Content uploaded by Laura J Holt
Author content
All content in this area was uploaded by Laura J Holt on Nov 19, 2014
Content may be subject to copyright.
409
T
he Diabetes Control and Complications Trial (DCCT) established intensive
insulin therapy as the standard of care for type 1 diabetes (TIDM). The adults and
adolescents enrolled in that trial benefited equally, with a 50% to 70% reduction in
the occurrence and progression of microvascular complications when receiving intensive
therapy compared with conventional therapy.
1,2
In addition, the benefits of intensive
insulin therapy were sustained 7 years subsequent to the end of the DCCT, as
demonstrated in the Epidemiology of Diabetes Interventions and Complications trial
(EDIC).
3-5
Although the adolescents, like the adults, receiving intensive therapy had a
similar risk reduction in complications, the adolescents did not achieve glycemic control
equivalent to the adults.
2
Adolescents in conventionally and intensively treated groups had
hemoglobin A1c levels approximately 1% higher than their adult counterparts. The elusive
nature of optimal glycemic control in youth with diabetes underscores the need for the
design, implementation, and evaluation of efficacious and translatable interventions
focused on optimizing glycemic control.
6,7
IMPACT OF AMBULATORY, FAMILY-FOCUSED TEAMWORK INTERVENTION ON
GLYCEMIC CONTROL IN YOUTH WITH TYPE 1 DIABETES
LORI
M. B. LAFFEL,MD, MPH, LAURA VANGSNESS,BA,ALEXA CONNELL,BA,ANN GOEBEL-FABBRI,PHD, DEBORAH BUTLER,MSW,AND
BARBARA
J. ANDERSON,PHD
Objective To evaluate an ambulatory, family-focused intervention aimed at optimizing glycemic control, minimizing diabetes-
related family conflict (DFC), and maintaining quality of life in youth with type 1 diabetes (T1DM).
Study design We randomly assigned 105 children and adolescents, 8 to 17 years of age, with T1DM for ≤6 years, to a family-fo-
cused teamwork (TW) intervention or to standard multidisciplinary diabetes care (SC). Patients in both study groups were seen
at 3- to 4-month intervals and were followed prospectively for 1 year. Measures of family involvement in diabetes tasks, DFC, and
quality of life were performed at baseline and after 1 year. Hemoglobin A1c was measured at each visit.
Results Patients (n = 100) completed follow-up, (50 in TW and 50 in SC). At entry, A1c was 8.4% ± 1.3% in TW and 8.3% ± 1.0%
in SC. After 1 year, A1c was 8.2% ± 1.1% in TW compared with 8.7% ± 1.5% in SC (P < .05). Both groups had similar frequencies
of blood glucose monitoring (BGM) and insulin dosing. Families exposed to the TW intervention maintained or increased family
involvement significantly more than families exposed to SC (P = .05). In multivariate analysis, the TW intervention and the daily
frequency of BGM significantly predicted A1c (R
2
= 0.17, P = .05). Despite increased family involvement, the TW group reported
no increase in DFC or decrease in quality of life.
Conclusions The ambulatory TW intervention prevented the expected deterioration in glycemic control seen with SC in youths
with T1DM of ≤6 years’ duration. Successful family involvement may assist in the preservation of health and the prevention of
long-term diabetes complications for youth with diabetes. (J Pediatr 2003;142:409-16)
From the Pediatric and Adolescent Unit,
the Behavioral Research and Mental
Health Section, the Genetics and Epi-
demiology Section, Joslin Diabetes Cen-
ter, Harvard Medical School, Boston,
Massachusetts.
Supported by a grant (DK-46887) from
the National Institute of Diabetes, Di-
gestive and Kidney Diseases, the Charles
H. Hood Foundation, and the Katherine
Adler Astrove Youth Education Fund.
Intervention modules were created by
the research team and are available by
request from the corresponding author.
Presented in part at the 62nd Annual
Meeting and Scientific Sessions of the
American Diabetes Association, San
Francisco, California, June 2002.
Submitted for publication July 2, 2002;
revision received Nov 4, 2002; accepted
Jan 3, 2003.
Reprint requests: Lori Laffel, MD, MPH,
Pediatric and Adolescent Unit, Joslin Di-
abetes Center, One Joslin Pl, Boston, MA
02215. E-mail: Lori.Laffel@joslin.har-
vard.edu.
Copyright © 2003 Mosby, Inc. All rights
reserved.
0022-3476/2003/$30.00 + 0
10.1067/mpd.2003.138
BGM Blood glucose monitoring
DCCT Diabetes Complications and Control Trial
DFC Diabetes-specific family conflict
EDIC Epidemiology of Diabetes Interventions and Complications Trial
T1DM Type 1 diabetes mellitus
Previous studies have identified the importance of de-
velopmentally appropriate family involvement around dia-
betes management for optimizing control and preventing
acute and chronic complications.
8,9
Interventions have fo-
cused on increasing family involvement in diabetes tasks to
encourage adherence as well as successful adaptation of the
child.
