Journal of Pediatric Psychology, Vol. 26, No. 8, 2001, pp. 513-523
© 2001 Society of Pediatric Psychology
Reliability and Validity of the Brief Behavioral Distress Scale: A Measure of Children's Distress During Invasive Medical Procedures
1 Kennedy Krieger Institute, 2 Johns Hopkins University School of Medicine, 3 University of Maryland Baltimore County
All correspondence should be sent to Cindy L. Tucker, Department of Behavioral Psychology, The Kennedy Krieger Institute, 707 N. Broadway, Baltimore, Maryland 21205. E-mail: tuckerc{at}kennedykrieger.org .
 |
Abstract |
|---|
 Top Abstract IntroductionMethodResultsDiscussionReferences |
|---|
Objective: To evaluate the reliability and validity of a new observational measure of children's procedure-related distress behaviors, the Brief Behavioral Distress Scale (BBDS), to provide clinicians with an efficient, economical alternative measure that does not depend on continuous interval coding.
Methods: Forty-eight randomly selected videotaped invasive medical procedures performed on children (ages 2 to 10 years) with chronic illness were coded with the BBDS and the Observation Scale of Behavioral Distress (OSBD). Reliability and validity analyses along with item analysis were conducted.
Results: Total distress scores of the BBDS were highly correlated with six of seven concurrent validity measures from multiple sources (i.e., OSBD, parent ratings, two nurse ratings, child self-report, and a physiological arousal measure, heart rate) (range r =.57-.76, p <.001-.0001). A robust association was found between the BBDS distress scores and OSBD total distress scores (r =.72, p <.0001). For two concurrent validity measures, the BBDS demonstrated stronger associations than did the OSBD. Interrater reliability was high for each BBDS distress behavior category.
Conclusions: Based on the findings reported, the BBDS is a reliable and valid measure of children's procedure-related distress with functional utility in both research and clinical settings.
Key words: children's pain; procedure-related distress; rating scale; invasive medical procedures.
|
Introduction |
|---|
|
TopAbstractIntroductionMethodResultsDiscussionReferences |
|---|
Children with chronic medical conditions are surviving in increasing numbers, yet their quality of life may be diminished by the need for repeated invasive medical procedures (Dahlquist, 1999
; Wallander & Varni,
1992). Painful needle sticks can cause significant and persistent
conditioned fear and anxiety for children, with little reported evidence for
habituation over time (Jay, Ozolins,
Elliott, & Caldwell, 1983). Procedure-related distress often
interferes with cooperation and jeopardizes the safe and timely delivery of
essential medical care. Medical personnel may choose to manage highly
distressed children with increased sedation, physical restraint, and, in some
cases, general anesthesia, with its inherent costs and risks and further
disruption of daily life (Anderson,
Zeltzer, & Fanurik, 1993). In a recent review, Powers
(1999) documented dramatic
improvements in the treatment of procedure-related distress in children over
the past two decades. Concurrently, observational assessment of children's
procedure-related distress also has improved.
From the seminal efforts on the development of the Procedure Behavior
Rating Scale (PBRS; Katz, Kellerman, &
Siegel, 1980) have emerged several related versions of
observational measures, including the Observation Scale of Behavioral Distress
(OSBD; Jay et al., 1983;
Elliott, Jay, & Woody,
1987) and the Child-Adult Medical Procedures Interaction Scale
(CAMPIS, CAMPIS-R; Blount et al.,
1989,
1997). The original version of
the OSBD (Jay et al.) remains the most commonly used outcome measure in
procedure-related distress treatment efficacy studies
(Powers, 1999). The OSBD
consists of 11 operationally defined behaviors associated with the pediatric
procedure-related distress constructs, anxiety and pain, with severity
assigned by expert-determined weights. The established strengths of the OSBD
have been widely reported.
The OSBD, however, is not without limitations. First, definitions of the
phases of the medical procedure (i.e., anticipatory, procedural, recovery)
vary considerably across studies (Blount,
Sturges, & Powers, 1990). Even when procedural phases are
consistently defined, differences or delays in execution of the needle
procedure, which often are independent of child behavior, can alter procedure
duration, thereby increasing or decreasing the number of observation intervals
and potentially deflating or inflating the resulting scores. Thus, continuous
interval coding systems using time-based phase definitions
(Dahlquist, Power, & Carlson,
1995; Jay et al.,
1983) may result in comparisons of nonequivalent procedural events
across subjects or within subjects across observations. Children may be
confronted with varying quantities and types of demands, blurring potentially
significant interactions relevant to treatment design. Alternatively, one can
focus on the essential demands for compliance placed on the child and record
the type and severity of distress exhibited in response to each specific
demand as a discrete step. Essential demands are the specific demands
for compliance sequentially placed on the child that are required for
procedure completion. Using this method, interventions aimed at distress
reduction in response to the specific demands of each step can be evaluated
and refined. This is consistent with Blount et al.
(1990) and Katz et al.,
(1980), who examined steps
specific to bone marrow aspiration (BMA), and Manne, Bakeman, Jacobsen, and
Redd (1993), who examined the
steps of venipunctures. Our research developed a measure that can be used
across a variety of needle stick procedures with a discrete-step format, which
affords greater measurement efficiency and more details on child behavior in
relation to specific social and environmental antecedents within the needle
stick routine.
