ERIC EJ792812: Reducing Parental Demand for Antibiotics by Promoting Communication Skills

2005·0.13 MB·English

by  ERIC

Reducing Parental Demand for Antibiotics
by Promoting Communication Skills
Stephen C. Alder, Eric P. Trunnell, George L. White, Jr., Joseph L. Lyon,
James P. Reading, Matthew H. Samore and Michael K. Magill
ABSTRACT
Antibiotic-resistant strains of bacteria are continuing to emerge as high rates of antibiotic use persist. Children are
among the highest users of antibiotics, with parents influencing physician decision-making regarding antibiotic
prescription. An intervention based on Social Cognitive Theory (SCT) to reduce parents’ expectations for antibiot-
ics in favor of communication with the physician was tested. A randomized factorial design study was conducted
testing a communication skills intervention against an information-based intervention, a combined intervention
and a control condition. Parents receiving the communication skills intervention reported higher efficacy to com-
municate with the physician (p = 0.021). Interaction between the communication skills and information interven-
tions was observed for specific treatment expectations prior to the visit (p = 0.049). The communication skills
intervention was protective against antibiotic prescribing (p = 0.042). Satisfaction was positively associated with
parents’ efficacy to communicate (p = 0.002) and when an antibiotic was not prescribed (p = 0.005).
Interventions using information to reduce antibiotic use may be based on the false assumption that such ap-
proaches are sufficient for motivating appropriate health practices. Using demonstrated behavior change approaches
to promote communication behaviors by parents might result in the dual benefits of reducing unwarranted antibi-
otic use and improving satisfaction with the clinic visit.
INTRODUCTION
The link between emerging antibiotic Stephen C. Alder, PhD, is with the Department Department of Family and Preventive Medi-
resistance and the overuse of antibiotics is of Family and Preventive Medicine, University cine, 375 Chipeta Way, Suite A, Salt Lake City,
well documented.1 High rates of antibiotic of Utah, 375 Chipeta Way, Suite A, Salt Lake UT 84108. James P. Reading, PhD, is with the
use persist, with antibiotic resistance con- City, Utah 84108; E-mail: salder@dfpm. Department of Family and Preventive Medi-
tinuing to emerge.2 Overprescribing of an- utah.edu. Eric P. Trunnell, PhD, is with Health cine, 375 Chipeta Way, Suite A, Salt Lake City,
tibiotics occurs most frequently in chil- Promotion and Education, HPER North, 250 UT 84108. Matthew H. Samore, MD, is with
dren.3 Patient expectations regarding S 1850 East Room 200, Salt Lake City, UT the Division of Clinical Epidemiology, School
antibiotics have been associated with in- 84108. George L. White, Jr., PhD, is with the of Medicine, 30 N 1900 E RM AC226. Michael
creased likelihood of antibiotics being pre- Department of Family and Preventive Medi- K. Magill, MD, is with the Department of Fam-
scribed.4,5 Further, when children are the cine, 375 Chipeta Way, Suite A, Salt Lake City, ily and Preventive Medicine, 375 Chipeta Way,
patients, the parents’ expectations influence UT 84108. Joseph L. Lyon, MD, is with the Suite A, Salt Lake City, UT 84108.
132 American Journal of Health Education — May/June 2005, Volume 36, No. 3
Stephen C. Alder, Eric P. Trunnell, George L. White, Jr., Joseph L. Lyon, James P. Reading, Matthew H. Samore and Michael K. Magill
the children’s treatment. When physicians their behaviors may be viewed as a func- to change parents’ expectations for antibi-
perceive a parental expectation that an tion of the subjective probability that their otics, linked with antibiotic-seeking behav-
antibiotic is needed for their child with a actions will successfully lead to the desired ior, to expectations for information that will
cough, they are more likely to prescribe outcome of resolution of the child’s illness help them best address their children’s ill-
an antibiotic with an associated diagnosis and the subjective value they place on the nesses, linked with effective communication
of bronchitis.6 importance of this outcome.14 Further, as practices. We tested two main hypotheses:
Parents often lack understanding regard- indicated by Self-efficacy, a parent’s belief— that participation in the intervention will
ing the appropriate use of antimicrobial or efficacy—in his or her ability to perform 1) improve parents’ self-reported ability to
therapy for upper respiratory infections.7 a behavior is predictive of engaging and per- communicate with their child’s medical
Interventions ameliorating parents’ expec- severing in that behavior. provider and 2) lead to a reduction in par-
tations for antibiotics may lead to a reduc- Using these behavioral principles to ents’ expectations for a specific treatment
tion in unwarranted antibiotic use. General guide interventions to promote parental for their child resulting from the clinic visit.
