Abstract
This study evaluates the impact of parent-provided distraction on children's responses (behavioral, physiological, parent, and self-report) during an IV insertion. Participants were 542 children, 4 to 10 years old, randomized to an experimental group that received a parent distraction coaching intervention or to routine care. Experimental group children had significantly less cortisol responsivity (p = .026). Children that received the highest level of distraction coaching had the lowest distress on behavioral, parent report, and cortisol measures. When parents provide a higher frequency and quality of distraction, children have lower distress responses on most measures.
Virtually all children undergo invasive medical procedures. Whereas some children only experience preventative immunizations, others require diagnostic tests and therapeutic treatments for serious illnesses. These experiences can provoke various levels of anxiety, fear, and pain in the child and a range of child behavioral responses from calm and controlled to panic and flailing. When a child is distressed, families and health care providers often experience anxiety, helplessness, and guilt. The procedure may then become technically more difficult (i.e., more attempts), further adding to everyone's discomfort. Experiencing stressful medical procedures in childhood can have long-term consequences, impacting the individuals' reaction to later painful events and acceptance of health care interventions in adulthood (von Baeyer, Marche, Rocha, & Salmon, 2004).
Distraction is a cognitive-behavioral intervention that is effective in reducing pain and distress for many children undergoing painful medical procedures (Kleiber & Harper, 1999; Uman, Chambers, McGrath, & Kisely, 2008). Distraction diverts attention from an adverse stimulus by redirecting attention to something else such as a book, toy, or nonprocedural talk. An effective distractor stimulates the senses, is developmentally appropriate, easily implemented, acutely engaging, and able to compete with negative stimuli to capture the child's attention (Cavender, Goff, Hollon, & Guzzetta, 2004). Evidence suggests that distraction may help the child to cope not only with the immediate medical procedure, but may also buffer memories of the experience so that the individual remembers less of the negative aspects, which may impact future responses to painful medical procedures (Cohen et al., 2001; Salmon, Price, & Pereira, 2002).
Young children generally need help or “coaching” to use distraction effectively. Professionals such as child life specialists, nurses, and psychologists typically provide distraction coaching, but the availability of these professionals to provide the intervention is limited in most practice settings. At the same time, many parents want to play an active role in the care of their children. Involving parents as distraction coaches for their children is a potential strategy to decrease child distress while managing limited professional resources and increasing parent involvement.
Parent distraction coaching can facilitate parent coping, promote child coping behaviors, reduce child distress, and compensate for ineffective and infrequent spontaneous distraction used by young children (Salmon & Pereira, 2002). It has been shown to increase quality of care and decrease the risk and cost of procedures (Christensen & Fatchett, 2002; Dahlquist, Pendley, Landtrip, Jones, & Steuber, 2002). Other studies have reported mixed results and vary by the distractor used, parent training method, and outcome measures. Blount and colleagues (1992) found children of parents trained to use distraction had less distress on two behavioral measures (the Child-Adult Medical Procedure Interaction Scale-Revised [Blount et al., 1997] and the Behavioral Approach-Avoidance and Distress Scale [Bachanas & Blount, 1996]), but found no significant differences on a third behavioral measure (the Observational Scale of Behavioral Distress–Revised [OSBD–R]; Jay & Elliott, 1984) or on self-report measures of fear and pain compared to children whose parents were not trained. Some found no differences in child distress or self-report of pain between trained parents and control groups (Cavender et al., 2004; Kleiber, Craft-Rosenberg, & Harper, 2001). Two limitations of these studies may contribute to the inconclusive findings. Previous studies often did not measure treatment fidelity (i.e., whether parents have implemented distraction coaching as intended) or address the possibility of a dose response (i.e., that children's responses to parent distraction coaching may be impacted by the quality and quantity of the distraction coaching provided by the parents).
The purpose of this study was to evaluate the impact of parent-provided distraction coaching on children's responses during an IV needle insertion. The hypotheses were as follows:
-
H1
Children of parents who were randomized to distraction coaching training respond with less distress to an IV insertion than children of parents in the control group.
-
H2
Children of parents who used higher levels of distraction coaching respond with less distress to an IV insertion than children of parents who use lower levels of distraction coaching.
Method
Design
A randomized clinical trial was used to compare the responses of children randomly assigned to the experimental group (i.e., parent distraction coaching training) or to a control group (i.e., routine care).
Participants
Children and their parents were invited to participate in this study from three Midwestern children's hospitals. Children 4 to 10 years of age undergoing a scheduled IV needle insertion for a diagnostic medical procedure were included in this study. From a developmental perspective, children in this age range rely on their parents for support in a stressful situation and have the cognitive skills to reliably answer study questions. Parents and children were excluded if they had developmental disabilities or English language skills that limited their ability to answer study questions. Children with cancers were not included due to treatment protocols that include central lines and intensive cognitive-behavioral interventions for serial procedures. One parent or guardian for each child agreed to participate as the child's support person.
