 |
|
 |
|
 |
 |
|
 |
 |
 |
 |
 |
 |
 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
PEDIATRICS Vol. 108 No. 2 August 2001, pp. 333-341
From the * Division of Adolescent Medicine, Children's Hospital
Medical Center, Cincinnati, Ohio; and Objective. Sexually active
adolescent girls have high rates of abnormal cervical cytology.
However, little is known about factors that influence intention to
return for Papanicolaou screening or follow-up. The aim of this study
was to determine whether a theory-based model that assessed knowledge,
attitudes, and behaviors predicted intention to return.
Methods. The study design consisted of a
self-administered, cross-sectional survey that assessed knowledge,
beliefs, perceived control over follow-up, perceived risk, cues for
Papanicolaou smears, impulsivity, risk behaviors, and past compliance
with Papanicolaou smear follow-up. Participants were recruited from a
hospital-based adolescent clinic that provides primary and subspecialty
care, and the study sample consisted of all sexually active girls and young women who were aged 12 to 24 years and had had previous Papanicolaou smears. The main outcome measure was intention to return
for Papanicolaou smear screening or follow-up.
Results. The enrollment rate was 92%
(N = 490), mean age was 18.2 years, 50% were
black, and 22% were Hispanic. Eighty-two percent of participants
intended to return. Variables that were independently associated with
intention to return included positive beliefs about follow-up (odds
ratio [OR]: 1.07; 95% confidence interval [CI]: 1.02-1.11),
perception that important others believe that the participant should
obtain a Papanicolaou smear (OR: 1.93; 95% CI: 1.38-2.74), perceived
control over returning (OR: 1.24; 95% CI: 1.06-1.46), and having
received cues to obtain a Papanicolaou smear (OR: 1.31; 95% CI:
1.08-1.60).
Conclusions. Analysis of this novel theoretical framework
demonstrated that knowledge and previous behaviors were not associated
with intention to return for Papanicolaou smear screening and follow-up
in this population of young women. However, modifiable attitudinal
components, including personal beliefs, perception of others' beliefs,
and cues to obtaining Papanicolaou smears, were associated with
intention to return.
Division of Adolescent/Young
Adult Medicine, Children's Hospital, Boston, Massachusetts.
ABSTRACT
Top
Abstract
Methods
Results
Discussion
References
Genital human papillomavirus (HPV) infection is the most
common sexually transmitted viral infection in the United States, with
a prevalence of approximately 50% in young sexually active women.1,2 These young women are at risk for abnormal
cervical cytology and cervical dysplasia,3-6 which may
progress to carcinoma in situ and invasive cervical cancer. National
data from the Centers for Disease Control and Prevention indicate that
of 31 569 women who were younger than 30 years and screened with
Papanicolaou smears between 1991 and 1993, 8.0% had atypical squamous
cells of undetermined significance (ASCUS), 9.4% had low-grade
squamous intraepithelial lesion (SIL), 2.1% had high-grade SIL, and
<0.1% had squamous cell cancer.7 Although HPV infection
frequently is transient in adolescents and the natural history of
cytologic abnormalities in adolescents is still being
elucidated,8 adolescents have a relatively high incidence
of abnormal cytology. A recent study demonstrated that the incidence of
SIL was higher in adolescents 10 to 19 years of age than in adult
women.9 In addition, it is likely that many adults with
high-grade lesions were infected as adolescents. Postulated mechanisms
for the high rates of abnormal cytology in young women include early
initiation of sexual intercourse, high number of sexual partners, high
incidence of sexually transmitted infections including HPV, high
incidence of smoking, and susceptibility of the adolescent cervix to
the acquisition of sexually transmitted infections and initiation of
carcinogenesis.10-14
The incidence and the mortality of cervical cancer have decreased
dramatically during the past 30 years as a result of a comprehensive national effort to screen women for cervical dysplasia using
Papanicolaou smears and improved treatment of carcinoma in situ and
early-stage cervical cancer. The 5-year survival rate of 96% for
localized cancer drops to 66% if the cancer has spread
regionally.15 Compliance with periodic Papanicolaou smear
screening, evaluation of abnormal Papanicolaou smears, and treatment of
precursor lesions correlate with decreased incidence and mortality of
cervical cancer.16,17 Despite this, the estimated rate of
noncompliance with follow-up appointments for abnormal Papanicolaou
smears ranges from 23% to 80%.1418-23 Noncompliance has been associated with younger age, lower educational level, unmarried status, Medicaid or no health insurance, better overall health, and lower grade cervical lesions.14,19,22
Despite the risk imposed by young age, little is known about the
cognitive, attitudinal, and behavioral factors that differentiate adolescent and young adult women who do and do not comply with recommendations for screening and follow-up. The factors that predict