recommendations for reducing parental communication as the means of acquiring Differences regarding antibiotic use and
demand for antibiotics include educating the needed medical care for achieving ill- receipt of an antibiotic prescription by
parents regarding the appropriate use of ness resolution for an ill child can reduce study group are explored. A secondary fo-
antibiotics and providing them with parental expectation for antibiotics.12,15,16,17 cus of this study regards parents’ satisfac-
guidelines to more successfully communi- Satisfaction with the clinic visit has been tion with the clinic visit. To address this fo-
cate with their children’s physicians.8 Suc- found to be associated with parents’ ability cus we also tested the hypotheses that
cessful communication includes helping to communicate with their child’s physician improvement in parents’ belief in their abil-
parents feel that they can successfully man- rather than whether an antibiotic is pre- ity to communicate with their children’s
age their child’s symptoms (i.e., fever, scribed.18 Enhancing communication be- medical providers will lead to improved sat-
cough, pain, etc.). tween parents and physicians may jointly isfaction with the clinic visit and that re-
Health behavior-change models have reduce unwarranted antibiotic prescribing ceipt of an antibiotic is not associated with
had limited use in the provision of primary and increase satisfaction with the clinic visit. higher satisfaction.
care.9 Typical efforts for addressing health The notion of promoting patient com-
METHODS
behavior change in primary care settings munication is not new. Past studies have
have been limited to approaches that are looked specifically at promoting patient Subjects
based on providing information. The un- question asking,19 expanding patients’ in- Two suburban primary care clinics with
derlying assumption behind these informa- volvement in their medical care,20 and the multiple providers in the Salt Lake City,
tion-based approaches is that the recipient link between physician-patient interactions Utah metropolitan area served as study sites.
will fully understand the information pre- and chronic disease outcomes.21 Results of A trained researcher was present at the clin-
sented, synthesize it and appropriately these studies have been generally favorable ics during a series of scheduled time blocks
relate it to his or her situation and, finally, in regards to health outcomes and patients’ from August to December 2000. During
align their behaviors. Instead, interventions assessment of the medical visit, yet no stud- these times, clinic intake staff invited par-
in primary care settings need to go beyond ies have specifically addressed the issue of ents of eligible patients to participate in the
the typical information-based approach by antibiotic treatment expectations. study. Parents were invited to participate if
utilizing proven health behavior-change We conducted a study to test a parent- their child was aged 1 to 10, presenting with
theories to direct health promotional efforts.10 focused intervention based on Expectancy- complaints of ear pain, sore throat, cough,
Social Cognitive Theory (SCT),11 includ- valence and Self-efficacy theories. The main congestion and/or fever, and had not re-
ing Expectancy-valence12 and Self-efficacy,13 intent of the tested intervention was to pro- ceived antibiotic therapy during the previ-
provides a useful theoretical framework for mote more effective interaction between ous two weeks.
addressing parental expectations of antibi- parents and physicians as a means of reduc- Eighty parents were enrolled and ran-
otics. In detail, components of Expectancy- ing unwarranted antibiotic prescribing. domized to one of four intervention con-
valence emphasize the role of outcome ex- Specifically, the communication interven- ditions, including 1) the intervention to
pectancies in changing behaviors. tion is designed to 1) promote communi- promote parental ability to communicate
Predicting that the likelihood of an indi- cation by the parent with the physician as with the medical provider (or communica-
vidual engaging in a specific behavior is the way to determine the best means of re- tion intervention), 2) an information-based
based on both the expectation that the solving the child’s illness and 2) promote antibiotic education intervention (or infor-
outcome will be achieved and the impor- the parent’s efficacy to effectively engage in mation-only intervention), 3) a combination
tance, or valence, placed on the outcome. a communicative dialogue with his or her of the communication promotion and an-
For parents seeking care for an ill child, child’s physician. Thus, the ultimate goal is tibiotic education interventions (or com-
American Journal of Health Education — May/June 2005, Volume 36, No. 3 133
Stephen C. Alder, Eric P. Trunnell, George L. White, Jr., Joseph L. Lyon, James P. Reading, Matthew H. Samore and Michael K. Magill
ceived educational information about an-
Figure 1. Talk to Your Doctor About Your Child’s Illness
tibiotics, antibiotic resistance, and the link
between misusing antibiotics and the
Questions recommended for parents when meeting with child’s physician:
emergence of antibiotic resistance. Included
in this intervention were the pamphlet
1. What is causing my child’s illness?
Antibiotics and Your Child, published by
2. When should I expect my child to feel better?
the Centers for Disease Control and Preven-
3. When should I call your office back if my child is not feeling better?
tion (available at: http://www.cdc.gov/
4. What can I do to help my child feel better at home?
drugresistance/ community//tools.htm),22 and
a fact sheet about antibiotics use and anti-
bined intervention) and 4) a control condi- on the role of the physician. The researcher
biotic resistance adapted from information
tion based on nutrition for children then modeled asking the recommended
obtained from the Use Antibiotics Wisely
(or control). questions to provide a vicarious experience
campaign. These materials include educa-
Interventions Used with Patients for the parent. Roles were then switched, tional information for the following topics:
with the researcher playing the part of the
Communication Intervention bacteria and viruses, resistant bacteria, how
physician and the parent taking on his or
Parents who were enrolled in the com- bacteria become resistant, when antibiot-
her own role creating an enactive mastery
munication intervention were asked to re- ics are and are not needed, and following
view four questions to be answered during experience. Following this interaction, the recommended instructions if an antibiotic
the clinic visit (Figure 1). These questions researcher provided specific feedback to the is prescribed.
were adapted from the Talk to Your Doctor parent regarding positive and negative as- The researcher introduced the topic of
approach used in the Use Antibiotics Wisely pects of the interaction as a means of ver- antibiotic resistance and explained to the
campaign conducted by the Maryland De- bal persuasion. parent the increasing seriousness of the
partment of Health and Mental Hygiene, Parents were then asked to describe phy- problem. The pamphlet and antibiotic
Johns Hopkins University School of Hy- sician characteristics that they perceived fact sheet were reviewed in detail. Empha-
giene and Public Health, and the Centers may make it difficult for them to ask the sis was place on the judicious use of antibi-
for Disease Control and Prevention (CDC). recommended questions. Top responses otics, with parents encouraged to apply this
These questions were designed to enable included physicians who were hurried information to their child’s current illness.
and those that did all the talking without
parents to obtain information about their Combined Intervention.
child’s illness during the clinic visit. Based allowing the parent to influence the inter- To assess potential interaction between
on Expectancy-valence, the researcher action. Two to three additional role-play- the communication and information-only
stated that the purpose for a clinic visit with ing exercises were conducted with the interventions, a combined intervention was
the physician was to obtain the necessary researcher acting the part of the physician included in this study. This intervention
information to address the child’s illness. with the difficult characteristics reported initially followed the protocol of the infor-
Parents were also asked to write any addi- by the parents. Following each role-play mation-only intervention and then, once
tional questions they had for their child’s exercise, specific feedback was given to the parent had been encouraged to use an-
provider during the clinic visit. the parent regarding his or her performance tibiotics for his or her child only when nec-
The next step in the intervention was to in successfully asking the questions and essary, the researcher transitioned to the
promote parents’ efficacy to be assertive in persisting until the answers were satisfac- communication intervention. Due to the
asking and receiving answers to the recom- torily provided. increased length of time needed to admin-
mended questions. Bandura15 identifies four The final step for the intervention was ister both interventions, the role-playing
principle sources from which self-efficacy to address the physiological and affective exercises were reduced when necessary.
beliefs are constructed, including: vicarious state of the parent. A simple breathing tech- Control: Nutrition for
experience (modeling), enactive mastery ex- nique of concentrating on three successive, Children Intervention.
