The Impact of Diabetes Education and Peer Support upon Weight and
Glycemic Control of Elderly Persons with Noninsulin Dependent
Diabetes Mellitus (NIDDM)
WILLETTA WILSON, PHD, AND CLARA PRATT, PHD
Abstract: We assessed diabetes education and peer support
interventions as facilitators of weight loss and glycemic control in a
community sample of 79 elderly persons with noninsulin-dependent
diabetes mellitus (NIDDM). Different groups received: education
only, education and peer support, and no treatment. Peer support
was higher in groups where it was actively facilitated. Weight loss
and reduction in level of glycemic control occurred within groups
receiving both diabetes education and peer support. (Am J Public
Health 1987; 77:634-635.)
Noninsulin-dependent diabetes mellitus (NIDDM) is the
most common type of diabetes.' The majority of individuals
with NIDDM are obese.2 Dietary modification, the treatment
of choice for obese persons with NIDDM,3 is rarely effec-
tive.7 The purposes of the current study were:
determine whether supportive behavior could be elicited
from elderly peers in a diabetes education class, and 2) to
determine the incremental effects upon weight loss and upon
the reduction in blood glucose levels when social support is
added to diabetes education.
Registration for the project was through senior citizen
centers or nutrition sites located in four largely rural counties
of Oregon. All participants had diabetes mellitus, were not
treated with insulin, and had been advised by their health care
provider to lose weight. Of the 79 subjects, 80 per cent were
female and 20 per cent male. The mean age was 68.2 (SD =
7.2) and initial mean assessments ofbody weight and level of
blood glucose were 165.7 lbs (SD = 36.2) and 57.9 nmoles per
fructose equivalent (SD = 14.1).
Three groups were formed: education only (E), (n = 19);
education and peer support (E+PS), (n = 32); and control
(C), (n = 28). All interventions and assessments were
conducted at the senior centers and nutrition sites through
which participants had registered. Because the sites were
located over a largely rural, four-county area, it was not
feasible to randomly assign participants to conditions, so
sites were randomized. There were three sites assigned to
Ten 60-minute education classes were conducted by a
registered dietitian. Basic concepts about diabetes and its
nutritional aspects served as the foundation for all sessions.
E groups received an additional hour of unstructured class
time at each meeting. Peer support sessions, also 60 minutes
Address reprint requests to Willetta Wilson, PhD, Research Scientist,
Oregon Research Institute, 1899 Willamette Street, Eugene, OR 97401. Dr.
Pratt is with Oregon State University. This paper, submitted to the Journal
April 24, 1986, was revised and acceptedforpublicationOctober 16, 1986.
© 1987 American Journal of Public Health 0090-0036/87$1.50
long, immediately followed for the E+PS groups. The peer
support facilitator was trained in group dynamics, and sought
to foster peer interaction, focusing on group behaviors
critical to self-help groups.8 For each intervention, eight
sessions were held weekly. The ninth and tenth sessions were
held during week 12 and week 16, respectively.
Glycosylated hemoglobin (GHb) was measured with a
colormetric assay procedure.9 Weight was measured with a
portable scale, calibrated with a standard weight. Level of
peer support was measured with an adaptation ofthe Arizona
Social Support Schedule.'0 Other psychosocial variables
were measured with an adaptation of the Diabetes Educa-
tional Profile." Assessments of all variables was made
preintervention, postintervention-1 (week 8), and postinter-
vention-2 (week 16).
There were no between-group differences in peer sup-
port, weight, or level of GHb at preintervention. At
postintervention-l mean peer support levels in group E+PS
(67.9, SD = 23.9) were higher than mean peer suport levels
reported for group E (50.8, SD = 24.9), (Xd = 17.1, 95%
CI:3.1, 31.1). Moderate weight loss and reduction of GHb
level occurred in group E+PS, but not in other groups at
postintervention-l (Table 1).
Our data indicate that peer support can be enhanced
among elderly persons attending diabetes education classes
and it is related to desired changes in health behavior. Group
E+PS experienced substantially greater weight loss and an
initial reduction in GHb level than group E.
The size ofour samples was small and the duration ofthe
interventions was less than optimal to show changes in level
of GHb. While GHb changes can be detected at eight-week
intervals,'2 a three-month interval may be better.'3"14 A one
year follow-up would have indicated whether the interven-
tions had a more lasting impact.
