This cluster randomized controlled study demonstrates that provider education and access to a CKD registry improved adherence to KDOQI guidelines for the primary process measure (PTH adherence) more than education alone. However, there was no consistent improvement in adherence to guidelines for secondary processes or intermediate clinical outcomes. Education was provided to clinicians in both the control and intervention clinic and improvement in guideline adherence was observed in both groups. A slight improvement in guideline adherence was also seen in a contemporaneous community cohort that was not involved in any study procedures. However, while there may have been a secular trend towards improved KDOQI guideline adherence, it was of lower magnitude than that observed in the study clinics.
Low adherence to KDOQI guidelines may be due to lack of awareness and familiarity with the guidelines . Given that these factors may be modifiable, researchers evaluated the impact of academic detailing and feedback to providers on guideline adherence . After one year, significant improvement was seen across multiple areas including diagnosis of CKD (from 38% to 70%) and evaluation of PTH, Vitamin D and phosphate (from 5% to 44%). These results are consistent with our finding of improved adherence in the control and intervention clinics, both of which received provider education. However, in the previous study, improvement plateaued, and even regressed slightly, two years later . The challenge will be to develop sustainable interventions that propel clinical care to expected levels of quality, not just improvement above a poor baseline performance. Provider education is a necessary component of any program to improve the care of patients with CKD but is likely not sufficient on its own.
In our study, providing access to an electronic CKD registry improved adherence to guideline recommended care for the primary process measure but not for secondary process measures or clinical outcomes. Previous qualitative research has identified information technology and disease registries as being critical factors in the success of quality improvement efforts . However, successful quality improvement programs are typically comprehensive and include project management, education, funding, partnerships, and health care delivery system reform . A recent study demonstrated no consistent improvement in quality of CKD care with a clinical decision support system embedded within the electronic medical record . Our quantitative study also suggests that advances in information systems, by themselves, are not sufficient for comprehensive quality improvement. Many providers and health care organizations are currently purchasing electronic health records and working to achieve “meaningful use”. Our results are important because they demonstrate that the ability to “generate lists of patients by specific conditions”, one of many stage 1 meaningful use criteria,  does not necessarily translate to actual use by providers or improvement in quality of care. Perhaps the meaningful use criterion needs to be modified to include not only the generation of lists but their actual use by providers.
The lack of improvement in guideline adherence in our study may be due to a lack of system redesign. Providers were not given protected time to access the CKD registry nor was there reorganization of the clinic to provide ancillary support or a multidisciplinary team that could utilize the registry. Physicians and other primary providers receive only minimal training in public health and little if any training in population or panel management. Medical education focuses primarily on the pathology of disease and the diagnosis and management of individual patients. Therefore, translating new information technologies, such as disease registries, into improvements in processes of care and clinical outcomes will require redesigning medical education to include population management. Additionally, delivery system redesign is needed to provide ancillary resources and modify the clinic work flow to capitalize on the wealth of information now available from electronic health records. It is likely that both provider education and system redesign will be required to fully realize the benefits of new information technologies.
This study has limitations that need to be considered. The intervention was limited to education and a disease registry and did not include other components of the chronic care model such as delivery system redesign, direct patient education, and self-management support . While this limited design may have contributed to a lack of improvement, it did allow for an independent evaluation of the impact of access to an electronic disease registry on processes and outcomes of care. Many of the clinical variables were not measured at the same time for all subjects and were instead measured using data collected over the entire 12 month study period. Because there were only two clinics, the effect of clinic could not be properly accounted for. While there were only 2 clinics, they did include various models of care delivery including primary care physicians, nurse practitioners, and supervised medicine residents. The duration of the study is limited and does not allow for an evaluation of long term effects of a chronic disease registry. The outcomes do not include any hard endpoints such as mortality or end-stage renal disease. Medicine residents may have been overwhelmed by the complex medical care required for patients with CKD and therefore unable to take advantage of the CKD registry. Finally, the study was quantitative in nature and does not include any qualitative assessment of providers’ interactions with the registry.
The main strength of this study is the use of a control clinic. Many studies of quality improvement efforts are limited by their use of historical controls. Use of a control clinic allows for a direct evaluation of the impact of access to a CKD registry on guideline adherence. In addition, we were able to monitor use of the CKD registry and demonstrate that most providers did not utilize the new technology. Finally, this study included a large number of patients and a diverse group of providers including internal medicine residents, physicians, and nurse practitioners.