perience (role playing), verbal persuasion normal breaths was introduced to the par- A control condition unrelated to antibi-
(verbal feedback from significant others), ent. The researcher practiced this technique otic use or parent-physician communica-
and physiological and affective states (emo- with the parents and directed them to use tion was included. Child nutrition was the
tional arousal). A short series of role-play- this exercise when the physician entered the focus of this intervention, which utilized
ing exercises was used to provide vicarious examination room. materials from the Food and Drug Admin-
and enactive mastery experiences. This ac- Information-only Intervention istration entitled TIPS for Using the FOOD
tivity began with the researcher taking on Parents randomized to the antibiotic GUIDE PYRAMID for Young Children 2 to
the role of a parent and the parent taking information-based intervention group re- 6 Years Old. Nutritional guidelines and ideas
134 American Journal of Health Education — May/June 2005, Volume 36, No. 3
Stephen C. Alder, Eric P. Trunnell, George L. White, Jr., Joseph L. Lyon, James P. Reading, Matthew H. Samore and Michael K. Magill
for achieving these guidelines provided in this questionnaire were altered to refer to a received the communication intervention,
the materials were discussed. parent’s satisfaction with his or her child’s either alone or in conjunction with the an-
Instrument care (e.g., ‘Those who provide my child’s tibiotic information intervention, reported
Study interventions were designed to be medical care sometimes hurry too much a significantly higher sense of efficacy to
fewer than seven minutes in length. Ques- when they treat my child’ rather than ‘Those communicate with their child’s provider
tionnaires were administered to participat- who provide my medical care sometimes (p=0.021) than those that did not receive
ing parents before receiving the interven- hurry too much when they treat me’). The this intervention. The antibiotic informa-
tion, following the intervention but prior following subscales were included in the tion intervention did not demonstrate im-
to seeing the physician, and after the clinic analysis: General Satisfaction, Technical provement in parents’ communication ef-
visit was concluded. Study questionnaires Quality, Interpersonal Manner, Communica- ficacy (p = 0.337). There was no interactive
tion and Time Spent With Doctor. Internal
were pre-tested with parents who had chil- effect between the communication inter-
consistency of these scales ranged from 0.61
dren of ages similar to those included in the vention and the antibiotic information in-
to 0.86.
study sample. tervention (p = 0.615).
In the baseline and post-intervention Data Analysis Expectation for Treatment
questionnaires, parents were asked to rate Data were entered into Microsoft Access Significant main effects for reduction in
their level of expectation for a specific 2000 and imported into SPSS 9.0 for Win- parents’ general expectation of treatment
therapy when visiting the physician. Re- dows for analysis. Descriptive statistics were for their child were not observed for either
sponse options ranged from almost always used to describe the study participants. the communication or antibiotic informa-
(1) to rarely (7). Differences between Demographic variables were compared tion interventions alone (Table 3). However,
baseline and post-intervention responses among the four treatment conditions to a significant interaction effect was present
were calculated and used as an indicator of identify where statistically significant dif- (p = 0.049) with the communication inter-
change in general treatment expectation. ferences existed. Where appropriate, these vention, leading to a reduction in expecta-
variables were included in multivariate
At the conclusion of the clinic visit, par- tion for treatment by parents when it was
analyses of outcomes between groups. Mul-
ents were asked to rate their own ability to implemented without the antibiotic infor-
tivariate analyses included factorial two-way
ask their child’s physician to respond to mation intervention (-0.45 versus 0.31)
analysis of covariance and multivariate lo-
questions about their child’s illness, to take compared to implementation with the
gistic regression. Associations between par-
time to explain the illness, and to discuss antibiotic intervention (-0.10 versus -0.05).
ents’ communication efficacy and satisfac-
the appropriateness of treatment options
Antibiotic Prescription
tion were tested using Spearman’s
before a treatment decision is made. Par-
Predictors of the final outcome of inter-
rank-order correlation. The University of
ents were also asked to rate their ability to
est, antibiotic prescription, are included in
Utah Institutional Review Board reviewed
express disagreement with their child’s phy-
Table 5. It is notable that the model reported
and approved this study.
sician. Parents were asked to respond to
does not control for signs/symptoms or di-
each of these four questions using 0 (cer-
RESULTS agnosis, and as such, should be considered
tainly cannot) to 100 (certainly can) scales.