Finally, because OSBD scores are derived from summing across all 11 distress behaviors, clinically relevant data about the type of distress being exhibited are lost. Conversely, as in this study, distress behaviors can be sorted conceptually into functional topographies based on the degree to which they affect procedure performance. Guided by the established OSBD severity weights, we classified the 11 distress behaviors into three categories of increasing intensity with respect to interference with medical procedures. Additional distress behaviors were added to complement the intent of each category (e.g., moaning, flinching, disruption of medical materials). A category for assessing children's coping responses during invasive procedures, active coping responses, also was included.
This article presents preliminary reliability and validity analyses of the Brief Behavioral Distress Scale (BBDS), a measure combining behavioral categories derived from the OSBD and an essential demands approach as an alternative for measuring child procedure-related distress and coping. We expected BBDS distress scores to be significantly positively correlated with OSBD total distress scores and concurrent validity measures of child distress (child pain ratings, parent ratings of child fear and pain, nurse ratings of child distress, and heart rate) and significantly negatively correlated with nurse ratings of child cooperation. Conversely, BBDS active coping responses scores were expected to correlate negatively with concurrent measures of child distress and positively with nurse ratings of child cooperation.
|
Method |
|---|
|
TopAbstractIntroductionMethodResultsDiscussionReferences |
|---|
Participants and Setting
Participants were a consecutive sample of 40 children diagnosed with a chronic illness and receiving repeated invasive medical procedures at a tertiary care pediatric hospital over a 2-year period. All participants were part of an ongoing longitudinal study designed to compare and evaluate the effects of different treatment approaches on children's procedure-related distress (NICHD, HD#35528, Behavioral Medicine Approaches to Pediatric Acute Pain, K. Slifer, P.I.). A member of the research team met with potential participants, explained the study procedures, and obtained informed consent from the parent and assent from the child. The 40 participants included 12 girls (30%) and 28 boys (70%) whose ages ranged from 2 years, 2 months to 10 years, 8 months (M = 6 years, 0 months). Sixty percent of the participants were Caucasian, 35% were African American, 2.5% were Hispanic, and 2.5% were Asian Indian. Twenty-one children receive one or more type of invasive procedure (i.e., implanted port access [PCA], venipuncture [IV], intramuscular injection [IM], subcutaneous injections [SQ]) for treatment of childhood cancer, specifically for childhood leukemia (57%) or solid tumor (43%). Nine children receive monthly chronic transfusions for treatment of a hematological disorder (i.e., sickle cell anemia, thalasemia). The remaining nine children receive an invasive procedure (i.e., PCA, IV, or SQ) for monitoring and treatment of a chronic medical condition (e.g., idiopathic thrombocytopenia, growth hormone insufficiency, immunodeficiency).
Procedure
We had an available archive of 360 observations videotaped over a 2-year
period. The 40 participants were observed at scheduled clinic visits and
followed from baseline, into nine treatment sessions (if so randomized) and
through a follow-up phase of approximately 6 months. The mean number of
observations collected per participant was 9.47, ranging from 2 to 27.
Videotaping began when the child entered the procedure room and continued for
3 to 4 minutes after the procedure. The filmer did not speak to the child once
videotaping began. The taped sessions were transcribed, coded, and scored
according to the procedure described by Jay et al.
(1983) for one of the outcome
measures, the OSBD. Procedure films ranged from 3 to 22 minutes (M =
10.0 min., SD = 6.4); specifically, M = 6.3 min.,
SD = 2.9 for IMs and SQs; M = 9.2 min., SD = 5.1
for PCAs; and M = 10.2 min., SD = 5.2 for IVs.
A random sample of 54 videotaped observations (15%) was selected for recoding with the BBDS. Of the nine BBDS coders, two coders had prior OSBD training, three coders had some familiarity with the OSBD and coding of pediatric invasive procedures, and four coders were unfamiliar with the OSBD and inexperienced in procedural distress coding. Coders were provided with approximately 1 hour of training on observational coding of pediatric procedural distress along with a review of the behavioral categories and operational definitions in the BBDS. As a group, coders reviewed three videotaped pediatric invasive procedures, discussed discrepancies, determined decision rules, and clarified behavioral definitions. A final version of the behavioral category definitions was provided to each coder with instructions for independent coding. Each coder was assigned six videotaped observations for recoding with the BBDS and given the following instructions: (1) view the procedure and indicate the sequence of each step observed on the coding form; (2) write in any additional steps observed and number as above; (3) during the duration of each observed step, indicate the occurrence of each of the four behavioral categories with a check in the appropriate box; and (4) if a step occurs more than once, combine multiple observations in the row listing the step so that all occurrences of a behavior during any performance of the step are included in the same box.
Due to scheduling conflicts, four coders completed only five observations.
