Journal of Pediatric Psychology, Vol. 27, No. 3, 2002, pp. 281-291
© 2002 Society of Pediatric Psychology
Hospital Emergency Rooms and Children's Health Care Attitudes
Memorial University of Newfoundland
All correspondence should be sent to Carole Peterson, Psychology Department, St. John's, Newfoundland Canada A1B 3X9. E-mail address: carole{at}mun.ca .
 |
Abstract |
|---|
 Top Abstract IntroductionMethodResultsDiscussionReferences |
|---|
Objective: To assess attitudes of children requiring hospital emergency room (ER) treatment for trauma injuries 5 years afterward to evaluate the long-term effect of treatment distress. For comparison, health care attitudes of a large random sample of children were assessed.
Method: Children (N = 139, 7-19 years old) recruited from the ER completed a health care attitude questionnaire. Comparable schoolchildren (N = 1,300) completed the same questionnaire, with the addition of a few questions asking about hospital contact. The ER-recruited group was part of a 5-year follow-up study, and at the time of initial recruitment, their parents had rated their children's degree of distress at both the time of injury and of ER treatment on a 6-point scale.
Results: For the ER-recruited sample, the degree of distress during ER treatment did not seem to have longterm effects on children's attitudes. For the random sample, contact with the ER, especially for a trauma injury, was related to children liking the ER more.
Conclusions: Although other research has shown that aversive medical experiences may negatively affect children's attitudes, these findings suggest that the nature of the medical contact is important in how children interpret medically induced pain, which is related to their attitudes.
Key words: hospital emergency rooms; attitudes; Children's Health Care Attitudes Questionnaire (CHCAQ); treatment distress.
|
Introduction |
|---|
|
TopAbstractIntroductionMethodResultsDiscussionReferences |
|---|
The attitudes that children develop toward health care may have long-lasting effects on their subsequent behavior related to health care. In particular, experiences involving considerable pain and distress may foster negative attitudes toward and subsequent avoidance of the health care system. In this study, the health care attitudes of children who required hospital emergency room (ER) treatment because of trauma injury were assessed 5 years later to evaluate whether their degree of distress during treatment was related to their health care attitudes several years later. For comparison, the health care attitudes of a large random sample of children were assessed, some of whom had also had experiences with the hospital ER.
Research from both the dental and medical domains points to potential
long-term effects from aversive health care experiences. Dental research has
repeatedly shown that children who have had painful experiences at the dentist
are more likely to become anxious about dental precedures (e.g.,
Davey, 1989
;
Liddell, 1990). In turn,
individuals who are more anxious about the dentist are more likely to have
irregular dental visits or in fact avoid the dentist, even though such
behavior may lead to more serious later dental problems (e.g.,
Vassend, 1993). Furthermore,
there seems to be a bidirectional relationship between dental pain and dental
anxiety (Woolgrave & Cumberbatch,
1986). Higher amounts of dental pain and aversive dental
experience seem to increase dental anxiety, and in turn, high dental anxiety
increases subsequent perceptions of dental pain.
It has often been assumed that a similar relationship between aversive experience and later attitudes occurs in the nondental health care domain. Such negative attitudes fostered by early aversive medical experiences can be detrimental to the individual in at least two ways. First, children who have been highly distressed by early experiences might find future medical events aversive and try to avoid them, thus influencing their likelihood of seeking necessary medical help in later life. Second, children who have been highly distressed by earlier medical experiences might react to later medical procedures with greater fear and perceived pain.
Some research supports the assumption that aversive medical experiences in
childhood lead to subsequent negative attitudes toward medical situations. For
example, traumatic medical experiences in childhood have been found to be
associated with subsequent avoidance of doctors
(Melamed, Robbins, & Fernandez,
1982; Quinton & Rutter,
1976), and children with previous negative medical experiences
were more behaviorally distressed during throat cultures
(Dahlquist et al., 1986). The
nursing literature also abounds with discussions of children's avoidance of
the nurse who uses the hypodermic needle
(Bush & Holmbeck, 1987).
Pate, Blount, Cohen, and Smith
(1996), when exploring the
predictors of adult functioning in medical situations, found that medical fear
and avoidance of medical care were partly predicted by experience of medical
fear as a child, and furthermore, the medical pain of these individuals as
adults was also predicted by their experience with pain during childhood.
(Pate et al. asked young adults to rate their childhood medical experiences.
It is not possible to tell whether childhood experience influenced adult
attitudes or adult attitudes influenced recollection of childhood experience.)
Other investigators have also suggested that children's attitudes toward
specific aspects of health care may play a significant role in influencing
children's approach or avoidance of medical events
(Gochman, 1985;
Jay, 1988).