adolescents' preventive health behaviors may differ from those of
adults because of developmental and psychological processes that are
specific to adolescence. Furthermore, although a large body of research
supports the utility of specific behavioral theories in predicting
health-related behaviors,24-27 complex behaviors (eg,
adherence to medical recommendations regarding Papanicolaou smear
screening) often cannot be explained by one theory. We therefore
developed a conceptual model that integrates several behavioral
theories to explain both intention to return and actual return for
Papanicolaou smear follow-up specifically in adolescent girls and young
women (Fig 1).28-35 The
significance of developing such models is that behavioral theory-based
interventions guided by these frameworks may be successful in
increasing intention to return and actual return for Papanicolaou smear
screening among adolescent girls and young women who are at risk for
cervical dysplasia.
View larger version (28K):
[in a new window]
Fig. 1.
Compliance model for intention to return and compliance with return
visit. The pathways indicated by solid lines represent the model tested
in this analysis.
The model that was tested in this study incorporates 4 theories that have been useful in predicting cancer prevention behaviors: the Theory of Planned Behavior,36 Health Belief Model,37 Social Cognitive Theory,38 and the Transtheoretical Model and Stages of Change.39 Each theory consists of a set of interrelated concepts, also termed constructs, that specify the relationships between specific variables to explain or predict health-related behaviors.40 For example, the Theory of Planned Behavior proposes that attitudes about performing a behavior, perceived beliefs of others regarding its performance, and perceived control over one's ability to perform it all affect intention to perform the behavior. Intention to perform the behavior, in turn, predicts actual behavior. A primary purpose of the Theory of Planned Behavior is to explain behavioral intention, and the success of the theory in explaining actual behavior is dependent on a number of factors, including the degree to which the behavior is under volitional control.40 In addition to the theories noted above, the conceptual model incorporates 2 other constructs hypothesized to predict follow-up intentions and behaviors: impulsivity and risk behaviors. The association between impulsivity and follow-up is supported by focus group and individual interview findings reported previously.34 The association between risk behaviors and follow-up is supported by the well-established clustering of behaviors that threaten health and well-being.41
This article summarizes the findings of the first phase of a longitudinal study designed to explore Papanicolaou smear follow-up by adolescent girls and young women. The overall goals of the study were to explore predictors of intention to return for Papanicolaou smear screening and follow-up, as well as the relationship between intention to return and actual return. Intention to return was chosen as the outcome measure in these analyses for the following reasons. First, behavioral intention is one of the most robust and consistent predictors of health-related behaviors. A meta-analysis that examined the utility of the Theory of Planned Behavior in predicting health-related behaviors demonstrated that the model explained 34% of the variance in behavior on average and that 66% of the explained variance in behavior was attributed to intention.24 Furthermore, the literature on the utility of these specific behavioral theories, particularly the Theory of Planned Behavior, in predicting adolescent behaviors is scarce. Little is known about the factors that have a direct impact on an adolescent's plan to perform a behavior. Finally, interventions to change behavior may be directed toward the immediate predictors of intention to perform a behavior, as often occurs in clinical settings, as well as toward the pathway between intention and behavior. This article therefore examines in detail the associations between cognitive, attitudinal, and behavioral factors and intention to return for Papanicolaou smear follow-up. The constructs on the left in Fig 1 are hypothesized to influence compliance with a return visit indirectly, through intention to return. The pathways indicated by solid lines represent the components of the model that were tested in this analysis. The specific hypotheses tested in this phase of the study are that the following variables are associated with intention to return: 1) knowledge about Papanicolaou smears and HPV, 2) positive personal beliefs and perceived beliefs of others about Papanicolaou smear follow-up, 3) high perceived control over follow-up, 4) health beliefs, including high perceived risk of abnormal Papanicolaou smears and cervical cancer and cues to obtaining a Papanicolaou smear, 5) communication with the provider regarding Papanicolaou smears, 6) low impulsivity, and 7) previous compliance with Papanicolaou smear follow-up.
