- Click here for explanation of classification scheme.

The purpose of the study is:

• To determine the frequency and effectiveness of behavorial goal choices in the self-management (SM) of diabetes
• To test goal-setting theory hypothesis that self-selection and behavorial specificity of goals are key to enhancing persistence.

• Type 2 diabetics
• Participants 25 years old or older
• English-speaking.

• Participants less than 25 years old
• Type I diabetics.

Recruitment

• The participants were recruited from 52 physicians in 30 practice settings
• The participants meeting the inclusion criteria were sent an invitation letter, a brochure and a postcard to return if they did not want to participate
• Patients were called and invited to participate.

Design

Physicians were stratified by size of practice and location (urban or rural), then randomized by the statistician.

Blinding Used

The was no blinding.

Intervention

• Participants arrived 30 to 35 minutes before the primary care visit. The goal-setting session was done during the participant's medical care visits and delivered through a computerized touch-screen CD-ROM session.
• Each participant completed a dietary intake and a physical activity assessment, which compared the participant's answer to recomended guidelines 
• The participants developed a SM action plan, which consisted of a behavior change goal in the area of diet or exercise 
• The participants selected specific activities to support the goal and identified barriers and chose strategies to overcome those barriers
• A care manager met for eight to 10 minutes with the participant to review the goals and discuss additional strategies
• Participants received one follow-up call two weeks after the initial visit to review progress and provide feedback on goal progression.

Statistical Analysis

1. One-way ANOVA for normally-distributed continuous variables. Age, fat intake, fruit and vegetable consumption, were used to compare the three groups: Fat reduction, fruit and vegetable consumption and physical activity.
2. Tukey multiple comparison adjustment was used to determine specific between-group-differences
3. Kruskall-Wallis test was used to examine the difference of non-normally-distributed variables such as the number of comorbid illness among participants within the three groups
4. X² test was used for dichotomous variables such as gender and marital status. The physical activity measures were completed using X².
5. F tests was used to compare outcome changes from baseline for all three groups 
6. The direct difference before and after goal-setting represented the changes in fat intake and fruit and vegetable consumption
7. Odds ratios were used to describe the changes in physical activity outcomes.

Timing of Measurements

Baseline and six months.

Dependent Variables

• Variable One: Fat intake, measured by the Block Dietary Data Systems (BDDS) "Fat Screener"
• Variable Two: Fruit and vegetable intake was measured by the seven-item BDDS "Fruit/Vegetable Screener"
• Variable Three: Physical activity was measured by a modified version of the Behavorial Risk Factor Surveillance System, 11-item questions.

Independent Variables

• Age
• Gender
• Ethnicity
• Education
• Annual income
• Marital status
• Diagnosed comorbid conditions.

Control Variables

• Feedback
• Frequency
• Self-selection
• Behavorial goals.

• Initial N: 469 participants randomized to intervention were included in this analysis
• Attrition (final N): 422 participants (5% of participants selected smoking cessation as the goal and were therefore excluded and 5% had no six-month follow-up data and were excluded)
• Mean age: 62 years.

Ethnicity

• 83% white non-Hispanic
• 10.8% Hispanic
• 5% other.

Other Relevant Demographics

• Female: 51%
• Married: 69%
• Less than high school education: 13%
• High school graduate: 26%
• One to three years of college: 32%
• Gollege or grad school: 29%
• Income under $10,000: 12%
• Income from $10,000 to $29,999: 27%
• Income from $30,000 to $49,999: 28%
• Income over $50,000: 33%.

Anthropometrics

Not described.

Location

Not mentioned.

Approximately 49% of the participants selected the physical activity goal, while approximately 24% selected the increase fruit and vegetables goal and 27% selected reduce fat goal. 

Goal selection did not differ by demographic or medical characteristics.

Table One: Patient Characteristics at Baseline

Patient Characteristics			Reduce Goals at Baseline (N=112)	Increase Fruit Goals at Baseline (N=100)	Increase Physical Activity at Baseline (N=210)	P-Value
Diet	Fat consumption	Mean	22.27	19.18	20.49	0.028
		SD	7.78	8.98	8.48
	Fruit and vegetable consumption	Mean	13.50	10.65	12.40	<0.0001
		SD	4.94	3.59	4.61
Physical Activity	Days of vigorous activity per 7 days		0.42	0.34	0.29	<0.0001
	Days of moderate activity per 7 days		0.49	0.45	0.37	<0.0001

 

Significant Differences at Baseline by Goal Selection Between Groups

• Reduce fat and increase fruit and vegetables (P=0.008)
• Reduce fat and increase fruit and vegetables (P<0.001) plus increase fruit and vegetables and increase physical activity (P<0.001)
• Reduce fat and increase fruit and vegetables (P<0.001) plus reduce fat and increase physical activity (P<0.001) plus increase fruit and vegetables and increase physical activity (P=0.016)
• Reduce fat and increase physical activity (P<0.001) plus increase fruit and vegetables and increase physical activity   (P<0.001).