suggestive. A significant protective effect is
Internal consistency of the items was as- Demographic characteristics of study
demonstrated for the SCT-based commu-
sessed using Cronbach’s α. The four items participants are reported in Table 1. Signifi-
had a high level of internal consistency (α cant differences existed among study groups nication intervention (OR = 0.171, p =
= 0.89). Internal consistency was improved in the percent of participating parents who 0.042), with parents’ increased anxiousness
when the item regarding the ability of par- were the primary caregiver for the child vis- and decreased coping tending toward an
ents to disagree with their child’s physician iting the clinic (p = 0.009), with marginally increased likelihood of antibiotic prescrip-
was dropped from the scale (α= 0.93). significant differences among the percent of tion (OR = 1.34, p = 0.06 and OR = 1.61,
Included in the final questionnaire was parents who were female (p = 0.087), the p = 0.07, respectively).
an adapted version of the Short-form Pa- hours per week the primary caregiver Satisfaction with the Clinic Visit
tient Satisfaction Questionnaire, developed worked outside the home (p = 0.054), and Parents’ sense of efficacy to communi-
by the RAND Corporation. This question- the hours per week the child was in school cate with their child’s physician was posi-
naire was designed to assess patient satis- or day care (p = 0.076). tively correlated with multiple dimensions
faction regarding various aspects of a clinic Communication Efficacy of satisfaction with the clinic visit. Signifi-
visit. The purpose of this assessment was to Results of the comparison between in- cant associations were observed for General
examine whether receipt of an antibiotic terventions regarding parents’ self-reported Satisfaction (p = 0.002), Interpersonal Man-
was associated with increased satisfaction efficacy to communicate with their child’s ner (p = 0.010), and Time Spent with Doc-
with the clinic visit by the parent. Items in provider are shown in Table 3. Parents who tor (p = 0.002). Satisfaction with the clinic
American Journal of Health Education — May/June 2005, Volume 36, No. 3 135
Stephen C. Alder, Eric P. Trunnell, George L. White, Jr., Joseph L. Lyon, James P. Reading, Matthew H. Samore and Michael K. Magill
Table 1. Study Participant Demographics
Intervention Groups
Antibiotic Parental
Topic Education Assertion Combined Control p
(n=20) (n=20) (n=20) (n=20)
Percent of female parents 80 70 100 80 0.087
Percent of parents who are 95 70 100 68 0.009
child’s primary caregiver
Percent of Caucasian parents 90 95 80 90 0.498
Age—Parent* 30.0 ( 6.5) 30.2 ( 6.7) 30.7 ( 6.4) 32.6 ( 9.0) 0.673
Age—Child* 3.2 ( 3.0) 4.6 ( 3.6) 3.9 ( 2.2) 3.7 ( 2.7) 0.545
Hours per week 25.2 (18.5) 26.1 (22.0) 10.7 (16.0) 18.3 (21.6) 0.054
primary caregiver works
outside the home*
Hours per week child is in 23.1 (15.6) 22.4 (16.2) 12.5 (15.4) 14.4 (15.8) 0.076
school or daycare*
Number of children in 2.2 ( 1.2) 2.5 ( 1.3) 1.9 ( 0.8) 2.3 ( 1.3) 0.443
family*
* Mean (Standard Deviation)
** Proportions tested using Pearson’s χ2 test; mean values tested using one-way analysis of variance.
visit was also compared between parents
Table 2. Parental Efficacy to Communicate with Child’s Physician*
whose children received an antibiotic pre-
Mean (S.D.) N F(df) p scription and those that did not. Parents
reported that all dimensions of satisfaction
Main Effects
were higher when an antibiotic was not pre-
Communication Assertion 93.47 ( 8.89) 40 5.600 (1,71) 0.021 scribed (Table 5).