| |
METHODS |
|---|
|
Top
Abstract
Methods
Results
Discussion
References
|
|---|
Participants
The target population consisted of a consecutive sample of all 558 females who were aged 12 to 24 years and seen in an urban, hospital-based adolescent clinic between October 1998 and June 1999 and who had a history of sexual intercourse and previous Papanicolaou smear(s) done at the hospital. Patients who were unable to complete a written questionnaire independently (N = 24) were ineligible. Of the 534 eligible patients, 44 declined participation because of competing time commitments or lack of interest. The study sample therefore consisted of 490 participants (enrollment rate: 92%). All participants provided informed consent, and the protocol was approved by the hospital Committee on Clinical Investigation. The data presented were obtained from a survey instrument administered to all participants on enrollment in the study.
Clinic Procedures and Data Collection
Although all participants attended the adolescent clinic, an estimated 10% of the participants had had 1 or more previous Papanicolaou smears done in the adolescent gynecology clinic or teen-tot clinic. All clinics are located in the same hospital, all samples are sent to the same laboratory, and results are available on the same computerized cytology result information system. Procedures to inform clinic patients of abnormal results and standards for screening and follow-up are as follows. Clinic patients are asked to return once a year for screening Papanicolaou smears after they become sexually active or at approximately age 18. Since 1993, all patients with abnormal Papanicolaou smear reports on specimens collected in these 3 clinics have been informed of the abnormality by letter. Patients with a first Papanicolaou smear demonstrating ASCUS were instructed to have a repeat Papanicolaou smear in 4 months. Patients with a second Papanicolaou demonstrating ASCUS or any Papanicolaou demonstrating low- or high-grade SIL were instructed to make 2 appointments, 1 to discuss the result and another for colposcopy. Patients with high-grade SIL also received a telephone call from 1 of 2 nurses in the adolescent clinic. Patients who missed their colposcopy appointments were sent a letter advising them to reschedule and were called by a nurse to help them reschedule. Patients who missed 2 colposcopy appointments were sent a registered letter and again called by a nurse.
The self-administered pencil-and-paper survey instrument was administered during the clinic visit and consisted of 116 items that assessed sociodemographic and health care characteristics as well as intention to return for follow-up, knowledge about Papanicolaou smears and HPV, attitudes about Papanicolaou smear screening and follow-up, impulsivity, and risk behaviors. Past compliance with Papanicolaou smear follow-up visits was determined with the use of computerized medical and laboratory reports. Sociodemographic characteristics, intention to return, and knowledge were assessed in part 1 of the survey. After part 1 was collected, participants completed part 2 of the survey, which assessed attitudes and beliefs about Papanicolaou smear screening and follow-up, impulsivity, and risk behaviors. Part 2 was preceded by a paragraph that defined a Papanicolaou smear and gave routine recommendations for screening and follow-up. The purpose of the introductory paragraph was to ensure that participants would have baseline knowledge on which to base their responses to questions about attitudes and beliefs; it was included because preliminary interviews and pilot tests of the survey indicated that knowledge about Papanicolaou smears and HPV was poor.