Few studies explicitly investigate the factors that lead to negative health
care attitudes in children, partly because there had been no good psychometric
measure of such attitudes until that developed by Joseph Bush and Grayson
Holmbeck (Bachanas & Roberts,
1995). They developed the Children's Health Care Attitudes
Questionnaire (CHCAQ), which measures children's attitudes toward health care
personnel, procedures, and settings along three dimensions: like-dislike,
approach-avoidance, and attributed effectiveness-ineffectiveness
(Bush & Holmbeck, 1987).
The CHCAQ includes a pain scale that asks children to assess the painfulness
of 17 stimuli (such as getting an injection). Studies that have used the CHCAQ
have shown that a number of factors contribute to children's health care
attitudes. Gender and age both play a role
(Bachanas & Roberts, 1995;
Bush & Holmbeck, 1987;
Hackworth & McMahon, 1991),
with boys tending to be more positive than girls, including liking health care
entities more, being less avoidant, and rating various procedures as less
painful. However, these studies have also had mixed findings about age
changes, with children showing both more positive and more negative attitudes
with age. Other factors that seem to play a role include mothers' health care
attitudes and children's perceived health locus of control, which mediates
their attitudes as well as their pain ratings
(Bachanas & Roberts, 1995;
Hackworth & McMahon,
1991).
Children's attitudes toward health care have also been related to their
behavior. Bachanas and Roberts
(1995) found that children
whose attitudes showed more avoidance and who rated health care entities as
less effective demonstrated more distress when their finger was pricked for
blood samples. That is, children's attitudes prior to a medical procedure
predicted the children's subsequent behavior a few minutes later. Of more
relevance to this study, prior experience with the health care system also
affected children's attitudes (Hackworth
& McMahon, 1991). Children who had immediate family members
who suffered from chronic or major illness had more negative attitudes toward
health care than did other children, and in particular they expressed more
avoidance. However, this study included only 55 children, ranging in age from
6 to 15 years, and there was no information about how recent the illnesses had
been, over how long a time they had been extended, nor the extent of the
children's direct exposure to the ill family member's physician visits or
medical treatment.
A potentially important yet unexplored issue is the nature of the child's
medical experiences. This may influence the attitudes that children develop as
well as their subsequent behavior. For example, children who have experienced
long-term major health problems or chronic illness may develop quite different
attitudes toward health care entities (personnel, procedures, and
institutions) than children who have experienced only short-term and less
serious health problems (Hackworth &
McMahon, 1991). Both of these groups in turn may have different
attitudes than children who simply visit the physician for well-child
check-ups, even though such check-ups include inoculations and blood samples.
For example, children who have experience with chronic illness seem to have
greater medical fears (Aho & Erickson,
1985), as well as more negative attitudes toward medical entities
(Hackworth & McMahon,
1991), but this may not be true for children who have other sorts
of aversive experiences with the medical system.
One medical setting in which children are often extremely distressed is the
hospital ER. Children are brought to the ER because of injury or illness
needing immediate medical assessment and treatment. Children sometimes display
considerable distress during treatment in the ER
(Peterson & Bell, 1996);
thus, this experience is a negative, aversive event. The question is whether
these negative experiences lead to the development of negative attitudes. This
question has not yet been answered, even though it could have important
implications for the treatment of children in the ER. For example, if high
levels of distress in the ER during treatment result in the development of
substantial negative attitudes toward health care components, it would be
important to work even harder to find ways of decreasing such distress or
alternatively to find ways of helping children cope with the aftermath of such
distress.
This study relates children's current health care attitudes to their
previous medical experience, and in particular their experience with hospital
ERs for two groups of children. The first group included children who had been
recruited from the ER of a children's hospital 5 years earlier for another
study (Peterson, 1999;
Peterson & Bell, 1996).
All children had been taken there for treatment of a trauma injury, mostly
broken bones or lacerations requiring suturing, and shortly afterward their
parents (who had been present during treatment) had been asked to rate their
child's degree of distress during medical treatment. The second group was a
large random sample of school-age children asked whether they had ever had
experience with a hospital, in particular, the ER. Both groups of children
filled out a health care attitude questionnaire. For the second group, we
could assess, in a representative sample, whether the presence or absence of
prior experience with hospitals or ERs affected children's health care
attitudes. For the first group, all of whom had had prior experience with ERs,
we could assess whether the degree of distress exhibited by the children
during ER treatment was related to their attitudes toward health care 5 years
later.