10-14
Furthermore, parent involvement with, and patient
adherence to, diabetes management tasks are greatest shortly
after diagnosis.
15
However, within 1 to 2 years after diagnosis,
patient and parent behaviors consistent with optimal manage-
ment often deteriorate,
15,16
as does glycemic control.
17,18
In the current study, we sought to design and evaluate a
family-focused intervention integrated into routine pediatric
diabetes office visits aimed at optimizing glycemic control
through negotiated family involvement in diabetes manage-
ment. Our intervention was targeted at maintaining family
teamwork around diabetes management with the goals of
minimizing diabetes-related family conflict, maintaining
quality of life, and increasing adherence to optimize glycemic
control in recently diagnosed patients with type 1 diabetes.
METHODS
Participants
This randomized, prospective study examined the im-
pact of the family-focused teamwork intervention on glycemic
control in 105 children/adolescents with T1DM who were pa-
tients in the Pediatric and Adolescent Unit at the Joslin Dia-
betes Center. Patient eligibility criteria included age 8 to 17
years; duration of T1DM longer than 2 months but less than
6 years; no concurring serious psychiatric or medical illness;
residence in New England or New York; at least one outpa-
tient medical visit at Joslin in the past year; and intention for
routine follow-up diabetes care at our center. Of the eligible
patients and families, 128 were approached and 105 (81%)
agreed to participate. Most patients (80%) who declined par-
ticipation cited that they had time constraints. Patients from
families declining participation were slightly older than par-
ticipating patients (13.6 ± 2.8 years versus 12.1 ± 2.3 years, re-
spectively) (P < .02), although there were no other significant
differences between groups.The Committee on Human Stud-
ies at the Joslin Diabetes Center approved the study protocol.
Procedures and Measures
Eligible patients were sent an introductory letter and, at
their next regularly scheduled diabetes care visit, each child
along with a parent met with a trained research assistant. After
we obtained written informed consent from the parent and as-
sent from the child, we randomly assigned the patient to one
of two study conditions: the teamwork intervention (TW)
group or the standard care (SC) group.
Patients were randomly assigned according to age (8-12
years; 13 to 16 years) and duration (0-2 years; 3-6 years).
Three families (2.8%) (two in SC and one in TW) discontin-
ued study participation because of insurance-related transfer
of care and thus were excluded from our analyses. After en-
rollment, two additional patients were excluded because of the
development of symptoms of a severe eating disorder in one
patient and the diagnosis of a learning disability in the other.
The final study sample included 100 families: 50 in the SC
group and 50 in the TW intervention group.
For each patient, a research assistant conducted a struc-
tured joint (patient and parent) interview at each quarterly di-
abetes visit to gather demographic information and to assess
the division of responsibility for diabetes management tasks in
the family during the preceding month. The complete inter-
view lasted about 15 minutes. Each child and a parent also
completed a set of questionnaires at baseline and 1 year later,
including the Peds QL,
19
the Diabetes Family Conflict
Scale,
20
and the Diabetes Family Responsibility Question-
naire.
8
An interval history, physical examination, and diabetes
adherence rating scale were also completed by the clinician at
each 3-month visit.
Teamwork Intervention Condition
The intervention for the families in the teamwork con-
dition focused on the importance of parent-child responsibili-
ty-sharing for diabetes tasks and ways to avoid conflict that
undermines such teamwork. At each visit, the research assis-
tant provided brief written materials on the module topic and
encouraged active family discussion as reinforcement. At the
conclusion of each session, the child, parent(s)/guardian(s),
and research assistant negotiated a responsibility-sharing plan.
This plan outlined who would be responsible for the different
tasks involved in insulin injections and the various compo-
nents of blood glucose monitoring. The research assistant em-
phasized the need for the family to work as a team, with the
parents offering hands-on and emotional support to the
child/adolescent. The four modules addressed the following
content areas: (1) communication around diabetes, especially
talking about blood glucose results within the family; (2)
meaning of Alc and explaining the need for parent-child
teamwork during the adolescent period; (3) response to blood
sugars and avoiding the “blame and shame cycle”; (4) sharing
the burden of diabetes tasks with family members and using a
logbook to problem-solve out-of-range values. These written
materials emphasized the multiple causes of high and low
blood glucose levels during childhood and adolescence; the
need for realistic expectations for blood glucose levels and be-
haviors; and the importance of maintaining parent involve-
ment with insulin injections and blood glucose monitoring.
[Note: Please see notation on title page for information on
how to obtain copies of materials from the authors.]
Standard Care Condition
Families randomly assigned to the SC condition partic-
ipated in a brief, joint parent-and-child interview quarterly
and completed questionnaires at the beginning and end of the
1-year study period. Adherence assessments and clinical data
were also gathered by the health care team for this group. In
contrast to the TW intervention, the research assistant did not
engage patients and families in the SC group in discussion
about family teamwork. Families in the SC condition received
the same educational materials as the TW group; however, the
410 Laffel et al The Journal of Pediatrics • April 2003
modules were given to the family all at once, as a package, only
after all patients and families completed the final set of ques-
tionnaires at the conclusion of the study. Both study groups
received equal attention with respect to the scheduling of fol-
low-up visits, contact between visits, and encouragement
around routine diabetes management. At the 1-year follow-
up, the teamwork intervention group had an average of 4.5 ±
0.6 medical visits per patient, whereas the standard care group
had an average of 4.6 ± 0.9 medical visits (not significant).
