Journal of Pediatric Psychology, Vol. 28, No. 6, 2003, pp. 383-392
© 2003 Society of Pediatric Psychology
The Illness Management Survey: Identifying Adolescents' Perceptions of Barriers to Adherence
The Children's Hospital of Philadelphia
All correspondence should be sent to Deirdre Logan, PhD, Department of Psychology, Children's Hospital of Philadelphia, 34th & Civic Center Blvd., Philadelphia, Pennsylvania 19104. E-mail: logan{at}email.chop.edu.
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionReferences |
|---|
Objective To develop a self-report measure of barriers to adherence and to evaluate its reliability and validity in a sample of adolescents with asthma. Methods The Illness Management Survey (IMS) was developed through item generation, expert panel review, and focus group administration. Adolescents with asthma (N = 152) completed the measure. Participants reported on perceived drawbacks to medication, risk-taking behavior, and social desirability tendencies. Providers rated adolescents' illness severity and adherence. Reliability and validity of the IMS were assessed, and factor structure was examined. Results The 27-item IMS shows high internal consistency (alpha =.87). Scores correlate with perceived medication drawbacks, risk taking, and self- and provider reports of adherence. Principal-components analysis indicates five domains of barriers, accounting for 52.4% of the variance: disease/regimen issues, cognitive difficulties, lack of social support/lack of self-efficacy, denial/distrust, and peer/family issues. Conclusions Preliminary data indicate that the IMS reliably and validly assesses perceived barriers to adherence within this sample of adolescents with asthma. It shows promise as a tool for identifying subgroups of nonadherent adolescents.
Key words: regimen adherence; adolescence; psychological assessment; chronic illness.
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Despite growing knowledge about adherence to pediatric medical regimens, the variables most crucial to successful interventions for improving adherence among adolescents with chronic illnesses remain elusive (Riekert & Drotar, 2000
).
Adherence is commonly defined as the "extent to which a
patient's behavior coincides with medical or health advice"
(Haynes, 1979). This
definition delineates a range of patient behaviors (taking medications
consistently, keeping clinic appointments, following special diets, making
lifestyle changes) and suggests that patients' choice of behaviors may or may
not align with medical recommendations. Studies have shown that poor adherence
results in increased morbidity and mortality, overuse of health care services,
and distorted results of medication trials
(Lemanek, Kamps, & Chung,
2001; Matsui,
2000, Quittner, Espelage, Ievers-Landis, & Drotar, 2000).
Approximately 50% of pediatric patients are considered nonadherent to
treatment recommendations (Rapoff &
Barnard, 1991). These rates vary widely with illness type,
severity, duration, regimen requirements, and developmental stage. Given that
adolescence is recognized as a period of increased risk for adherence problems
(La Greca, 1990), this study
seeks to identify adolescents' own perceptions of what makes adherence so
difficult to maintain at their stage of development.
In developing a measure of barriers or obstacles that interfere with
adherence, ecologically based theoretical models that have been proposed to
explain illness behaviors—such as the Children's Health Belief Model
(Bush & Ianotti, 1990) and
Hanson's (1992) systemic model
for youth with diabetes—were used. Compared with more individually based
frameworks, these models suggest a prominent role for systemic factors.
Consistent with the Children's Health Belief Model, the theoretical foundation
for this study focuses on the role of perceived barriers to adherence and
views caregivers as important environmental influences. This is augmented with
a multi-systemic view as described by Hanson
(1992), incorporating
influences on perceived barriers to adherence that are exerted by family
(e.g., support, resources, family members' perceptions of the illness), peers
(e.g., peer relationships), and the medical team (e.g., relations with patient
and family, perceptions of severity, communication skills).
Because the primary influences on adherence are likely to differ across
individuals (Bauman, 2000), the
goal of this study was to develop a measure that could identify specific
barriers to adherence for individual adolescents. Currently, comprehensive
assessment tools to elucidate adolescents' perceptions of potential barriers
to adherence are lacking. This is information that could aid clinicians in
estimating patients' level of risk for nonadherence. The few existing measures
that accomplish this aim are limited to a specific illness, or even to one
aspect of a particular regimen (e.g.,
Glasgow, McCaul, & Schafer,
1986; Kyngäs, Kroll &
Duffy, 2000; Schlunt et al.,
1996). The present study will advance the research in this area by
developing an assessment technique that covers a wide spectrum of barriers.
The many illness-specific factors that can potentially influence adherence
present challenges to adopting a broad, non-illness based view of adherence
behaviors. However, noncategorical approaches to the study of adherence are
needed (Quittner et al., 2000).
