Surveying uveitis specialists—a call for consensus

In the early 1990s, former Michigan supreme court Justice Thomas Brennan became disillusioned with popular law school rankings and so decided to survey 100 academics, judges, and lawyers on his own, asking them to rank a list of ten schools he provided. He used a composite index similar in structure, but different in content, to those used by mainstream surveyors, such as U.S. News & World Report. As expected, many of the big name schools—Harvard, Yale, Stanford—made it to the top of the list. Penn State, as Brennan recalled, “[Was] about in the middle of the pack. Maybe fifth among the 10 schools listed.” There was one small problem, however. Penn State had no law school at the time. Brennan had included it to make a point: surveys are limited by both the quality of the questions asked and by how familiar respondents are with the subject being surveyed [1, 2]. Similar surveys using questionnaires are employed commonly in medicine and, whenwell-designed and implemented, can provide valuable information regarding perceptions, preferences, and practice patterns—a finger on the pulse, if you will, of current treatment approaches [3, 4]. Almost without exception, variation is the rule in such surveys. Engstrom and colleagues, for example, published in 1991 the results of a survey of 62 treating uveitis specialists in the American Uveitis Society (AUS) regarding what were then current practices in the management of ocular toxoplasmosis [5]. While a sizable proportion of respondents viewed poor vision (<20/200), marked or severe vitreous inflammation, and zone 1 lesion location [6] as absolute indications for antimicrobial therapy, the extent to which experts reached agreement regarding when to initiate treatment was by no means absolute. Moreover, the lack of consensus was even more striking when considering which agent(s) to choose. Specifically, experts reported use of ten different antimicrobial drugs in more than four different combinations, with approximately one third recommending a combination containing pyrimethamine, folinic acid, sulfadiazine, and prednisone. Holland and Lewis updated this survey in 2002, publishing the results gathered from 79 respondents who evaluated and managed patients with ocular toxoplasmosis [7]. Similar “absolute treatment indicators” were identified, but this time 11 different agents were used in an equal number of combinations, including some not mentioned in the 1991 survey. Once again, the most common approach involved the combined use of pyrimethamine, folinic acid, sulfadiazine, and prednisone—although more than two thirds recommended other single or combination agent treatment regimens. Similar trends and variations in treatment approach were observed when uveitis specialists were surveyed in 2011 by Wakefield and associates [8]—fully 20 years after the original ocular toxoplasmosis practice survey was published. Even greater variation regarding indicators of treatment exists among nonuveitis specialists [9]. One thing well-designed surveys of medical practitioners can do particularly well is identify limitations and deficiencies in the level and understanding of current clinical knowledge, E. T. Cunningham Jr. California Pacific Medical Center, San Francisco, CA, USA