Instruments
Multiple standardized instruments were used to collect data on child, parent, and environmental variables in this study (McCarthy et al., in preparation). Only instruments measuring family demographic variables and child characteristic variables that were significant for this article's hypotheses and analyses are included here. A measure of the level of parent distraction coaching (quality and quantity) provided by the parents was used to assess treatment fidelity, addressing a limitation noted in previous studies. Children's responses to the IV insertion included behavioral, physiological, and child-reported and parent-reported measures of child pain and distress.
Measures of demographic variables and child characteristics
Variables that influenced a child's response to distraction coaching included child age, gender, coping style, behavior factors, and learning disabilities. Parents completed a demographic questionnaire to obtain information on child and parent age, gender, race, and education. To assess preferred coping style, children were asked if they would prefer to watch or look away from having a common childhood injury (i.e., a scraped knee) washed and bandaged. Due to the complexity of other child coping measures, we developed this single question for this study. A similar approach to assessing coping style preference in young children has been successfully used in other research (Bernard, Cohen, McClellan, & MacLaren, 2004). Child behavior was assessed with the Pediatric Behavior Scale–30 (PBS–30), a 30-item screening scale adapted from the PBS (Lindgren & Koeppl, 1987; McCarthy, Lindgren, Mengeling, Tsalikian, & Engvall, 2002). The PBS–30 covers four major areas of behavioral and emotional adjustment: aggression–opposition, hyperactivity–inattention, depression–anxiety, and physical complaints. Subscales include 15 to 27 items that yield t scores of 40 to 90. Previous studies report internal consistency coefficients for the four general factors as 0.83, 0.87, 0.80, and 0.73, respectively (Lindgren & Koeppl, 1987; McCarthy et al., 2002). Whether a child had a learning disability was measured with a single parent report question included in the demographic questionnaire.
Parent-provided distraction
Performance of distraction coaching by the parents was measured with the Distraction Coaching Index (DCI; Kleiber, McCarthy, Hanrahan, Meyers, & Weathers, 2007). The DCI is a behavioral observation scale that measures the frequency and quality of distraction coaching. Videotapes of parent behavior during the child's medical procedure are divided into 10-sec intervals for analysis. Frequency of coaching is a percentage determined by the number of intervals in which distraction coaching is evident divided by the total number of intervals. Frequency of coaching was calculated twice—first for the 2 min preceding the IV insertion (pre-procedural) and then from the initial stick and for 2 min following the IV insertion (procedural). Quality of distraction coaching is assessed using five quality indicators, scored on 4-point scales for each of the two time periods (i.e., pre-procedural and procedural). Total DCI scores range from 0 (novice) to 40 (expert). Construct validity was established by comparing DCI scores of expert distraction coaches (child life specialists), parents trained to be distraction coaches, and untrained parents. All comparisons were significant at p < .0001 (Kleiber et al., 2007). In this study, interrater agreement was established at a 90% threshold before coding began on study data. For the 106 tapes that were reviewed for assessing interrater agreement, overall agreement on whether distraction was used in a tape segment was 96%. Adjusted for chance agreement, this corresponds to a kappa coefficient of 0.95 (95% confidence interval [CI] = 0.93, 0.96). The interrater agreement for the ratings on the quality of the distraction had a weighted kappa coefficient of 0.84 (95% CI = 0.81, 0.86). The computed DCI derived from the frequency and quality ratings had an intraclass correlation of 0.99. The interrater coefficient of variation for the DCI was 8.6%.
Children's responses
Behavioral distress was measured with the OSBD–R (Jay & Elliott, 1984), which consists of eight behavioral categories indicative of anxiety and pain in children (e.g., cry, scream, flail, and restrain). Time samplings of the child's behavior are recorded, and each behavioral category is weighted to indicate the intensity of distress (e.g., scream is weighted more than cry). A total distress score is calculated by adding together the weighted values of each of the behaviors at each interval. Higher scores reflect more distress. Jay and Elliott reported an internal consistency of 0.72; internal consistency for this study was 0.76.
Interrater reliability training for scoring the OSBD–R was carried out using videotapes from a previous study of children undergoing IV insertion. After group training sessions, pairs of new and experienced observers coded 12 to 15 videotapes simultaneously and independently until they achieved interrater agreement of greater than 90%. Research personnel blind to group assignment coded videotapes of children's behavior during the IV procedure. Throughout data analysis, a second behavioral coder scored every fifth videotape with an interrater agreement for specific behaviors of 99.3%; and for total scores, the intraclass correlation was 0.98. Interrater coefficient of variation for the total OSBD–R score was 15% (95% CI = 14%, 17%).
Physiological stress response was measured with salivary cortisol. Four salivary cortisol samples were obtained from each child—two during the clinic visit and two on a normal weekend baseline day. The first clinic sample, collected upon entering the clinic before the IV insertion, reflected the child's stress response to being in the clinic; and the second clinic sample, collected 20 min after the IV insertion, reflected stress in response to the IV insertion. Baseline samples were collected at times that matched those collected in the clinic. For example, if the first clinic cortisol was collected at 9 a.m., then the first baseline sample was collected at 9 a.m. Cortisol responsivity as an outcome was measured by the percentage of change in cortisol levels between the matched clinic and baseline cortisol samples. Information was obtained on factors that might impact cortisol levels, such as steroid medications, diagnoses, and dietary intake; and these were controlled for in data analyses. Explicit methods for collection and analyses of salivary cortisol samples and reliability of the assay are reported elsewhere (Hanrahan, McCarthy, Kleiber, Lutgendorf, & Tsalikian, 2006; McCarthy et al., 2009).