Intervention studies aimed at improving rates of adherence typically have
not addressed whether different subgroups of nonadherers might benefit from
interventions tailored to specific areas of difficulty. Conceptualizing
adherence on a continuous and fluid spectrum has advantages with regard to
understanding intricacies and changes over time
(Lemanek et al., 2001;
Quittner et al., 2000). Several researchers suggest that delineating
nonadherence subtypes is a fruitful area for further research efforts
(La Greca, 1990;
Byron, 1998), yet few studies
have attempted to accomplish this goal. One study
(Koocher, McGrath, & Gudas,
1990) identified three typologies of nonadherence in youth with
cystic fibrosis (i.e., inadequate knowledge, psychosocial resistance, and
educated nonadherence), but the authors did not subject their typology to
empirical validation. Developing an empirically supported and valid measure
for classifying subgroups or typologies of nonadherent adolescents, based on
the barriers that contribute most heavily to their nonadherent behaviors,
would therefore advance this area of inquiry. Developing a multidimensional
assessment device with a solid base in the literature on barriers to adherence
can enable us to design interventions tailored to specific subgroups of
nonadherent adolescents.
The aims of the present study were therefore twofold. The first aim was to develop and validate a brief measure, the Illness Management Survey (IMS), designed to assess perceptions of barriers to adherence and to be used easily in clinical settings. This aim incorporates the processes of instrument development, reliability assessment, and validation. The second aim was to determine, with the use of the IMS, whether chronically ill adolescents could be classified into typologies based on their perceptions of the relative importance of specific types of barriers to adherence. We hypothesized that subgroups could be classified by their responses to this self-report assessment. Based on a comprehensive review of the pediatric adherence literature, we proposed that the following domains (described under Methods, below) would capture these subgroups: (a) disease and regimen factors (e.g., pain/discomfort, hassles, physical changes); (b) cognitive factors (e.g., confusion, memory, knowledge of illness); (c) intrapersonal and developmental factors (e.g., autonomy issues, perceived immortality, denial/minimizing, pessimism/optimism, self-efficacy); (d) family and medical system influences (e.g., parent guidance/shared responsibility, closeness, relationship to medical team; and (e) peer influences (e.g., wishes not to appear different from friends, exposure to other teens with similar illnesses).
Patients with asthma were selected as the target population for this stage
of measure development for several reasons: (a) Asthma is the most common
pediatric chronic illness (Creer &
Bender, 1995); (b) There are well-documented high rates of
nonadherence among this group despite widespread availability of effective
treatments (Bender, Milgrom, Rand, &
Ackerson, 1998; Milgrom et
al., 1996); and (c) Due to the complexity of treatment regimens,
the requirements of both continuous maintenance and crisis treatment, and the
intermittent and variable course of the illness, asthma patients are in great
need of continued study to understand adherence decisions and facilitate
adaptive illness management (e.g.,
Spector, 2000). At this
development stage we chose not to limit participation to adolescents
identified as nonadherent, so that we could obtain input from adolescents
representing the full spectrum of adherence-related attitudes and
behaviors.
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Initial Measure Development
As a first step in the process of measure development, the pediatric adherence literature was reviewed in detail. Correlates of nonadherence that were supported by previous empirical studies were identified and grouped into categories used in prior research (e.g., La Greca & Schuman, 1995
;
Matsui, 2000) and consistent
with the Children's Health Belief Model
(Bush & Ianotti, 1990).
Disease and regimen factors include the duration and course of illness,
symptom severity, pain and discomfort, type and complexity of regimen,
efficacy of regimen, and side effects (e.g.,
Bond, Aiken, & Somerville,
1992; Brownbridge &
Fielding, 1994). Patient factors include (a) developmental
characteristics such as peer issues, perceived injustice, level of
independence/autonomy, and a sense of immortality; (b) cognitive factors such
as knowledge of the disease and regimen, memory, and confusion; and (c)
intrapersonal or attributional tendencies such as mood, wishes to deny or
minimize illness, pessimism, locus of control, and self-efficacy (e.g.,
Brownbridge & Fielding,
1994; Gudas, Koocher, &
Wypij, 1991; Schlunt et al.,
1996). Family factors include parental support, shared
responsibility for regimen, family cohesion, and family dysfunction (e.g.,
Anderson, Auslander, Jung, Miller, &
Santiago, 1990; Miller-Johnson
et al., 1994; Rapoff,
1999). Other systems also play a role in adherence; for instance,
qualities of the relationship between the patient/family and the medical team,
physician's communication skills, and perceptions of the physician as warm and
approachable also are important influences (e.g.,
Lemanek et al., 2001;
Rapoff, 1999). This review
yielded a list of empirically supported barriers to adherence among
adolescents with chronic illness. Questionnaire items were written to reflect
each barrier. The items were purposely general enough to pertain to a range of
chronic illnesses. Agreement with each statement was indicated with a 5-point
Likert-type scale, with a range of 1 = strongly disagree to 5 = strongly
agree. Next, five experts in the area of medical regimen adherence (pediatric psychologists and pediatricians) reviewed the original list of 80 potential items. These experts rated each item for its relevance to the aim of assessing barriers to adherence among adolescents. They also gave feedback regarding which of the hypothesized domains of nonadherence (i.e., disease/regimen, cognitive, intrapersonal, family/medical systems, peer influences) they thought the item tapped. Based on the expert panel review, one of our hypothesized domains (family/medical systems) was divided into separate family and medical systems categories. Items were retained if the average rating for the item's relevance as an adherence barrier exceeded 3.0 on a 5-point scale and if 4 out of 5 experts agreed on which factor the item represented. Using this system, 55 items were retained for the pilot questionnaire. The exact wording of some items was refined based on specific comments from the review panel.