Our hypotheses are as follows. First, contact with the ER, which is normally aversive, will be associated with negative attitudes. We hypothesized that those children in the random sample who had received hospital treatment would have more negative attitudes toward the health care system than those who had not had such contact with hospitals. Second, the more stressful the contact, the more negative their attitudes toward the health care system. In terms of the sample of children who had been recruited from the ER, we hypothesized that children who were more distressed during ER treatment would have more negative attitudes than children who were less distressed during ER treatment.
|
Method |
|---|
|
TopAbstractIntroductionMethodResultsDiscussionReferences |
|---|
Trauma Group
A total of 139 children (77 boys and 62 girls) had been recruited from the ER of the only children's hospital in Newfoundland, Canada, 5 years earlier. Their current mean age was 12.50 years (SD = 3.44, range = 7-19). They were mostly white, from mixed socioeconomic backgrounds, and lived in one of two nearby cities or in the surrounding communities. The children had experienced trauma injuries that included broken bones, lacerations requiring suturing, dog bites, second degree burns, and crushed fingers requiring drainage. At the time of injury, children were between 2 and 13 years of age, and at the time of this study, they were between 7 and 19 years old.
At the time of initial recruitment, all families of injured children who
were of the appropriate ages were approached in the ER and asked to be part of
a long-term study of children's memory for trauma injuries. The majority of
approached families (85%) agreed to participate. No incentives were offered
for participation. They were then visited at home, and both parents and
children were interviewed about their recall of the injury and subsequent
treatment, and then children were reinterviewed after 6 months, 1 year, and 2
years. The study was described and signed consent was obtained at initial
recruitment and at home visits, as approved by the Memorial University of
Newfoundland Faculty of Science Human Research Ethics Committee and the
Faculty of Medicine Human Investigation Committee. (See
Peterson, 1999, and
Peterson & Bell, 1996, for
reports on children's memory of these events.) At the time of the initial
interview, parents were also asked to rate their child's degree of distress at
both the time of injury and at the time of ER treatment. The rating scale
ranged from 1 (not stressed at all, not upset) to 6 (highly distressed,
extremely upset). Five years after recruitment in the ER, families were
contacted again and revisited. Children were asked to fill out a questionnaire
about health care attitudes and were interviewed once again about their recall
of target events, although the memory data are not included in this report.
(See Peterson & Whalen,
2001.)
Random Sample
Sampling was done based on age and geographic distribution. The current age
of the trauma group was used, not their age at the time of the trauma (mean
age of sample = 10.71, SD = 2.96, range = 6-19.) The geographic
distribution of the trauma group children was used to establish the sampling
frame, which was the list of schools and grades in the three regions (St.
John's, Mt. Pearl, and surrounding communities). Stratified multistage
sampling was used to first select schools in a region and then to select half
of the grades within a school. When the research assistant visited a selected
school, she determined how many classes were in each of the selected grades.
She then randomly selected (via random numbers) the appropriate classes to be
included.
After the random sample was selected and ethical approval obtained from the Faculty of Science Human Research Ethics Committee, permission to conduct the research was obtained from the school board. School principals were then contacted and a meeting arranged to select the classes in the schools. Permission requests that described the study were sent home with children in the selected classes. Public service announcements about the survey were made in the media, and parents were encouraged to ask their children for the permission slips. This led to an excellent rate of response from the schools and from the parents. Only 2 schools of the original 35 sampled did not allow the survey. These were replaced with two other schools randomly selected from the remaining pool. There were rarely more than one or two students in any class who did not return the signed permission slip. A total of 1,300 schoolchildren (610 boys and 690 girls) participated.
No information about religion, socioeconomic status (SES), or ethnic origin was collected from the children for several reasons. First, it is not clear that young children could accurately report their family's SES. In terms of ethnic origin, 97% of the population in the catchment area is classed by Statistics Canada as "nonimmigrant." The entire visible minority population in the area is only 1.38% of the population (1996 Census—Population statistics for St. John's Metropolitan Area). Finally, whereas in the United States variables such as access to medical care, health insurance, income, race, or religious affiliation might be related to sampling biases, in Canada in general, and in Newfoundland in particular, this is not likely. As there is no direct cost for health care in Canada, all families, regardless of SES, have access to the same health care. Further, the Janeway Hospital is the only children's hospital in the region. Thus, any child within 150 miles of the hospital suffering trauma injury would be sent to Janeway, whether he or she presented at a physician's office or at the ER of another hospital.
As demographic information was not collected from the children, we could
only compare the random sample to the trauma group on the basis of the
sampling frame, geographic distribution. A chisquare test indicated that more
children came from St. John's and fewer from the surrounding communities than
expected based on the trauma group, (
2[2] = 49.14, p
<.001). However, a second chisquare test indicated that the random sample
did not differ significantly from expectation based on the proportions in the
population from which the sample was drawn (2[2] = 2.74,
p >.05).
Questionnaire
Two modified versions of the CHCAQ developed by Bush and Holmbeck
(1987) were used to assess
children's health care attitudes. Version 1 of the CHCAQ was used for children
between the ages of 7 and 10 years. Six items were added at the beginning of
the questionnaire. The first two items asked children their age and gender.
