HRS: Do clinical data registries improve care?
The public reporting of healthcare data, particularly in the cardiovascular field, has been in the spotlight as the National Cardiovascular Data Registries (NCDR) among others, aim to improve quality of care. And during a presentation May 4 at the annual Heart Rhythm Society meeting, Stephen C. Hammill, MD, from the Mayo Clinic in Rochester, Minn., said that these types of registries will have many benefits.
Already within NCDR's National ICD Registry are 750,000 records depicting more than 90 percent of device implants. These data include information about leads, complications and pediatric implants, among others.
Hammill called the quarterly benchmarking a strength to these registries because “you are able to see what your results for each data element and how it compares to hospitals of a similar size and others in the U.S."
“Registries are here to help us measure how well we are managing evidence-based medicine,” Hammill said.
Is all the data reported evidence-based?
A recent study found that 22.5 percent of ICD implants reported to the registry were non-evidence-based. Meaning that these included implants for patients who had:
Because the number of non-evidence based implants was high, Hammill said that there were some concerns that the figures were both overestimated and underestimated.
For those who over estimated the issue, Hammill said that they were quick to point the figure at inaccurate data within the registry, while those who underestimated the problem said that “cherry picking” could have been involved. However, Hammill was quick to say that due to the fact that reporting is beginning to be mandated, he doubts clinicians would inaccurately report data publicly.
However, Hammill argued that non-evidence based implants are often appropriate it just remains a question as to how often.
Hammill said he was uncertain what the ratio of non-evidence based implants to evidence-based outcomes should be but agreed that currently there is “a wide variation in the community on how we practice evidence based implantation.”
He urged that it will be imperative that the data elements collected in these registries are up to par and accurate.
Do these types of registries improve care?
Hammill offered that this will all depend on the data and that “a registry with feedback like this provides an important opportunity to improve care.”
He did offer, however, that administrative data has had its problems. These problems stem from the fact that coding may not be supported by a clinical record and create bias toward maximizing reimbursement, there were problems distinguishing comorbidities from complications and that these types of data do not take into account that clinicians are treating a sicker patient population.
However, he said that in the future NCDR goals and registries will help:
Lastly, Hammill said that public reporting “should occur at the appropriate level of accountability and include a formal process for evaluating the impact of the program on the quality and cost of healthcare including the assessment of unintended consequences such as, will be start to avoid treating the most complex population of patients now that these outcomes data are public?"
Already within NCDR's National ICD Registry are 750,000 records depicting more than 90 percent of device implants. These data include information about leads, complications and pediatric implants, among others.
Hammill called the quarterly benchmarking a strength to these registries because “you are able to see what your results for each data element and how it compares to hospitals of a similar size and others in the U.S."
“Registries are here to help us measure how well we are managing evidence-based medicine,” Hammill said.
Is all the data reported evidence-based?
A recent study found that 22.5 percent of ICD implants reported to the registry were non-evidence-based. Meaning that these included implants for patients who had:
- An MI within 40 days;
- CABG within three months;
- NYHA Class IV; or
- Newly diagnosed heart failure (HF).
Because the number of non-evidence based implants was high, Hammill said that there were some concerns that the figures were both overestimated and underestimated.
For those who over estimated the issue, Hammill said that they were quick to point the figure at inaccurate data within the registry, while those who underestimated the problem said that “cherry picking” could have been involved. However, Hammill was quick to say that due to the fact that reporting is beginning to be mandated, he doubts clinicians would inaccurately report data publicly.
However, Hammill argued that non-evidence based implants are often appropriate it just remains a question as to how often.
Hammill said he was uncertain what the ratio of non-evidence based implants to evidence-based outcomes should be but agreed that currently there is “a wide variation in the community on how we practice evidence based implantation.”
He urged that it will be imperative that the data elements collected in these registries are up to par and accurate.
Do these types of registries improve care?
Hammill offered that this will all depend on the data and that “a registry with feedback like this provides an important opportunity to improve care.”
He did offer, however, that administrative data has had its problems. These problems stem from the fact that coding may not be supported by a clinical record and create bias toward maximizing reimbursement, there were problems distinguishing comorbidities from complications and that these types of data do not take into account that clinicians are treating a sicker patient population.
However, he said that in the future NCDR goals and registries will help:
- Promote quality improvement;
- Report performance measures with scientific validility; and
- Standardize data elements and uniform submission process across all public reporting programs.
Lastly, Hammill said that public reporting “should occur at the appropriate level of accountability and include a formal process for evaluating the impact of the program on the quality and cost of healthcare including the assessment of unintended consequences such as, will be start to avoid treating the most complex population of patients now that these outcomes data are public?"