The IMS was subsequently administered to a focus group (N = 6) of adolescents with asthma. The group consisted of 3 males and 3 females with a mean age of 16.2 years. Ethnically, the group was 50% white, 33% African American, and 17% Latino. These teens were identified for participation by their primary asthma care providers (e.g., pulmonology nurse practitioner, allergy physician) based on providers' opinions of patients who would give helpful feedback about adherence issues. Parental consent and adolescent assent were obtained via telephone prior to the focus group. Adolescents' written and verbal feedback on the questionnaire supported the measure's face validity and did not identify any additional barriers that adolescents believed were missing from the measure. Additional minor wording revisions were made upon recommendation of the focus group participants.
Participants
One hundred fifty-two adolescents (91 males, 61 females) aged 11-18 years
were recruited from outpatient pulmonary and allergy clinics at a large
pediatric hospital. Exclusion criteria were (a) inability to comprehend
written English at the fifth-grade reading level and (b) asthma diagnosed
within the past year. Participants represented a range of racial and ethnic
backgrounds, including white (50.7%), African American (39.3%), Latino (2%),
Asian American (1.3%), biracial (3.3%), and other (3.3%). Ninety-nine (65.1%)
of the participants reported that their parents were married, 26 (17.1%) came
from divorced families, 3 (2%) had a parent who was widowed, and 19 (12.5%)
reported that their parents had never been married. The age at which
participants were diagnosed with asthma ranged from birth to 16 years with a
mean of 6.25 years.
Procedures
The study was approved by the hospital's institutional review board.
Eligible adolescents and their parents were approached during outpatient
clinic visits. Study staff explained the study and obtained written parental
consent and adolescent assent/consent. Adolescents then completed the battery
of questionnaires, requiring an average of 25-30 minutes. Families received no
financial or other compensation for participation. Three families who were
approached refused to participate, and five adolescents failed to complete the
full battery of questionnaires after enrolling, resulting in a response rate
of 95%. A random subset of the total sample (n = 100, or two thirds
of the sample) was recontacted by mail 4 weeks after clinic visit to complete
the IMS a second time for the purpose of establishing test-retest reliability.
Response rate for this was low; of the 100 participants contacted, 31
responded.
Measures
Demographic and Illness Information Form
This form gathered background information (e.g., gender, age, ethnicity,
family status, socioeconomic status) as well as information pertaining to
participants' illness (e.g., diagnosis, duration, limitations). Perceived
asthma severity was assessed with a single item asking teens to indicate
whether asthma was mild, moderate, or severe. Adolescents also were asked to
report their own perceived level of adherence to their medical regimen by
responding to single items covering (a) adherence to maintenance medication,
(b) management of asthma attacks, (c) avoidance of environmental irritants,
and (d) adherence to scheduled medical appointments. Likert-type scales were
used with responses ranging from never to always or
almost every day (depending on the construct being measured).
Marlowe-Crowne Social Desirability Scale-Short Form
(Reynolds, 1982)
This 13-item true-false questionnaire is commonly included in studies using
self-report measures in order to control for social desirability biases in
response patterns. The short form has adequate validity and reliability when
compared with the standard 33-item form
(Reynolds, 1982).
Perceptions of Asthma Medication Scale (PAM;
DePaola, Roberts, Blaiss, Frick, &
McNeal, 1997)
This questionnaire assesses perceptions of the benefits and drawbacks of
asthma medication. The 15-item Drawbacks subscale from the children's version
(CPAM-D) is used in this study. The measure uses a 5-point Likert-type
response scale, with potential total scores ranging from 15 to 75. The CPAM
has demonstrated test-retest reliability of .81 for a group of 9-15 year old
participants and a Cronbach's alpha of .85 for the drawbacks subscale
(DePaola et al., 1997).
Correlations between the IMS and the CPAM-D will be one indicator of the IMS's
validity, since both measures assess barriers to medication use (although the
IMS also targets other aspects of the medical regimen).