The remaining items were four yes/no questions in which children indicated
whether they had an experience in a hospital ER due to illness or injury,
whether they had stayed overnight in the hospital, and whether they had had an
operation. The remainder of the questionnaire presented the questions that had
been in the original questionnaire in a slightly different format. (In the
original questionnaire, the word "shots" was used for injections.
We changed this to "needles," a word used in the local community
for injections.) Three groups of multiple-choice questions focused on the
attitudinal dimensions of like-dislike, attributed
effectiveness-ineffectiveness, and approach-avoidance. For the like-dislike
questions, an array of five faces ranging from a smile to a frown were
presented. Children were given directions at the top of the page that told
them to circle the letter that they agreed with the most and to use the
pictures to help them with their choices. The same array and instructions were
used for the attributed effectiveness questions. We believed that children at
this age would be better able to indicate their feelings toward these
questions in terms of concrete symbols, such as faces, as opposed to abstract
symbols, such as the plus-minus signs presented in the original questionnaire.
For the approach-avoidance questions, an array of yes and no symbols was
presented next to each question, and next to each corresponding answer, rather
than just presented at the top of the page, as in the original questionnaire.
In addition to the attitude questions, the CHCAQ pain ratings scale was
presented, with five computer-drawn thermometers indicating different levels
of pain. Children were instructed to rate each procedure (e.g., "Getting
a needle in your arm") in terms of how much pain they felt or would feel
on a scale of 1 (no pain) to 5 (the worst pain you have ever felt), using the
thermometers. The pain rating score was intended to measure children's
reported overall sensitivity to pain. Therefore, the pain ratings score was
each child's mean response to the 17 procedures. If a child did not rate a
procedure, the rating score was the mean response to the items he or she did
rate. Very few children omitted any items. The items most frequently omitted
asked about getting stitches and waking up after an operation.
The second version of the questionnaire was administered to young people between the ages of 11 and 19 years of age. This version was identical to version 1 except for the removal of the pictures and graphic symbols in the like-dislike, attributed effectiveness-ineffectiveness, and approach-avoidance questions.
Questionnaire Analysis
As noted, questions about attitudes toward the health system could be
divided into three sections: liking, efficacy, and approach questions. Scoring
on the liking and efficacy questions was reversed so that a high score
indicated more liking and more effectiveness. (Liking scores were reversed so
that direction was consistent with the other questions. However, we realize
that none of the children likely really felt positively about the health
activity, such as having an operation.) A high approach score meant that if he
or she needed the procedure the child would not try to avoid it. We first
examined the reliability of the questions in each group. The Cronbach's alpha
reliability for the liking items was.72, for efficacy items.83, and for
approach items also.83. (The alpha for the entire scale was.48, indicating
that the items were not all measuring the same construct.) As the reliability
indicated that the items in each set were measuring a similar construct, we
summed the items together to create three variables with these names.
We expected that the attitudes most likely to be influenced by contact with the health system would be those directly related to the point of contact. We divided the attitude questions into two sets, those most likely to be influenced by contact and those less likely to be influenced by contact, which we labeled "noncontact" questions. Contact items were those asking about operations and the ER. Noncontact questions included those asking about hospitals in general, dentists, medicine, needles, nurses, and finger pricking to draw blood. Cronbach's alpha for the seven noncontact liking items was.69, for the seven noncontact efficacy items.75, and for the seven noncontact attraction items.79. Because the reliability indicated that the items in each set were measuring a similar construct, we summed the items together within each group to create six variables that we labeled C(contact)-liking, C-efficacy, C-approach, NC(noncontact)-liking, NC-efficacy, and NC-approach.
|
Results |
|---|
|
TopAbstractIntroductionMethodResultsDiscussionReferences |
|---|
In this section we will first look at the differences between the trauma group and random samples. We will then look at how contact with the health system affects all children's attitudes toward the system. Finally, in the trauma group, we will look at the impact of the amount of stress on children's attitudes.
Comparison of Trauma Group and Random Sample
There was no difference in the mean age in the two groups, whether they had
been operated on, or whether they had to stay in the hospital overnight. The
lack of difference between the two groups on operations and overnight stays
suggests that the trauma group is not made up of children who are predisposed
to ailments for which hospitalization is required.
Children in the trauma group were much more likely to report having gone to
the ER for injury (84%) than were children in the random sample (50.3%),
(2(1) = 59.75, p <.01). As the trauma group was
selected because they had been at the ER for injury, this finding might be
considered as validating children's understanding of the questions. However,
since all of the trauma children had been to the ER for an injury, it appears
that a few of them did not understand the question or did not recall the
visit.
A z test was used to compare the attitudes of children in the trauma group with those of children in the random sample. The only variables on which the two differed was liking for the ER (z = 2.34) and liking for an operation (z = 2.82). Children in the trauma group liked the ER less (random = 2.81, trauma = 2.56), and liked having an operation less (operation: random = 2.01, trauma = 1.72) than did children in the random sample. The difference in liking for the ER is still significant if we limit the comparison to children in the random sample who had contact with the ER (M = 2.89, z = 2.77). Given the lack of demographic differences between the two samples, we combined them to look at the impact of contact with the health system on health care attitudes. Because there were differences in liking scores between the two samples, we included sample as an independent variable for liking for the ER and for an operation.