Measures of Parental Involvement in T1DM
Management Tasks
We assessed parent involvement in diabetes care by
using two separate measures. In the first, current insulin and
blood glucose monitoring (BGM) routines were ascertained
from the structured, joint (parent and child) interview devel-
oped by the authors
13
to assess in detail the division of respon-
sibility within families during a typical day for two tasks:
insulin injections and BGM. The quantitative system for cod-
ing this interview has been previously reported.
13
Because the
system involved coding components of behavior into higher-
order combinations of involvement, reliability of this coding
system was checked independently throughout the study by
two trained research assistants. Interrater reliability for both
the insulin routine and BGM scores ranged from 85% to 98%.
The second measure to assess parent involvement in di-
abetes care was the Diabetes Family Responsibility Question-
naire.
8
This 17-item self-report scale was completed by both
the patient and one parent. Scores could range from 17 (indi-
cating child has complete responsibility) to 51 (parent(s) has
complete responsibility).
Measures of Adherence to Diabetes Management
Tasks
We assessed adherence to various diabetes tasks with a rat-
ing scale (modified from Jacobson et al
21
) that was completed
independently by the child’s clinician during the medical visit to
reflect adherence behaviors over the previous 3 to 4 months.
Measure of Child Quality of Life
We assessed the patient’s health-related quality of life
by using the PedsQL, which measures the child and parent’s
perceptions of the child’s comprehensive quality of life.
19
The
PedsQL uses a 5-point Likert scale and a total quality of life
score was obtained by averaging the scores from all items on
the instrument. Scores could range from 0 to 100 (highest
quality of life).
Measure of Diabetes-Related Family Conflict
Each child and a parent completed the Diabetes Family
Conflict Scale to assess the degree of family conflict in 19
management tasks.
20,22
The level of diabetes-specific conflict
in the family was rated on a 3-point scale: 1 (never argue), 2
(argue a fair amount), and 3 (always argue). We chose to sum
the number of items in which any level of conflict was ac-
knowledged (2 or 3). Scores could range from 0 to 19, with 19
indicating conflict on all items.
Physical Data and Glycemic Control
The medical record of each patient provided data on
growth, pubertal development (by Tanner staging), insulin
therapy, and diabetes management. Blood sampling from pa-
tients at each visit provided the measure of glycemic control
(A1c) (reference range, 4.0% to 6.0%) by high-performance
liquid chromatography (Tosoh 2.2, Tosoh Corp, Foster City,
Calif ) according to the DCCT standards.
Statistical Analysis
Statistical analyses were performed with SAS version 8
for Windows (SAS Institute, Cary, NC). Means ± SD values
are presented unless otherwise indicated. Analyses included
paired and unpaired t tests, analysis of variance, Pearson bi-
variate correlations,χ
2
analysis, and multivariate analysis. In
general, univariate analyses were examined first. The complex
relations among the developmental, behavioral, and biological
variables called for multivariate analyses to control for poten-
tial confounders.
RESULTS
Patient Characteristics
Taken together, the patients had a mean age of 12.1 ±
2.3 years at baseline, a mean duration of diabetes of 2.7 ± 1.6
years, and an average A1c of 8.36% ± 1.7%. The two groups
(Table I) were comparable at entry with respect to age, dura-
tion of diabetes, sex distribution, and parent socioeconomic
status. At baseline, youths in the SC group weighed slightly
more and, as a result, had a slightly greater body mass index
(BMI) than youths in the TW intervention group. However,
after 1 year, both groups had similar measures of height,
weight, and BMI, supporting greater growth and develop-
ment among TW participants than among SC participants.
The TW group had a higher percentage of patients from sin-
gle-parent families (separated/divorced/widowed) at baseline
as compared with the SC group (22% vs 8%).
Diabetes-Related Characteristics at Baseline and One-
Year Follow-Up
In both groups at baseline, 46% to 48% of patients re-
ceived 3 or more injections of insulin per day, and 66% were
checking their blood glucose levels at least 4 times per day
(Table II). After 1 year, in both the TW (68%) and SC (60%)
groups, more patients were receiving 3 or more injections of
insulin daily or were receiving pump therapy. Despite the in-
creased intensity of insulin therapy, there was a trend toward
decreased blood glucose monitoring over the year, with 58%
and 60% of SC and TW patients, respectively, checking their
blood glucose levels 4 or more times daily. There was no sig-
nificant difference in the daily dosing of insulin (U/kg per
day), although the total daily dose of insulin increased, as ex-
pected, in both groups, as the result of growth and pubertal
development of all study patients during the 1 year of follow-
up. Finally, at baseline, both groups had similar levels of
glycemic control, with A1c equal to 8.4% ± 1.3% and 8.3% ±
Impact of Ambulatory, Family-Focused Teamwork Intervention on Glycemic
Control in Youth With Type 1 Diabetes
411
1.0% in the TW and SC groups, respectively. After 1 year of
follow-up, the TW intervention group showed no deteriora-
tion in glycemic control, A1c of 8.2% ± 1.1%, as compared
with the SC group, A1c of 8.7% ± 1.5% (P < .05) (Table II).