Adolescent Risk Taking Survey (ARTS;
Alexander et al., 1990)
This six-item scale assesses risk-taking propensity among adolescents and
is included to gauge the relationship between nonadherent attitudes and
general risk-taking behavior. It was developed in a sample of young
adolescents and found to have good reliability and validity. The 3-point
response scale yields total scores of 6-18. Scale alpha coefficients were.78
and.80 at separate administrations, and item-total correlations remained
stable over a 1-year period (see Alexander
et al., 1990).
Clinician Assessment of Adherence
The primary asthma care providers rated participants' current level of
adherence to three realms of asthma care—maintenance medication,
response to asthma attacks, and preventive efforts (e.g., avoiding
environmental irritants). A 4-point scale was used to classify adherence in
each area. This clinician rating scale was modeled after that of Van Sciver,
D'Angelo, Rappaport, and Woolf
(1995), which was shown to be
a reliable method of obtaining clinician adherence ratings. Clinicians also
indicated their judgments of participants' level of asthma severity (mild,
moderate, or severe). Detailed clinician reports of adherence were not
collected at this stage; rather, this global clinician assessment provided
preliminary data regarding the relationship between adolescent self-reports
and clinician judgment of medical adherence.
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Following descriptive information about the study measures, results are reviewed with attention to the steps involved in measure development. These include item analysis and other scale-reduction techniques, assessment of reliability and validity, and factor analysis to determine the subscale structure that emerged within the questionnaire.
Descriptive Information
Mean (SD) scores were: Marlowe-Crowne social desirability
scale-short form, 6.5 (3.2); CPAM-D, 30.6 (10.0); ARTS, 9.3 (2.3). See
Table I for descriptive data
related to provider and self-reports of perceived disease severity and
adherence.
|
Scale Reduction
Given our aim of developing a brief and clinically useful measure, we began
our data analyses by examining the original 55-item pilot questionnaire for
items that could be eliminated without detracting from the scale's useful
properties. Each item was examined to determine whether it elicited a range of
responses. Four items had insufficient distribution, with a significant skew
toward one end of the continuum, and were therefore eliminated. Next,
corrected item-total correlations were examined. Items were eliminated if they
failed to correlate with the total scale score at a level of r = .25
or greater. Using this criterion
(DeVellis, 1991), 11
additional items (with correlations ranging from r = .01 to
r = .24) were eliminated.
Correlations between individual IMS items and total scores on the
Marlowe-Crowne measure of social desirability were examined next. We wanted to
develop a scale that was not overly influenced by social desirability, a
challenging problem in the area of adolescent self-report on adherence
(Bender, Milgrom, & Wamboldt,
2000). Therefore, we eliminated items that varied heavily with
Marlowe-Crowne total scores. Four items that correlated with Marlowe-Crowne
scores at r = .25 (directly or inversely) or greater were omitted
from the scale at this stage. Combined, these reduction techniques yielded a
36-item scale. Each remaining item had a range of responses. Total scale score
showed no significant skew pattern (skewness = .14, standard error = .20).
Reliability
The IMS's reliability was assessed through several methods. First,
Cronbach's alpha was calculated to indicate internal consistency. The 36-item
scale had an alpha of .89, indicating strong internal consistency (and
potential for further item reduction). Test-retest reliability also was
assessed. Although the aim was to obtain these data on one third of the
sample, the response rate resulted in 21% of the total sample available for
test-retest analysis. Compared with those who did not respond, those who
returned their test-retest questionnaires were significantly more likely to be
from two-parent homes [t(59.8) = 2.7, p < .01)] and from
white rather than nonwhite families [t(52.8) = 3.9, p <
.001]. Based on this subset of the full sample, we obtained a test-retest
reliability correlation of r =.88 for the IMS total score at the
first and second administrations.
Validity
Correlations between the IMS and other measures of adherence attitudes
(CPAM-D) and behaviors (self-report and provider ratings) are provided in
Table II. Higher IMS scores
correlated positively with stronger perceptions of drawbacks to medications.
IMS scores also showed significant inverse relationships with both self- and
provider-reported adherence (i.e., higher IMS scores indicated less adherence
per both sets of reporters). The positive correlation between the IMS and the
ARTS indicates a moderate relationship between adolescents' perceptions of
barriers to adherence and self-reports of general risk-taking behavior.
|
Factor Structure
After establishing the psychometric properties of the IMS, a
principal-components analysis (PCA) with Varimax rotation was performed. The
joint criteria of eigenvalues >1 and Cattell's elbow criteria on the scree
plot (DeVellis, 1991;
Kim & Mueller, 1978)
indicated that three to eight factors could explain the structure of the IMS.
Examination of the factor analysis results suggests that a five-factor
solution is most interpretable.
