Impact of communicating familial risk of diabetes on illness perceptions and self-reported behavioral outcomes: a randomized controlled trial.

OBJECTIVE: To assess the potential effectiveness of communicating familial risk of diabetes on illness perceptions and self-reported behavioral outcomes. RESEARCH DESIGN AND METHODS: Individuals with a family history of diabetes were randomized to receive risk information based on familial and general risk factors (n = 59) or general risk factors alone (n = 59). Outcomes were assessed using questionnaires at baseline, 1 week, and 3 months.
RESULTS: Compared with individuals receiving general risk information, those receiving familial risk information perceived heredity to be a more important cause of diabetes (P < 0.01) at 1-week follow-up, perceived greater control over preventing diabetes (P < 0.05), and reported having eaten more healthily (P = 0.01) after 3 months. Behavioral intentions did not differ between the groups.
CONCLUSIONS: Communicating familial risk increased personal control and, thus, did not result in fatalism. Although the intervention did not influence intentions to change behavior, there was some evidence to suggest it increases healthy behavior.

Prevention of type 2 diabetes is especially important for people with a positive family history of diabetes because family history is one of the strongest risk factors (1). Individuals with a positive family history have difficulty understanding the causes of diabetes (2), underestimate their risk (3), and are less likely than those without a family history to believe that diabetes is preventable (4). Family history information might be used to raise awareness of individual risk and thereby positively influence preventive behaviors to reduce the risk (5). However, the belief that diabetes is determined mainly by genetic predisposition may prevent individuals from engaging in risk-reducing behavior as a result of fatalism (2,6,7). The aim of this study was to assess the potential effectiveness of communicating familial risk of diabetes on illness perceptions and self-reported behavioral outcomes.

RESEARCH DESIGN AND METHODS

In 2007, a randomized trial was conducted among individuals who were at risk for diabetes and had participated in a diabetes screening program 5 years earlier (8). People (n = 233; age ≤75 years) with self-reported family history (one or more first-degree relatives) and the highest diabetes risk scores on a symptom-risk questionnaire (8) were invited. Exclusion criteria were as follows: being diagnosed with diabetes and not understanding Dutch. The VU University Medical Center Ethical Committee approved the protocol.

Participants were randomly assigned by computerized and concealed block randomization to receive risk information based on familial risk and general risk factors (intervention group) or based on general risk factors alone (control group) during a personal consultation with a researcher (M.P.) at a Diabetes Research Centre. Five-year diabetes risk was estimated using a validated Diabetes Risk Test (9) and communicated to each participant using a graphical bar chart. In the intervention group alone, a family tree was constructed, familial risk was discussed, and the multifactorial character of diabetes was explained, indicating the nature of the risk in the bar chart. All participants received information on diabetes, including preventive measures.

Sample size calculation was performed on intention-to-change behavior (diet, physical activity, and diabetes testing). With a mean ± SD difference of 2.00 ± 1.6 in the intervention group compared with 1.00 in the control group for 80% power (P < 0.05), 41 individuals per group were needed. Outcome measures were assessed at baseline and at 1-week and 3-month follow-up and included behavioral intentions, self-reported behaviors, illness perceptions (causal beliefs, perceived consequences of diabetes, and personal control over preventing diabetes), perceived susceptibility to diabetes, worry about diabetes risk, and psychological well-being (Table 1). The effect of the intervention on outcome measures was investigated using ANCOVA for follow-up measurements with baseline measures as covariates.