Two observations could not be coded because of film quality. This resulted in
a set of 48 recoded procedures that reflected the distribution of procedures
in the overall sample: 20 IVs, 20 PCAs, 5 IMs, and 3 SQs. The final set of 48
observations included procedures from 87.5% (35) of the total 40 participants,
with multiple observations (across phases) randomly selected from 13 (37%) of
the 35 participants in the sample. Thirty-one of the procedures occurred
during baseline, 12 during treatment, and 5 during follow-up. The participants
had been randomly assigned to either a distraction
(Pringle et al., in press),
counter-conditioning (Slifer, Babbitt,
& Cataldo, 1995), or nondirective attention control
intervention group. During each of the nine treatment sessions, participants
and their caregivers met with a therapist for 20 minutes prior to scheduled
invasive procedures.
A topical cream mixture of lidocaine and prilocaine, eutectic mixture of local anesthetics, (EMLA®, Astra Pharmaceutical Products, Inc., Westborough, MA 01581) was used to numb the needle site prior to 30 procedures. Intradermal lidocaine was used prior to two of the IV procedures. The remaining 16 procedures were performed without the use of any pharmacological agent.
Measures
Brief Behavioral Distress Scale
The BBDS was adapted from the behavioral observation method described by
Slifer et al. (1995), who used
it to measure children's cooperation and procedure-related distress with
discrete component steps involved in several types of invasive medical
procedures. By breaking a procedure into discrete steps, one can make
observations relative to the specific demands for compliance sequentially
placed on the child during the procedure. These steps constitute the essential
demands of the invasive procedure (Blount
et al., 1990). A variety of needle stick procedures was observed
and analyzed into a series of steps essential to procedure completion. Twelve
possible steps were identified. Steps were defined by their onset and offset,
without regard to sequence or duration. Observers recorded the occurrence or
nonoccurrence of the target behaviors during each step of the procedure.
Descriptions and operational definitions of target behaviors for each category
are presented below.
Noninterfering Distress Behaviors. This category includes behaviors indicative of mild or partially regulated procedural distress. The child displays behaviors that do not delay or interfere with the performance of the procedure such as crying, whining, groaning, moaning, negative vocalizations (i.e., "Stop," "I don't want it," "Wait, I'm not ready yet") that are not above conversational level, or seeking physical comfort from another person or object.
Potentially Interfering Distress Behaviors. These behaviors indicate moderate intensity distress that is being less well regulated by the child and that do not necessarily disrupt performance of the procedure but may delay, lengthen, or make it more difficult to perform. This category includes behaviors such as screaming, yelling, shouting, negative vocalizations (i.e., "Stop," "I don't want it," "Wait, I'm not ready yet") that are above conversational level, obvious tensing of muscles, facial contortions, jaw clenching, tooth grinding or gritting, tightly clenched fists (not in response to being told to "make a fist"), or flinching.
Interfering Distress Behaviors. These behaviors result in the procedure being halted, delayed, or performed using physical restraint and reflect a severe degree of distress and protest with little to no ability (or effort) by the child to regulate his or her behavior. This category includes attempted or completed behaviors such as escape or avoidance (i.e., getting up from seat, leaving room, leaning or pulling away from nurse, resisting or moving out of position, hiding, or blocking the procedure site), disruptions of medical materials, or physical aggression.
Active Coping Response Without Verbal Delay. When a health care provider attempts to perform a step in an invasive procedure, the child engages in an observable coping response such as playing with a toy or activity by pressing buttons, pointing to, or verbalizing about it; counting aloud; singing; using focused breathing or relaxation strategies; audible praying; initiating or engaging in a nonmedical conversation (beyond single-word responses); viewing a preferred videotape; making positive coping statements (i.e.; "It will be over soon," "It didn't hurt"); asking questions about the procedure (e.g., "What are you doing?" "When will it be done?"); or holding, stroking, or patting an object (i.e., blanket, Teddy, pacifier) or person. Active coping response is not scored for any step in which the behavior is accompanied by a verbal delay statement, such as, "Don't do it yet, I need to pray first," or "Wait until I am ready to count." Otherwise, active coping may be scored simultaneously with any distress behavior.
Asking questions about the procedure was included based on findings that
information seeking has been significantly negatively correlated with
high procedural anxiety or distress by some investigators
(Katz et al., 1980;
Manne et al., 1993;
Peterson & Toler, 1986).
This is a departure from the OSBD and the CAMPIS scales, both of which include
information seeking as a distress behavior
(Blount et al., 1989;
Jay et al., 1983).
Notwithstanding this disagreement in the literature, asking questions was
included here as an observable though not necessarily effective effort by the
child to cope, unless accompanied by a verbal delay statement.
The total number of steps occurring during each procedure was noted (M = 9 steps, range = 6-12). The number of steps in which a behavior occurred was summed and divided by the total number of steps, then multiplied by 100 to yield percentage of steps with occurrence scores for each behavioral category. The number of steps in which behaviors from any of the three distress behavior categories occurred (i.e.; noninterfering, potentially interfering, and interfering) was summed and divided by the total number of steps multiplied by three (number of distress categories), then multiplied by 100 to obtain the Total Distress Score. Use of percentage scores allowed for a common metric for comparing scores from procedures with different number of steps. The Active Coping Response Score was treated as a separate subscale. Coding of the four behavioral categories of the BBDS generally took 10 to 15 minutes per observation, with coding being completed in one or, infrequently, two passes of the videotaped procedure.
