, comanagement is not a significant risk factor.”

Dr. Abbott conceded that the wording of the paper may have lent itself to confusion. “There were some statements in the paper that alluded to comanagement being a risk factor,” he said. “But what we found when we looked at this more closely was that it was all related to volume. As defined in the beginning of the paper, risk factors were important, but not necessarily statistically significant.”

Dr. Abbott emphasized that the study’s examination of malpractice suits was designed to gather information for insurance purposes, not to make determinations about patient care.

“The reader has to understand that we did this from an insurance perspective,” he said. “So if we insure two doctors, and doctor A does 1,000 cases and Dr. B does 100 cases, the risk for insuring the high-volume doctor will be greater. It has nothing to do with per-case risk, so the high-volume doctor may in fact have a lower percentage risk. We don’t comment on that at all. It is merely an absolute number from an insurance perspective.”

Impact of study

According to Dr. Probst, the study is likely to have far-reaching implications not only for optometrists, but ophthalmologists as well. “What this paper does is effectively blacklist certain types of refractive surgeons, particularly those who are high-volume, male and who comanage,” he said. “We are currently in a malpractice crisis in the United States; litigation claims are going through the roof, and there are certain areas of the United States where it is difficult for doctors to get insurance at all.”

According to Dr. Probst, this crisis can only be worsened by the data contained in Dr. Abbott’s study. “The danger with this paper — especially in light of the misinterpretation of the data — is that insurance companies might take this at face value,” he said. “Insurance companies could not only raise the rates, but refuse to cover refractive surgeons who fall into these ‘risk’ categories.”

Without malpractice insurance, Dr. Probst added, a surgeon is essentially incapacitated. “They really couldn’t practice,” he said. “So this is coming under the radar as an academic exercise, looking at the risks of a lawsuit, but really, it is basically hindering surgeons who want to comanage.”

Optometrists also are concerned about potential damage the study could do to both their livelihoods and reputations.

“At the end of the article, the researchers state that the study should help improve the overall quality of care provided to refractive surgery patients, by identifying and addressing these risks,” said Louis Phillips, OD, FAAO, president of the Optometric Council on Refractive Technology. “But the study didn’t look at quality of care. If you are going to say patients receive better quality of care or inferior quality of care because they are comanaged, your study should evaluate quality of care.”

For Your Information:

• Richard L. Abbott, MD, is a clinical professor of ophthalmology at the University of California, San Francisco. He can be reached at 10 Kirkham St., K-301, San Francisco, CA 94143; (415) 502-6265; fax: (415) 502-7418; rabbott@itsa.ucsf.edu. He is also a paid consultant to OMIC.
• J. James Thimons, OD, is a Primary Care Optometry News Editorial Board member who practices in Fairfield, Conn. He can be reached at 75 Kings Highway Cutoff, Fairfield, CT 06475; (203) 366-8000; fax: (203) 334-2401; jim.thimons@tlcvision.com.
• Michael D. Twa, OD, MS, FAAO, is a researcher at The Ohio State University College of Optometry. He can be reached at 338 West 10th Ave., Columbus, OH 43218-2342; (614) 292-5586; fax: (614) 292-4949; mtwa@optometry.osu.edu.
• Louis E. Probst, MD, is the medical director for TLC Laser Eye Centers Chicagoland, Windsor, Ontario; and Madison, Wis. He can be reached at 4 Westbrook Corp. Center, Ste. 111, Westchester, IL 60154; (708) 562-2020; fax: (708) 562-4785.
• Louis Phillips, OD, FAAO, is president of the Optometric Council on Refractive Technology and practices at SightLine Laser Eye Center in Sewickley, Pa. He can be reached at 2591 Wexford-Bayne Road, Ste. 104, Sewickley, PA 15143; (724) 933-5588; fax: (724) 933-6051; louisjpod@aol.com.