This significant difference in A1c occurred despite greater
growth and development among participants in the TW
group.
Family Behavior Characteristics at Baseline and One-
Year Follow-Up
At baseline, patients and parents in the TW interven-
tion and SC groups reported similar levels of family involve-
ment, based on the Diabetes Family Responsibility
questionnaire.
8
Both groups also had similar child and parent
reports of diabetes-specific family conflict as measured by the
Diabetes Family Conflict scale.
20
Finally, general health-re-
lated quality of life measured by the PedsQL
19
portrayed sim-
ilar levels of quality of life in the two study groups from both
parent proxy and child reports. These measures remained
comparable between the groups after 1 year. In addition, there
were no significant within-group changes in any of these pa-
rameters from baseline to 1-year follow-up (Table III).
In order to better characterize parent involvement in di-
abetes management both at baseline and during the one-year
follow-up, we examined parent involvement in blood glucose
monitoring and insulin injections as measured by the struc-
tured joint interview (BGM score and Insulin Injection score).
These interviews provided a microanalysis of the level of par-
ent involvement in these fundamental management tasks. In
order to simplify comparisons, we divided the BGM and In-
sulin Routine scores at the midpoints to form two new cate-
gories of “minimal” and “maximal” involvement. As shown in
Figure 1, A,approximately 60% of patients in both the TW
and SC groups had maximal parent involvement in blood glu-
cose monitoring and injections at baseline.
We created a more comprehensive measure of parent in-
volvement at one-year by examining the change from baseline
in family involvement from the Diabetes Family Responsibil-
ity questionnaire and from the routine BGM and insulin in-
jection scores. Since one of the main goals of our intervention
was to prevent the deterioration in parent involvement with
diabetes management tasks that often occurs during the first
few years after the diagnosis of diabetes as well as during early
adolescence, we examined the percentage of patients in the
TW and SC groups that exhibited an increase, maintenance,
or decrease in their ratings of parent involvement on these
measures. As displayed in Figure 1, B,a significantly higher
percentage of TW families (30%) increased or maintained
family teamwork, while only 14% of SC families maintained
or increased involvement (χ
2
= 3.73, df = 1, P = .05). In other
words, we found that patients exposed to the family teamwork
intervention were 2.6 times more likely to increase or main-
tain family involvement during the one-year intervention pe-
riod than did patients exposed to standard care (P = .05).
Impact of Teamwork Intervention on Glycemic
Control
Our teamwork intervention specifically targeted in-
creased parent involvement with blood glucose monitoring
and insulin dosing. The TW intervention aimed to support
adherence to these management tasks through increased posi-
tive family involvement at a time when both parent involve-
ment and glycemic control tend to deteriorate in a pediatric
population due to adolescence and longer duration of diabetes.
As shown in Table II, whereas the A1c in the TW and SC
groups did not differ at baseline, the Alc in the TW group did
not deteriorate as did the Alc in the SC group after one-year
of follow-up (t = –2.01, df = 98, P < .05). In order to confirm
the link between the teamwork intervention and the preven-
tion of deterioration in glycemic control after one-year, we
performed a multivariate analysis. In the regression model (R
2
= 0.17, P = .04), controlling for age, duration of diabetes, and
adjustments to the diabetes treatment program (including
changes in both daily insulin dosage (U/kg) and BGM fre-
quency), assignment to the TW intervention remained the
only significant predictor of change in A1c over the year (P =
.04). After controlling for these potentially confounding co-
variates, the change in A1c after one year was 0.5% less (better
control) in the TW intervention group than in the SC group.
In another multivariate model, we examined the predictors
of A1c after the one-year study period. Assignment to the TW
intervention (P = .04) and the frequency of BGM (P = .05) were
the only significant predictors of A1c (model: R
2
= 0.17, P = .05),
after controlling for age, duration, gender, and daily insulin
dosage (U/kg). In the model, the adjusted mean A1c value for
the TW intervention group was 8.8% while the adjusted mean
A1c for the SC group was 9.3%. In this model, more frequent
blood glucose monitoring predicted lower A1c levels (Fig 2).