Effect of Contact on Attitudes Toward the Health System
Our first hypothesis was that children who had contact with the hospital
would have more negative attitudes toward the health system than would
children who had not had hospital contact. We used stepwise multiple
regression to assess the relative contribution of the independent variables
(contact with hospital, person variables such as age, gender, pain
sensitivity, and sample) to the dependent variables (liking for ER and
operation, efficacy of ER and operations, approach to ER and operations, and
NC-liking, NC-efficacy, and NC-approach). Responses to questions about
hospital contact were used to divide the random sample into two groups,
children who had contact with the ER and those who had not. Children who had
contact with the ER were divided into those who had contact for an injury and
those who had contact for illness.
Liking. The only nonpersonality variables contributing significantly to the variance of liking for the ER were effect of having been to the ER for an injury and the sample. Having been to the ER for an injury was significant, F(1, 1,011) = 4.56, p <.033, and accounted for.4% of the variance (see Table I). Those who had contact with the ER for an injury liked it more (M = 3.16) than do those who have not had such contact (M = 3.34, t[1,043] = -2.22, p <.03). Although this is not a major contributor to the variance, its importance is in the direction it takes. Contrary to our first hypothesis, contact with the ER led to more rather than less liking. Sample also contributed significantly, F(1, 1,012) = 3.91, p <.05, and accounted for.3% of the variance. As noted, children in the trauma group liked the ER less than did children in the random sample, regardless of whether the children in the random sample had contact with the ER.
|
Having had an operation had a significant effect, F(1, 1,394) = 64.55, p <.001, and accounted for 4.4% of the variance in liking for operations (see Table II). This was followed by pain sensitivity, F(1, 1,393) = 51.29, p <.001, and accounted for 3.4% of the variance in liking for operations. Those who had an operation likel it more (M = 1.32) than do those who have not (M = 0.79, t[893] = 7.86, p <.001). Similarly, those who are more sensitive to pain are less likely to like having an operation (r [1,418] = -.20, p <.001). Sample also accounted for a significant amount of the variance, F(1, 1,390) = 7.59, p <.01, with children in the random sample (M = 3.99) liking an operation more than children in the trauma group (M = 4.22, t[1,430] = -2.77, p <.03).
|
Age was significant, F(1, 1,366) = 104.08, p <.001, and accounted for 7.1% of the variance in liking for other aspects of the health care system (see Table III). This was followed by pain sensitivity, F(1, 1,365) = 102.88, p <.001, and accounted for 6.5% of the variance in liking for other aspects of the health care system. Older children were less likely to like having contact with other aspects of the health care system (r [1,395] = -.27, p <.001). Those who are more sensitive to pain are less likely to like to have contact with other aspects of the health care system (r [1,390] = -.18, p <.001). Ever having been to the ER for an injury also contributed significantly F(1,1,364) = 5.67, p <.05, and accounted for.4% of the variance. (A similar analysis was done with stress at injury and at treatment added to the group of independent variables [using only the trauma group]. The only dependent variable to which stress [at treatment] was a significant contributor was liking for an operation, F[1, 115] = 4.18, p <.05, to which it contributed 3.5% of the variance.)
|
Efficacy. Ratings of efficacy are not influenced by either contact with the health care system or personal variables. None of the independent variables was significant. All of the independent variables together only accounted for a small percentage of the variance in efficacy ratings for going to the ER when injured (0.9%), for going to the ER when ill (0.1%), and less than 0.1% of the variance for efficacy ratings for an operation. The independent variables accounted for a similarly small efficacy rating for other aspects of the health system (3%).
Approach. The effect of pain sensitivity was significant, F(1,1,398) = 29.74, p <.001, and accounted for 2.0% of the variance in willingness to go to the ER if necessary. (See Table IV.) This was followed by having been to the ER for an illness, F(1, 1,397) = 20.67, p <.001, and accounted for 1.4% of the variance in willingness to approach the ER. Those who were more sensitive to pain were less willing to go to the ER, (r [1,422] = -.14, p <.001). Those who had been to the ER for an illness were more willing to go to the ER (M = 1.09) than were those who had not been to the ER for an illness (M =.90, t[1,421] = 2.96, p <.005). Again, contrary to our hypothesis, contact with the health system led to more positive, rather than more negative attitudes. Ever having been to the ER for an injury also contributed significantly F(1, 1,396) = 6.40, p <.05, and accounted for.3% of the variance.