Nevertheless, community-dwelling persons with
NIDDM were studied, in contrast to the more typical and less
generalizable studies of young insulin-dependent diabetic
patients evaluated while they are hospitalized or at a diabetes
summer camp.15"16Moreover, group E+PS facilitated reduc-
tions in weight and GHb, at least on a short-term basis. This
finding warrants additional investigation ofpeer supportas an
adjunct to diabetes education and other types of patient
education. Including a group facilitator on the diabetes
education team might increase costs, but the increased
efficacy of diabetes education should compensate for the
expenditure. With diabetes' annual total economic impactof
at least $14 billion,'7 we should be able to afford investigating
procedures that might improve the prognosis for the 5.8
million individuals with diabetes.'8
A6JPH May 1987, Vol. 77, No. 5
PUBLIC HEALTH BRIEFS
TABLE 1-Difference Scores and Confidence Intervals for Preintervention, Postintervention-1, and
Postinterventlon-2 Scores within Groups [Education only (E), Education and Peer Support (E+PS),
and Control (C)]
Group E (N = 19)
-1.2 lbs. (-2.6, +0.2)
-0.5 lbs. (-1.8, +0.5)
Group E+PS (N = 32)
-3.1 nmoles (-7.7, +1.5)
+1.1 nmoles (-1.1, +3.2)
-5.5 lbs. (-6.8, -4.2)
+0.5 lbs. (-0.4, +1.4)
Group C (N = 28)
-3.7 nmoles (-5.7, -0.5)
+2.0 nmoles (+0.9, +3.0)
+0.6 lbs. (-1.1, +1.8)
-0.7 nmoles (-2.5, +11.1)
NOTE: Postintervention-2 data are not reported for group C.
anmoles per fructose equivalent.
This work was support by a grant from the Andrus Foundation.
1. Harris MI: Classification and diagnostic criteria for diabetes and other
categories of glucose intolerance. In: National Diabetes Data Group:
Diabetes in America. NIH Pub. No. 85-1468. Washington, DC: Govt
Printing Office, 1985.
2. National Diabetes Data Group: Classification and diagnosis ofdiabetes and
other categories of glucose intolerance. Diabetes 1979; 28:1039-1057.
3. Third Annual Report of the National Diabetes Advisory Board: NIH Pub.
No. 80-2072. Washington, DC: Govt Printing Office, 1980.
4. Cohen F, Lazarus R: Coping with the stress of illness. In: Stone G, Cohen
F, Adler N (eds): Health Psychology. San Francisco: Jossey Bass, 1979.
5. Speers M, Turk D: Diabetes of self-care, knowledge, beliefs, motivation
and action. Patient Educ Couns 1982; 3:144-149.
6. Watts F: Behavioral aspects of the management of diabetes mellitus:
education, self-care, and metabolic control. Behav Res Ther 1980;
7. Wylie-Rosett J: Development of new educational strategies for persons
with diabetes. J Am Dietet Assoc 1982; 81:268-271.
8. Levy L: Processes and activities in groups. In: Lieberman M, Borman L
(eds): Self-help Groups for Coping with Crisis. San Francisco: Jossey
9. Parker K: Improved colormetric assay for glycosylated hemoglobin. Clin
Chem 1981; 27:669-672.
10. Barrera M: Social support in the adjustment of pregnant adolescents:
assessment issues. In: Gottlieb B (ed): Social Networks and Social
Support. Beverly Hills: Sage Publications, 1981.
11. Diabetes Education Profile Project: Diabetes Educational Profile: Users'
Manual. Ann Arbor: University of Michigan, 1980.
12. Huisman W, Kuijken JPAA, Tan-Tjiong HL, Duurkoop EP, Leijnse B:
Unstable glycosylated hemoglobin in patients with diabetes mellitus.
Clinica Chimica Acta 1982; 118:303-309.
13. Pecoraro RE, Graf RG, Halter JB, Beiter H, Porte D Jr: Comparison of a
colormetric assay for glycosylated hemoglobin with ion-exchange chro-
matography. Diabetes 1979; 28:1120-1125.
14. Aleyassine H, Gardiner RJ, Tonks DB, Koch P: Glycosylated hemoglobin
in diabetes mellitus: correlations with fasting plasma glucose, serum lipids,
and glycosuria. Diabetes Care 1980; 3:508-514.
15. Hauser ST, Pollets D: Psychological aspects ofdiabetes mellitus: a critical
review. Diabetes Care 1979; 2:227-232.
16. Johnson SB: Psychosocial factors in juvenile diabetes: a review. J Behav
Med 1980; 3:96-116.
17. Entmacher PS, Sinnock P, Bostic E, Harris MI: Economic impact of
diabetes. In: National Diabetes Data Group: Diabetes in America. NIH
Pub. No. 85-1468. Washington, DC: Govt Printing Office, 1985.
18. Harris MI: Prevalence of non-insulin-dependent diabetes and impaired
glucose tolerance. In: National Diabetes Data Group: Diabetes in Amer-
ica. NIH Pub. No. 85-1468. Washington, DC: Govt Printing Office, 1985.
AJPH May 1987, Vol. 77, No. 5