Results from the International Cataract Surgery Outcomes Study

Abstract It is widely accepted that cataract extraction with intraocular lens implantation is a highly effective and successful procedure. However, quality assessments and studies of effectiveness should still be undertaken. As with any surgical treatment modality, complications may occur, leading to suboptimal outcomes, additional health costs and deterioration in patients' functional capacity. International variation in clinical practice patterns and outcomes can serve as important pointers in the attempt to identify areas amenable to improvements in quality and cost‐effectiveness. Once demonstrated, similar clinical results obtained in different health care systems can improve the level of confidence in a clinical standard against which the quality of care can be evaluated. The International Cataract Surgery Outcomes Study was established in 1992. The objective of this international comparative research project was to compare cataract management, outcomes of surgery and quality of care in four international sites. The study was conducted in the 1990s, since when many developments and refinements have emerged within cataract surgery. The actual figures reported in this thesis may no longer be of specific relevance as a decade has passed since their collection. However, the research questions and methods used in the study are still highly important and justify the publication of this report. The report deals with problems related to quality assessment, benchmarking, and the establishment and design of nationwide clinical databases − issues that are currently the focus of much attention. Moreover, the problems related to cross‐national comparisons are increasingly relevant as more international databases are established. The study makes suggestions on how to report and compare objective as well as subjective criteria for surgery. The issue of how to report subjective criteria is a particular subject of current discussion. Four sites with high‐quality health care systems were examined in this study: the USA, Denmark, the Province of Manitoba (Canada), and Barcelona (Spain). The design of the international research programme was based on methods developed by the US National Cataract Surgery Outcomes Study conducted by the US Cataract Patients Outcomes Research Team. The International Cataract Surgery Outcomes Study comprised three separate studies: a survey of ophthalmologists; a prospective cohort study, and a retrospective register‐based cohort study. The survey study was based on data generated by a self‐administered questionnaire completed by ophthalmologists in the four study areas. The questionnaire examined routine clinical practice involving patients considered for cataract surgery, and included questions on anaesthesia, monitoring and surgical techniques. The prospective cohort study was a large‐scale, longitudinal observational study of patients undergoing first‐eye cataract surgery in each study site. Patients were sampled consecutively from multiple clinics and followed for 4 months postoperatively. The retrospective cohort study was based on the Danish National Patient Register and claims data from the USA. This study could not be carried out in Barcelona or Manitoba as no suitable administrative databases were available. The papers based on register databases deal with retinal detachment and endophthalmitis but are not included in this thesis as the material was previously reported in my PhD thesis. The application of the studies was highly co‐ordinated among the four sites and similar methods and instruments were used for data collection. The development of the data collection strategy, questionnaires, clinical data forms and data analyses were co‐ordinated through weekly telephone conferences, annual in‐person conferences, correspondence by mail or fax, and the exchange of sas programs and data files via the Internet. The survey study was based on responses from 1121 ophthalmologists in the four sites and results were presented in two papers. Within the previous year the participating ophthalmologists had performed a total of 212 428 cataract surgeries. With regard to preoperative ophthalmic testing, the present study reveals that refraction, fundus examination and A‐scanning were performed routinely by most surgeons in all four sites. Other tests were reported to be performed routinely by some surgeons. It is unclear why any surgeon would use these other tests routinely in cataract patients with no ocular comorbidity. It appears that if this recommendation from the US Clinical Practice Guidelines Panel was broadly accepted, the use of these procedures and costs of care could be reduced, especially in Barcelona, the USA and Canada. Restricted use of medical screening tests was reported in Denmark. If this restricted screening were to be implemented in the USA, Canada and Barcelona, it would have significant resource implications. The most striking finding concerned the difference in monitoring practice between Denmark and each of the other three sites. In Denmark, monitoring equipment is seldom used and only occasionally is an anaesthesiologist present during cataract surgery. By contrast, in the other study sites, the presence of an anaesthesiologist using monitoring equipment is the norm. Adopting the Danish model in other sites would potentially yield significant cost savings. The results represent part of the background data used to inform the decision to conduct the two large‐scale, multicentre Studies of Medical Testing for Cataract Surgery. The current study is an example of how surveys of clinical practice can pinpoint topics that need to be examined in randomized clinical trials. For the second study, 1422 patients were followed from prior to surgery until 4 months postoperatively. Preoperatively, a medical history was obtained and an ophthalmic examination of each patient performed. After consent had been obtained, patients were contacted for an in‐depth telephone interview. The interview was repeated 4 months postoperatively. The interview included the VF‐14, an index of functional impairment in patients with cataract. Perioperative data were available for 1344 patients (95%). The 4‐month postoperative interview and clinical examination were completed by 1284 patients (91%). Main reasons for not re‐evaluating patients were: surgery was cancelled (3%); refusal to participate (2%); lost to follow‐up (1%), and death or being too sick (1%). The results have been presented in several papers, of which four are included in this thesis. One paper compared the preoperative clinical status of patients across the four sites and showed differences in both visual acuity (VA) and VF‐14 measures. The VF‐14 is a questionnaire scoring disability related to vision. The findings suggest that indications for surgery in comparable patients were similar in the USA and Denmark and were more liberal than in Manitoba and Barcelona. The results highlight the need to control for patient case mix when making comparisons among providers in a clinical database. This information is important when planning national databases that aim to compare quality of care. A feasible method may be to use one of the recently developed systems for case severity grading before cataract surgery. In another paper, perioperative clinical practice and rates of early complications following cataract surgery were compared across the four health care systems. Once again, the importance of controlling for case mix was demonstrated. Significant differences in clinical practice patterns were revealed, suggesting a general trend towards slower diffusion of new medical technology in Europe compared with North America. There were significant differences across sites in rates of intra‐ and early postoperative events. The most important differences were seen for rates of capsular rupture, hyphaema, corneal oedema and elevated pressure. Rates of these adverse events might potentially be minimized if factors responsible for the observed differences could be identified. Our results point towards the need for further research in this area. In a third paper, 4‐month VA outcomes were compared across the four sites. When mean postoperative VA or crude proportions of patients with a visual outcome of < 0.67 were compared across sites, a much poorer outcome was seen in Barcelona. However, higher age, poorer general health status, lower preoperative VA and presence of ocular comorbidity were found to be significant risk factors associated with increased likelihood of poorer postoperative VA. The proportions of patients with these risk factors varied across sites. After controlling for the different distributions of these factors, no significant difference remained across the four sites regarding risk of a poor visual outcome. Once again the importance of controlling for case mix was demonstrated. In the fourth paper, we examined the postoperative VF‐14 score as a measure of visual outcomes for cataract surgery in health care settings in four countries. Controlling for case mix was also necessary for this variable. After controlling for patient case mix, the odds for achieving an optimal visual function outcome were similar across the four sites. Age, gender and coexisting ocular pathology were important predictors of visual functional outcome. Despite what seemed to be an optimal surgical outcome, a third of patients still experienced visual disabilities in everyday life. A measure of the VF‐14 might help to elucidate this issue, especially in any study evaluating the benefits of cataract surgery in a public health care context.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.