Based on our PCA results, 5 additional items were omitted because they had no factor loadings of .40 or greater. The factor analysis was then rerun with 31 items. After initial inspection of the revised rotated component matrix, 4 final items were deleted for lack of conceptual integrity with the factors on which they loaded; deletion of these items did not alter the measure's factor structure. The principal component analysis was rerun, resulting in a final five-factor solution accounting for 52.4% of the variance in the responses (see Table III). The resulting 27-item scale has a Cronbach's alpha of .87. Sample mean for the 27-item scale was 58.9 (SD = 13.6). Two items had loadings of.40 or above on two separate factors each. The remaining 25 items each had distinct single-factor loadings.
|
Empirical results suggest some revisions to our hypothesized factor
structure. Factor 1, labeled disease/regimen/medical systems, contains seven
items (
= .78). Factor 2, labeled cognitive difficulties, contains
seven items ( = .79). Our hypothesized intrapersonal/developmental
domain, along with some of the more relationship-focused aspects of our
hypothesized medical systems domain, formed the basis of the next two factors.
Factor 3 is a six-item factor, labeled lack of social support/lack of
self-efficacy ( = .74), with a focus on beliefs that others fail to
help the adolescent manage his/her regimen, and that the teen's own behaviors
(e.g., following the regimen) do not affect the illness. Factor 4, labeled
denial/distrust, centers on the wish to deny the presence of illness or the
possibility of negative consequences to nonadherence and beliefs that health
care providers were unhelpful. It contains five items ( = .65).
Finally, Factor 5, labeled peer/family issues, has only four items ( =
.56). The factors were significantly intercorrelated (see
Table III).
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
This study provides initial psychometric data on the IMS. The measure development process described herein yielded a brief self-report assessment of adolescents' perceived barriers to medical regimen adherence, which was piloted among adolescents with asthma. At this stage in its development, the IMS appears to be a psychometrically sound tool for gauging individuals' perceptions of the number and intensity of barriers to their adherence. Additional research may demonstrate that the IMS also can identify subgroups of nonadherers based on the types of barriers they view as most challenging.
Regarding the demographic characteristics of our sample, IMS total scores
were significantly related only to age of respondents, with older teens
reporting more barriers. It is well known that older adolescents have more
difficulty with regimen adherence than do younger adolescents (e.g.,
Kyngäs, 1999). This
finding may have implications for the development of interventions to improve
adherence, because it suggests that developmental or age influences play a
significant role. Intervening earlier may therefore serve a protective
function in preventing adolescents from becoming nonadherent. IMS scores bore
no other statistically significant relationships to adolescent or family
demographic characteristics.
Participants who perceived their asthma as more severe had higher IMS scores. Provider perceptions of greater disease severity also related to higher IMS scores. Teens with more difficult-to-manage asthma may encounter more barriers because their illness requires more work. Alternatively, it may be that teens who identify more barriers to care are less adherent, therefore they experience more symptoms, leading to perceptions of greater illness severity. The study did not incorporate an objective marker of adherence, but it is important to consider that unless adolescents themselves perceive their illness to be severe, they may be unmotivated to change behaviors or to work toward decreasing barriers to adherence. In future research, it will be useful to examine possible mediating influences on the relationship between perceived illness severity and adherence attitudes, such as family influences.
This new instrument shows evidence of concurrent validity in its
correlations with other measures of barriers to adherence (perceptions of
medication drawbacks) and general risk-taking behavior. There is some promise
of predictive validity in the significant associations between IMS total
scores and adolescents' self-report and provider ratings of patients' actual
adherence behaviors. Rates of nonadherent behaviors reported by our
participants and providers are consistent with published self-report and
provider-report rates of nonadherence among this population
(Bender et al., 1998;
Spector, 2000). Although
neither self-nor provider report of adherence behavior can be assumed to gauge
patients' adherence behavior accurately
(Rapoff, 1999), these results
represent a good first step toward establishing the IMS as a tool to identify
one aspect of risk for nonadherence (i.e., perceived barriers).
Preliminary factor analytic findings yield a factor structure that departs
somewhat from the originally hypothesized typology of adolescent nonadherence.
Based on theory and past findings, we originally proposed six rationally
derived domains—disease and regimen factors, cognitive factors,
intrapersonal/developmental factors, family systems influences, medical
systems influences, and peer influences—that we believed would account
for the major barriers to adherence. The empirically derived factor structure
of the IMS suggests that grouping barriers on the basis of the different
ecological systems in the adolescent's life may not be the most useful method
for classifying subgroups of nonadherers. Rather, it may be more accurate to
group barriers based on internal processes, such as adolescents' cognitive
skills, tendency toward denial, and level of pessimistic thinking (i.e.,
perceptions of low social support and low self-efficacy). Some environmental
influences, however, such as disease and regimen characteristics and
peer/family influences, do emerge as important contextual forces. It is also
interesting to note that family influences were a smaller component than we
expected, with only one family-focused item—"My family members
don't understand what it's like to live with my illness"—loading
on the peer/family factor. This may reflect the normative developmental
tendency among adolescents to emphasize peer relationships over family
relationships (e.g., Hartup,
1996) and for peer relationships to link to health-risk behaviors
(La Greca, Prinstein, & Fetter,
2001).