DISCUSSION
Our family teamwork intervention demonstrated a sig-
nificant difference in glycemic control after one year compared
412 Laffel et al The Journal of Pediatrics • April 2003
Ta b le I. Baseline patient characteristics by study
condition (n = 100)
Study condition
Teamwork Standard
intervention care
(n = 50) (n = 50)
Age (y), mean ± SD 11.9 ± 2.4 12.2 ± 2.2
Duration (y), mean ± SD 2.7 ± 1.6 2.7 ± 1.6
Sex (% male) 54 52
Height (cm) 149.6 ± 15.3 154.2 ± 15.0
Weight (kg) 45.0 ± 13.6 51.3 ± 15.7
Body mass index 19.7 ± 3.2 21.2 ± 3.9
Single parent families (%) 22 8
Mother’s occupational code
*
3.4 ± 1.4 3.5 ± 1.5
Father’s occupational code
*
2.3 ± 1.4 2.4 ± 1.2
*Occupational code: 1 = major professional (physician, lawyer); 3 =
skilled worker; 6 = unemployed/retired.
to standard multidisciplinary care in a relatively recent onset
group of youth with T1DM. The A1c in the teamwork group
did not deteriorate and was significantly lower than the A1c
in the standard care group after one year. Multivariate analy-
ses indicated that the significant difference in A1c resulted
from the teamwork-mediated family involvement in diabetes
management tasks. Furthermore, although parent involve-
ment was maintained or increased significantly over one year
in the teamwork families, there was no increase in diabetes-
related family conflict or any negative impact on quality of life
as reported by patients and parents receiving the teamwork in-
tervention.
The baseline levels of parent involvement in daily in-
sulin injections and blood glucose monitoring were identical
in the two groups. This level of parent involvement was main-
tained or increased in the patients randomized to the team-
work intervention compared to those receiving standard care.
More than twice as many families in the teamwork group in-
creased or maintained family involvement in the tasks of dia-
betes management after one-year of follow up compared to
the standard care group. This observation is particularly strik-
ing in that there were more single parent families among the
family teamwork participants at baseline than among the stan-
dard care participants. It has been suggested that children in
single-parent families may be at risk for disengagement from
parent support in diabetes care tasks.
23,24
Thus, the impact of
the teamwork curriculum on establishing positive family be-
haviors in such a high-risk group is notable.
Impact of Ambulatory, Family-Focused Teamwork Intervention on Glycemic
Control in Youth With Type 1 Diabetes
413
Ta b le II. Diabetes-related data by study condition at baseline and one year according to group (n = 100)
Baseline 1 Year
Teamwork Standard Teamwork Standard
intervention care intervention care
(n = 50) (n = 50) (n = 50) (n = 50)
Insulin (U/kg per d), mean ± SD 0.9 ± 0.3 0.9 ± 0.3 1.0 ± 0.3 1.0 ± 0.3
Injections per d (%)
252543240
340424448
484168
CSII 0 0 8 4
Frequency of BGM/d (%)
0-2 6 8 10 14
328263028
458604048
5+ 8 6 20 10
A1c (%), mean ± SD
*
8.4 ± 1.3 8.3 ± 1.0 8.2 ± 1.1 8.7 ± 1.5
*At 1-year follow-up, there was a significant difference in A1c between TW and SC groups (t = –2.01, df = 98, P < .05).
Ta b le III. Family behavior characteristics at baseline and one year according to group (n = 100)
Baseline 1 Year
Teamwork Standard Teamwork Standard
intervention care intervention care
(n = 50) (n = 49) (n = 50) (n = 50)
Diabetes-related conflict (0-19)
Child 4.9 (3.8) 4.6 (3.8) 4.3 (3.7) 3.8 (3.8)
Parent 3.2 (3.4) 3.4 (3.3) 3.1 (3.9) 2.8 (2.9)
Diabetes family responsibility (17-51)
Child 35.1 (6.2) 34.1 (4.9) 34.4 (4.8) 33.6 (5.6)
Parent 38.1 (4.9) 38.0 (5.6) 37.3 (5.0) 36.4 (5.3)
Child quality of life (0-100)
Child 83.5 (9.2) 82.3 (12.9) 85.3 (9.9) 84.9 (12.0)
Parent 81.9 (12.8) 78.9 (15.7) 81.1 (13.2) 81.9 (12.4)
Data are mean (SD) scores (NS). Survey data missing on one family from SC group at baseline.
In addition to assignment to the teamwork intervention,
analyses indicated that frequency of blood glucose monitoring
was the only other significant predictor of A1c at follow-up.
The strong relationship between frequency of BGM and A1c
remains consistent with our previous reports and with find-
ings from other investigators.
13,25,26
Our observations suggest
that an intervention targeting parent involvement around dia-
betes management tasks, highlighting the importance of posi-
tive, nonjudgmental communication around BGM,
contributed to the maintenance of family involvement and
was the foundation for the success of this office-based inter-
vention’s impact on A1c.
The intervention modules were designed for ease of ad-
ministration and translatability to other pediatric care settings.
While research assistants administered the modules in 15 to
20 minutes sessions, either before or after the routine diabetes
visit, the interventions are self-explanatory. They utilize lan-
guage that supports successful behaviors around complex dia-
betes management tasks within families. This model of
successful integration of ambulatory-based interventions
could be distributed by office staff and may be suitable for the
care of pediatric patients with other chronic diseases that, like
diabetes, require the family to share the burden of intensive,
home-based treatments.