|
The effect of pain sensitivity was significant, F(1, 1,393) = 51.40, p <.001, and accounted for 3.6% of the variance in willingness to undergo an operation if needed (see Table V). This was followed by having had to stay overnight in the hospital, F(1, 1,392) = 19.85, p <.001, and accounted for 1.3% of the variance in willingness to undergo an operation if needed. Children who were more sensitive to pain were less willing to undergo an operation (r [1,417] = -.18, p <.001). Children who had stayed overnight in the hospital were more willing to have an operation (M = 2.94) than were those who had not stayed overnight in the hospital (M = 2.58, t[1,085] = 4.63, p <.001). Finally, having had an operation was significant, F(1, 1,391) = 6.47, p <.05, and accounted for.5% of the variance. Children who had an operation were more willing to have an operation if needed (M = 2.95) than were children who had never had an operation (M = 2.58, t[1,425] = 4.77, p <.001). Again, contrary to what might be expected, contact led to more, not less willingness to approach the system if necessary.
|
The effect of pain sensitivity was significant, F(1, 1,381) = 57.63, p <.001, and accounted for 4.0% of the variance in willingness to approach other aspects of the health care system (see Table VI). This was followed by age, F(1, 1,380) = 27.45, p <.001, which accounted for 1.9% of the variance in willingness to approach other aspects of the health care system. Children who were more sensitive to pain were less willing to approach other aspects of the health care system (r [1,404] = -.20, p <.001). Similarly, older children are less willing to approach other aspects of the health care system (r [1,408] = -.07, p <.01).
|
Liking and Approach. A reasonable assumption is that willingness to approach the health care system would be related to liking for a particular aspect of the system. This is true for children's attitudes toward the ER in the random sample, but it is not for those in the trauma group. In the random sample, the more a child liked the ER, the more willing he or she is to approach the ER (r [916] =.38, p <.001). However, in the trauma group the relationship between liking for the ER and approach was not significant (r [127] =.04, p >.10).
On the other hand, liking and approach to having an operation were related in both groups. The more a child in the random sample liked an operation, the more he or she is willing to have an operation if one is needed (r [1,286] =.45, p <.001). This was also true of children in the trauma group (r [138] =.30, p <.001).
Stress and Health Care Attitudes
Stress measures at time of injury and at time of treatment were only
available for children in the trauma group. There was no difference in the
stress parents reported for males and females at the time of injury
(t[133] =.55, p >.10). The relationship between age and
stress at time of injury was not significant (r [135] = -.13,
p >.10). Stress at time of treatment was related to age
(r [134] = -.44, p <.01), and to pain sensitivity
(r [120] =.29, p <.01). Age and pain sensitivity were
also correlated (r [124] = -.13, p <.01). However, the
relationship between stress at time of treatment and pain sensitivity became
marginal (rpartial [117] =.16, p <.10), when
age was partialled out of the correlation. Thus, younger children exhibited
more stress at time of treatment than did older children, but stress was not
directly related to pain sensitivity.
Our hypothesis was that the more stressful the contact, the more negative children's attitudes would be toward the health care system. To evaluate the effect of children's degree of distress at the time of both injury and ER treatment, we correlated the two stress measures with liking, efficacy, and approach. Surprisingly, neither stress at time of injury nor stress at time of ER treatment was significantly related to liking, efficacy, or approach. In other words, children who had been highly distressed at the time of their injury or subsequent ER treatment did not have more negative attitudes toward health care entities than did children who had experienced lower or minimal amounts of distress. Children's distress in the ER (or at the injury-causing accident that brought them to the ER) appeared to have no measurable long-term effect on their health care attitudes at all. Thus, our second hypothesis was not supported either; there was no indication that greater distress led to more negative attitudes toward the health care system.
|
Discussion |
|---|
|
TopAbstractIntroductionMethodResultsDiscussionReferences |
|---|
Our hypotheses were that contact with the ER, which is normally aversive, would be associated with later negative attitudes by children toward health care entities. And the more stressful the contact, the more negative their attitudes would be. To our surprise, neither hypothesis was supported. For the random sample, we found that contact with the ER was associated with children liking the ER more, not less. Although the children in the trauma group liked the ER less than did those in the random sample, the degree of distress that children experienced in the ER while being treated for an injury did not appear to be related to their attitudes toward the ER 5 years later.
Considerable earlier research has found an association between aversive
encounters with the health care system and negative attitudes, especially in
the dental domain. For example, children who experience more distress in the
dentist's office are more likely to dislike dentists, and these attitudes
subsequently influence their behavior toward dentists and dental procedures
and increases the likelihood of subsequent dental avoidance
(Davey, 1989;
Liddell, 1990;
Vassend, 1993). As another
example, children with chronic illness who have multiple medical visits to
treat or manage the illness also have more negative attitudes toward medical
entities (Hackworth & McMahon,
1991). Other researchers have also noted associations between
aversive medical experience and subsequent negative attitudes (e.g.,
Melamed et al., 1982;
Pate et al., 1996;
Quinton & Rutter,
1976).