Child self-report of pain during the IV insertion was measured with the Oucher Scale. This standardized instrument was administered to the child shortly after the IV needle stick. The Oucher Scale assesses pain intensity in children aged 3 to 12 years and includes two separate scales (Aradine, Beyer, & Tompkins, 1988). One scale is a series of six photographs showing a child in varying degrees of discomfort, and is used by younger children. Children who are able to identify the larger of two numbers use the vertical numeric scale (0–10) that is printed next to the faces. Discriminant validity has been demonstrated by investigating relations between the Oucher Scale and two fear scales for children (Beyer & Aradine, 1986).
Parent report of child distress (PRCD) was measured with one item from the Perception of Procedures Questionnaire (PPQ; Kazak, Penati, Waibel, & Blackall, 1995). The PPQ was originally developed to measure parents' perceptions of procedural distress during pediatric spinal taps and bone marrow aspiration. Permission was obtained to revise the tool for use in this research with children having IV insertions. The revised PPQ consists of nine Likert-type questions, anchored with “not at all” to “extremely,” measuring parent involvement, parent distress, child distress before the procedure, and child distress during the procedure. Three experts who work with children during painful procedures evaluated the revised tool for clarity and content validity. For the hypotheses in this study, one question from the PPQ was used to determine child distress from the parent's perspective: “How distressed was your child today during the IV procedure?”
Intervention Training
In families randomized to the experimental group, one parent was asked to volunteer to be the child's main support person (coach) in the treatment room during the IV procedure. The designated parent received the multifaceted 15-min intervention on how to provide distraction coaching, which includes educational materials, a video, and discussion with a research assistant. The parent and child then selected distractors such as books, toys, and video games. Questions were answered, and practice with the distractors was encouraged.
Families randomized to the control group received routine care in the clinic, and the parent was encouraged to do whatever he or she normally did to help the child during a medical procedure. Distraction items were clearly visible in the treatment room and available to the parent, but there was no direction to use them.
Procedures
The institutional review boards at each data collection site approved this study. Protocols and procedures were developed for consistency of recruitment across sites. Families of children with scheduled appointments requiring an IV insertion were sent a letter explaining the study prior to their clinic visit. At the time of the clinic visit, consent and assent were obtained, salivary cortisol was obtained from the child, and families were randomly assigned to the control or experimental group. A topical lidocaine analgesic was applied to two potential IV sites on the child by clinic staff per clinic protocol. Parents and children answered the study questions, and teaching was done with parents in the intervention group during the 30 to 60 min that the topical analgesic was allowed to dwell.
The IV insertion procedure took place as usual in the clinic setting. Clinic personnel were blind to treatment condition and were requested to allow the parent to be the main support person during the procedure. The child and parent were videotaped from the time the child was placed on the exam table to the time the clinic staff indicated that the IV catheter was secured in place. Four minutes of video (2 min pre-procedural and 2 min procedural) were coded at a later time using the OSBD–R and the DCI.
Following IV insertion, children completed the Oucher Scale, and parents completed the PRCD. Children provided a second salivary cortisol sample 20 to 30 min after the IV insertion, and families were given cortisol collection material and instructions to obtain the child's baseline samples at home. Each family received $30.00 compensation for participating in the study.
Analysis
H1: Control and experimental group comparisons
The ordinal variables Oucher and PRCD were compared between the control and experimental groups using the Wilcoxon rank-sum test. Test of normality of the distribution of OSBD–R scores showed a non-normal distribution, so the Wilcoxon rank-sum test, which is a nonparametric test, was also used for group comparisons. Cortisol values were log transformed prior to analysis to normalize the data distribution. Linear mixed-model analysis for repeated measures was applied to the log-transformed values to test for differences in mean change from baseline (home to clinic) between control and experimental groups (i.e., cortisol responsivity).
H2: Levels of distraction and child responses
Child responses were compared by levels of parent-provided distraction as defined by the DCI. To examine the effect of the level of parent-provided distraction, the participants in the study from both the control and experimental groups were combined and regrouped by the parents' DCI score. Based on reported data from DCI development (Kleiber et al., 2007), DCI scores were classified into three levels: < 10 (low-level distraction), 10 ≤ 20 (moderate level distraction), and ≥ 20 (high-level distraction). The Kruskal–Wallis test was used to compare the Oucher Scale, PRCD, and the OSBD–R among the children grouped by level of the DCI. Linear mixed-model analysis for repeated measures was used for the comparison of cortisol responsivity.