Intention to return was assessed with the following question: "How sure or unsure are you that you will return for your next Pap smear?" Responses, on a 5-point Likert scale, were "you are very sure," "you are somewhat sure," "you are neither sure nor unsure," "you are somewhat unsure," and "you are very unsure."
Knowledge about Papanicolaou smears was measured with the use of 5 items with a possible score range of 0 (no correct items) to 5 (all correct items). Knowledge about HPV was measured with the use of a 6-item scale with a possible score range of 0 (the participant had never heard of HPV) to 6 (the patient had heard of HPV and answered all other items correctly). See Table 1 for psychometric characteristics of scales.
|
The items that assessed attitudes were divided into the following 5 domains: behavioral beliefs, normative beliefs, perceived behavioral control, health beliefs, and communication with the provider, all regarding Papanicolaou smears. Behavioral beliefs about Papanicolaou smear follow-up specific to this population were elicited through individual interviews and focus groups. Responses were measured with the use of a 5-point Likert scale ranging from "strongly agree" to "strongly disagree," and responses were summed to create a scale score. Normative beliefs are defined by Ajzen36 as an individual's perception of important others' beliefs about outcomes attributed to a behavior. Normative beliefs regarding the participant's mother, friends, best friend, boyfriend, and doctor were elicited by asking, "How do you think the following people feel about your getting a Pap smear?" Responses were on a 5-point Likert scale ranging from "you definitely should get a Pap smear" to "you definitely should not get a Pap smear." Responses were summed to create a scale score. Perceived behavioral control (one's perceived ability to perform a behavior) was measured with the use of 2 items. Responses to, "How easy or difficult would it be for you to come in for your next Pap smear?" were measured on a 5-point Likert scale ranging from "extremely easy" to "extremely difficult." Responses to, "I have total control over whether I come in for my next Pap smear," were measured on a 5-point Likert scale ranging from "strongly agree" to "strongly disagree." Responses to the 2 questions were summed for analysis.
Health beliefs measured included perceived risk and cues to action. Perceived benefits and barriers were included in the assessment of behavioral beliefs. Perceived susceptibility was measured with the use of the following 3 items: "How likely is it that you would have an abnormal Pap smear?" "If you had a Pap smear which was abnormal but left untreated, how likely is it that it would become cancer?" and, "How likely is it that you will get cancer of the cervix sometime in your life?" Responses were on a 5-point Likert scale ranging from "extremely likely" to "extremely unlikely." Perceived severity was measured with the use of the following 3 items (responses to each item were in the form of 5-point Likert scales with the extremes indicated in parentheses): "How serious a health problem would an abnormal Pap smear be for you?" ("the most serious" to "not at all serious"), "If you get cervical cancer in the future, it's not a big problem because it is easy to treat and to cure" ("strongly agree" to "strongly disagree"), and, "If you get cervical cancer in the future, how likely is it that you would die from it?" ("extremely likely" to "extremely unlikely"). Cues to action were measured with the use of 6 yes/no items that assessed whether the participant had heard that she should have a Papanicolaou smear from specific sources: a doctor or a nurse, her mother, a friend, the clinic, health/sexuality education classes at school, or the media/Internet. Cues were summed to create an index score.
Communication with the provider was assessed with the use of a 7-item scale, derived from individual interviews and focus groups conducted previously with the same clinic population34 as well as the work of Freed et al42 and Ginsburg et al.43 The scale was designed to assess key elements of the 3 components of doctor-patient communication44: mutual exchange of information, medical decision making, and creation of an interpersonal relationship. Beliefs about communication with the provider were measured by asking the participant how likely it is on a 5-point Likert scale (ranging from "will definitely not happen" to "will definitely happen") that the doctor will display a particular communicative behavior, such as explaining the participant's condition to her. Scoring was performed by summing the results.
Behavioral variables included measures of impulsivity, risk behaviors and associated health outcomes linked to the development of abnormal Papanicolaou smears or cervical cancer (age of first sexual intercourse, number of lifetime sexual partners, contraception use at last sexual intercourse, condom use at last sexual intercourse, smoking in the past 30 days, history of pregnancy, history of any sexually transmitted infection, and history of an abnormal Papanicolaou smear) and past compliance with Papanicolaou smear follow-up.