Observational Scale of Behavioral Distress (OSBD)
Overt child distress was measured using the OSBD
(Jay et al., 1983). Written
transcripts of all vocalizations occurring during the filmed procedure were
produced to improve interrater reliability. From the videotape and
transcription, observers coded for the presence or absence of 11 child
distress behaviors (e.g., crying, verbalizing fear) during 15-second
continuous intervals. Distress scores were calculated by summing the number of
15-second intervals during which each distress behavior occurred and
multiplying that number by an expert-determined intensity weight from 1 to 4
reflecting the severity of the behavior. A Total Distress Score was
calculated by summing the weighted distress scores and dividing by the total
number of 15-second intervals for the procedure. Transcription of videotapes
took approximately 2 hours to produce. Coding of the OSBD usually took 30 to
45 minutes, generally requiring three to four passes of the videotape. For
this study, 18% of the 360 videotaped sessions were independently coded by a
second observer, yielding a mean kappa coefficient of.76.
Interrater reliabilities reported in the literature for the OSBD are
uniformly good. Percentage agreement scores range from 72% to 99% for total
distress (Jay et al., 1983),
and kappa coefficients of.90 and.87 have been reported respectively for the
procedural and recovery phases (Gonzalez et al., 1989). OSBD distress scores
have been significantly correlated with other indices of children's procedural
pain and distress, such as parent and nurse ratings of child distress,
children's ratings of fear, and physiological measures of distress
(Elliott et al., 1987;
Jay et al., 1983;
Jay, Elliott, Katz, & Siegel,
1987). The OSBD also has been shown to be sensitive to changes in
children's procedural distress after psychological intervention (e.g.,
Dahlquist, 1992;
Jay et al., 1987;
Pringle et al., in press).
Child Self-Report Ratings of Pain or Discomfort
Immediately after the invasive procedure, children were asked by research
staff to indicate the degree to which they experienced discomfort. Three- and
4-year-olds were assessed by means of the Oucher Scale
(Beyers, Villarruel, & Denyes,
1992), a posterlike instrument that produces ordinal data ranging
from 0 to 5 when only the photographic scale is used. These six photographs
are available in Caucasian, African American, and Hispanic versions. Studies
of the convergent, discriminant, and construct validity of the Oucher indicate
good to excellent validity (e.g., r =.70-.98) and test-retest
reliability at a moderate level (e.g., r =.54-.72). As directed in
the manual, a Piagetian seriation task was performed prior to use of the
Oucher with 3- and 4-year-olds to determine adequate cognitive understanding
and maturity for task completion. No self-report ratings of discomfort were
obtained for children under the age of 3 years (n = 3) or aged 3 or 4
years who failed the cognitive screening test (n = 2).
Children who were 5 years old or older were assessed using the Coloured
Analogue Scale (CAS; McGrath et al.,
1996). The CAS assesses children's degree of subjective discomfort
by adjusting a sliding plastic marker along a visual analogue continuum that
ranges from "most pain" to "no pain" and varies in
color, area, and length. McGrath et al.
(1996) reported adequate
content and construct validity as well as test-retest reliability with
children 5 to 16 years of age. In order to make the Oucher data comparable to
the ordinal data obtained with the CAS, which range from 0 to 10, we
transformed the Oucher data by multiplying each rating by 2. The
transformation does not guarantee equivalency from a psychometric standpoint,
but does allow for assessment of children in the younger age group.
Eight children (M = 66 months, range = 39-106), did not respond when asked to rate their discomfort. These children were observed to remain distressed and withdrawn from clinic staff and researchers following the procedure, thereby likely accounting for their lack of response. In total, self-report data for 35 children ages 38 to 119 months (M = 73) were available for analysis.
Parent Ratings of Child Fear and Pain
The parent present during the observed invasive medical procedure was asked
just prior to the procedure to rate his or her child's degree of fear about
the impending procedure using a Visual Analogue Scale (VAS). End-points of the
10-centimeter scale were labeled Not Afraid and Very Afraid.
Immediately following the procedure, the parent was asked to rate the degree
to which his or her child experienced pain during the invasive procedure. The
endpoints of this 10-centimeter VAS were labeled No Pain and Most
Pain (Powers, Blount, Bachanas,
Cotter, & Swan, 1993).
Nurse Ratings of Child Distress and Cooperation
The nurse or other health care provider performing the invasive medical
procedure was asked to use a VAS to rate the child's degree of observed
distress and degree of cooperation during the just completed procedure. The
end-points of the 10 centimeter scale for distress were labeled Not
Distressed and Very Distressed, and for cooperation, Not
Cooperative and Very Cooperative
(Powers et al., 1993).