How can we account for our different results for the children in the random sample? We believe our results suggest that the precise nature of a child's contact with the health care system is an important factor, not simply whether that contact was aversive or not. One can assume that aversive contact was the norm for children from the random sample who had had to be treated at the ER, but as a group they liked the ER more than did those children who had not had such contact.
Part of the explanation of why our results for this sample are so different may reside in the relationship of the pain-inducing ER event to the entire sequence of events that children experience. When children are taken to the ER for a trauma injury, they have already been injured. They have already experienced the sudden pain and fear that accompanies bone fractures, lacerations, burns, crushed fingers, and dog bites. Thus, when they arrive in the ER, they are already distressed. There, they are repeatedly told by parents and medical personnel that their injury will be treated and that they will be better after this treatment. Even though the treatment may temporarily hurt, nonetheless all children were treated as outpatients and sent home, and they all probably felt much better soon after their treatment. In other words, the sequence of events was that children experienced sudden pain first as a result of an injury, went to the hospital ER because of that pain and the injury (where they often experienced additional pain during treatment), but as a result of this visit, the pain was soon alleviated. In contrast, children who experience dental pain usually feel good prior to their dental visit, and the pain is often caused by the dentist's procedures. So children often leave feeling worse than when they arrived. Similarly, although they often do not feel well prior to visiting the physician or hospital, children with chronic illness generally do not experience a dramatic improvement after treatment. Type of contact and sequence of children's feeling-states are both variables that should be explored in future research.
Liking for the health care system is related to children's willingness to approach it. For children in the random sample, liking for the ER was related to willingness to go to the ER if needed. There was no relationship for children in the trauma group. On the other hand, when asked about operations, a topic about which they had not been interviewed over the 5 years, their liking was related to willingness to approach in both the trauma and the random sample.
In a number of ways, our findings support prior research, which has
suggested that the relationship between children's experience with the medical
system and their subsequent attitudes is complex
(Bachanas & Roberts, 1995;
Bush & Holmbeck, 1987;
Hackworth & McMahon, 1991).
First, although attitudes of like/dislike were influenced by contact with the
ER, attitudes of approach/avoidance were influenced instead by personal
variables such as pain sensitivity. In terms of efficacy, none of our
variables influenced children's attitudes. Second, the lack of a relationship
between the distress of children in the trauma group during their ER visit and
their subsequent attitudes toward the ER suggests that one cannot simply
consider the amount of distress experienced by children in a medical/dental
setting. Rather, the larger event within which the medical/dental pain occurs
is important. Third, attitudes are related to points of contact or the context
of contact with the medical system. Thus, children's attitudes about the ER
were associated with having had contact with the ER, but this did not
generalize to attitudes about medical personnel or procedures in other medical
settings. Fourth, similar to the findings of Bachanas and Roberts
(1995), Bush and Holmbeck
(1987), and Hackworth and
McMahon (1991), we found that
personal variables such as the age of the child and pain sensitivity played a
significant role (although gender did not).
This study has a number of limitations. First, the effect sizes for almost all comparisons are small, which tempers the clinical utility of the findings. Also, a child-completed questionnaire is the only outcome measure of the children's health care attitudes. It would be important to extend this type of research to include measures of children's behavior when undergoing medical examination or treatment and even children's willingness to go to the doctor or the hospital. The number of visits by the children to the hospital was not controlled. It is not known whether children in the trauma group visited the ER again in the intervening 5 years since their initial recruitment, or how many times (and for what) they visited the ER. Furthermore, the only measure of distress is a rating scale provided by the parent. For the random sample, there is no demographic information, information on the frequency of hospital visits, the nature of those visits, or how long ago they occurred. All of these factors may play a role. For example, in the absence of demographic information, we cannot tell if there were substantive differences between the trauma group and the random sample.
The different circumstances in which the two groups responded to the questionnaire might have influenced findings, as one group filled it out at home whereas the other group filled it out in the classroom. In addition, a number of alternative explanations may help account for why the ER-recruited children's distress in the ER had no apparent measurable long-term effect on attitudes. Five years is a long time delay, and any effect may have dissipated over this length of time or may have been superceded by more recent ER experiences. Or the measures we used may not have adequately assessed the constructs of interest. It is also possible that children did not recall how much distress they had experienced 5 years earlier during ER treatment, although children seemed to accurately recall whether they had cried.
An important future direction for research is to explore the relationship
between memory and attitudes. When interviewed, the trauma group of children
clearly had extensive and largely accurate recall about their injury and
hospital treatment 5 years earlier
(Peterson & Whalen, 2001).