Table 1

Outcomes of the ANCOVA analyses at baseline and at 1-week and 3-month follow-up*

	Intervention group			Control group			P
	Baseline	1-week follow-up	3-month follow-up	Baseline	1-week follow-up	3-month follow-up	Baseline and 1-week follow-up†	Baseline and 3- month follow-up†
n	54	53	46	53‡	48	47
Behavioral intentions (scale 1–7)§
Healthy diet	2.9 ± 2.0	4.3 ± 2.4	—	3.2 ± 1.8	4.3 ± 2.3	—	0.79	—
Physical activity	3.3 ± 2.2	4.1 ± 2.2	—	3.5 ± 1.9	4.2 ± 2.2	—	0.72	—
Test for diabetes	3.7 ± 2.3	4.1 ± 2.3	—	3.1 ± 2.4	4.4 ± 2.1	—	0.67	—
Health behavior (scale 1–7)‖
Healthy diet	3.6 ± 2.2	—	4.9 ± 1.7	3.9 ± 1.9	—	4.0 ± 2.2	—	0.01
Physical activity	3.9 ± 2.1	—	5.2 ± 1.8	3.6 ± 1.9	—	4.4 ± 2.2	—	0.08
Causal beliefs (scale 1–5)¶
Heredity	4.0 ± 0.8	4.4 ± 0.6	4.0 ± 0.9	3.7 ± 1.0	4.0 ± 0.7	3.8 ± 0.8	0.007	0.72
Lifestyle	3.9 ± 0.8	4.6 ± 0.6	4.1 ± 0.9	3.7 ± 0.9	4.3 ± 0.8	3.8 ± 0.9	0.12	0.30
Perceived consequences (scale 1–5)#	2.9 ± 0.7	2.8 ± 0.7	2.7 ± 0.7	2.7 ± 0.8	3.0 ± 0.6	3.0 ± 0.6	0.02	0.06
Personal control (scale 1–5)**	3.7 ± 0.8	4.2 ± 0.6	4.0 ± 0.6	3.8 ± 1.0	3.9 ± 0.8	3.7 ± 0.8	0.06	0.03
Perceived susceptibility (scale 1–7)††	3.3 ± 1.3	4.0 ± 1.2	3.2 ± 1.3	3.1 ± 1.3	3.6 ± 1.3	3.3 ± 1.4	0.16	0.86
Diabetes risk worry (scale 1–7)‡‡	2.7 ± 1.4	3.0 ± 1.5	2.4 ± 1.4	2.7 ± 1.7	3.2 ± 1.8	2.8 ± 1.6	0.65	0.21
Psychological well-being (scale 1–5)§§
PANAS positive	3.1 ± 0.7	3.2 ± 0.7	3.2 ± 0.7	3.0 ± 0.7	3.1 ± 0.6	3.1 ± 0.8	0.69	0.43
PANAS negative	1.7 ± 0.6	1.6 ± 0.6	1.6 ± 0.5	1.8 ± 0.6	1.7 ± 0.7	1.7 ± 0.6	0.70	0.73
STAI	1.9 ± 0.6	1.9 ± 0.6	1.9 ± 0.6	2.0 ± 0.6	1.9 ± 0.6	1.8 ± 0.5	0.93	0.29

Data are means ± SD.

*Unadjusted analyses are presented because the predefined variables did not affect the outcome of the trial.

†P is based on ANCOVA.

‡Baseline measures were unavailable for one person in the control group.

§Intention to eat more healthily (at least two pieces of fruit and 200 g of vegetables a day and low–saturated fat nutrition) or to increase physical activity (at least 30 min of moderate activity at least 5 days a week) within the following month was assessed (completely applicable to me [1] through completely inapplicable to me [7]).

‖Participants were asked to indicate whether they had changed their behavior in the previous 3 months (completely disagree [1] through completely agree [7]).

¶Participants were asked to indicate the extent to which they believed that a given cause could be a cause of diabetes (definitely not [1] through definitely [5]), based on the Revised Illness Perception Questionnaire (IPQ-R) ((13)). A heredity subscale was comprised of two items: “heredity, diabetes runs in the family” and “predisposition” (α = 0.62). A lifestyle subscale was comprised of three items: “unhealthy diet or eating habit,” “lack of physical activity,” and “being overweight” (α = 0.75).

#Perceived consequences were assessed using a 6-item scale (α = 0.86), based on the IPQ-R ((13)).

**Personal control over developing diabetes was assessed using the following single item: “There is a lot I can do to prevent getting diabetes” (completely disagree [1] through completely agree [5]), based on the IPQ-R ((13)).