The sample of adolescents who participated in this phase of measure
development consisted of chronically ill youth representing a range of
adherence attitudes and behaviors. Although all participants had a single
chronic illness (asthma), further pilot work may reveal that the measure is
adaptable to adolescents across illnesses. The IMS represents an important
contribution to the field of adherence research given its potential
applicability across illness types, its face validity, resistance to social
desirability in its final item pool, and strong internal consistency. The IMS
also shows promise as a tool for classifying subgroups of adolescent
nonadherers based on the prominent factors teens report as central to their
attitudes toward regimen adherence. It is important to note that the
intercorrelations among the IMS subscale scores suggest that the typologies
elicited by the IMS do not represent vastly divergent subgroups of
nonadherers. Rather, use of this measure can elucidate the relative
importance of different domains of adherence barriers for individuals. Put
another way, the IMS offers data regarding which barriers to adherence are
primary influences on individuals' behaviors and which are in turn secondary
(Bauman, 2000).
The study findings have implications for research on adolescent nonadherence. In future studies, the IMS can be used to quantify the extent of barriers to adherence perceived by adolescents. It may therefore contribute to the assessment of adolescents' risk for nonadherence. Additionally, it can be used as a measure of intra-individual change in attitudes toward adherence over time, such as before and after participation in adherence-focused interventions. With further development to verify its factor structure, the IMS also can be used to identify subgroups of nonadherent teens to participate in targeted interventions designed to reduce specific barriers to adherence or to modify teens' perceptions of these barriers.
Clinically, the IMS offers health care providers an efficient method to determine the issues that adolescents perceive as the greatest barriers to their regimen adherence. This knowledge can aid providers in their attempts to reduce the hurdles adolescents encounter in living with a chronic illness. It may also enhance providers' abilities to engage parents and other sources of social support to work collaboratively toward the goal of increasing adolescents' adherence.
The implications of our findings should be considered within the limitations of the study. As previously noted, the sample was limited to asthma patients. Our next step in developing the measure is to validate its use with other illness groups. Study design relied upon a clinic-based sample. Teens who did not receive care in the specialty clinics or who failed to show for appointments were missed by this recruitment strategy, thus introducing some potential bias into the findings, especially given that clinic attendance is one indicator of treatment adherence. The ethnic characteristics of our sample (i.e., the low proportions of Latino and Asian American teens) limit generalizability of the findings and highlight the need for additional development in samples with larger representations of these minority groups.
The study did not include an objective measure of adherence such as electronic measurement of medication use, limiting our ability to form conclusions about the relationships between the IMS and actual adherence behaviors. However, for the purposes of this development study (where overall level of adherence was of less interest than adolescents' perceptions of barriers to adherence), we felt that obtaining self- and provider-report data was adequate. Future work should compare IMS response patterns with objective markers of adherence. The low response rate for test-retest reliability analysis is an additional limitation. Our test-retest results should be interpreted with caution, given that participants who returned their forms differed from the full sample in some known ways (ethnic composition and family structure) and may represent a group more adherent to their medical regimens.
It is important to underscore that the findings presented here represent
the development phase of the instrument. Further work is necessary to confirm
the factor structure that emerged in the current sample. Our sample was the
minimum size acceptable for a factor analysis given the number of items and
domains we examined (McCallum, Widaman,
Zhang, & Hong, 1999); replicating these findings within a
larger sample is an important future step. It would also be informative to use
the measure in a sample of identified nonadherent adolescents to determine
whether it functions in a similar manner within a more behaviorally
homogeneous sample.
In summary, the IMS represents a promising tool for identifying adolescents' perceptions of the barriers to their medical regimen adherence. It can serve as a brief screening device in clinical settings to help providers gain more comprehensive understanding of how adolescents manage their chronic illnesses. In the future we hope that the IMS will assist researchers and clinicians in designing interventions tailored to specific areas of adherence difficulties.
|
Acknowledgments |
|---|
We thank Nathan Blum, MD, Stacie Isenberg, PsyD, Alice Kahle, PhD, Anne Kazak, PhD, and Paul Robins, PhD, for assistance in measure development; Melissa Alderfer, PhD, for statistical consultation; R. Clayton, MD, J. Ginsberg, CRNP, J. Helm, CRNP, N. Pawlowski, MD, and M. Winston, CRNP, for facilitating data collection; and the patients and families of CHOP Allergy and Pulmonology clinics.