No Communication Assertion 86.28 (17.63) 39
Antibiotic Education 91.19 (11.98) 40 0.935 (1,71) 0.337 DISCUSSION
No Antibiotic Education 88.62 (16.37) 39 The results of this study demonstrate
that information-only approaches to address-
ing antibiotic overuse are not sufficient, and
Interaction Effects 0.225 (1,71) 0.615
in some respects may have no bearing on
Antibiotic Education the intended outcome of improving appro-
Communication Assertion 95.13 ( 8.28) 20 priate use. In this study, behavior change
No Communication Assertion 87.25 (13.91) 20 theory was used as a basis for designing and
No Antibiotic Education implementing an alternate approach to
Communication Assertion 91.81 ( 9.37) 20 antibiotic reduction. The results reported
No Communication Assertion 85.26 (21.21) 19 signify that moving beyond information-
based approaches may be warranted for
clinic-based interventions directed at
* Results based on a factorial two-way analysis of covariance. Covariates include: hours per week
patients or parents of patients.
primary caregiver works outside home, hours per week child spends in school or child care, number
The results reported in Table 2 support
of children in family and respondent’s age. Due to missing data, one subject in the control
condition is not included in this analysis. the first hypothesis that the described
communication intervention leads to
136 American Journal of Health Education — May/June 2005, Volume 36, No. 3
Stephen C. Alder, Eric P. Trunnell, George L. White, Jr., Joseph L. Lyon, James P. Reading, Matthew H. Samore and Michael K. Magill
improvement in parents’ sense of efficacy
Table 3. Change in Parents’ Expectation of
for communicating with their child’s phy-
Treatment from Pre-intervention to Post-intervention*
sician. The improvement was similar
whether the communication intervention
Mean (S.D.) N F(df) p
was administered alone or in conjunction
Main Effects
with the information-based intervention.
However, use of the communication inter- Communication Assertion -0.28 (0.75) 40 2.465 (1,67) 0.121
vention in the absence of the information-
No Communication Assertion 0.11 (1.39) 35
only intervention led to a greater reduction
Antibiotic Education -0.08 (0.98) 39 0.272 (1,67) 0.604
in parents’ expectation of specific therapy
prior to visiting with the medical provider No Antibiotic Education -0.11 (1.24) 36
(Table 3).
Interaction Effects 4.018 (1,67) 0.049
The provision of information when
working to promote healthy practices is Antibiotic Education
pervasive. The success of such efforts may Communication Assertion -0.10 (0.85) 20
be greatly enhanced by intervening to im-
No Communication Assertion -0.05 (1.13) 20
prove their skills to engage in the desired
No Antibiotic Education
behaviors, either directly or indirectly. A
Communication Assertion -0.45 (0.61) 19
general flaw in the common practice in
clinic settings of providing information No Communication Assertion 0.31 (1.66) 16
alone may be that the ability of individuals
*Results based on a factorial two-way analysis of covariance. Covariates include: hours per week
to take information and make the necessary
primary caregiver works outside home, hours per week child spends in school or child care, number
translations into action is generally overes-
of children in family and respondent’s age.
timated. The results of this study support Due to missing data, one subject in the Communication Intervention and four subjects in the control
providing more explicit direction regarding condition were not included in this analysis.
behaviors needed to achieve the goals of the
educational effort.
Table 4. Multivariable Logistic Regression
One of the primary rationales given for
Model of Antibiotic Prescription*
providing antibiotics is that of patient (or
parent) satisfaction. The belief is that if an Odds Ratio
antibiotic is not given when it is expected, (n=75) 95% Confidence Interval p
the patient will be dissatisfied and seek
Communication Assertion 0.171 0.031, 0.934 0.042
medical care elsewhere. The results of this
Antibiotic Education 0.398 0.082, 1.924 0.252
study do not support this link. The posi-
tive correlation between efficacy to com- Interaction: Communication 4.146 0.432, 39.810 0.218
municate and multiple dimensions of sat- Assertion x Antibiotic
Education
isfaction with the clinic visit indicate that
the interaction with the medical provider Number of Days Child Has 0.944 0.822, 1.083 0.408
may play a substantial role in how satisfied Been Ill
the patient or parent is with the care re- How Well Parent is Coping 1.608 0.958, 2.699 0.073
ceived. Satisfaction was higher when an an- With Child’s Illness
tibiotic was not prescribed (Table 5). More (Negatively Scaled)
investigation is needed to better understand How Anxious Parent is 1.335 0.983, 1.812 0.064
the association between failure to receive an About Child’s Illness
antibiotic and higher satisfaction with the Age of Child 0.879 0.711, 1.087 0.233
clinic visit.
