Dietmate - Computerized Diet and Exercise Intervention for High Blood Pressure

Computerized Diet and Exercise Intervention for High Blood Pressure

Summary of Phase I Results
Albert Jerome, Ph.D.
PICS, Inc., Reston, VA

This research was supported by SBIR grant R43 HL 53170 from the National Heart, Lung, and Blood Institute. Please do not quote without permission.


Twenty-one subjects between the ages of 21 and 60 (12 men and 9 women) participated in the study. Subjects were recruited from the greater DC metropolitan area through newspaper advertisements and announcements posted at local government agencies and other worksites. One-hundred seventy-seven people called in response to the ads; 48 met all of the preliminary screening criteria, which were as follows:

  • SBP >130 and < 179 or DBP >85 and < 109 (average from two assessments at least 7 days apart);

  • BMI > 26 and < 40;

  • not currently taking any medication that could affect blood pressure (no blood pressure medication within the last three months);

  • not pregnant or lactating;

  • not suffering from any health disorder that would preclude participation in the study;

  • not exercising more than two times per week; and

  • having written consent from a physician to participate.

Thirty-eight eligible volunteers were scheduled for an initial assessment. Six volunteers had blood pressure too low to meet the study requirements; 3 volunteers were started on blood pressure medication when visiting their doctor for permission to participate; and 8 volunteers either canceled, or did not show up for, their first appointment. Of the 21 subjects who began the study, 15 (71%) were Caucasian, 5 were Black (24%) and 1 was Asian (5%). Subjects ranged in age from 30 to 59 (M=44.0, SD=8.4). Mean pretreatment data were as follows: weight = 209.7 lb. (SD=29.3); SBP = 136.1 (SD=9.0); DBP = 91.4 (SD=5.8); resting pulse = 77.1 (SD=10.4); and minutes per week of exercise in prior 4 weeks = 35.5 (SD=40.9). No participants were smokers.


Demographics; history of BP, diet, weight, and exercise; and medical screening. During the pretreatment assessment, subjects completed a questionnaire that covered demographics; history of blood pressure levels and treatment; weight and dieting history; exercise habits; medication use; and medical conditions that might preclude participation.

Blood pressure. During each visit, blood pressure was assessed a minimum of 3 times using a standard sphygmomanometer and cuff. Subject position and criteria for cuff size and diastolic reading followed recommendations by the American Society for Hypertension (1992). If necessary, additional readings were taken until the last two readings showed differences of < 5 mm Hg for both SBP and DBP. Single SBP and DBP scores were computed for each assessment by dropping the first reading and averaging all subsequent readings. Pretreatment BP scores were computed by averaging scores from the two pretreatment assessments.

Other physical measures. Weight, height, and resting pulse were assessed during the pretreatment assessment and repeated (excepting height) at the posttreatment assessment. Compliance measures. During the posttreatment assessment, subjects were asked to rate on 1 - 7 scales how closely they followed various aspects of the program, e.g., dietary, exercise, record keeping, etc.

Product ratings. Ratings of overall satisfaction, satisfaction with specific program components, likelihood of using the product long-term, and likelihood of purchasing the product were rated on 1 - 10 Likert scales during the posttreatment assessment. A brief structured interview was also conducted to assess what was liked or disliked about the program (both computer and written materials) and ideas for program modifications. Data uploaded from computers. The CANE-H computers stored detailed daily and weekly summary information on intake of total calories, calories from fat, calories from saturated fat, cholesterol and sodium. These data were uploaded at the end of the study.


Telephone screening. Initial screening was done via telephone by the project research associate (RA). The general outline and procedures of the study were described, and subjects who appeared to meet the study criteria (outlined above) were scheduled for individual appointments.

Pretreatment assessments. Two pretreatment assessments were conducted at least one week apart. At the first pre-treatment assessment, subjects were provided with detailed information concerning the study, informed consent was obtained, and the measures described above were taken. Subjects were also provided with a description of the study and a permission slip to have signed by their doctor.