Autonomic Physical Arousal Measure
To obtain an indicator of autonomic nervous system arousal, the child's
heart rate (HR) was measured using a portable electrocardiogram (ECG) quality
event recorder (Heart Access PlusTM, Paceart Associates, L. P.,
Fairfield, NJ 07004) attached to the child's chest via two neonatal-sized
adhesive electrodes, (e.g., Huggables®). A 10-second ECG recording was
obtained by depressing the activator button on the device at five
predetermined sampling points during each observed procedure. Mean HRs across
three of those time sampling points were analyzed for this study:
specifically, (1) anticipatory HR, when the injection site is cleansed or
covering tape or dressing is removed; (2) procedural HR, when the needle first
enters the skin; and (3) recovery HR, 30 seconds after the needle exits the
skin or tape is applied to secure the needle at the insertion site. ECG
recordings then were downloaded to a computer database where they were stored
and analyzed for HR readings (in beats per minute). HR data were missing for
13 (27%) cases due to high levels of artifact on the ECG recordings or child
or parent refusal.
|
Results |
|---|
|
TopAbstractIntroductionMethodResultsDiscussionReferences |
|---|
Interrater Reliability
For 20 randomly selected observations, two observers independently recorded the occurrence of target behaviors for each step of the procedure. Pearson product-moment correlation analyses conducted between the pairs of observers' BBDS scores yielded an r (19) of.87 (range across categories =.85-.93). Percent agreement scores (agreement/ agreement plus disagreement x 100) for BBDS total distress scores ranged from 65% to 100%, (M = 87%, SD = 9.75). Percent agreement scores for individual behavioral categories were as follows: 93% (SD = 11.26) for interfering distress, 85% (SD = 17.07) for potentially interfering distress, 84% (SD = 14.93) for noninterfering distress, and 86% (SD = 21.05) for active coping responses. Kappa coefficients were.87 for interfering distress,.68 for potentially interfering distress,.68 for noninterfering distress,.74 for BBDS total distress, and.72 for active coping responses.
Item Analysis
We conducted an item analysis on the four behavioral categories comprising
the BBDS. Criteria for retaining a behavioral category were (1) occurrence of
behaviors in at least 10% of the sample, (2) correlation coefficients of +.3
or greater between each distress category and the total distress score, (3)
correlation coefficients of +.3 or greater with at least one other distress
category, and (4) kappa coefficients of at least.6
(Elliott et al., 1987;
Nunnally, 1978).
Interfering distress behaviors occurred in 27 (56.3%) of 48 procedures and
correlated significantly with BBDS total distress (r [47] =.79,
p <.0001) and potentially interfering distress (r [47]
=.71, p <.0001). Behaviors from the potentially interfering
distress category occurred in 36 (75%) of the procedures and correlated
significantly with BBDS total distress (r [47] =.90, p
<.0001) and non-interfering distress (r [47] =.41, p
<.004). Noninterfering distress behaviors were coded for 42 (88%) of the
procedures and correlated significantly with BBDS total distress (r
[47] =.69, p <.0001). No significant correlation was found between
interfering and noninterfering distress behaviors (r [47] = -.225,
p =.125). Cronbach's coefficient alpha test of internal consistency
revealed a coefficient of.71 for BBDS total distress. This is comparable with
the coefficient alpha of.72 reported for the OSBD by Elliott et al.
(1987).
The behavioral category, active coping responses, was coded for 30 (62.5%) of the procedures. Coping responses correlated with BBDS total distress scores in the expected direction (r [47] = -.29, p <.046) but did not produce significant correlations with any of the three distress behavior categories (range r = -.20 to -.27, p <.173 to.063). Given the lack of significant negative correlation with any of the three BBDS distress behavior category scores, the active coping responses category was not retained as part of the BBDS.
Validity
Comparisons were performed to determine the construct and concurrent
validity of the BBDS as compared to the frequently used and well-established
measure of child behavioral distress, the OSBD.
Table I displays the Pearson
product-moment correlation coefficients between concurrent validity measures
and both BBDS and OSBD total distress scores as well as BBDS distress behavior
scores. Bonferroni corrections (.05/11 = p <.005) were used to set
alpha levels for significance to reduce the risk of Type I error. Results
indicate that BBDS distress scores were significantly positively correlated
with OSBD total distress. Both BBDS and OSBD total distress scores were
significantly associated with all concurrent validity measures with one
exception, the parent rating of child's fear of procedure.
|
Interfering distress behaviors were significantly positively correlated with three of five concurrent validity measures of child distress. Potentially interfering distress behaviors were significantly positively correlated with four of five concurrent validity measures of child distress. Noninterfering distress behaviors were significantly positively correlated with two of five concurrent validity measures of child distress. All three distress behaviors were significantly negatively correlated with nurse reports of child cooperation.
When the correlation coefficients for BBDS and OSBD total distress scores with the concurrent validity measures were compared (see Table I), individual r to z transformations indicated that BBDS total distress scores were more strongly related to nurse ratings of both child distress and cooperation than were OSBD total distress scores (z [0,1] = 2.39, p <.008; z [0,1] = 2.14, p <.016, respectively). Correlation coefficients for the four remaining validity measures (i.e., child self-report rating of pain, parent rating of child's fear, parent rating of child's pain, mean heart rate) were essentially identical.
A Pearson product-moment correlation coefficient revealed no association
between child's age and observed distress as measured by BBDS total distress
scores (r [30] = -.19, p <.302). This finding is in
contrast to the significant age effects typically found using the OSBD and
other similar observational measures of child procedure-related distress
(e.g., Blount, Landolf-Frische, Powers, & Sturges, 1991;
Jay et al., 1983;
Katz et al., 1980).