In addition, the effect of the DCI on child response was tested, adjusting for the effect of various child variables that were either found to be significantly associated with child response or significantly different among the DCI groups at p < .10. Those variables were child age, gender, coping style, behavior factors, and learning disabilities. For the OSBD–R, this was done using multifactor regression analysis, with level of the DCI as the primary independent variable of interest and various child variables as covariates. For the PRCD, which has a skewed distribution with > 40% of the participants with a 1 to 2 level of distress, the scale was categorized into three levels: low (1–2), moderate (3–4), and high (5–7) distress; and the test of the effect of the DCI, adjusting for the child covariates, was done using the generalized logit model. The distribution of the Oucher Scale response was also skewed, with 60% responding in the 0 to 3 range, 20% in the 4 to 6 range, and 20% in the 7 to 10 range. With small frequencies in the mid- and high ranges, the Oucher Scale was divided into low (0–3) versus moderate to high (4–10), and logistic regression analysis was performed to test for the DCI effect, adjusting for child characteristics. Linear mixed-model analysis for repeated measures that included child characteristics as covariates was used for the comparison of cortisol responsivity.
Results
Participants
Families were recruited into this study over a 3-year period (2003–2006). From 720 families invited to participate, 542 children—262 (48%) girls and 280 (52%) boys—with a mean age of 6.95 years (SD = 1.90) participated in this study, for an overall participation rate of 75%. Figure 1 summarizes the families invited to participate and the numbers enrolled, excluded, and analyzed. Four children were withdrawn from the study for the following reasons: limited English, child refusal after consent, parent not allowed by staff into the treatment room, and lack of time to collect sufficient data. A total of 174 (24%) families declined participation, most often due to parents feeling overwhelmed by the clinic visit.
FIGURE 1.
Participants.
Parent coaches were most frequently White mothers (87%), with some college education (70%). Children were most often from the gastrointestinal clinic (45%) or from nephrology (17%), endocrinology (12%), pulmonary (8%), and cardiology (3%). The remaining 12% were from other clinics, primarily diagnostic radiology at one study site (77%). The most common diagnostic tests that children in the study had were endoscopy, renal scan, bronchoscopy, and pituitary stimulation tests. For 81% of children, parents reported < 10 experiences with painful medical procedures. More than 50 painful medical experiences were reported for 5% of children. Parents reported most of the children (78%) had previously had an IV insertion, and about one half (51%) were informed about the IV 24 hr or more in advance. There were no significant differences between groups for the demographic variables including child gender and race or parent gender, race, or educational level. Table 1 summarizes the demographic information of the children and their parents.
TABLE 1.
Participant Demographics
| Variable | Intervention (n = 249) |
Control (n = 293) |
Total (n = 542) |
pa | |||
|---|---|---|---|---|---|---|---|
| n | % | n | % | n | % | ||
| Child gender | .814 | ||||||
| Male | 130 | 52.21 | 150 | 51.19 | 280 | 51.66 | |
| Female | 119 | 47.79 | 143 | 48.84 | 262 | 48.34 | |
| Child race | .487 | ||||||
| White | 208 | 83.53 | 245 | 83.62 | 453 | 83.58 | |
| Hispanic | 2 | 0.80 | 5 | 1.71 | 7 | 1.29 | |
| Black | 18 | 7.23 | 14 | 4.78 | 32 | 5.90 | |
| Other | 21 | 8.43 | 29 | 9.90 | 50 | 9.23 | |
| Parent gender | .492 | ||||||
| Male | 31 | 12.55 | 31 | 10.65 | 62 | 11.52 | |
| Female | 216 | 87.45 | 260 | 89.35 | 476 | 88.48 | |
| Parent race | .472 | ||||||
| White | 215 | 87.40 | 262 | 90.34 | 477 | 88.99 | |
| Hispanic | 3 | 1.22 | 4 | 1.38 | 7 | 1.31 | |
| Black | 19 | 7.72 | 13 | 4.48 | 32 | 5.97 | |
| Other | 9 | 3.66 | 11 | 3.79 | 20 | 3.73 | |
| Education | .778 | ||||||
| High school | 74 | 30.08 | 84 | 28.97 | 158 | 29.48 | |
| College or more | 172 | 69.92 | 205 | 71.03 | 378 | 70.52 | |
Chi-square or Wilcoxon rank-sum test.
H1: Control and experimental group comparisons
Child responses (OSBD–R, PRCD, Oucher Scale, and cortisol responsivity) were compared between the control and the experimental groups (see Table 2). Significant group differences were found only in cortisol responsivity. For cortisol responsivity, the experimental group on average over the two clinic times (before IV insertion and in response to the IV insertion) had a significantly lower mean increase from baseline in the child's cortisol level (p = .026) relative to the control group. Although mean cortisol responsivity before and during IV insertion showed a lower response for the experimental group at both periods, the difference between control and experimental group before IV insertion was not statistically significant (p = .118), with a slightly greater difference observed during IV insertion (p = .051). This indicates that the stress response (cortisol responsivity) was, on average, less in children whose parents had distraction training.
TABLE 2.
