A project co-funded by the European Union and
the European Society of Cataract & Refractive Surgeons
ET: What are the potential research benefits of building an international cataract and refractive surgery outcomes registry? Aren't randomised clinical trials enough to evaluate new procedures and technologies?
Prof Lundström: The really important thing about registries is they are supposed to collect data on all cases so they show you the results of treatments in the real world.You don't get this with randomised
trials.To have the highest predictive power with randomised trials you need strict exclusion criteria. But when you keep out a lot of different patients this also means the
trial results are not generalisable. In the registries we can see what really happens when we apply a technique to all kinds of patients.We can also see it over a much
longer time, which can be very important. The orthopaedics registries have found that certain implants that worked well at first started to give problems after three
or four years.You can't have a clinical trial go on for so long.
ET: Clinical trial data are collected under controlled circumstances, which is a major factor contributing to the reliability of their findings. How do registry data compare in terms of the strength of evidence they produce?
Prof Lundström: The strength of evidence in a registry depends on many factors.There must be reasonably large amounts of data, and you have to make sure that it is not just selected cases – you have to be sure that critical parameters like complications are represented in the data. If the registries have a high coverage of cases, if they have systems for validating data, and if they are multicentre or
national registries the strength of the evidence they produce becomes stronger and stronger. In a general discussion about different levels of evidence, with prospective randomised trials
at the top and expert opinion at the bottom, registry evidence is getting closer in reputation to the clinical trial.
ET: All over the world purchasers of healthcare services, whether they are
national health plans, insurance companies or individuals, are looking for quality data to guide their buying decisions.Are you concerned at all that registry data might be used to rank providers for this purpose?
Prof Lundström: Registry data are not very suitable for this kind of use. If you use registry data in aggregate for calculating risk or complication rates, they are very robust. But if you want to rank units or
rank surgeons and there is a certain percentage of missing data, the confidence intervals become so large the data are almost meaningless. If there is just a three per cent loss of data and we calculate based on my hospital number being 45 out of 70, that is misleading because the actual result could be anywhere between number five and number 70.Therefore, it is extremely dangerous to rank if you don't have almost 100 per cent data. Registries also may not take into account patient mix. If you have complicated cases, you give
them to the most skilled, senior surgeons, so your best surgeon may end up with your highest vitreous loss rate because he takes all the hardest cases. I think they may bring forward the possibility of using the
data to rank us, but I do not think they will force us because they are well aware of the difficulties.The data may be good enough to assign centres to categories, but they are not good enough to precisely rank
units or individuals. In Sweden, the authorities are not interested in going below the clinic level, or at least that is what they are saying now.
ET: It sounds like much of the success of the EUREQUO project will depend on getting a high level of cooperation from the participating countries and surgeons.What are the prospects of gaining such cooperation?
Prof Lundström: It seems there is a movement under way; it is coming more and more. In some countries it is the initiative of professionals, in others it is the
health authorities that are pushing for it, but everyone wants this data.There are registries in Holland and Slovenia and Malaysia and Perth in Western Australia. They do not yet have the same coverage that we do in Sweden, but they are working toward it.There is a lot of enthusiasm for it in southern Europe and
in the east.They have incentive to show that they have the same standards in place and produce the same outcomes as other countries.The support of national health authorities does help. In Sweden every
private cataract surgeon must participate to get paid. But the national board of health in Denmark set up a registry that failed to reach sufficiently high coverage, so the authorities stepped in and cut the funding. Some countries will be quicker than others, but as more participate that
will stimulate the others to join.
ET: When the ESCRS European Cataract Outcomes Survey (ECOS) moved from a paper system to on online entry system, participation dropped because clinics could not export data directly from their systems to the ECOS database. Do you anticipate similar issues with EUREQUO, and if so,
how will they be addressed?
Prof Lundström: There will be a number of improvements to make the system more user-friendly. One of the goals of the three year development grant from the EU is to develop data export tools that will integrate directly into electronic medical record systems so that centres will not have to double-enter data. The software will reside in the computers so participants will not even have to be connected to the Internet.We will most likely pick up coverage quicker in areas where electronic record systems are the standard.
ET: Past attempts to keep registries current also have encountered difficulties keeping up with rapidly changing technology. How do you see EUREQUO addressing this challenge?
Prof Lundström: It is difficult to decide what should be in the database given the speed technology changes. Cataract
surgery still is to correct a medical disability, but it is increasingly for refractive purposes as well. If you focus on one indication for surgery but not the other you can miss important data.We try to
include all that we think is important, but it is not entirely scientific; you cannot include all parameters.The important thing is that important indicators of quality of care are
identified and collected. I hope that this European registry will somehow stimulate a discussion about how an audit should be conducted, what data are needed and what it should look like. One drawback of previous refractive registry systems is that when surgeons were asked to give data for
a national audit, the data being collected was not enough. Individual countries must decide what should be in an audit, but if EUREQUO could set a standard for what
ought to be included in the audit this would increase the value of the system for surgeons. My hope is that this project is moving us in the right direction. It will take time, of course.
mats.lundstrom@ltblekinge.se
This web-site arises from the project EUREQUO which has received funding from the European Union, in the framework of the Public Health Programme