Frequency of Occurrences of Distress Behaviors During Essential
Procedural Demands
Table II displays the
frequency of occurrence of each procedural step along with the distribution of
occurrence of distress behaviors from each category. Interfering distress
behaviors occurred most frequently during positioning of the child and
exposing the injection site, needle manipulation, and needle insertion.
Potentially interfering distress behaviors occurred most frequently during
needle insertion and manipulation. Noninterfering distress behaviors occurred
most frequently during needle manipulation, securing the needle with tape, and
positioning and exposing the site. The steps, gauze or bandage application,
replacement of clothing, and needle removal produced the fewest occurrences of
behavioral distress.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMethodResultsDiscussionReferences |
|---|
Based on the results of this initial study, we found the BBDS to be a reliable and valid measure of children's observable procedure-related distress with some unique features in relation to a well-established children's distress measure, the OSBD. BBDS total distress scores were highly correlated with six of seven concurrent validity measures from multiple sources. A robust association was found between BBDS and OSBD total distress scores.
For two concurrent validity measures, nurse ratings of child cooperation and distress during procedures, the BBDS demonstrated stronger associations than did the OSBD. Improved validity with nurse ratings may be related to the conceptual distress categories employed by the BBDS. These categories specifically relate to the degree to which the child's distress behavior may interfere with completion of a necessary medical procedure. The ability to complete a procedure without delays or physical restraint is a highly salient feature for nurses, which the BBDS distress categories are uniquely designed to measure.
Validity for the three BBDS distress behavior categories was supported with
all significant correlations in the expected direction. BBDS distress
behaviors were most strongly associated with OSBD total distress scores and
nurse reports of child distress and cooperation. Neither BBDS total distress
scores nor any of the distress behavior scores were significantly associated
with parent reports of children's fear prior to the procedure. This is
consistent with lower associations between parent ratings and observed
distress reported by Manne, Jacobsen, and Redd
(1992). Overall, the more
overt distress behaviors, interfering and potentially interfering, were more
strongly associated with a greater number of concurrent validity measures as
compared to non-interfering distress behaviors, which may reflect behaviors
considered by raters to be nondisruptive, justifiable expressions of
discomfort or protest (Slifer et al.,
1995). The physical nature of the escape behaviors comprising the
interfering distress category likely accounts for the particularly strong
association found with the physiological marker, heart rate.
No significant correlations were found between the category, active coping
responses, and the OSBD or other concurrent validity measures. The operational
definition of this category was purposely broad to account for all observable
efforts employed by children to cope with procedural distress. Blount et al.
(1997) discussed this approach
in contrast to their own more conservative definition of coping behaviors as
only those inversely correlated with distress. The dilemma revolves around the
logical paradox in which distress produces the need for coping, while
simultaneously "coping" implies alleviation of distress. Broader
definitions including attempts to cope that may or may not
successfully ameliorate distress are by definition more likely to result in no
or positive associations with observed distress
(Manne et al., 1993;
Siegel & Smith, 1991). In
this study, the behavioral category, active coping, was determined by item
analysis not to meet the criteria for retention in the BBDS. The lack of
significant association with validity measures further supports this decision.
In future studies, this positive behavioral category could be redefined to
include performance of responses on which the child has received specific
training, thereby serving as a measure of treatment integrity rather than of
child coping, which is better assessed by measures specifically designed to
assess child coping and its associated factors, such as the CAMPIS-R
(Blount et al., 1997).
No association was found between child's age and degree of observed
procedure-related distress in contrast to the findings of other studies (e.g.,
Blount et al., 1991;
Jay et al., 1983;
Katz et al., 1980). Carr,
Lemanek, and Armstrong (1998)
attributed reported age difference in observed distress, in part, to the
overrepresentation of "behaviors typical of younger children (e.g.,
crying, screaming, flailing)" (p. 305) in observational distress
measures rather than to actual differences in experienced distress. By adding
two types of distress behaviors more commonly observed in older children,
flinch and groan, LaBaron and Zeltzer (1984) found that age differences were
eliminated. The distress categories of the BBDS, which include these
behaviors, may work to reduce the age bias typically reported for other
observational measures.
The use of an essential demands approach for measuring child distress offers the BBDS some advantages compared to continuous interval coding of observations. In clinical practice, analysis of distress in response to essential procedural demands allows clinicians to examine areas of particular difficulty for individual children, targeting interventions to the specific social and medical stimuli associated with each discrete procedural step. For instance, the broadest and most disruptive distress behaviors among this sample were observed during positioning for the procedure and exposure of the injection site. This essential demand has not been identified previously as a particularly distressing demand for young children undergoing invasive procedures. Yet, it likely represents a stimulus that, for many children, signals the impending aversive procedure. This step may be missed as a focus for specific intervention due to the practice of defining the anticipatory phase by a standard number of intervals prior to site cleaning or some other specified antecedent event. An unforeseen delay between positioning or site exposure and the event defining the start of the procedural phase would result in the loss of clinically important observational data related to these often distressing antecedent demands.
Blount et al. (1990)
examined distress and coping in response to the steps of a BMA and found that
quantity and type of distress behavior differed significantly across phases of
the procedure. In this study, similar peaks and decreases in distress behavior
appeared at corresponding steps in the procedures. These results, combined
with previous reports revealing demand-specific variations in the levels and
types of child distress observed, suggest continuing research to evaluate
interventions specific to distress at these predictably critical points in
needle stick procedures.