Child Response by Control or Experimental Group
| Distress Response | Control | Experimental | p |
|---|---|---|---|
| OSBD–R | n = 287 | n = 244 | |
| Median (IQR) | 1.63 (0.45–5.08) | 1.71 (0.40–4.79) | .737a |
| PRCD—Frequency (%) | n = 290 | n = 247 | |
| 1–2 | 118 (41%) | 105 (43%) | .546a |
| 3–4 | 44 (15%) | 52 (21%) | |
| 5–7 | 128 (44%) | 90 (36%) | |
| Oucher Scale—Frequency (%) | n = 278 | n = 240 | |
| 0–1 | 101 (36%) | 89 (37%) | .639a |
| 2–3 | 77 (28%) | 56 (23%) | |
| 4–6 | 56 (20%) | 45 (19%) | |
| 7–8 | 12 (4%) | 14 (6%) | |
| 9–10 | 32 (12%) | 36 (15%) | |
| Cortisol responsivityc | n = 123 | n = 115 | |
| Average over clinic times | 50.8% ± 9.4% | 23.4% ± 7.9% | .026b |
| In clinic before the IV | 34.2% ± 10.2% | 13.3% ± 8.8% | .118b |
| In response to IV | 69.4% ± 13.9% | 34.5% ± 11.4% | .051b |
Note. OSBD–R = Observational Scale of Behavioral Distress–Revised; IQR = interquartile range; PRCD = parent report of child distress.
Kruskal–Wallis test.
Test of mean contrast from repeated measures analysis.
Cortisol responsivity is the percentage of change between cortisol levels obtained at the same time on two different days (baseline day and clinic day).
H2: Levels of distraction and child responses
Table 3 shows the child characteristics among the children grouped by level of the DCI. These are the characteristics described in the analysis section that were significantly different among the DCI groups at p < .10. Table 4 presents comparisons of child responses on the OSBD–R, PRCD, Oucher Scale, and cortisol among the three parent DCI groups, not adjusted for any child characteristics. Children that received the highest level of distraction (DCI ≥ 20) had the lowest OSBD–R, the smallest proportion of high (5–7 level) PRCD, and the smallest mean change in cortisol level from baseline to clinic compared to the other two DCI groups.
TABLE 3.
Child Characteristics by Dose of Distraction per the DCI Score
| Variable | DCI < 10 (N = 299) |
DCI 10 ≤ 20 (n = 97) |
DCI ≥ 20 (n = 132) |
pa |
|---|---|---|---|---|
| Median age (IQR) | 7 (6–9) | 7 (5–8) | 6.5 (5–8) | .003 |
| PBS–30: Median inattention (IQR) | 3 (1–4) | 2 (1.5–4) | 3 (1–5) | .072 |
| PBS–30: Median impulsivity (IQR) | 3 (2–5) | 3 (2–5) | 4 (3–5) | .049 |
| PBS–30: Median hyperactivity (IQR) | 3 (1–5) | 2 (1–5) | 3 (2–6) | .020 |
| Coping (watch), frequency (%) | 179/282 (63%) | 44/87 (51%) | 65/127 (51%) | .020 |
| Learning disability, frequency (%) | 37/296 (12%) | 7/96 (7%) | 5/131 (4%) | .013 |
Note. DCI = Distraction Coaching Index; IQR = interquartile range (25th–75th percentile); PBS–30 = Pediatric Behavior Scale–30 items.
Chi-square test was used for frequency (%) comparison, and the Kruskal–Wallis test was used for median (IQR).
TABLE 4.
Child Response by Dose of Distraction per the DCI Score
| Distress Response | DCI < 10 | DCI (10 ≤ 20) | DCI ≥ 20 | p |
|---|---|---|---|---|
| OSBD–R | n = 299 | n = 97 | n = 132 | |
| Median (IQR) | 1.58 (0.42–5.20) | 2.67 (0.88–5.50) | 0.98 (0.33–3.98) | .017a |
| PRCD—Frequency (%) | n = 297 | n = 96 | n = 130 | |
| 1–2 | 130 (44%) | 28 (29%) | 57 (44%) | .092a |
| 3–4 | 41 (14%) | 23 (24%) | 30 (23%) | |
| 5–7 | 126 (42%) | 45 (47%) | 43 (33%) | |
| Oucher Scale—Frequency (%) | n = 286 | n = 89 | n = 130 | |
| 0–1 | 111 (39%) | 27 (30%) | 46 (35%) | .411a |
| 2–3 | 74 (26%) | 25 (28%) | 31 (24%) | |
| 4–6 | 54 (19%) | 21 (23%) | 24 (18%) | |
| 7–8 | 10 (3%) | 6 (7%) | 10 (8%) | |
| 9–10 | 37 (13%) | 10 (11%) | 19 (15%) | |
| Cortisol responsivityc | n = 133 | n = 41 | n = 59 | |
| Average over clinic times | 52.0% ± 9.1% | 42.8% ± 15.5% | 5.0% ± 9.5% | .003b |
| In clinic before the IV | 38.3% ± 10.0% | 25.6% ± 16.2% | −5.1% ± 10.3% | .016b |
| In response to IV | 67.0% ± 13.2% | 62.4% ± 23.2% | 16.2% ± 13.9% | .036b |
Note. DCI = Distraction Coaching Index; OSBD–R = Observed Scale of Behavioral Distress–Revised; IQR = interquartile range; PRCD = parent report of child distress.
Kruskal–Wallis test.
Test of mean contrast from repeated measures analysis.