27,28
The importance of family involvement in diabetes man-
agement tasks has been reinforced with research on the influ-
ence of family involvement across various areas of childhood
development. In particular, children and adolescents whose
parents are involved in their academic and daily pursuits have
significantly less school failure, teen pregnancy, teen drug use,
and adolescent adjustment reactions.
29
In addition, the im-
portance of parental involvement for maintaining consistent
insulin administration and avoiding recurrent diabetic ke-
toacidosis has been well documented in the literature.
30,31
Other investigators in pediatric diabetes research have also
highlighted the value of family involvement. Both cognitive
behavior therapy as well as family-based therapy approaches
have been utilized to encourage positive interactions among
family members and reduce family conflict, which may im-
pede the successful implementation of an intensive diabetes
treatment program.
12,15,32
Our study successfully implemented an office-based
family teamwork intervention that appeared efficacious after
one year compared to standard multidisciplinary diabetes care.
In our prior studies,
14
we learned that families who participat-
ed in an “Attention Control Condition,” consisting of tradi-
tional diabetes education materials and equivalent time with a
research assistant, did not differ significantly from the Stan-
dard Care group in biological, behavioral or family outcome
variables. In the present study, we chose not to include an At-
414 Laffel et al The Journal of Pediatrics • April 2003
Fig 1. A, Parent involvement in blood glucose monitoring and in-
jections according to group at baseline. At study entry, patients ran-
domly assigned to the TW intervention had levels of parental
involvement in diabetes management tasks similar to patients ran-
domly assigned to SC. B, Change in family involvement with DM
tasks during follow-up according to group. Patients randomly as-
signed to the TW intervention had significantly more family in-
volvement in diabetes management tasks after 1 year compared
with patients randomly assigned to SC (P = .05). Family involve-
ment was measured from the structured joint parent and child in-
terview
13
and from the Diabetes Family Responsibility
Questionnaire.
8
Patients in the TW group had 2.6 times the chance
of increasing/maintaining family involvement compared with patients
in the SC group after 1 year (P = .05). **χ
2
= 3.73, df = 1, P = .05
Fig 2. Blood glucose monitoring predicts A1c. In a multivariate
model, after controlling for age, duration of diabetes, sex, and daily
insulin dose (U/kg per day), assignment to the teamwork interven-
tion and frequency of blood glucose monitoring were significant
predictors of A1c (R
2
= 0.17, P = .05). More frequent blood glucose
monitoring (P = .05) predicted lower A1c levels.
tention Control Group and, thus, increase our power to find
differences with this low-intensity, ambulatory intervention.
In the current investigation, we were able to achieve a mean
A1c of 8.2% ± 1.1% in the family teamwork intervention, a value
similar to the mean of 8.1% ± 1.2% that was achieved in the in-
tensively treated group (n = 92) of adolescents in the DCCT.
2
Our patients had similar durations of diabetes to those adoles-
cents enrolled in the primary prevention cohort of the DCCT (n
= 125). Indeed, it is remarkable that our low-intensity, low-cost,
office-based intervention achieved similar glycemic success to the
DCCT, which included weekly phone calls by trained research
nurses and monthly follow-up medical visits.
33
In our study, the
observation that between 32-40% of patients were receiving only
two daily insulin injections after one year of follow-up suggests
that additional opportunities exist to intensify insulin therapy in
these patients to optimize control. It should be noted, however,
neither the number of injections per day nor units/kg per day of
insulin were predictors of glycemic control in a multivariate
model.
Interpretations of our findings warrant caution. Our
study cohort consisted of patients with diabetes of ≤6 years’
duration, suggesting an opportunity to affect behaviors early
within the course of diabetes. Additional studies are needed to
evaluate if patients with longer durations of diabetes, who may
have established negative family behavior patterns that may
be more resistant to change, would benefit from this low-in-
tensity, brief ambulatory intervention. Furthermore, it is im-
portant to evaluate this cohort in the future, especially as the
patients approach later adolescence, to see if the impact of this
intervention persists in maintaining glycemic control. The
generalizability of our findings to families from heterogeneous
ethnic and racial groups also needs to be established. More-
over, future interventions may need to be intensified to achieve
the revised A1c target of <7%, as advocated by the American
Diabetes Association.
34
To attain this goal, approaches may
include intensive treatment innovations such as continuous
glucose monitors, pen injectors, and new insulin analogs that
require sophisticated understanding and family support for
implementation.
Maintaining successful family involvement appears key
to achieving optimal glycemic control in youths with diabetes
to preserve health and prevent future complications. Incorpo-
rating family teamwork in diabetes management tasks is pos-
sible with written modules implemented by trained research
assistants working within a multidisciplinary pediatric dia-
betes practice. With the current emphasis on intensive dia-
betes management in the context of limited health care
resources, our findings suggest the efficacy of this office-based
intervention and point to the value of further study and as-
sessment of its translatability to other healthcare settings.