Received July 30, 2002; revision received October 14, 2002; accepted November 13, 2002
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Alexander, C. S., Kim, Y. J., Ensminger, M., Johnson, K. E., Smith, B. J., & Dolan, L. J. (1990). A measure of risk taking for young adolescents: Reliability and validity assessments. Journal of Youth and Adolescence, 19, 559-569.[CrossRef]
Anderson, B. J., Auslander, W. F., Jung, K. C., Miller, J. P.,
& Santiago, J. V. (1990). Assessing family sharing of
diabetes responsibilities. Journal of Pediatric
Psychology, 15,
477-492.
Bauman, L. J. (2000). A patient centered approach to adherence: Risks for nonadherence. In D. Drotar (Ed.), Promoting adherence to medical treatment in chronic childhood illness: Concepts, methods, and interventions (pp. 71-93). Mahwah, NJ: Lawrence Erlbaum Associates.
Bender, B., Milgrom, H., Rand, C., & Ackerson, L. (1998). Psychological factors associated with medication nonadherence in asthmatic children. Journal of Asthma, 35, 347-353.[ISI][Medline]
Bender, B. G., Milgrom, H., & Wamboldt, F. S. (2000). Measurement of treatment nonadherence in children with asthma. In D. Drotar (Ed.), Promoting adherence to medical treatment in chronic childhood illness: Concepts, methods, and interventions (pp. 153-171). Mahwah, NJ: Lawrence Erlbaum Associates.
Bond, G. G., Aiken, L. S., & Somerville, S. C. (1992). The health belief model and adolescents with insulin-dependent diabetes mellitus. Health Psychology, 11, 190-198.[CrossRef][ISI][Medline]
Brownbridge, G., & Fielding, D. M. (1994). Psychosocial adjustment and adherence to dialysis treatment regimes. Pediatric Nephrology, 8, 744-749.[CrossRef][ISI][Medline]
Bush, P. J., & Ianotti, R. J. (1990). A children's health belief model. Medical Care, 28, 69-86.[CrossRef][ISI][Medline]
Byron, M. (1998). Adherence to treatment in children. In: L. B. Myers & K. Midence (Eds.), Adherence to treatment in medical conditions (pp. 161-189). Amsterdam, Netherlands: Harwood Academic Publishers.
Creer, T. L., & Bender, B. G. (1995). Pediatric asthma. In M. C. Roberts (Ed.), Handbook of pediatric psychology (2nd ed., pp. 219-240). New York: Guilford Press.
DePaola, L. M., Roberts, M. C., Blaiss, M. S., Frick, P. J., & McNeal, R. E. (1997). Mothers' and children's perceptions of asthma medication. Children's Health Care, 20, 265-283.[CrossRef]
DeVellis, R. F. (1991). Scale development: Theory and applications. Newbury Park, CA: Sage.
Hanson, C. L. (1992). Developing systemic models of the adaptation of youths with diabetes. In A. M. La Greca, L. J. Siegel, J. L. Wallander, & C. E. Walker (Eds.), Stress and coping in child health (pp. 212-241). New York: Guilford Press.
Hartup, W. W. (1996). The company they keep: Friendships and their developmental significance. Child Development, 67, 1-13.[ISI][Medline]
Haynes, R. B. (1979). Introduction. In R. B. Haynes, D. W. Taylor, & D. C. Sackett (Eds.), Compliance in health care (pp. 1-7). Baltimore: Johns Hopkins University Press.
Glasgow, R. E., McCaul, K. D., & Schafer, L. C. (1986). Barriers to regimen adherence among persons with insulin-dependent diabetes. Journal of Behavioral Medicine, 9, 65-77.[CrossRef][ISI][Medline]
Gudas, L. J., Koocher, G. P., & Wypij, D. (1991). Perceptions of medical compliance in children and adolescents with cystic fibrosis. Developmental and Behavioral Pediatrics, 12, 236-242.
Kim, J. O., & Mueller, C. W. (1978). Factor analysis: Statistical methods and practical issues. Newbury Park, CA: Sage Publications.
Koocher, G. P., McGrath, M. L., & Gudas, L. J. (1990). Typologies of nonadherence in cystic fibrosis. Developmental and Behavioral Pediatrics, 11, 353-358.
Kyngäs, H. A. (1999). Compliance of adolescents with asthma. Nursing & Health Sciences, 1, 195-202.
Kyngäs, H. A., Kroll, T., & Duffy, M. E. (2000). Compliance in adolescents with chronic diseases: A review. Journal of Adolescent Health, 26, 379-388.[CrossRef][ISI][Medline]
La Greca, A. M. (1990). Issues in adherence with
pediatric regimens. Journal of Pediatric Psychology,
15,
423-436.
La Greca, A. M., Prinstein, M. J., & Fetter, M. D.
(2001). Adolescent peer crowd affiliation: Linkages with
health-risk behaviors and close friendships. Journal of Pediatric
Psychology, 26,
131-144.