Cortisol responsivity is the percentage of change between cortisol levels obtained at the same time on two different days (baseline day and clinic day).
The effect of level of distraction coaching, as measured by the DCI, on child response was further examined adjusting for the effects of child variables (age, inattention, impulsivity, hyperactivity, coping preference, and learning disability) in a multifactor model for each of the child response variables. Differences based on the OSBD–R, PRCD, and cortisol remained significantly different, whereas the Oucher Scale results were nonsignificant. For the OSBD–R, the adjusted effect of the DCI showed that, relative to the group with DCI ≥ 20, mean OSBD–R was significantly higher in those with DCI < 10 by 39% ± 3% (p = .002) and in those with DCI 10 ≤ 20 by 45% ± 4% (p = .004). From the logistic regression model for PRCD, for children with DCI < 10 compared to DCI ≥ 20, the adjusted odds ratio for child distress of 5 to 7 (relative to 1–2) is 1.84 (95% CI = 1.01, 3.33; p = .046). For children with DCI 10 ≤ 20 compared to children with DCI ≥ 20, the adjusted odds ratio is 2.39 (95% CI = 1.13, 5.03; p = .022). For cortisol responsivity, the linear mixed-model analysis for repeated measures with covariates added showed a significant difference in level of responsivity between those with DCI ≥ 20 (5.4% ± 9.8%) and those with DCI < 10 (51.4% ± 9.5%; p = .005). The logistic regression models for the Oucher Scale showed no significant effect of the DCI (p = .986).
Discussion
The first hypothesis, that children whose parents received brief training in distraction coaching would be less distressed than control group children, was not supported by three of the four child response measures. Only cortisol responsivity was found to differ between the two groups, with children in the experimental group displaying less physiological reactivity. One explanation for this is control group contamination. In many studies, as in this one, families are randomly assigned to groups, and all participants in the experimental group receive the same intervention training. When evaluating an intervention through use of experimental and control groups, it is expected that the control group will not use the intervention provided to the experimental group. Our data show that some parents in the control group performed high levels of distraction coaching, as measured with the DCI, without the training intervention. Some of the parents in the control group had previous experience with using distraction, some had observed health professionals use distraction with their children during previous encounters in the clinic, and some naturally incorporated distraction into supporting their children, resulting in control group contamination. In this study, the control group parents were not restricted from using distraction; they were told to assist their children as they would normally. Likewise, there were parents in the treatment group that used no or very little distraction coaching, sometimes due to environmental factors, such as interference by the health care providers, captured in the videotape analysis. The role of the environment on successful parent coaching needs further study, and may indicate a need to provide interventions targeted to support the use of parent coaching.
It was evident early in this study that the quality and frequency of distraction provided by parent coaches varied across parents in both groups. In previous studies, researchers did not typically report whether the parents provided distraction coaching as intended by the training intervention. Incorporating measures of treatment fidelity, strategies to monitor and enhance the reliability and validity of behavioral interventions (Borrelli et al., 2005), into research methods is key to maintaining internal and external validity and interpreting the results of behavioral intervention research. Five categories of treatment fidelity have been identified—those that focus on the provider of the intervention (design of study, training providers, and treatment delivery) and those that focus on the recipient of the intervention (treatment receipt and treatment enactment; Bellig et al., 2004; Borrelli et al., 2005). In this study, particular attention was given to treatment delivery and evaluating whether the parents provided the intervention or distraction as intended by the training. The DCI, a behavioral coding of quality and frequency of distraction coaching, gave researchers the ability to evaluate the parent's coaching and assess treatment fidelity.
The second hypothesis proposed that children whose parents provided a higher level of distraction coaching would be less distressed than those whose parents provided poor quality distraction. Data were regrouped based on parent DCI scores for this analysis. H2 was supported by three of the four outcome measures of child response, with only the Oucher Scale not demonstrating a significant difference. After adjusting for child characteristics in the DCI groups, H2 was still supported by the same three out of four outcome measures. Children of parents who used the highest level of coaching had lower scores on the behavioral (OSBD–R) and physiological (cortisol responsivity) measures, and had the lowest percentage of children in high distress per parent report (PRCD). There were no differences in child report of pain when analyzed by level of distraction coaching. A potential explanation for this finding is that all of the children received a topical analgesic prior to the IV insertion, so needle pain may have been minimal for most of the children.
This study provides further support for the important design strategy of using multiple outcome measures of child distress response to provide a broader view of children's responses to painful medical procedures. This study and previous studies on parent-provided distraction typically used multiple outcome measures of child response (Blount et al., 1992; Cavender et al., 2004; Dahlquist et al., 2002; Kleiber et al., 2001). In this study, we found the range of Oucher Scale scores, the most direct measure of child perception of pain, was narrow and suggests that, overall, there was a low level of pain in this group. A child may appear calm, as measured by the OSBD–R, in response to a painful procedure but experience internal stress, as measured with cortisol. Another child may display behavioral distress (OSBD–R) but report that the painful procedure did not really hurt very much (Oucher Scale). This indicates the need to carefully clarify the conceptual and operational definition of distress used in a study.