Impulsivity was assessed with the use of a scale derived from the work of Eysenck and Eysenck.45,46 The subscale that we used measured "narrow impulsiveness," a construct specific for impulsive traits. The wording of the original items was changed slightly to reflect American language (eg, "queue" was changed to "line"). Versions of this subscale have been tested and normed in both adults and adolescents.47 The scale consisted of 13 yes/no items and was scored by summing the responses that indicated higher impulsivity.
Past compliance with Papanicolaou screening and follow-up was assessed with the use of computerized medical and cytology records. Past compliance, which was derived from practices standard in the clinic, was defined as having kept an appointment for 1) routine Papanicolaou smear within 15 months of the previous test, 2) follow-up Papanicolaou smear within 6 months of a smear that demonstrated ASCUS, 3) colposcopy within 2 months of a second smear that demonstrated ASCUS, and 4) colposcopy within 2 months of a smear that demonstrated low- or high-grade SIL or carcinoma.
Statistical Methods
Intention to return and the knowledge scales were analyzed as dichotomous variables. For the item that assessed intention to return, responses of "very sure" or "somewhat sure" were categorized as "intends to return." Responses of "neither sure nor unsure," "somewhat unsure," and "very unsure" were categorized as "does not intend to return." The item was dichotomized in this way because responses to this item were highly skewed (the great majority of respondents reported that they were sure that they would return) and for theoretical reasons: responses to Likert-type questions frequently are dichotomized for analysis into neutral through negative responses and positive responses. Analyses also were performed with the use of the intention-to-return variable as a continuous variable and dichotomized as "very sure" versus all other responses, and results were similar in terms of significant findings. Scales that assessed knowledge about Papanicolaou smears and HPV also were dichotomized into 0 (none of the items was answered correctly) and greater than or equal to 1, because the scales were highly skewed and the majority of participants received a score of 0. Individual items that composed the scales that measured behavioral beliefs, normative beliefs, and perceived behavioral control were analyzed as ordinal variables. Individual health belief items that assessed risk were dichotomized to represent high versus low susceptibility and severity (eg, those who reported that they were "extremely likely" or "likely" to develop an abnormal Papanicolaou smear were categorized as "high susceptibility"). Summary scales that measured beliefs about Papanicolaou smear follow-up, normative beliefs, perceived behavioral control, communication with the provider, impulsivity, and cues to action were analyzed as continuous variables.
Associations between dichotomized or categorical variables and
intention to return were determined with the use of
2 tests. Associations between ordinal
variables or non-normally distributed continuous variables and
intention to return were determined with the use of Wilcoxon rank-sum
tests. Associations between continuous scale scores and intention to
return were analyzed with the use of Student's t tests if
the scale scores were normally distributed and Wilcoxon rank-sum tests
if the scale scores were not normally distributed. Variables that were
associated with intention to return at P < .05 were
entered into a stepwise logistic regression procedure to identify
variables that were independently associated with intention to return.
Adjusted odds ratios and 95% confidence intervals are reported for the
model, both unadjusted and adjusted for age, race, and insurance
status.
| |
RESULTS |
|---|
|
Top
Abstract
Methods
Results
Discussion
References
|
|---|
Baseline characteristics of the study sample are presented in Table 2. Approximately 50% of the participants were younger than 18 years, 50% were black, 22% were Hispanic, and 52% had Medicaid insurance. Twenty-seven percent had a history of at least 1 abnormal Papanicolaou smear, and 39% of those who were expected to return for follow-up had not complied with previous appointments for follow-up Papanicolaou smears or colposcopy. Eighty-two percent intended to return for Papanicolaou follow-up. Papanicolaou smear and HPV knowledge scores were low: 68% scored 0 on the Papanicolaou smear knowledge subscale, and 75% scored 0 on the HPV knowledge subscale.
