454 Child Health Psychology
illness or illness symptoms, perceived severity of symptoms,
and costs (perceived barriers) versus bene“ts of engaging in
treatment interact to predict adherence (Janz & Becker,
1984). Bond, Aiken, and Somerville (1992) illustrate the
complexity and inconsistency of “ndings regarding the
health belief model. They applied the model to adolescents
with IDDM, adding perceived cues (onset or change in ill-
ness symptoms) to engage in adherence behaviors. They
reported that perceived cues were most associated with ad-
herence such that when symptoms were experienced, adoles-
cents with IDDM were more likely to seek medical help.
Bene“ts-costs were also associated with adherence. For
threat (perceived susceptibility and severity), adherence was
highest in the case of low threat and high bene“ts-costs but
metabolic control was highest in the case of high threat and
high cues.
Research examining the effectiveness of interventions to
improve treatment adherence is complicated by a prolifera-
tion of indirect and direct approaches to measuring adherence
and related questions regarding the reliability and validity of
the measures used (La Greca & Schuman, 1995). Indirect
measures are patient and parent reports of adherence behav-
iors or patient, parent, and physician ratings of adherence.
These measures tend to overestimate adherence. Direct or ob-
jective measures of adherence such as observation of patient
engagement in procedures, drug assays, and pill counts have
been subjected to criticism due to observer bias and individ-
ual differences in metabolism. An additional complication is
the lack of one-to-one correspondence between following
prescribed treatment and positive medical outcome; that is, as
the health beliefs model suggests, why engage in compli-
cated, time-consuming, and sometimes painful treatments
when the payoff in terms of illness symptoms is unclear?
This not only contributes to a lack of adherence but also
makes physical health an unreliable outcome measure for in-
terventions targeting adherence. It has been suggested that
multiple measures of adherence, direct and indirect, be em-
ployed and that patients• speci“c adherence behaviors be
compared to treatment prescribed speci“cally to them (not a
general treatment regimen for the condition under study; La
Greca & Schuman, 1995).
Research with pediatric samples supports the effective-
ness of education for parents and children, that is, improving
knowledge and skills in carrying out treatment for chronic
pediatric illnesses (Delamater et al., 1990). Coping skills
training or training in approaches to problem solving has also
been effective in work with children with chronic illness be-
cause it allows the patient to address his or her unique barri-
ers to adherence (Delamater et al., 1990; Satin, La Greca,
Zigo, & Skyler, 1989). Satin et al. (1989) studied adolescents
with IDDM who participated in multiple family groups only
or multiple family groups and simulation of treatment re-
quirements by parents. Multiple family groups concentrated
on problem-solving barriers to adherence. A control group
received no intervention. Findings showed that compared to
the control group, both treatment groups had better metabolic
control and maternal reports of adherence for up to six
months.
Concrete measures, such as increased medical supervi-
sion, token reinforcement, and parental praise for keeping
appointments (Finney, Lemanek, Brophy, & Cataldo, 1990;
Greenan-Fowler, Powell, & Varni, 1987), are effective in
improving adherence. Da Costa, Rapoff, Lemanek, and
Goldstein (1997) reported the outcome of an intervention
program involving parent and child education and token re-
inforcement for taking medication using a case study, a
withdrawal design with two children with moderate to
severe asthma. Results indicated that use of token reinforce-
ment was helpful in improving medication adherence; how-
ever, maintenance was problematic once the reinforcement
was withdrawn, and the impact on pulmonary function was
not clear.
To summarize, multicomponent intervention programs
appear to be most effective in improving adherence to pre-
scribed treatments because they can target each family•s bar-
riers to adherence (La Greca & Schuman, 1995). However, it
is dif“cult to determine which components of these programs
lead to their effectiveness. It is recommended that the effec-
tiveness of interventions with particular patients and their
families be assessed in prospective studies that employ pri-
marily objective measures of adherence.INDICATED INTERVENTIONSTraumatic Brain Injury in ChildrenTraumatic brain injury (TBI) is the leading cause of death and
permanent disability in children and adolescents (Guyer &
Ellers, 1990), affecting between 185 and 230 per 100,000
children under 15 years of age (Kraus, 1995). TBI is not
evenly distributed in the population. That is, some children
and adolescents appear to be at increased risk for injury, in-
cluding those with behavior and/or learning problems, high
risk taking, or disadvantaged families who monitored their
behavior less prior to injury (Kraus, Rock, & Hamyari, 1990;
Ylvisaker, 1998). In turn, these risk factors impact the avail-
ability of supports necessary after the injury.
The impact of TBI on society is profound in terms of
“nancial cost, family stress, disruption in school, and the