Study limitations, which decreases generalizability, should be noted. The sample was predominately White, middle class, well–educated, and Midwestern. Mothers primarily participated as coaches, limiting understanding of the role of fathers as distraction coaches. This convenience sample primarily had previous IV or other procedural experience; results may differ with children with acute illnesses that are more naive to health care. Findings are limited, likewise, to preschool and young school-age children.
In summary, distraction is a valuable intervention for helping children cope with medical procedures, and teaching parents to be distraction coaches is a logical use of resources. Previous research has established that professionals are effective distraction coaches, but that parents may not be as effective in providing this cognitive-behavioral intervention. This study provides further information on the impact of parents as distraction coaches on the child's distress response. However, there are more questions that need to be answered. Why are some parents effective distraction coaches and others not? How do the health professionals present during the procedure impact the parent's success? Is there a way to identify parents who can coach and those families that need professional assistance prior to a procedure? These and other questions need to be addressed in order to provide the most effective distraction coaching to the child.
Implications for Practice
Results of this study indicate that when parents provide high-quality distraction coaching, children have lower distress responses on most measures of child response. These results support parent-provided distraction as a valuable intervention for children undergoing painful procedures. However, the results of this study also indicate that, with basic training, not all parents are able to provide a level or dose of distraction that is effective in decreasing all aspects of child distress. Some children, even with high-quality distraction coaching, still display distress responses. Further research is needed to understand the factors that contribute to a parent's ability be a distraction coach and those that identify which children will respond positively to parent distraction coaching. Elements of effective training programs for parents, appropriate modes of delivery, and methods for translating these results into routine pediatric practice are needed.
Acknowledgments
This study was funded by R01 Grant No. NR05269-01A2 from the National Institute for Nursing Research to Ann Marie McCarthy.
Contributor Information
Ann Marie McCarthy, College of Nursing, University of Iowa, Iowa City, IA.
Charmaine Kleiber, College of Nursing, University of Iowa, Iowa City, IA.
Kirsten Hanrahan, University of Iowa Health Care, Iowa City, IA.
M. Bridget Zimmerman, Department of Biostatistics, University of Iowa, Iowa City, IA.
Nina Westhus, School of Nursing, Saint Louis University, St. Louis, MO.
Susan Allen, Emergency Department, Pediatric Intensive Care Unit, Internal Resource Team, Blank Children's Hospital, Des Moines, IA.
References
- Aradine CR, Beyer JE, Tompkins JM. Children's pain perception before and after analgesia: A study of instrument construct validity and related issues. Journal of Pediatric Nursing. 1988;3:11–23. [PubMed] [Google Scholar]
- Bachanas PJ, Blount RL. The Behavioral Approach-Avoidance and Distress Scale: An Investigation of reliability and validity during painful medical procedures. Journal of Pediatric Psychology. 1996;21:671–681. doi: 10.1093/jpepsy/21.5.671. [DOI] [PubMed] [Google Scholar]
- Bellig AJ, Resnick B, Minicucci DS, Ogedegbe G, Ernst D, Borrelli B, et al. Enhancing treatment fidelity in health behavior change studies: Best practices and recommendations from the NIH Behavior Change Consortium. Health Psychology. 2004;23:443–451. doi: 10.1037/0278-6133.23.5.443. [DOI] [PubMed] [Google Scholar]
- Bernard RS, Cohen LL, McClellan CB, MacLaren JE. Pediatric procedural approach-avoidance coping and distress: A multitrait-multimethod analysis. Journal of Pediatric Psychology. 2004;29:131–141. doi: 10.1093/jpepsy/jsh016. [DOI] [PubMed] [Google Scholar]
- Beyer JE, Aradine CR. Content validity of an instrument to measure young children's perceptions of the intensity of their pain. Journal of Pediatric Nursing. 1986;1:386–395. [PubMed] [Google Scholar]
- Blount RL, Bachanas PJ, Powers SW, Cotter MC, Franklin A, Chaplin WC, et al. Training children to cope and parents to coach them during routine immunizations: Effects on child, parent, and staff behaviors. Behavior Therapy. 1992;23:689–705. [Google Scholar]
- Blount RL, Cohen LL, Frank NC, Bachanas PJ, Smith AJ, Manimala MR, et al. The Child-Adult Medical Procedure Interaction Scale-Revised: An assessment of validity. Journal of Pediatric Psychology. 1997;22:73–88. doi: 10.1093/jpepsy/22.1.73. [DOI] [PubMed] [Google Scholar]