However, children did differ in the extensiveness and accuracy of their
recall, and memory variables may be important contributors to subsequent
attitudes. Research exploring the relationship between memory and attitudes is
currently underway in our laboratory. Similarly, the effect of parental
reminders about the ER visit should also be examined, as these types of
reminders may cue memories, in either a positive or negative direction.
This research has clinical implications. Most important, contact with the health care system that effectively alleviates child distress seems to be associated with more positive attitudes toward the health care system. Especially for children in the random sample, the effective treatment of the painful injuries that had necessitated ER treatment was associated with children liking the ER more, despite the distress caused by treatment in the ER. Perhaps clinicians would profit by considering ways in which children's contact with the health care system could be perceived as effective by children. In summary, the nature of children's contact with the health care system, not just whether it was painful or aversive, may be an important variable influencing children's health care attitudes.
|
Acknowledgments |
|---|
Preparation of this article was supported by grants from the Janeway Foundation Research Grant Program and Memorial University's Vice-President's SSHRC General Grant, as well as Grant OGP0000513 from the Natural Sciences and Engineering Research Council of Canada (to C. Peterson). We thank Jo-Ann Keats, who collected all the data from the school sample; the schools and teachers who allowed us into their classrooms; and the Janeway Hospital and their Emergency Room staff. We also thank all the recruiters, interviewers, and transcribers who participated. And, finally, we thank the parents and children who so willingly participated.
Received September 22, 2000; revision received February 28, 2001; revision received July 14, 2001; revision received September 21, 2001; accepted September 26, 2001
|
References |
|---|
|
TopAbstractIntroductionMethodResultsDiscussionReferences |
|---|
Aho, A. C., & Erickson, M. T. (1985). Effects of grade, gender, and hospitalization on children's medical fears. Journal of Developmental and Behavioral Pediatrics, 6, 146-153.[ISI][Medline]
Bachanas, P. J., & Roberts, M. C. (1995) Factors
affecting children's attitudes toward health care and responses to stressful
medical procedures. Journal of Pediatric Psychology,
20, 261-275.
Bush, J. P., & Holmbeck, G. N. (1987). Children's
attitudes about health care: Initial development of a questionnaire.
Journal of Pediatric Psychology,
12, 429-443.
Dahlquist, L. M., Gil, K. M., Armstrong, F. D., DeLawyer, D. D., Green, P., & Wuori, D. (1986). Preparing children for medical examinations: The importance of previous medical experience. Health Psychology, 5, 249-259.[ISI][Medline]
Davey, G. C. L. (1989). Dental phobias and anxieties: Evidence for conditioning processes in the acquisition and modulation of a learned fear. Behaviour Research and Therapy, 27, 51-58.[ISI][Medline]
Gochman, D. S. (1985). Family determinants of children's concepts of health and illness. In D.C. Turk & R. D. Kerns (Eds.), Health, illness and families: A life-span perspective (pp. 23-50). New York: Wiley.
Jay, S. M. (1988). Invasive medical procedures: Psychological intervention and assessment. In D. K. Routh (Ed.), Handbook of pediatric psychology (pp. 401-425). New York: Guilford Press.
Hackworth, S. R., & McMahon, R. J. (1991). Factors
mediating children's health care attitudes. Journal of Pediatric
Psychology, 16,
69-85.
Liddell, A. (1990). Personality characteristics versus medical and dental experiences of dentally anxious children. Journal of Behavioral Medicine, 13, 183-194.[ISI][Medline]
Melamed, B. G., Robbins, R. L., & Fernandez, J. (1982). Factors to be considered in psychological preparation for surgery. In D. Rough & M. Woolrich (Eds.), Advances in behavioral pediatrics (pp. 73-113). New York: JAI Press.
Pate, J. T., Blount, R. L., Cohen, L. L., & Smith, A. J. (1996). Childhood medical experience and temperament as predictors of adult functioning in medical situations. Children's Health Care, 25, 281-298.
Peterson, C. (1999). Children's memory for medical emergencies: Two years later. Developmental Psychology, 35, 1493-1506.[ISI][Medline]
Peterson, C., & Bell, M. (1996). Children's memory for traumatic injury. Child Development, 67, 3045-3070.[ISI][Medline]
Peterson, C., & Whalen, N. (2001). Five years later: Children's memory for medical emergencies. Applied Cognitive Psychology, 15, 1-18.
Quinton, D., & Rutter, M. (1976). Early hospital admissions and later disturbances of behavior: An attempted replication of Douglas' findings. Developmental Medicine and Child Neurology, 18, 447-459.[ISI][Medline]
Vassend, O. (1993). Anxiety, pain and discomfort associated with dental treatment. Behaviour Research and Therapy, 31, 659-666.[ISI][Medline]
Woolgrave, J., & Cumberbatch, G. (1986). Dental anxiety and regularity of dental attendance. Journal of Dentistry, 14, 209-213.[ISI][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||