|
Tables 35 demonstrate univariate associations between predictor variables and intention to return. Demographic and health care characteristics associated significantly with intention to return were age >18 years, white race, and private insurance. Knowledge about Papanicolaou smears and HPV was not associated with intention to return (Table 3). In contrast, most individual items that composed scales and indices that measured attitudes and beliefs about Papanicolaou smear follow-up were associated significantly with intention to return (data not shown but are available from the first author). Twelve of 14 behavioral beliefs about Papanicolaou smear follow-up were associated with intention to return. Behavioral beliefs that were positively associated with intention to return included the following: returning would enable her to find out that something is wrong that she cannot see, give her peace of mind, help her take control of her health, and help protect her health. Behavioral beliefs that were negatively associated with intention to return included the following: the procedure would be painful or embarrassing, she doesn't want to look for trouble, she is afraid her parents might find out, she doesn't have a consistent provider, her provider doesn't communicate, she doesn't have time, and she can't get transportation. All normative beliefs (perceived beliefs of mother, friends, best friend, boyfriend, and doctor) and both items that assessed perceived control over returning (ease of returning and control over returning) were associated significantly with intention to return. Two of 6 health beliefs regarding risk (perceived susceptibility to the development of cervical cancer and perceived severity of cervical cancer in terms of ease of cure), 4 of 6 cues to action (doctor, mother, letter or telephone call from clinic, and school health/sexuality classes), and all 7 beliefs about communication with the provider were associated significantly with intention to return.
|
|
All scales or indices (Table 4) that measured attitudes about Papanicolaou smear follow-up were associated with intention to return, including behavioral beliefs about Papanicolaou smear follow-up, normative beliefs, perceived behavioral control, cues to action, and communication with the provider. In addition, 2 dichotomized attitudinal variables that assessed perceived susceptibility and perceived severity were associated with intention to return. Those who perceived themselves to be highly susceptible to an abnormal Papanicolaou smear progressing to cervical cancer were more likely to intend to return (85%) than those who did not (75%; P = .008). Those who perceived cervical cancer to be severe (unlikely to be able to be treated or cured) were more likely to intend to return (85%) than those who did not (77%; P = .043). Impulsivity was associated negatively with intention to return (Table 4), and history of an abnormal Papanicolaou smear was associated positively with intention to return (Table 5). However, risk behaviors, related health outcomes other than an abnormal Papanicolaou smear, and past compliance with Papanicolaou smear follow-up were not significantly associated with intention to return.
|
The independent variables that composed the final logistic regression model included beliefs about Papanicolaou smear follow-up, normative beliefs, perceived behavioral control, susceptibility to an abnormal Papanicolaou smear progressing to cancer, severity of cervical cancer in terms of effectiveness of treatment or cure, cues to action, communication with provider, impulsivity, and history of an abnormal Papanicolaou smear. Logistic regression analysis (Table 6) demonstrated that the variables that were independently associated with intention to return included scales that measured beliefs about Papanicolaou smear follow-up, normative beliefs, perceived behavioral control, and cues to action. Each increase of 1 point on the scale that measured beliefs about Papanicolaou smear follow-up was associated with a 7% increase in the odds that a participant would intend to return for follow-up. Each increase of 1 point on the scale of normative beliefs was associated with a 93% increase in the odds that a participant would intend to return. Each increase of 1 point on the scale of perceived control over returning was associated with a 24% increase in the odds that a participant would intend to return. As an example, for individuals who differed by 1 point on the scale of perceived control over returning, the odds ratio was 1.24. For those who differed by 2 points on the scale, the odds ratio increased to 1.53. Each additional cue to obtain a Papanicolaou smear increased by 31% the odds that a participant would intend to return. Controlling for age, race, and insurance status did not affect which variables were significantly associated with intention to return.