- Borrelli L, Sepinwall D, Bellig AJ, Breger R, DeFrancesco C, Sharp EL, et al. A new tool to assess treatment fidelity and evaluation of treatment fidelity across 10 years of health behavior research. Journal of Consulting and Clinical Psychology. 2005;73:852–860. doi: 10.1037/0022-006X.73.5.852. [DOI] [PubMed] [Google Scholar]
- Cavender K, Goff MD, Hollon EC, Guzzetta CE. Parents' positioning and distracting children during venipuncture: Effects on children's pain, fear, and distress. Journal of Holistic Nursing. 2004;22:32–56. doi: 10.1177/0898010104263306. [DOI] [PubMed] [Google Scholar]
- Christensen J, Fatchett D. Promoting parental use of distraction and relaxation in pediatric oncology patients during invasive procedures. Journal of Pediatric Oncology Nursing. 2002;19:127–132. doi: 10.1177/104345420201900404. [DOI] [PubMed] [Google Scholar]
- Cohen LL, Blount RL, Cohen RJ, Ball CM, McClellan CB, Bernard RS. Children's expectations and memories of acute distress: Short- and long-term efficacy of pain management interventions. Journal of Pediatric Psychology. 2001;26:367–374. doi: 10.1093/jpepsy/26.6.367. [DOI] [PubMed] [Google Scholar]
- Dahlquist LM, Pendley JS, Landtrip DS, Jones CL, Steuber C. Distraction intervention for preschoolers undergoing intramuscular injections and subcutaneous port access. Health Psychology. 2002;21:94–99. [PubMed] [Google Scholar]
- Hanrahan K, McCarthy AM, Kleiber C, Lutgendorf S, Tsalikian E. Strategies for salivary cortisol collection and analysis in research with children. Applied Nursing Research. 2006;19(2):95–101. doi: 10.1016/j.apnr.2006.02.001. [DOI] [PubMed] [Google Scholar]
- Jay SM, Elliott C. Behavioral observation scales for measuring children's distress: The effects of increased methodological rigor. Journal of Consulting & Clinical Psychology. 1984;52:1106–1107. doi: 10.1037//0022-006x.52.6.1106. [DOI] [PubMed] [Google Scholar]
- Kazak AE, Penati B, Waibel MK, Blackall GF. The Perception of Procedures Questionnaire: Psychometric properties of a brief parent report measure of procedural distress. Journal of Pediatric Psychology. 1995;21:195–207. doi: 10.1093/jpepsy/21.2.195. [DOI] [PubMed] [Google Scholar]
- Kleiber C, Craft-Rosenberg M, Harper DC. Parents as distraction coaches during IV insertion: A randomized study. Journal of Pain and Symptom Management. 2001;22:851–861. doi: 10.1016/s0885-3924(01)00316-5. [DOI] [PubMed] [Google Scholar]
- Kleiber C, Harper DC. Effects of distraction on children's pain and distress during medical procedures: A meta-analysis. Nursing Research. 1999;48(1):44–49. doi: 10.1097/00006199-199901000-00007. [DOI] [PubMed] [Google Scholar]
- Kleiber C, McCarthy AM, Hanrahan K, Meyers L, Weathers N. Development of the Distraction Coaching Index. Children's Health Care. 2007;36:219–235. [Google Scholar]
- Lindgren SD, Koeppl GK. Assessing child behavior in a medical setting: Development of the Pediatric Behavior Scale. In: Prinz RJ, editor. Advances in behavioral assessment of children and families. III. Greenwich, CT: JAI; 1987. pp. 57–90. [Google Scholar]
- McCarthy AM, Hanrahan K, Kleiber C, Zimmerman MB, Lutgendorf S, Tsalikian E. Normative salivary cortisol values and responsivity in children. Applied Nursing Research. 2009;22(1):54–62. doi: 10.1016/j.apnr.2007.04.009. [DOI] [PMC free article] [PubMed] [Google Scholar]
- McCarthy AM, Kleiber C, Hanrahan K, Zimmerman MB, Westhus N, Allen S. Factors explaining children's responses to intravenous needle insertions. doi: 10.1097/NNR.0b013e3181f80ed5. Manuscript in preparation. [DOI] [PMC free article] [PubMed] [Google Scholar]
- McCarthy AM, Lindgren S, Mengeling M, Tsalikian E, Engvall J. Effects of diabetes on learning in children. Pediatrics. 2002;109:135. doi: 10.1542/peds.109.1.e9. [DOI] [PubMed] [Google Scholar]
- Salmon K, Pereira JK. Predicting children's response to an invasive medical investigation: The influence of effortful control and parent behavior. Journal of Pediatric Psychology. 2002;27:227–233. doi: 10.1093/jpepsy/27.3.227. [DOI] [PubMed] [Google Scholar]
- Salmon K, Price M, Pereira JK. Factors associated with young children's long-term recall of an invasive medical procedure: A preliminary investigation. Journal of Developmental & Behavioral Pediatrics. 2002;23:347–352. doi: 10.1097/00004703-200210000-00008. [DOI] [PubMed] [Google Scholar]
- Uman LS, Chambers CT, McGrath PJ, Kisely S. A systematic review of randomized controlled trials examining psychological interventions for needle-related procedural pain and distress in children and adolescents: An abbreviated Cochrane review. Journal of Pediatric Psychology. 2008;33:842–854. doi: 10.1093/jpepsy/jsn031. [DOI] [PMC free article] [PubMed] [Google Scholar]
- von Baeyer CL, Marche TA, Rocha EM, Salmon K. Children's memory for pain: Overview and implications for practice. Journal of Pain. 2004;5:241–249. doi: 10.1016/j.jpain.2004.05.001. [DOI] [PubMed] [Google Scholar]