La Greca, A. M., & Schuman, W. B. (1995). Adherence to prescribed treatment regimens. In M. C. Roberts (Ed.), Handbook of pediatric psychology (2nd ed., p. 55-83). New York: Guilford Press.
Lemanek, K. L., Kamps, J., & Chung, N. B. (2001).
Empirically supported treatments in pediatric psychology: Regimen adherence.
Journal of Pediatric Psychology,
26,
253-275.
Matsui, D. M. (2000). Children's adherence to medication treatment. In D. Drotar (Ed.), Promoting adherence to medical treatment in chronic childhood illness: Concepts, methods, and interventions (pp. 135-151). Mahwah, NJ: Lawrence Erlbaum Associates.
McCallum, R. C., Widaman, K. F., Zhang, S., & Hong, S. (1999). Sample size in factor analysis. Psychological Methods, 4, 84-99.
Milgrom, H., Bender, B., Ackerson, L., Bowry, P., Smith, B., & Rand, C. (1996). Noncompliance and treatment failure in children with asthma. Journal of Allergy and Clinical Immunology, 98, 1051-1057.[CrossRef][ISI][Medline]
Miller-Johnson, S., Emery, R. E., Marvin, R. S., Clarke, W., Lovinger, R., & Martin, M. (1994). Parent-child relationships and the management of insulin-dependent diabetes mellitus. Journal of Consulting and Clinical Psychology, 62, 603-610.[CrossRef][ISI][Medline]
Quittner, A. L, Espelage, D. L., Ievers-Landis, C., & Drotar, D. (2002). Measuring adherence to medical treatments in childhood chronic illness: Considering multiple methods and sources of information. Journal of Clinical Psychology in Medical Settings, 7, 41-54.
Rapoff, M. A. (1999). Adherence to pediatric medical regimens. New York: Kluwer Press.
Rapoff, M. A., & Barnard, M. U. (1991). Compliance with pediatric medical regimens. In J. A. Cramer & B. Spilker (Eds.), Patient compliance in medical practice and clinical trials. New York: Raven Press, 73-98.
Reynolds, W. M. (1982). Development of reliable and valid short forms of the Marlowe-Crowne Social Desirability Scale. Journal of Clinical Psychology, 38, 119-125.
Riekert, K. A., & Drotar, D. (2000). Adherence to medical treatment in pediatric chronic illness: Critical issues and answered questions. In D. Drotar (Ed.), Promoting Adherence to Medical Treatment in Chronic Childhood Illness: Concepts, Methods, and Interventions (pp. 3-32). Mahwah, NJ: Lawrence Erlbaum Associates.
Schlunt, D. G., Rea, M., Hodge, M., Flannery, M. E., Kline, S., Meek, J., et al. (1996). Assessing and overcoming situational obstacles to dietary adherence in adolescents with IDDM. Journal of Adolescent Health, 19, 282-288.[CrossRef][ISI][Medline]
Spector, S. (2000). Noncompliance with asthma therapy: Are there solutions? Journal of Asthma, 37, 381-388.[ISI][Medline]
Van Sciver, M. M., D'Angelo, E. J., Rappaport, L., & Woolf, A. D. (1995). Pediatric compliance and the roles of distinct treatment characteristics, treatment attitudes, and family stress: A preliminary report. Journal of Developmental and Behavioral Pediatrics, 16, 350-358.[ISI][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  K. A. Hommel, C. M. Davis, and R. N. Baldassano Medication Adherence and Quality of Life in Pediatric Inflammatory Bowel Disease J. Pediatr. Psychol., March 11, 2008; (2008) jsn022v1. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. L. Quittner, A. C. Modi, K. L. Lemanek, C. E. Ievers-Landis, and M. A. Rapoff Evidence-based Assessment of Adherence to Medical Treatments in Pediatric Psychology J. Pediatr. Psychol., September 10, 2007; (2007) jsm064v1. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. E. Simons and R. L. Blount Identifying Barriers to Medication Adherence in Adolescent Transplant Recipients J. Pediatr. Psychol., August 1, 2007; 32(7): 831 - 844. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. Kae Kintner Testing the Acceptance of Asthma Model With Children and Adolescents West J Nurs Res, June 1, 2007; 29(4): 410 - 431. [Abstract] [PDF]
|
|
|
|
|
|
A. C. Modi and A. L. Quittner Barriers to Treatment Adherence for Children with Cystic Fibrosis and Asthma: What Gets in the Way? J. Pediatr. Psychol., September 1, 2006; 31(8): 846 - 858. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P-A Michaud, J-C Suris, and R Viner The adolescent with a chronic condition. Part II: healthcare provision Arch. Dis. Child., October 1, 2004; 89(10): 943 - 949. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||