Second pretreatment assessment and beginning treatment. After subjects obtained their physician's permission, they were asked to return for a second visit to our office (approximately 7 to 10 days later), to confirm their blood pressure readings and to pick up their treatment materials. Blood pressure readings for the two pretreatment assessments were averaged, and subjects whose averaged readings fell in the acceptable range were allowed to participate in the study. Subjects were given brief instructions on how to use the computer and how to get started with the program. The same physical measures were repeated during this visit, accepting height. During this second pretreatment visit subjects were also assigned to one of two exercise programs: (1) walking only, or (2) unrestricted exercise, based on their risk profile for heart disease. This was determined by a High Risk Interview administered by the RA. Subjects who fell in the "high risk" category were asked to sign a waiver that they would limit their exercise to walking only unless completing a stress test by their physician. They were also given a Walking Kit developed by the Stanford Research Center (1994). Posttreatment assessment. Eight weeks after beginning treatment, subjects were contacted and scheduled for individual assessments. The RA assessed blood pressure, resting pulse, and weight; administered the compliance and satisfaction and ease of use measures; and conducted the structured interview. Data from the computers were uploaded to a PC database.


Treatment duration. Results are reported for 16 subjects who completed the final evaluation. Mean duration of treatment was 58.2 days (SD=7.4). Five subjects were unavailable for posttreatment assessment. One subject withdrew because of medical complications (unrelated to treatment) and the remaining four subjects are scheduled for their final visit. However, because of scheduling conflicts, we were unable to complete follow-up in time to include them in this summary. Thus there were no treatment "dropouts," other than one subject who withdrew for medical reasons.

Changes in blood pressure. Changes in systolic and diastolic blood pressure, weight, and resting pulse were examined by dependent measures t-tests, the results of which are summarized in table 1 below.

As can be seen by examining this table, significant changes occurred in both SBP and DBP during the eight week intervention. The magnitude and distribution of changes in blood pressure can be seen more clearly by examining table 2 below.

Changes in eating habits.Table 3 below provides a comparison of changes in the frequency of habits from pretreatment to end of treatment. Data are self report of the number of days per week that subjects ate breakfast, lunch, dinner, and snacks. As the data in Table 3 indicate, subjects increased the frequency with which they ate breakfast and lunch by one day per week and decreased their snacking frequency. These changes are consistent with the goals of the nutritional intervention: to promote eating three balanced meals per day and limiting snacks.

Nutritional data. Table 4 below provides nutritional information uploaded from the HHCs. Data are mean values for the entire treatment period. Nutritional targets are also provided to allow an assessment of degree of compliance with the intervention. As can be seen from examining this table, subjects exhibited a high degree of compliance with the dietary prescription provided by the CANE-H computers.

Exercise. Subjects were prompted to exercise at least 3 times per week during the treatment period. Data in Table 5 provide a comparison of changes in exercise activity from the pretreatment period to the last 2 weeks of treatment. Data were collected via self-report at pretreatment and end of treatment assessments. As can be seen by examining this table, subjects significantly increased exercise activity during the treatment period. Number of exercise sessions per week more than doubled, and total minutes expended exercising more than tripled. Likewise, subjects exhibited a nearly 60% increase in the length of each exercise session. Exercise frequency data uploaded from the computer were essentially identical to these data: mean exercise sessions recorded for the last week of treatment was 4.0 (SD=2.0) and for the previous week 3.5 (SD=1.8). The HHCs also stored data on total caloric expenditure from exercise. Means for the last two weeks of treatment were 1304 (SD=1050) and 1625 (SD=1345), respectively.

Program ratings. Results of program ratings (1 - 10 scales) were as follows:

  • overall satisfaction with the program (M = 8.1, SD = 2.2);

  • likelihood of using the program for 12 months or more (M = 7.1, SD = 2.7); and

  • likelihood of purchasing the product (M = 6.3, SD = 3.0)

(Note: this question was phrased as likelihood of purchase if they were not given one, because subjects were allowed to keep their computers at the end of the study).

Table 6 below shows distributions of these three measures.

Summary and conclusions

Results of this Phase I SBIR study demonstrate clearly the feasibility of using a computerized, self-management diet and exercise intervention for the reduction of high blood pressure. Each of the four primary dependent variables, SBP, DBP, weight, and resting pulse, exhibited significant decreases during the 8-week intervention. Subjects also demonstrated a high degree of compliance with the intervention and made positive changes in eating habits by increasing the frequency of eating breakfast and lunch and decreasing snacking. Positive changes in exercise activity were also demonstrated: exercises sessions per week, total minutes expended exercising, and average minutes per session all increased significantly from pretreatment to the last two weeks of treatment. Finally, subjects provided very positive ratings of their overall satisfaction with the program, probability of long-term use, and likelihood of purchasing the product.