Respondent’s Sex 6.963 0.727, 66.743 0.092
This study was designed to test whether
(Female = 1)
promoting parental communication effi-
cacy will lead to a reduction of expectation * Analysis is based on multivariable logistic regression – antibiotic prescription is coded as 1, no
for antibiotics, with trained research per- antibiotic prescription is coded as ‘0.’ Due to missing data, one subject in the Communication
Intervention and four subjects in the control condition were not included in this analysis.
sonnel administering the interventions. For
American Journal of Health Education — May/June 2005, Volume 36, No. 3 137
Stephen C. Alder, Eric P. Trunnell, George L. White, Jr., Joseph L. Lyon, James P. Reading, Matthew H. Samore and Michael K. Magill
Table 5. Parent Satisfaction with Clinic Visit and Whether an Antibiotic was Prescribed
Was an Antibiotic Prescribed?
Yes No
Mean (S.D.) Mean (S.D.) t (df) p
General Satisfaction 3.74 (0.83) 4.22 (0.65) 2.868 (76.07) 0.005
Technical Quality 3.50 (0.67) 4.03 (0.62) 3.657 (77.00) <0.001
Interpersonal Manner 4.01 (0.66) 4.32 (0.51) 2.273 (77.00) 0.026
Communication 3.50 (0.78) 4.04 (0.59) 3.472 (75.51) 0.001
Time Spent with Doctor 3.54 (0.87) 3.86 (0.81) 1.687 (77.00) 0.096
such approaches to be practical, existing during the clinic visit. Research is also improving antibiotic prescribing is high.
clinical personnel will need to provide the needed to better understand the link By working with patients and parents to
information and training. Further, the use between antibiotic prescription and satis- engage in communication behaviors and
of trained research personnel to adminis- faction with the clinic visit, including to seek an interaction-oriented medical
ter the interventions likely resulted in more whether the reported results translate to consultation rather than a specific treat-
strict adherence to the designed interven- similar interventions administered by ment, the dual benefits of reducing unwar-
tions than would have occurred using clinic clinic personnel. ranted antibiotic use and improved satis-
personnel as a routine part of delivering faction with clinical care may be gained.
care. Research is needed in order to deter- CONCLUSIONS Based on the findings presented, further
mine whether the same effects are present Existing approaches to patient/parent- research is merited, especially regarding
when using clinic personnel in the day-to- physician interaction may be based on the development of sound health promotional
day clinical context and how to efficiently false assumption that providing informa- approaches that can be efficiently and
implement the interventions in such a set- tion is sufficient to motivating appropriate effectively implemented within the primary
ting. Also, diagnoses and better evaluation health practices. The health promotional care delivery system rather than by research
of the appropriateness of antibiotic pre- approach used in this study moves beyond study personnel.
scription need to be included in future re- an information-only based approach that
search in order to better understand the is often the means of providing education REFERENCES
benefit gained by intervention. in primary care settings. Helping patients 1. Livermore DM. Bacterial resistance: ori-
Even though assignment was random- or parents to communicate in the clinic gins, epidemiology, and impact. Clin Infect Dis.
ized, differences exist between treatment setting may provide a substantial improve- 2003; 36: S11–S23.
groups. Research is needed to verify that ment in the provision of medical care, 2. Finkelstein JA, Metlay JP, Davis RL, Rifas-
differences in outcomes are due to the study including the reduction of unwarranted Shiman SL, Dowell SF, Platt R. Antimicrobial
interventions and not the characteristics of antibiotic prescribing. use in defined populations of infants and young
the participating parents. Future studies Current practices in primary care set- children. Arch Pediatr Adolesc Med. 2000; 154:
that include larger samples will be better tings regarding education generally are lim- 395–400.
suited to adjust for potentially confound- ited to providing information to patients, 3. Nyquist AC, Gonzales R, Steiner JF, Sande
ing factors that may have influenced study especially in regards to the provision of MA. Antibiotic prescribing for children with
outcomes. acute care. The results of this study provide colds, upper-respiratory tract infections, and
Items used in study questionnaires did promising evidence that better utilization bronchitis. JAMA. 1998; 279: 857–877.
