This study addresses the challenges associated with implementation of clinical practice guidelines (CPG's) and is motivated by our interest in gaining insight regarding the following general research questions about CPG implementation:

A. Can physician adherence to complex CPGs be promoted by use of a hand-held computerized decision support tool providing patient-specific recommendations, documentation, and drug dosing assistance? B. Will increased adherence to CPGs reduce variation in management by age, gender and race/ethnicity such that disparities in healthcare are reduced or eliminated? C. What are the cost implications of using PDA-based technology to promote CPG adherence?

This randomized, controlled, unblinded, practice-based trial will address these research questions by testing the following hypotheses in a 2 year behavioral intervention period:

1. The absolute proportion of patients that is treated appropriately with respect to lipid-lowering drug therapy within 4 months after testing will be increased by a net of at least 9% by the intervention as measured in baseline and follow-up independent cross-sectional samples of eligible patients (primary endpoint).
2. The absolute proportion of patients that is treated to the appropriate low density lipoprotein cholesterol (LDL-C) goal during follow-up of the baseline cohort of eligible patients is increased by a net of at least 12% by the intervention (secondary endpoint).
3. The proportions of eligible patients that are appropriately screened, risk-stratified, and counseled regarding therapeutic lifestyle changes are increased by the intervention (tertiary endpoints).
4. The intervention effect in subgroups defined by disease status (CVD, diabetes or neither), age, gender, and race/ethnicity reduces any disparities observed at baseline (exploratory analyses).
5. In addition, we will estimate the marginal cost effectiveness of the intervention for the primary endpoint.

The aims were modified in Year 1 to include an attention control group to enable evaluating and testing the impact of strategies to improve adherence to the recently released JNC 7 guideline by testing the following hypotheses:

1. The absolute proportion of patients that is treated appropriately with respect to blood pressure lowering drug therapy will be 10% greater in intervention practices than in comparison practices as measured in follow-up independent cross-sectional samples of eligible patients (primary endpoint).
2. The absolute proportion of patients that is treated to the appropriate blood pressure goal during follow-up will be 10% greater in the intervention practices (secondary endpoint).
3. The intervention effect in subgroups defined by disease status (CVD, diabetes or neither), age, gender, and race/ethnicity reduces any disparities observed at baseline (exploratory analyses).
4. In addition, we will estimate the marginal cost effectiveness of the intervention for the primary endpoint.

BACKGROUND:

Many investigators have identified "gaps" in the quality of care delivered in routine practice in the U.S. According to the Institute of Medicine (IOM) report "Crossing the Quality Chasm: A New Health System for the 21st Century," the "gap" between evidence and practice might be better termed a "chasm". High blood cholesterol was identified as one of 15 high priority conditions for Department of Health and Human Services (DHHS) initial focus. In addition, the report stresses the importance of increased use of information technology to support clinical and administrative processes. Six major aims are spelled out in the IOM report "health care should be safe, effective, patient-centered, timely, efficient and equitable." We have developed an intervention that is grounded in this philosophy. Specifically, we are focusing on a guideline that promotes the timely use of safe and effective interventions in a patient-tailored manner. A hand-held computerized decision support tool will be a core component of the intervention to support consistent and efficient medical decision-making. Although our goal is to improve overall adherence with the chosen guideline, we expect that the quality improvement process will result in increased equity, reflected in reduced disparities with respect to appropriate treatment between patient subgroups defined by gender, race/ethnicity, and age.

DESIGN NARRATIVE:

GLAD Heart is a randomized trial. The study intervention (academic detailing on ATPIII and a hand-held computerized decision support tool that provides patient-specific, ATPIII-based risk stratification and treatment recommendations for cholesterol management) and an attention control intervention (academic detailing on JNC7 and automated blood pressure devices) will be delivered at the level of randomization which is the practice. All practices (N=64) will receive usual care, consisting of guideline dissemination, baseline performance feedback, continuing education, and patient education materials. Outcomes will be assessed based on patient level data from chart reviews (N 20,000) of adult patients eligible for cholesterol measurement. To assess ATPIII-related outcomes, we will compare changes in baseline and follow-up chart abstraction between the two arms. To assess JNC7-related outcomes, we will compare differences in follow-up chart abstraction only between the two arms. Appropriate analytic methods will be used to analyze these correlated data. Impact evaluation for both arms will be conducted via baseline and follow-up surveys of participating providers. Process evaluation will be conducted by tracking use of the PDA-based tool and monitoring appropriate use of the automated blood pressure devices. Other tracking information will be collected to ensure intervention and data collection success.

Inclusion/exclusion criteria are based on practice characteristics. Practices are considered the study participants. However, because of the individual nature of the intervention tool, individual healthcare providers in the participating practices will be asked to participate voluntarily in the intervention and data collection. We anticipate approximately 320 healthcare providers to be included in these 64 practices. The research study will not directly intervene on patients of participating healthcare providers. However, patient charts will be assessed to evaluate changes in the practice's screening and management behaviors. Patient records will be included in this project in proportion to patient representation in participating primary care practices in NC and in proportion to their rates of cholesterol screening.

Potentially eligible patients for the chart abstraction sample will be adults, ages 21 through 84 years, seen in the participating primary care practices during the baseline (July 1, 2001 - June 30, 2003) or follow-up (March 1, 2004 - February 28, 2006) periods. We are examining two-year periods to increase the proportion of patients that has been screened. We can include "year 0" (July 1, 2001 - June 30, 2002), because ATPIII was published in May 2001. Although the guidelines do not exclude persons older than 84 years, at present we propose to restrict the evaluation sample to the aforementioned age group, due to acknowledged controversies regarding the appropriate preventive strategies in the very old.

Practice Inclusion Criteria:

Primary Care Practices (Internal Medicine or Family Medicine) Within 3-hour driving radius of WFUSM At least 50% of practice's providers willing to participate in study Willingness to accept randomization assignment Willingness to participate in chart audit Have a computer suitable for connectivity with the PDAs for installation of software upgrades and for printing of patient-specific reports for medical record documentation purposes

Practice Exclusion Criteria:

Academic practices

Provider Inclusion Criteria:

Practicing healthcare provider affiliated with a primary care practice participating in GLAD Heart.

Willingness to participate

Provider Exclusion Criteria:

None

Potentially eligible patients for the chart abstraction sample will be adults, ages 21 through 84 years, seen in the participating primary care practices during the baseline (July 1, 2001 - June 30, 2003) or follow-up (March 1, 2004 - February 28, 2006) periods. We are examining two-year periods to increase the proportion of patients that has been screened. We can include "year 0" (July 1, 2001 - June 30, 2002), because ATPIII was published in May 2001. Although the guidelines do not exclude persons older than 84 years, at present we propose to restrict the evaluation sample to the aforementioned age group, due to acknowledged controversies regarding the appropriate preventive strategies in the very old.