Legal Research Databases to Study Rare Perioperative Outcomes in Patients with Obstructive Sleep Apnea

Malpractice claims are a high-value means to identify cases of rare severe perioperative morbidity and mortality. Although malpractice claims represent only a subset of patient injuries, biased by more severe permanent adverse outcomes and substandard care, they provide a cost-effective means to study infrequent serious injuries. They are also useful to stimulate and inform future research to improve patient safety. The Anesthesia Closed Claims Project has long utilized this approach to the study of adverse events in anesthesiology and pain medicine1 and now contains >10,500 claims against anesthesiologists. The Anesthesia Closed Claims database is generated by practicing, board-certified anesthesiologists who travel to participating medical liability insurance companies, where they review and systematically collect information from medical records, detailed expert witness reviews, as well as the legal and insurance company documents and evaluations. Several other sources of closed malpractice claims exist, including the National Practitioner Data Bank,2 medical liability insurance company data (e.g., National Health Service Litigation Authority in England,3,4 the PIAA5 [formerly the Physician Insurers Association of America]), and legal research databases6 (e.g., WestlawNext, Westlaw, LexisNexis, and LexisNexis Academic). The National Practitioner Data Bank contains required reports on adverse licensing actions and medical malpractice settlements against physicians and dentists, for use by hospital and state licensing boards. PIAA collects claims that are voluntarily submitted by member companies, representing approximately 28,000 closed claims per year (approximately 62% of the estimated 45,000 claims per year).5 The PIAA database is particularly useful for trends over time in malpractice settlements across specialties in the United States. Westlaw and WestlawNext (www.westlaw.com) are electronic legal research services edited by Thompson Reuters, with >40,000 databases of case law, state and federal statutes, and public records, representing all areas of law (e.g., intellectual property, immigration, bankruptcy, energy/environment, and health care law). LexisNexis and LexisNexis Academic (www.lexisnexis.com) are very large electronic databases for legal and public records (e.g., briefs, motions, libraries of statutes, case judgments), containing >5 billion documents for all areas of law. In this issue of Anesthesia & Analgesia, Fouladpour et al.7 studied medicolegal consequences of severe perioperative injuries in patients with known or suspected obstructive sleep apnea (OSA). They utilized the above online legal research electronic databases to identify cases and abstract basic patient and legal information available in these legal records. Of 77,630 medicolegal cases in these databases from 1991 to 2010, the authors identified >3000 cases as being possibly related to OSA. After review, the authors found 50 cases of adults (>18 years) with OSA who had postoperative complications. On further review, half of these cases were excluded because either they were not surgical cases or the injury was a surgical complication unrelated to OSA. During this 20-year time span, Fouladpour et al.7 identified only 24 cases of severe adverse perioperative outcomes related to OSA, with >80% of these occurring after 2000 and 41% after 2006. The patients were predominantly men (63%), middle aged, with a known diagnosis of OSA, undergoing elective surgical procedures. Approximately half of the complications occurred in the operating room or the postanesthesia care unit and were often related to difficult intubation or premature extubation. Approximately half of the adverse events occurred on surgical floors (46%) and were often related to respiratory arrest in an unmonitored setting. Immediate adverse outcomes were death (46%), anoxic brain injury (46%), and upper airway complications (8%), with death ultimately occurring in 71% of the cases. Verdicts favored the plaintiffs in 58% of cases, with very large financial settlements (mean, $2.5 million; range, $650,000–$7.7 million, not adjusted for inflation). The study highlighted that perioperative complications related to OSA are increasingly reported as the central contention of malpractice suits, with potentially large settlements. The study also points to the need for further research to reduce these severe perioperative complications in patients with OSA. The legal research databases provide a very different source of medicolegal information than the Anesthesia Closed Claims Project. As major sources of US legal cases, the legal research databases are used by attorneys in building their cases, estimating awards, and identifying experts. It is unclear exactly how the cases are submitted to the databases. Court decisions, as public records, are submitted. However, we do not know whether attorneys also submit cases because they may have little incentive to systematically collect data and enter the data into the database. In addition, it is unclear whether submission by plaintiff or defense attorneys is more common. The databases may be useful for advertising legal services (e.g., plaintiff attorneys may have reason to submit their high award cases). Hence, the legal databases are biased to court cases or other high settlements. As the authors acknowledge, the bias for court rulings in the legal databases underestimates the true medicolegal burden of OSA. The National Practitioner Data Bank also has this same bias because it includes claims with a settlement made on behalf of a named physician but does not include claims for which the physician is dropped or the lawsuit is discontinued. Most lawsuits are settled out of court. For instance, in the Anesthesia Closed Claims database of 10,546 