††Perceived susceptibility was assessed using the mean score of three items (α = 0.88): “How likely do you think it is that you will get diabetes within the next 5 years?” (very likely [1] through very unlikely [7]); “Based on your feelings, how big is the chance of you getting diabetes within the next 5 years?” (very low [1] through very high [7]); and “In your opinion, what is the chance of you getting diabetes compared with an average man/woman your age?” (a lot lower [1] through a lot higher [7]).

‡‡Participants were asked to indicate their feelings when thinking about their chance of getting diabetes using a 7-point rating scale for two worry items (fear, worry) (α = 0.86).

§§The Positive (α = 0.88) and Negative (α = 0.84) Affect Schedule (PANAS) ((14)) was used to assess general mood states, and the Dutch short form of the State-Trait Anxiety Inventory (STAI) ((15)) (α = 0.87) was used to measure anxiety at the time of assessment.

RESULTS

Of 233 participants invited, 187 (80%) responded to the invitation and 118 (51%) agreed to participate and were randomly assigned (n = 59 in each group) (see supplementary Figure A1, available in an online appendix at http://care.diabetesjournals.org/cgi/content/full/dc08-1049/DC1). Ten individuals did not receive the consultation and were excluded. Participants were Dutch Caucasian. Mean ± SD age at baseline was 67.1 ± 5.3 years; 43% were men; 5% completed higher vocational training or university; mean ± SD BMI was 28.3 ± 4.3 kg/m2; and 52 and 31% reported having high blood pressure and high cholesterol, respectively. The median number of first-degree relatives was 1 (range 1–7). At baseline, there were no significant differences in participant characteristics between the groups.

For all variables used in our analyses, 10 and 18% of the data were missing at 1-week and 3-month follow-up, respectively. There were no differences at baseline in outcome variables between participants with missing data at follow-up and those for whom complete data were obtained.

The intervention had no effect on behavioral intentions (Table 1). People who had received the intervention reported having eaten more healthily than those in the control group in the previous 3 months (P = 0.01). Being more physically active showed a marginal significant difference (P = 0.08). There was a significant increase in perceiving heredity as a cause of diabetes in the intervention group (P < 0.01) compared with the control group at 1 week. Perceived consequences of diabetes increased in the control group and slightly decreased in the intervention group at 1 week (P = 0.02). The intervention group perceived greater personal control over preventing diabetes than the control group at the 3-month follow-up (P = 0.03), an effect that was of borderline significance after 1 week (P = 0.06). Communicating familial risk information did not affect perceived susceptibility, worry, or psychological well-being.

CONCLUSIONS

Our study shows that an intervention in which familial risk of diabetes is communicated did not result in fatalism and actually led to increased perceived control over preventing diabetes. Although at 1 week both groups had increased their intentions to change their health behavior, participants receiving familial risk information reported having eaten more healthily 3 months after the consultation. A possible explanation might be that familial risk information, being more novel and more personally relevant, was better retained. In line with a recent cross-sectional study (10), our study suggests that informing people of their risk of diabetes attributable to their family history could increase their engagement in risk-reducing behaviors. In addition, our results and others (11) show that discussing familial diabetes risk does not adversely affect psychological well-being.

Although an earlier theory-based behavioral intervention aimed at increasing physical activity of people at familial risk of diabetes was no more effective than information given in an advice leaflet (12), it is promising that some positive results of communicating familial risk in our minimal design were found. Both groups received a personal consultation differing only in the type of risk information (familial vs. general risk information) that was given. This study, though small, is one of the first to examine this issue. Because the measures of behavior and personal control were based on single items and the measures of behavior were self-reported, the effects of the intervention must be considered tentative. Additionally, participants were recruited from a previous diabetes screening study, thereby limiting generalization.

More robust trials are needed to confirm these findings, using objective measures of health-related behavior in larger samples. More research is also needed in the area of risk communication and fatalistic attitudes, particularly with the introduction of more genetic information available in addition to family history.