Table Two: Comparison Between Baseline and Six Months for Reduce Fat Goal (27% of Participants)

Variables	Baseline (N=112) Mean and SD	Six Months (N=92) Mean and SD	Statistical Significance of Group Difference
Fat Consumption	22.27 (7.78)	18.42 (7.99)	P<0.0001
Fruit and Vegetable Consumption	13.50 (4.94)	14.12 (5.07)	P=0.0967
Vigorous Activity	0.42	0.46	P=0.4594
Moderate Activity	0.49	0.55	P=0.0917

Table Three: Comparison Between Baseline and Six Months for Increase Fruits and Vegetable Goal (24% Participants)

Variables	Baseline (N=100)	Six Months (N=90)	Significance
Fat Consumption	19.18 (8.99)	16.84 (8.75)	P=0.0029
Fruit and Vegetable Consumption	10.65 (3.59)	12.70 (4.11)	P<0.0001
Vigorous Activity	0.34	0.36	P=0.288
Moderate Activity	0.45	0.48	P=0.4515

Table Four: Comparison Between Baseline and Six Months for Increase Physical Activity Goal (49% Participants)

Variables	Baseline (N=210)	Six Months (N=169)	Significance
Fat Consumption	20.49 (8.49)	19.09 (7.80)	P=0.0058
Fruit and Vegetable Consumption	12.40 (4.61)	12.86 (5.20)	P=0.0587
Vigorous Activity	0.29	0.37	P=0.0002
Moderate Activity	0.37	0.42	P=0.0156

Other Findings

• P-value for comparing changes of outcome among the three goal-selection groups
  • Fat consumption: P=0.00088
  • Fruit and vegetable consumption: P=0.075
  • Vigorous activity: P=0.1634
  • Moderate activity: P=0.6921.
• Goal selection did differ by reports of baseline behavior. Participants who selected increase physical activity, reduce fat consumption or increase fruit and vegetable consumption were significantly and respectively less active, consumed more dietary fat and ate fewer fruits and vegetables.
• Participants who selected a reduced-fat goal showed significantly larger decrease than did those that selected physical activity or fruit and vegetable goals
• Participants who selected a fruit and vegetable goal showed signficant changes in fruit and vegetable consumption
• Participants who selected a physical activity goal demonstrated significant changes in days of moderate and vigorous physical activity.

The study investigated the goal-setting process and the six-month behavorial changes. The results supported the hypothesis that the participants' active participation in selecting and setting goals would lead to the selection of a goal from the area that could use the most improvement and the goal that was most personally appropriate.

The second hypothesis, that participants would demonstrate the greatest amount of change relative to their self-selected goal behavior was only partially supported. Those who selected reduce fat intake as a goal reduced fat intake to a significantly greater degree than did those who selected physical activity or fruit and vegetable consumption goals. Although the statistically significant differences in physical activity and fruit and vegetable consumption by goal-selection were not found, these groups had the greatest change from baseline to six months. The lack of significance could be due to the fact that the participants identified an initial baseline goal but in actuality changed other behaviors without recording them.

This study demonstrated that goal-selection and specification, based on information related to multiple health behaviors, can be completed via computer assistance.

Government:

HHS

• The author clearly stated the purpose of the study which was;
  • To determine the frequency and the effectiveness of behavorial goal choices in the self-managment of diabetes
  • To test the goal-setting theory hypothesis that self-selection and behavorial specificity of goals can persist over a six-month period.
• Significant differences by goal-selection were reported at baseline. The author explained the effect of the goal-selection on changes in behavior at baseline and six months later. Significant differences as well as those that only approached significance were explained.
• The author identified the study limitations:
  • The methodology had limits, therefore, the study results should be interpreted with caution
  • The study objective of self-selection of behavorial goals were met, however the goals were assessed by self-report. The author identified the limitations of self-report in the discussion.
  • The author recognized the limited number of ethnic minorities in the study, but also acknowledged that the sample was representative of the region
  • The statistician and the staff were not blinded to the intervention.
• Dropouts were not adequately explained. The group that selected increase physical activity, had a 20% drop-out rate at six months.  
• The author did not present any data on control participants who did not set goals and were not assessed on behavior change outcomes. It would have been informative for the author to present the baseline information for the control group. Baseline demographics, diet and physical activity on the control group would have added information for comparison between the intervention group and the control group.
• No information related to length of time patients had diabetes was provided.