To test the effect of a computerized support tool to enhance brief physician-delivered health lifestyle counseling to patients with increased metabolic risk factors during two usual care visits.

• 18 to 75 years old
• Waist circumference more than 35 inches for women or more than 40 inches for men or BMI higher than 25kg/m² and two or more features of metabolic syndrome:
  • TG higher than 150mg per dL
  • HDL less than 40mg per dL for men and less than 50mg per dL for women
  • Fasting glucose of 200mg per dL or higher
  • BP 135/85mm Hg or higher
• Use of medications to control any of the above.

• Substance use or abuse
• Severe arthritis or medical conditions limiting physical activity
• Recent myocardial infarction or stroke
• Peripheral vascular disease.

Recruitment

Clinic staff reviewed records to verify eligibility and contacted, screened and enrolled potential participants sequentially without consideration for their interest in losing weight. 

Design

RCT between two clinics.

Blinding Used

Implied with measurements.

Intervention

Computer expert system generated four to five page individualized, tailored reports providing feedback addressing participant-identified barriers to improving physical activity and diet. At baseline the subjects brought two to three dietary and physical activity goals they wanted to achieve based on the report. They were then given a 30-page planning guide that provided information on preventing diabetes and achieving dietary and physical activity goals. The computer also provided the patient's physician with a one-page summary of the patient's assessment and provided the physician with patient-specific counseling recommendations based on a motivational interviewing approach. 

Statistical Analysis

Analysis for non-similarity between groups was carried out using hypothesis testing. A one-tail T-test was performed and results were considered significant if there was a less than 5% chance that the variation between groups occurred due to random chance alone.   

Timing of Measurements

Baseline, six months (intervention only) and 12 months. 

Dependent Variables

• Height and weight using a digital scale
• Waist circumference at the superior iliac crest with tape measure parallel to the floor at the end of a relaxed expiration. All measurers were trained by one investigator.
• BP
• Fasting blood for glucose, insulin and lipids
• Physical activity by validated seven-day recall
• Energy intake by combining the Block and McPherson FFQs.

Independent Variables

• Computer expert system generated four- to five-page individualized, tailored report providing feedback addressing participant-identified barriers to improving physical activity and diet. At baseline the subjects brought two to three dietary and physical activity goals they wanted to achieve based on the report. They were then given a 30-page planning guide that provided information on preventing diabetes and achieving dietary and physical activity goals. The computer also provided the patient's physician with a one page summary of the patient's assessment and provided the physician with patient-specific counseling recommendations based on a motivational interviewing approach. The subjects had a brief study-related visit at six months to reassess their goals and review their goal sheet with their physician, who reinforced their lifestyle change goals. 
• The control group was given a packet of health education materials addressing diabetes, diet and exercise before a baseline usual care visit. They had no additional intervention about their specific goals for weight or physical activity other than what they might receive during usual care. 

• Initial N: 279 
• Attrition (final N): 263 (83 males, 180 females)
• Age: 50.0±11.79 years for controls and 49.2±13.04 years for intervention
• Ethnicity:
  • 50.8% white, 45.4% Hispanic and 5% other in controls
  • 50.4% white, 42.9% Hispanic and 9% other in intervention
• Anthropometrics: Similar in sex, economic status, disease risk status, age, weight, BMI, waist circumference, lipids and fasting glucose and insulin.
• Location: Pueblo, CO.

Key Findings

• Intervention patients lost 1.5kg (1.6% weight change) vs. a mean weight gain of 0.15kg (0.16% weight change for controls (P=0.002)] 
• 26.5% of intervention sustained a weight loss of 5% or more of total body weight at 12 months vs. 8.5% of controls [OR=3.86; (P=0.002)] and 7.5% sustained a weight loss of 10% or more at 12 months vs. 2,3% of controls [odds ratio=3.4; (P = 0.024)]
• Intervention had a significant decrease in waist circumference (-2.2cm) vs. a gain in controls (+1.5cm, P=0.01)
• Baseline BMI, sex, blood serum lipid levels, BP, blood glucose and age did not predict success
• Intervention increased self-reported physical activity by a mean of 613 metabolic equivalent unit-minutes per week vs. a 123 metabolic equivalent unit-minutes per week for controls (P=0.15)
• Intervention reduced self-reported energy (kcal) by a mean of 1,495kcal per week (14.9%) vs. a 843kcal per week (6.6%) reduction for controls (P=0.02)
• No significant differences between the two groups were found for changes in fasting insulin, fasting glucose, TC, LDL-C, HDL-C, TG or systolic and diastolic BP
• In subjects who lost 5% or more of body weight compared to those who did not, significant differences were found:
  • Change in TC (P=0.03)
  • Change in fasting glucose (P=0.01)
  • Change in TG (P=0.05)
• No significant differences were found between genders or racial/ethnic groups in any of the outcomes.   

A brief computer-based intervention designed to increase the dialogue between patients and clinicians about behavioral goals can lead to increased 12-month weight loss.