We acknowledge contributions of the Pediatric Team at the Joslin Di-
abetes Center: Joan Mansfield, MD, Alyne Ricker, MD, Cindy
Pasquarello, RN, BSN, CDE, Kristen Rice, RN, BSN, CDE, Kath-
leen Walsh, RN, CDE, Louise Crescenzi, and the pediatric endocrine
fellows; we also acknowledge the research assistance of Abigail Mans-
field and Amanda Fisher and the computer/statistical expertise of
Linda Ficociello.
REFERENCES
1. DCCT Research Group. The effect of intensive treatment of diabetes
on the development and progression of long-term complications in insulin-
dependent diabetes mellitus. N Engl J Med 1993;329:977-86.
2. DCCT Research Group. Effect of intensive diabetes treatment on the
development of long-term complications in adolescents with insulin-depend-
ent diabetes mellitus. J Pediatr 1994;125:177-88.
3. Retinopathy and nephropathy in patients with type 1 diabetes four years
after a trial of intensive therapy. The Diabetes Control and Complications
Trial/Epidemiology of Diabetes Interventions and Complications Research
Group. N Engl J Med 2000;342:381-9.
4. White NH, Cleary PA, Dahms W, Goldstein D, Malone J, Tamborlane
WV. Beneficial effects of intensive therapy of diabetes during adolescence:
outcomes after the conclusion of the Diabetes Control and Complications
Trial (DCCT). J Pediatr 2001;139:804-12.
5. Effect of intensive therapy on the microvascular complications of type 1
diabetes mellitus. JAMA 2002;287:2563-9.
6. Drash AL. The child, the adolescent, and the Diabetes Control and
Complications Trial. Diabetes Care 1993;16:1515-6.
7. Wolfsdorf JI. Improving diabetes control in adolescents. Diabetes Care
1999;22:1767-8.
8. Anderson BJ, Auslander WF, Jung KC, Miller JP, Santiago JV. Assessing
family sharing of diabetes responsibilities. J Pediatr Psychol 1990;15:477-92.
9. Wysocki T, Taylor A, Hough BS, Linscheid TR, Yeates KO, Naglieri
JA.Deviation from developmentally appropriate self-care autonomy. Associa-
tion with diabetes outcomes. Diabetes Care 1996;19:119-25.
10. Grey M, Boland EA, Davidson M, Yu C, Sullivan-Bolyai S, Tambor-
lane WV. Short-term effects of coping skills training as adjunct to intensive
therapy in adolescents. Diabetes Care 1998;21:902-8.
11. Grey M, Boland EA, Davidson M, Li J, Tamborlane WV. Coping skills
training for youth with diabetes mellitus has long-lasting effects on metabolic
control and quality of life. J Pediatr 2000;137:107-13.
12. Wysocki T, Harris MA, Greco P, Bubb J, Danda CE, Harvey LM et al.
Randomized, controlled trial of behavior therapy for families of adolescents
with insulin-dependent diabetes mellitus. J Pediatr Psychol 2000;25:23-33.
13. Anderson BJ, Ho J, Brackett J, Finkelstein D, Laffel L. Parental involve-
ment in diabetes management tasks: relationships to blood glucose monitor-
ing adherence and metabolic control in young adolescents with
insulin-dependent diabetes mellitus. J Pediatr 1997;130:257-65.
14. Anderson BJ, Brackett J, Ho J, Laffel LM. An office-based intervention
to maintain parent-adolescent teamwork in diabetes management. Impact on
parent involvement, family conflict, and subsequent glycemic control. Dia-
betes Care 1999;22:713-21.
15. Grey M, Cameron ME, Lipman TH, Thurber FW. Psychosocial status
of children with diabetes in the first 2 years after diagnosis. Diabetes Care
1995;18:1330-6.
16. Northam E, Anderson P, Adler R, Werther G, Warne G. Psychosocial
and family functioning in children with insulin-dependent diabetes at diag-
nosis and one year later. J Pediatr Psychol 1996;21:699-717.
17. Kovacs M, Goldston D, Obrosky DS, Iyengar S. Prevalence and predic-
tors of pervasive noncompliance with medical treatment among youths with
insulin-dependent diabetes mellitus. J Am Acad Child Adolesc Psychiatry
1992;31:1112-9.
18. Hanson CL, Henggeler SW, Burghen GA. Model of associations be-
tween psychosocial variables and health-outcome measures of adolescents
with IDDM. Diabetes Care 1987;10:752-8.
19. Varni JW, Seid M, Rode CA. The PedsQL: measurement model for the
pediatric quality of life inventory. Med Care 1999;37:126-39.
20. Rubin RR, Young-Hyman D, Peyrot M. Parent-child responsibility and
conflict in diabetes care [abstract]. Diabetes 1989;38(suppl 2):28.
21. Jacobson AM, Hauser ST, Lavori P, Wolfsdorf JI, Herskowitz RD, Mil-
ley JE, et al. Adherence among children and adolescents with insulin-depend-
ent diabetes mellitus over a four-year longitudinal follow-up, I: the influence
of patient coping and adjustment. J Pediatr Psychol 1990;15:511-26.