|
| |
DISCUSSION |
|---|
|
Top
Abstract
Methods
Results
Discussion
References
|
|---|
This article describes the development and initial testing of a new conceptual model designed to explain cervical cancer screening and follow-up in adolescent girls and young women. A critical issue in the study of adolescent behaviors is whether theories of adult behavior can help predict or explain adolescent behavior. Our findings suggest that constructs tested primarily in adult populations also are useful in understanding adolescent intention to return for Papanicolaou smear follow-up. Although knowledge, risk behaviors, and past compliance with appointments for Papanicolaou smear screening and follow-up were not associated with intention to return as hypothesized, all attitudinal components of the model as well as lower impulsivity were associated with intention to return in univariate analyses. Positive beliefs about Papanicolaou smear follow-up, strong normative beliefs regarding follow-up, high perceived behavioral control for follow-up, and cues to action for obtaining a Papanicolaou smear were associated independently with intention to return for Papanicolaou smear follow-up. These data therefore extend the findings of previous investigations that examined the utility of specific behavioral theories28-32,48 in predicting cervical cancer screening intentions and behaviors in adult women. First, the findings demonstrate that a conceptual model composed of several behavioral theories predicts intention to return for Papanicolaou smear follow-up; second, they suggest that constructs derived from these behavioral theories are generalizable to an adolescent population.
Central implications of these findings are that interventions guided by behavioral theories may promote intended or actual return for Papanicolaou smear follow-up by adolescent girls and young women who are at risk for cervical dysplasia. Our findings suggest that patient attitudes about screening and follow-up, rather than knowledge or past behavior, are associated with intention to return. Although attitudes may be more difficult to change than knowledge, theoretically they are modifiable. Components of interventions designed to increase intention to return for Papanicolaou smear follow-up might include enhancing positive attitudes about follow-up, such as protecting one's health or discovering an asymptomatic problem, emphasizing that providers and parents believe that Papanicolaou screening and follow-up is important, increasing young women's sense of control over returning, and instituting clinic reminders or providing information on Papanicolaou smear screening and follow-up in school health and sexuality classes. Strategies to educate providers about appropriate recommendations for screening and follow-up and provide guidance as to how they might improve their communication with young women also may be valuable. Effective interventions may need to use multiple but consistent messages delivered in a variety of ways to target the model components identified in this study as important for intention to return for Papanicolaou smear screening and follow-up.
An important consideration in the utilization of these findings for interventions is that the theoretical constructs that predicted intention to return may not explain actual return in a subset of adolescents49; some adolescents do not keep appointments despite positive intentions.50 Although >80% of the participants in our study intended to return, actual return rates are likely to be substantially lower23; thus, sociodemographic or attitudinal factors may mediate or moderate the pathway between intention to return and actual return. Sheeran and Orbell48,49 have begun to examine the relationship between intention and behavior and to propose interventions that may enhance compliance in women with high intentions to return for Papanicolaou smear screening. In addition, the pathways may change in a full model including data on actual return. Whether adolescent attitudes have an impact on actual return indirectly, through intention, or directly will have implications for interventions. Future studies therefore are needed to address the following questions, to design evidence-based, effective interventions to enhance follow-up. To what extent does intention to return predict actual return in adolescent girls and young women, and what factors influence the pathway between intention and behavior? Do interventions designed to change attitudes enhance intention to return and actual return rates? In adult populations, interventions aimed at changing the attitudinal predictors of intention to return (involving reminder letters, telephone counseling, educational programs, educational brochures, transportation incentives, and tracking systems) have enhanced actual compliance for screening and follow-up Papanicolaou smears, but the effectiveness of these interventions in adolescents remains to be tested.51-56
There are several limitations to this study. First, the model was tested in an urban, hospital-based sample with a relatively high proportion of black and Hispanic young women and may perform differently in other settings and populations. Second, it was necessary to create several new measures for this survey because there were no available instruments specific to knowledge and attitudes about Papanicolaou smears in adolescents. Third, the scales that measured knowledge, susceptibility, and severity were skewed and therefore were dichotomized for analysis, which may have limited our ability to detect significant associations between the variables and intention to return. Finally, previous compliance with Papanicolaou smear follow-up recommendations did not predict intention to return in the future. There are a number of potential explanations for this finding, including the possibility that participants answered the item that assessed intention to return in what they perceived to be a socially desirable way. It also may reflect an increasing sense of responsibility for one's health that these participants are acquiring as they become older.
