Do Clinical Practice Guidelines Improve Processes or Outcomes in Primary Care?

Military Medicine, Mar 2005 by Lesho, Emil P, Myers, Cris P, Ott, Monica, Winslow, Constance, Brown, Joan E

Background: Clinical practice guidelines (CPGs) are common, but it is not clear whether they improve care. Methods: Quality indicators for processes and outcomes of care were obtained from a computerized system-wide database by patient administration and utilization management personnel unaware of this study and without connection to or interests in guideline implementation. These indicators were compared before and after guideline implementation. Results: After the asthma CPG, nebulizer treatments, emergency department visits, and admissions decreased significantly (p

Introduction

Guideline-based medical practice pervades the delivery of health care.1-7 As of 1997, there were more than 2,500 clinical practice guidelines (CPGs) available, and the number continues to increase.8 Mandatory implementation of and compliance with an increasing number of CPGs are also becoming more common,1,6,9 although the effectiveness of many CPGs in improving care remains uncertain.1-3,10 A Cochrane systematic review of interventions to improve the care of diabetic patients recently concluded that the effect of guidelines for diabetes mellitus was unclear.4 Although experts acknowledge the uncertainty of the effectiveness of CPGs, the experts encourage their implementation.11,12 Studies that showed that CPGs improved the process or outcomes of care have been criticized as lacking strength or having minimal impact.10 Furthermore, guidelines may be less effective in changing or improving practice in the primary care setting.8,13

The purpose of this study was to determine what effect asthma, diabetes, and tobacco CPGs have on processes of care and outcomes in a primary care setting.

Methods

Design, Setting, and Patients

The investigation was a before-and-after study in a primary care department of a managed care organization with approximately 68,000 beneficiaries of all ages. For the asthma portion of the study, only patients 6 years of age or older were included. For the diabetes and tobacco portions, only patients 18 years of age or older were included in the study.

Guideline Intervention

Evidenced-based CPGs derived from the National Heart, Lung, and Blood Institute and the American Diabetes Association were implemented through a series of required lectures for all primary care providers and the distribution of "tool kits" to all points of care. The tool kits consisted of full-sized and pocket-sized laminated management algorithms and other reference materials for each provider, along with educational materials for patients. Before the final implementation, local tailoring was performed using the Delphi method.6 No decision point or recommendation in the guidelines that was based on good evidence was tailored. Guideline adherence was later encouraged at regular departmental performance-improvement meetings. Three separate guidelines were studied, to determine their impact on process and outcome measurements, i.e., asthma, diabetes, and tobacco cessation. All guidelines are available on the World Wide Web (http://www.qmo.amedd.army.mil).

Data Extraction

The Composite Health Care System is a computerized database that links scheduling (including all primary and consultative appointments), pharmacy, radiology, laboratory, and all other ancillary health care services throughout the entire military health care system. Physician orders for all outpatient services, including diagnostic tests and prescriptions, must be entered through this system. Patient visits for all outpatient care, including emergency department care, are recorded. Use of all ancillary health care services is also recorded. This results in the automatic generation of a computerized medical record that can be easily queried to determine such things as utilization, productivity, laboratory values, X-ray results, clinic appointments, and type and number of prescriptions. Patient and provider profiles can also be easily generated. For example, one can go to the database and query for all patients who were given prescriptions for insulin and who had protein in their urine. One can then see whether these patients were given prescriptions for an angiotensin-converting enzyme (ACE) inhibitor.

Personnel from the patient administration, information management, performance improvement, and utilization management departments, all of whom had no connection to the study or interest in guideline implementation, were asked to query the Composite Health Care System database to provide all of the data for this study, with two exceptions. Whether a patient was told to stop smoking was available only by review of the handwritten progress notes of the provider, and inpatient admissions were obtained from hospital admission records using the International Classification of Diseases, 9th Revision, codes.

Asthma Process and Outcome Measurements

Three process and three outcome measurements were used to assess the impact of the asthma CPG. These were the number of requests to respiratory service for nebulized albuterol treatments for asthma exacerbations, the number of emergency department visits, and the number of hospital admissions for asthma exacerbations 1 year before and 1 year after guideline implementation. For the subgroup of patients with persistent asthma, the percentage given appropriate long-term controller medications, the percentage receiving regular measurements of lung function tests (spirometry at each office visit), and the percentage receiving in-depth asthma education before and after guideline implementation were also compared.