22. Steinberg EB. Predictors of diabetes-specific conflict among parents and
children with insulin-dependent diabetes mellitus (IDDM) [abstract]. Dis-
sertation Abstracts International 1999;60[2-B]:859.
Impact of Ambulatory, Family-Focused Teamwork Intervention on Glycemic
Control in Youth With Type 1 Diabetes
415
23. Jacobson AM, Hauser ST, Lavori P, Willett JB, Cole CF, Wolfsdorf JI,
et al. Family environment and glycemic control: a four-year prospective study
of children and adolescents with insulin-dependent diabetes mellitus. Psy-
chosomat Med 1994;56:401-9.
24. Auslander WF, Bubb J, Rogge M, Santiago JV. Family stress and re-
sources: potential areas of intervention in children recently diagnosed with di-
abetes. Health Soc Work 1993;18:101-13.
25. Levine BS, Anderson BJ, Butler DA, Brackett J, Laffel L. Predictors of
glycemic control and short-term adverse outcomes in youth with type 1 dia-
betes. J Pediatr 2001;139:197-203.
26. Nathan DM, McKitrick C, Larkin M, Schaffran R, Singer DE.
Glycemic control in diabetes mellitus: have changes in therapy made a differ-
ence? Am J Med 1996;100:157-63.
27. Quittner AL, Drotar D, Ievers-Landis C, Slocum N, Seidner D, Jacob-
sen J. Adherence to medical treatments in adolescents with cystic fibrosis: the
development and evaluation of family-based interventions. In: Drotar D, edi-
tor. Promoting adherence to medical treatment in chronic childhood illness.
Mahwah (NJ): Lawrence Erlbaum; 2000. p. 383-407.
28. Armstrong FD, Harris LL,Thompson W, Semrad JL, Jensen MM, Lee
DY, et al. The Outpatient Developmental Services Project: integration of pe-
diatric psychology with primary medical care for children infected with HIV.
J Pediatr Psychol 1999;24:381-91.
29. Resnick MD, Bearman PS, Blum RW, Bauman KE, Harris KM, Jones
J, et al. Protecting adolescents from harm. Findings from the National Longi-
tudinal Study on Adolescent Health. JAMA 1997;278:823-32.
30. Glasgow AM, Weissberg-Benchell J, Tynan WD, Epstein SF, Driscoll
C, Turek J, et al. Readmissions of children with diabetes mellitus to a chil-
dren’s hospital. Pediatrics 1991;88:98-104.
31. Golden MP, Herrold AJ, Orr DP. An approach to prevention of recurrent
diabetic ketoacidosis in the pediatric population. J Pediatr 1985;107:195-200.
32. Grey M, Davidson M, Boland EA, Tamborlane WV. Clinical and psy-
chosocial factors associated with achievement of treatment goals in adoles-
cents with diabetes mellitus. J Adolesc Health 2001;28:377-85.
33. DCCT Research Group. Implementation of treatment protocols in the
Diabetes Control and Complications Trial. Diabetes Care 1995;18:361-76.
34. American Diabetes Association. Standards of medical care for patients
with diabetes mellitus. Diabetes Care 2003;26:S33-S50.
416 Laffel et al The Journal of Pediatrics • April 2003
50 Years Ago in The Journal of Pediatrics
CHANGES IN BLOOD COAGULATION FACTORS DURING THE FIRST WEEKS OF LIFE
Owen CA, Hurn MM. J Pediatr 1953;42:424-8
These authors conducted an elegant study of the coagulation system in term newborn infants. Fifty years ago, it was under-
stood that the blood coagulation system of neonates was different from that of adults, and that vitamin K was necessary for
blood clotting. However, the blood coagulation “cascade” had not yet been described. These investigators were trying to deter-
mine which factors other than prothrombin (factor II) were responsible for bleeding risk during the first few days of life. First,
they found that labile factor (what is now called factor V) was present in neonatal plasma at adult levels. Stable factor (now
known as factor VII) was reduced at birth and rose during the first few days of life or after administration of vitamin K. Pro-
thrombin levels were also reduced in the first few days of life but did not normalize with administration with vitamin K. The
authors concluded that the prothrombin time, thought previously to measure only the concentration of prothrombin, actually
measured the activation of prothrombin to thrombin and was dependent on factors of factor VII and factor V. Eventually, the
importance of factor X was established as well. The prothrombin time continues today as the primary screening test for abnor-
malities in the tissue factor/factor VII dependent pathway for coagulation. The coagulation system has become increasingly
complex, but we still do not have a complete understanding of the remarkable differences in hemostasis between infants and
adults. More research is necessary in this important area.
Janna M. Journeycake, MD
Division of Hematology-Oncology
Department of Pediatrics
University of Texas Southwestern Medical Center at Dallas
Dallas, TX 75390-9063
YMPD178
10.1067/mpd.2003.178