not undergo formal validity and reliability of theoretically based health promotion will 4. Holmes JM, Metlay J, Holmes WC,
assessments. While the Short-form Patient lead to more appropriate care. Linking pa- Mikanatha N. Developing a patient intervention
Satisfaction Questionnaire is a standardized tient and health education professionals to reduce antibiotic overuse. AMIA Annu Symp
questionnaire, altering the items to focus on with those in the primary care medical de- Proc. 2003; 864.
the experience of parents visiting medical livery system to better incorporate ap- 5. Mangione-Smith R, McGlynn EA, Elliott
clinics for their children needs to be evalu- proaches that will lead to more effective in- MN, Krogstad P, Brook RH. The relationship
ated. More research is needed to establish teraction between medical providers and between perceived parental expectations and
standardized measures consistent with the patients is an area that needs further con- pediatrician antimicrobial prescribing behavior.
constructs included in this study, especially sideration. Pediatrics. 1999; 103: 711–718.
regarding expectation of medical treatment The cost of failing to make progress in 6. Vinson DC, Lutz LJ. The effect of parental
138 American Journal of Health Education — May/June 2005, Volume 36, No. 3
Stephen C. Alder, Eric P. Trunnell, George L. White, Jr., Joseph L. Lyon, James P. Reading, Matthew H. Samore and Michael K. Magill
expectations on treatment of children with a cations of Social Learning Theory. New York, N.Y.: 1982: 263–292.
cough: a report from ASPN. J Fam Pract. 1993; Praeger; 1982. 18. Hamm RM, Hicks RJ, Bemben DA. An-
37: 23–27. 13. Bandura A. Self-Efficacy: The Exercise of tibiotics and respiratory infections: are patients
7. Mainous AG, Zoorob RJ, Oler MJ, Haynes Control. New York: W.H. Freeman and Com- more satisfied when expectations are met? J Fam
DM. Patient knowledge of upper-respiratory pany; 1997. Pract. 1996; 43: 56–62.
infections: implications for antibiotic expecta- 14. Rosenstock IM, Strecher VJ, Becker MH. 19. Roter DL. Patient participation in pa-
tions and unnecessary utilization. JAMA. 1997; Social learning theory and the Health Belief tient-provider interaction: The effects of patient
45: 75–83. Model. Health Educ Q. 1988; 15: 175–183. question asking on the quality of interaction,
8. Friemuth V, Linnan HW, Potter P. Com- 15. Maddux JE. Expectancies and the social satisfaction and compliance. Health Education
municating the threat of emerging infections to cognitive perspective: basic principles, pro- Monographs. 1977; 5: 281–315.
the public. Emerg Infect Dis. 2000; 6: 337–347. cesses, and variables. In: Kirsch I, ed. How ex- 20. Greenfield S, Kaplan S, Ware JE. Expand-
9. Elder JP, Ayala GX, Harris S. Theories and pectancies shape experiences. Washington, D.C.: ing patient involvement in care. Ann Intern Med.
intervention approaches to health behavior American Psychological Association; 1999: 1985; 102: 520–528.
change in primary care. Am J Prev Med. 1999; 17–40. 21. Kaplan SH, Greenfield S, Ware JE. As-
17: 275–284. 16. Nader NR. Improving the practice of pe- sessing the effects of physician-patient interac-
10. Trunnell EP, Alder SC, Wise J, Yih R. Us- diatric patient education: a synthesis and selec- tions on the outcomes of chronic disease. Med
ing self-efficacy to improve parental decisions tive review. Prev Med. 1985; 14: 688–701. Care. 1989; 27: S110–S127.
regarding antibiotic use. 2003 (Submitted). 17. Feather NT. Human values and the 22. Centers for Disease Control and Preven-
11. Bandura A. Social foundations of thought prediction of action: an expectancy-valence tion. Antibiotic resistance: a new threat to you and
and action. Englewood Cliffs, NJ:Prentice- analysis. In: Feather NT, ed. Expectations and your family’s health. Available at: www.cdc.gov/
Hall; 1986. action: expectancy-value models in psychology. ncidod/dbmd/antibioticresistance. Accessed
12. Rotter JB. The Development and Appli- Hillsdale, NJ: Lawrence Erlbaum Associates; March 3, 2005.
American Journal of Health Education — May/June 2005, Volume 36, No. 3 139