The BBDS is offered as a potential alternative to more complex continuous interval coding systems. Further, we propose that the BBDS should be useful as a coding system during the actual performance of the procedure. To test this proposition, validity and reliability analyses of the BBDS must be performed in real time as well as on videotaped observations and the data compared. Should the findings of this study not be replicated in future research with in vivo coding, the time savings of coding the BBDS from videotape would nevertheless be significant in daily practice. In this study, coding a procedure from videotape using the BBDS took approximately one-third to one-quarter the amount of time required to code the same procedure with the OSBD (10 to 15 minutes vs. 30 to 45 minutes, respectively). When including time for transcribing vocalizations from the videotapes, the time savings of using the BBDS increase appreciably from 150 minutes (2 hours for transcribing plus 30 to 45 minutes for coding) for OSBD coding to 10 to 15 minutes for BBDS coding, a tenfold reduction. All of the OSBD coders and half of the BBDS coders in this study were moderately to well-experienced in the observation and/or coding of pediatric invasive procedures. This fact may limit the generalizability of these time savings for the BBDS when used by inexperienced coders in a clinical setting.
Although providing support for the reliability and concurrent validity of the BBDS, this preliminary investigation has several limitations. First, all the children studied received treatment at the same medical facility, albeit from personnel staffing five different pediatric specialty clinics. Child participants were diagnosed with a variety of chronic conditions ranging from mild and nonlethal to serious and life-threatening. The procedures studied did not include those typically reported to cause the greatest discomfort for children, BMAs and lumbar punctures. Children from ages 2 to 10 years were included, spanning several cognitive stages of development. It is uncertain if the findings here would be replicated if children from different medical facilities, with more homogeneous chronic conditions, receiving other procedures, and from different age ranges were studied. Several additional studies with improved methodology and larger samples will add to information on the potential utility of the BBDS. For instance, future validity research should avoid the use of repeated procedures and focus on procedures occurring during only one condition (i.e., baseline). Hypotheses regarding coder experience and ease of use should be tested. Operational definition revision, and reliability and validity testing of the coping responses category are needed prior to further use.
The OSBD continues to be the most frequently used outcome measure in the
procedure-related distress literature
(Powers, 1999). However, it is
costly in terms of time, staffing, and training, with a complicated coding and
scoring system heavily reliant on vocalization content and subtle indices of
distress. As the developers state, the OSBD may be most useful for researchers
who are well-funded with access to ample resources (e.g., staff and equipment
for intensive analyses of videotapes) and interested in detecting subtle
differences in individual distress behaviors
(Elliott et al., 1987). The
same holds true for the continuous-interval coded CAMPIS-R that allows for
complex sequential analyses of staff- and parent-child interactions in terms
of distress and coping promotion (Blount et
al., 1997). However, for researchers with modest resources or
clinicians who wish to use session-by-session data to evaluate treatment
effectiveness, the BBDS appears to afford an effective and economical
alternative for measuring children's procedure-related distress based on these
preliminary reliability and validity findings.
|
Acknowledgments |
|---|
We thank the parents and children who agreed to participate in this study, the staff of the Johns Hopkins Pediatric Oncology, Hematology, Endocrine, and Immunology services for allowing us to work with their patients; Michael F. Cataldo, PhD, for his administrative support, and Adrianna Amari for her helpful comments on an earlier draft of this manuscript. We thank Lisa Hilley, Wendy Sulc, and the other research assistants of the Pediatric Psychology Lab at the University of Maryland Baltimore County for OSBD coding, and Adrianna Amari, Melissa Beck, Suzanne Busby, Tanya Diver, Stephanie Eischen, Alana Kane, Sharon McDonnell, Christopher Sarampote, and Ana Sobel for their assistance with data collection and coding. This study was supported in part by the National Institute of Child Health and Human Development, Grant#HD 35528, K. Slifer, P.I.
Received June 30, 2000; revision received December 7, 2000; revision received March 15, 2001; accepted April 20, 2001
|
References |
|---|
|
TopAbstractIntroductionMethodResultsDiscussionReferences |
|---|
Anderson, C. T. M., Zeltzer, L. K., & Fanurik, D. (1993). Procedural pain. In N. L. Schechter, C. V. Berde, & M. Yaster (Eds.), Pain in infants, children and adolescents (pp. 435-458). Baltimore, MD: Williams & Wilkins.
Beyers, J. E., Villarruel, A. M., & Denyes, M. (1992). The creation, calibration, continuing development of the Oucher: A measure of pain intensity in children. Journal of Pediatric Nursing, 7, 335-346.[Medline]
Blount, R. L., Cohen, L. L., Frank, N. C., Bachanas, P. J., Smith,
A. J., Manimala, M. R., & Pate, J. T. (1997). The Child-Adult
Medical Procedure Interaction Scale—Revised: An assessment of validity.
Journal of Pediatric Psychology,
22, 73-88.