Despite these limitations, our data provide a starting point for future theoretically grounded observational and interventional studies of young women's intentions and behaviors regarding Papanicolaou smear screening and follow-up visits. Ensuring compliance with cervical cancer prevention recommendations in young women is a complex issue and ultimately is likely to require comprehensive interventions that not only focus on changing adolescent attitudes and intentions but also involve education of health care providers and the support of public health agencies.57
| |
ACKNOWLEDGMENTS |
|---|
This study was supported by the Deborah Munroe Noonan Memorial Fund and by Project Nos. 6 T71 MC 00009-09 and 5 T71 MC 00001-24 from the Maternal Child Health Bureau, Health Resources and Services Administration, US Department of Health and Human Resources.
We thank Victoria Chiou, BA, and Rebekah Kaplowitz, MD, for their assistance with data collection and interpretation; Lawren Daltroy, PhD, and Karen Emmons, PhD, for their thoughtful contributions to survey development and analysis; and Jonathan Ellen, MD, MPH, and Lorraine Freed, MD, MPH, for their major contribution to the development of the communication scale.
| |
FOOTNOTES |
|---|
Received for publication Sep 27, 2000; accepted Dec 21, 2000.
Address correspondence to Jessica A. Kahn, MD, MPH, Division of Adolescent Medicine, Children's Hospital Medical Center, 3333 Burnet Ave, Cincinnati, OH 45229. E-mail: kahnj1{at}chmcc.org
| |
ABBREVIATIONS |
|---|
HPV, human papillomavirus; ASCUS, atypical squamous cells of undetermined significance; SIL, squamous intraepithelial lesion.
| |
REFERENCES |
|---|
|
Top
Abstract
Methods
Results
Discussion
References
|
|---|
behavior changes: the short and
the long of it.
Prev Med
1993;
22:676-688 [CrossRef][Medline]This article has been cited by other articles:
 |
|
 |
|
  K. A. Robb, A. Miles, and J. Wardle Perceived Risk of Colorectal Cancer: Sources of Risk Judgments Cancer Epidemiol. Biomarkers Prev., April 1, 2007; 16(4): 694 - 702. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. D. Wright, J. S. Rader, R. Davila, M. A. Powell, D. G. Mutch, F. Gao, and R. K. Gibb Human papillomavirus triage for young women with atypical squamous cells of undetermined significance. Obstet. Gynecol., April 1, 2006; 107(4): 822 - 829. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. D. Wright, R. M. Davila, K. R. Pinto, D. F. Merritt, R. K. Gibb, J. S. Rader, D. G. Mutch, F. Gao, and M. A. Powell Cervical Dysplasia in Adolescents Obstet. Gynecol., July 1, 2005; 106(1): 115 - 120. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. P. Tercyak, M. Nicolas, T. Councill, S. Prahlad, K. L. Taylor, and A. T. Shad Brief Report: Health Beliefs Among Survivors of Childhood Cancer J. Pediatr. Psychol., July 1, 2004; 29(5): 397 - 402. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. A. Kahn, E. Goodman, B. Huang, G. B. Slap, and S. J. Emans Predictors of Papanicolaou Smear Return in a Hospital-Based Adolescent and Young Adult Clinic Obstet. Gynecol., March 1, 2003; 101(3): 490 - 499. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Research charities try pharma industry trick. Ann. Onc., February 1, 2002; 13(2): 179 - 180. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
||
|
||
Home | My Pediatrics | Journal Information | Current Issue | Past Issues | Subscriptions & Services | Contact Us Click here for faster international access © 2001 American Academy of Pediatrics. All rights reserved. |
||
|
||
