Effects of computerized guidelines for managing heart disease in primary care

From Clinfowiki
Jump to: navigation, search

Effects of computerized guidelines for managing heart disease in primary care

Patient safety is appropriately a major concern in the provision of health care today. Many errors leading to patient morbidity and mortality can be linked to the failure to follow evidence-based guidelines. These types of errors (i.e. not providing the recommended care) are referred to as errors of omission. It is believed that errors of omission may be more responsible for patient illness and injury than errors of commission (i.e., doing the wrong thing.)

Incorporating an evidence-based reminder system into a computerized medical records is touted as one mechanism for reducing the number of errors of omission. This theory is based on the idea that there is too much information for a clinician to keep track of and simple alerts would serve as reminders.

This paper describes a randomized, controlled study designed to look at the effects of computer- based cardiac care suggestions in an ambulatory primary care setting experienced in the use of electronic health records. 706 patients with confirmed diagnosis of Congestive Heart Failure and /or Ischemic Heart Disease were enrolled and followed over a period of one year. During this time 2,609 cardiac care suggestions were made to the clinicians. Outcome variables included adherence to the care suggestions by clinicians, patient perceived health related quality of life and exacerbations of heart disease. The analysis of these outcomes showed no difference in clinician adherence to guidelines (23 vs. 22 %.) There was also no difference in the patient perceived quality of life, medication compliance, health care utilization or costs or satisfaction with care.

Comments

The study did not find the computerized cardiac care suggestions effective. Similar studies by the same group and in the same primary care setting had found improved compliance with computerized outpatient and inpatient preventative care and drug monitoring guidelines. Suggested reasons for the failure of this intervention may have been the interpretation by providers that the alerts were intrusive or too time consuming. There was no requirement to enter a reason for ignoring the alerts and they could easily be closed with an escape key. The primary care providers using the guidelines were not involved in their development- they were defined by cardiologists and may have been impractical in the busy, inner-city primary care setting. Providers were not rewarded in any way for following the alerts.

This study does not mean that computerized alerts are not valuable, but it does remind us how important the design, customization, implementation and provider involvement is in using these systems. They will not be automatically followed.

References