claims, 88% of the claims involved a lawsuit and 12% were anesthesiologist reports of adverse events to their malpractice insurer. Of all lawsuits in the database, 22% had a trial verdict, 55% had a pretrial settlement, and 22% were dropped or discontinued. These data put the 24 cases described by Fouladpour et al.7 into perspective as representing only a small portion of OSA medical claims. This bias also explains the very large awards to plaintiffs. Another disadvantage of the legal databases is the lack of collection of detailed medical data. Legal documents briefly summarize the legal allegations and arguments, but do not provide specific information concerning the medical aspects of the case. The National Practitioner Data Bank and insurance company information (including the National Health Service Litigation Authority) also lack detailed medical record information. Attorneys often base legal arguments on a little more than a 50% probability rather than on scientific certainty. Allegations may not represent the true event that happened to the patient. The legal argument bias most noticeably arises during cases of postoperative neuropathies alleged to be because of malpositioning during surgery and regional block. On detailed review of medical records and expert witness analysis, the etiology of these events may instead be because of patient disease, correlating factors, or postoperative events. An example of this bias is use of older legal data in ischemic optic neuropathy after prone spine surgery, which alleged pressure on the globe because of malpositioning and later alleged hypotension and anemia as the etiology for the blindness, whereas recent scientific literature suggested instead that other factors are involved in the etiology.8 In contrast to the above databases, the Anesthesia Closed Claims Project systematically collects detailed medical information from medical records concerning the adverse events. Because of this inability to review medical records, Fouladpour et al.7 were not able to describe medical information concerning the management of the airway, anesthetic drugs, and the extubation criteria. Information concerning postoperative acute pain management, nursing checks, and other medical information were also lacking. In contrast, a recent Anesthesia Closed Claims analysis of 92 patients with postoperative opioid-induced respiratory depression found that most cases occurred within 24 hours of surgery, with multiple prescribers, concurrent administration of non-opioid sedating medications, and inadequate nursing assessments or response common among the cases.9 The Anesthesia Closed Claims Project has not formally studied injuries and liability in patients with OSA. However, a data search of 913 claims for procedural anesthesia (n = 827) and acute pain management (n = 104) collected between 2010 and 2014 revealed 117 claims that involved patients presenting with OSA. Nearly half of these claims represented adverse events related to OSA (difficult intubation [n = 7], premature extubation [n = 9], airway obstruction [n = 5], and inadequate oxygenation and ventilation [n = 29], similar to the findings of Fouladpour et al.7 Ninety-two percent of the claims with OSA-related adverse events were death (n = 46) or severe brain damage (n = 18). Of these 50 claims, 34 (68%) were settled, 1 (2%) went to trial, and 15 (30%) were dropped or discontinued. Thirty-five (75%) were paid with a median payment of $522,500 (range, $32,999–$3,045.003, adjusted for inflation). Patient registries are even more useful than malpractice claims to collect specific medical record information related to disease processes and rare outcomes. These registries can be generated from a variety of sources, including malpractice claims. Cases collected can then be used in case-control studies to investigate factors that were associated with the adverse outcome. This approach allowed the study of the risk factors for the rare event of blindness after prone spine surgery.8 To study the perioperative risks of OSA, the Society of Anesthesia and Sleep Medicine and the Anesthesia Quality Institute have partnered to create the OSA Death and Near Miss Registry.10 The goal of this registry is to identify common themes or factors associated with OSA-reported adverse perioperative events, including the level of monitoring used, understanding of why the event occurred, and provide insight for future studies for the perioperative care of patients with OSA.10 Cases can be submitted to this registry online (www.asaclosedclaims.org) under the tab for the OSA Death and Near Miss Registry. In summary, legal research databases are low-cost screening tools to initially investigate injury and liability associated with specific diseases or procedures. However, they are biased toward settlements and lack medical record information. DISCLOSURES Name: Karen B. Domino, MD, MPH. Contribution: This author helped prepare the manuscript. Attestation: Karen B. Domino approved the final manuscript. Conflicts of Interest: Karen B. Domino is the Director of Anesthesia Closed Claims Project, funded by the Anesthesia Quality Institute and American Society of Anesthesiologists, and Committee Member of the Society of Anesthesia and Sleep Medicine (SASM) Obstructive Sleep Apnea Death and Near Miss Registry. Name: Karen L. Posner, PhD. Contribution: This author helped prepare the manuscript. Attestation: Karen L. Posner approved the final manuscript. Conflicts of Interest: Karen L. Posner is the Project Manager of Anesthesia Closed Claims Project, funded by the Anesthesia Quality Institute and American Society of Anesthesiologists, and Committee Member of the Society of Anesthesia and Sleep Medicine (SASM) Obstructive Sleep Apnea Death and Near Miss Registry and receives partial funding from the SASM for work on this Registry. This manuscript was handled by: David Hillman, MD.