Blount, R. L., Corbin, S. M., Sturges, J. W., Wolfe, V. V., Prater, J. M., & James, L. D. (1989). The relationship between adults' behavior and child coping and distress during BMA/LP procedures: A sequential analysis. Behavior Therapy, 20, 585-601.[ISI]
Blount, R. L., Landolf-Fritsche, B., Powers, S. W., & Sturges,
J. W. (1991). Differences between high and low coping children
and between parent and staff behaviors during painful medical procedures.
Journal of Pediatric Psychology,
16, 795-809.
Blount, R. L., Sturges, J. W., & Powers, S. W. (1990). Analysis of child and adult behavioral variations by phase of medical procedure. Behavior Therapy, 21, 33-48.
Carr, T. D., Lemanek, K. L., & Armstrong, D. (1998). Pain and fear ratings: Clinical implications of age and gender differences. Journal of Pain Symptom Management, 15, 305-313.
Dahlquist, L. M. (1992). Coping with aversive medical treatments. In A. M. LaGreca, L. J. Siegel, J. L. Wallander, & E. Walker (Eds.), Stress and coping in child health (pp. 345-376). New York: Guilford.
Dahlquist, L. M. (1999). Commentary on
"Treatments that Work in Pediatric Psychology: Procedure-Related
Pain." Journal of Pediatric Psychology,
24, 153-154.
Dahlquist, L. M., Power, T. G., & Carlson, L.
(1995). Physician and parent behavior during invasive pediatric
cancer procedures: Relationships to child behavioral distress.
Journal of Pediatric Psychology,
20, 477-490.
Elliott, C. H., Jay, S. M., & Woody, P. (1987). An
observation scale for measuring children's distress during medical procedures.
Journal of Pediatric Psychology,
12, 543-551.
Gonzales, J. C., Routh, D. K., Saab, P. G., Armstrong, F. D.,
Shifman, L., Guerra, E., & Fawcett, N. (1989). Effects of
parent presence on children's reactions to injection: Behavioral,
physiological, and subjective aspects. Journal of Pediatric
Psychology, 14,
449-462.
Jay, S. M., Elliott, C. H., Katz, E., & Siegel, S. E. (1987). Cognitive-behavioral and pharmacological interventions for children's' distress during painful medical procedures. Journal of Consulting and Clinical Psychology, 55, 860-865.[ISI][Medline]
Jay, S. M., Ozolins, M., Elliott, C. H., & Caldwell. S. (1983). Assessment of children's distress during painful medical procedures. Health Psychology, 2, 133-147.
Katz, E. R., Kellerman, J., & Siegel, S. E. (1980). Behavioral distress in children with cancer undergoing medical procedures: Developmental consideration. Journal of Consulting and Clinical Psychology, 48, 356-365.[ISI][Medline]
LeBaron, S., & Zeltzer, L. K. (1984). Assessment of acute pain and anxiety in children and adolescents by self-reports, observer reports, and a behavior checklist. Journal of Consulting and Clinical Psychology, 52, 729-738.[ISI][Medline]
Manne, S. L., Bakeman, R., Jacobsen, P., & Redd, W. H. (1993). Children's coping during invasive medical procedures. Behavior Therapy, 24, 143-158.
Manne, S. L., Jacobsen, P. B., & Redd, W. H. (1992). Assessment of acute pediatric pain: Do child self-report, parent, and nurse ratings measure the same phenomenon? Pain, 48, 45-52.[ISI][Medline]
McGrath, P. A., Seifert, C. E., Speechley, K. N., Booth, J. C., Stitt, L., & Gibson, M. C. (1996). A new scale for assessing children's pain: An initial validation study. Pain, 64, 435-443.[ISI][Medline]
Nunnally, J. C. (1978). Psychometric Theory, 2nd ed. New York: McGraw-Hill.
Peterson, L., & Toler, S. (1986). An information seeking disposition in child surgery patients. Health Psychology, 5, 343-358.[ISI][Medline]
Powers, S. W. (1999). Empirically supported treatments
in pediatric psychology: Procedure-related pain. Journal of
Pediatric Psychology, 24,
131-145.
Powers, S. W., Blount, R. L., Bachanas, P. J., Cotter, M. W., &
Swan, S. C. (1993). Helping preschool leukemia patients and their
parents cope during injections. Journal of Pediatric
Psychology, 18,
681-695.
Pringle, B., Hilley, L., Gelfand, K., Dahlquist, L. M., Switkin, M., Diver, T., Sulc, W., & Eskenazi, A. (in press). Decreasing children's distress during needle sticks and maintaining treatment gains over time. Clinical Psychology in Medical Settings.
Siegel, L. J., & Smith, K. E. (1991). Coping and adaptation in children's pain. In J. P. Bush & S. W. Harkins (Eds.), Clinical and research issues from a developmental perspective (pp. 149-170). New York: Springer-Verlag.
Slifer, K. J., Babbitt, R. L., & Cataldo, M. D. (1995). Simulation and counterconditioning as adjuncts to pharmacotherapy for invasive pediatric procedures. Developmental and Behavioral Pediatrics, 16, 133-141.
Wallander, J. L., & Varni, J. W. (1992). Adjustment in children with chronic physical disorders: Programmatic research on a disability-stress-coping model. In A. M. LaGreca, L. J. Seigel, J. L. Wallander, & E. Walker (Eds.), Stress and coping in child health (pp. 279-298). New York: Guilford.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||