Journal of Pediatric Psychology. Vol. 21. No. 2, 1996. pp. 195-207

The Perception of Procedures Questionnaire:Psychometric Properties of a Brief Parent ReportMeasure of Procedural Distress1

Anne E. Kazak,2 Biancamaria Penati, Mary Katherine Waibel, andGeorge F. BlackallThe Children's Hospital of Philadelphia

Received March 17. 1995: accepted August 15. 1995

Reported the reliability and validity of the Perception of Procedures Question-naire (PPQ), a 19-item parent-report measure developed to assess child andparent distress related to lumbar punctures and bone marrow aspirates in thediagnosis and treatment of childhood cancer. PPQ data from 140 mothers and 96fathers of children and adolescents with leukemia in a first remission wereanalyzed separately. Factor analyses yielded five factors for mothers and fathers:Parent Satisfaction; Child Distress: During; Child Distress: Before; Parent Dis-tress; and Parent Involvement. Internal consistency (Cronbach's alpha) was highfor the total score and the five factor scores as were interrater reliabilitiesbetween mothers and fathers. Validity was determined using the Parenting StressIndex-Short Form, the Pediatric Oncology Quality of Life Scale, and parent andnurse ratings during procedures. Factors 2 and 3, assessing child distress, showstrong associations with the validation measures and support the distinctionbetween distress before and during procedures. This developing scale is recom-mended for use in the assessment and evaluation of child and parent procedure-related distress in pediatric oncology.

KEY WORDS: childhood cancer; assessment, medical procedures; pain; distress; parent perceptions.

'This project was supported by Grant CA57917 from the National Cancer Institute to Anne E. Kazak.The authors appreciate the support of Patricia Brophy, Chair of the Analgesia Protocol for Proce-dures in Oncology (APPO) Workgroup, and the other members of the workgroup who participatedin the development of the PPQ and data collection: Marc Blumberg. Bret Boyer, Tabitha Callaway,Kenneth Covelman, Ruth Daller, Kelly Johnson. Charles Scher. and Sam Scott.

2AII correspondence should be addressed to Anne E. Kazak, Division of Oncology, The Children'sHospital of Philadelphia. 324 S. 34th St.. Philadelphia, Pennsylvania 19104^*399.

195

0H6-8693/96/040O-OI95JO9 50/0 O 1996 Plraam PublHhmj Cotporaion

196 Kazak, Ptnati, Waibd, and Blackall

As part of their diagnostic workup and treatment regimens, children with leuke-mia undergo repeated invasive procedures such as lumbar punctures (LPs) andbone marrow aspirates (BMAs). These procedures are painful and frightening forchildren, although behavioral interventions are helpful in reducing immediatedistress (see Rape & Bush, 1994, for a recent review). Indeed, survivors ofchildhood cancer recall disturbing memories of procedures after treatment ends,with these memories contributing to symptoms of posttraumatic distress (Stuber,Christakis, Houskamp, & Kazak, in press). These data are even more striking forparents, with 39.7% of mothers and 33.3% of fathers reporting severe levels ofposttraumatic stress symptoms up to 12 years after treatment ends (Stuber et al.,in press).

With regard to the assessment of procedural distress, there are three generalapproaches: (a) observational measures within the procedural context; (b) childreport measures of pain and anxiety, and (c) parent report measures of childdistress. Each has advantages and disadvantages which are discussed briefly.3

The measure presented in the current paper, the Perception of Procedures Ques-tionnaire (PPQ), is introduced as an easily administered parent-report assessmenttool that provides data on both child and parent distress. The PPQ assesses theimmediate procedural context while also reflecting maternal and paternal percep-tions of general satisfaction with and involvement in their child's leukemiatreatment.

Research on procedural distress has emphasized the use of observationalmeasures that yield important data on patients, parents, and staff (e.g., Blount etal., 1989; Elliott, Jay, & Woody, 1987; Manne, Bakeman, et al., 1992). Thesemeasures have provided detailed objective data regarding what happens duringthe circumscribed procedural context. They have been utilized predominantly inresearch on interventions related to procedural distress and are relatively expen-sive and labor intensive, as procedures must be observed or taped, and coded.

To assess pain and distress during medical procedures, several child mea-sures use developmentally appropriate strategies such as the Oucher scale (Be-yer, Denyes, & Villarruel, 1992) and visual analog and other developmentallyrelevant scales (McGrath, 1990). Difficulties include having children distinguishbetween anxiety and pain and the effects of amnestic and/or sedative propertiesof medications which may have been administered.

Few parent report measures for child distress during procedures exist. TheBehavioral Upset in Medical Patients-Revised (BUMP-R; Rodriguez & Boggs,1994) is an efficient instrument that quantitatively investigates behavioral upsetin hospitalized pediatric patients. However, it does not assess emotional distress

-The scope of papers reviewed is limited to measures thai have been used in pediatric oncology andwhich pertain to medical procedures. Although scales used with other pediatric samples and moregeneral assessment procedures (e.g., quality of life) may be applicable, a review of these approachesis beyond the scope of the present paper.

Parent Report of Procedures 197

nor provide the specificity needed to comprehend the invasive medical proce-dures experienced by children and adolescents with leukemia. While parentreport measures obviously reflect parent perspectives and often fail to be highlyassociated with child report (Manne, Jacobsen, & Redd, 1992), they do providevaluable information. Parents know their children well and are ultimately respon-sible for helping their child through distressing and painful procedures. Thus,their opinions of their child's comfort, and their own level of distress, areimportant markers in assessing the extent to which procedures are accomplishedin a manner that maximizes the child's and parent's comfort.

Our goal was to develop a brief, reliable, and valid parent report measurewhich would provide ratings of mothers' and fathers' perceptions of their chil-dren before and during procedures and reports of parents' own distress. We alsobelieved that perceptions of the procedural context should be considered withbroader components of families' experiences over the course of treatment forchildhood cancer. Thus, we included parent ratings of satisfaction with theirchild's care and input regarding parents actual and desired level of involvementin procedures. Although mothers most frequently accompany their children dur-ing pediatric oncology treatments, we felt it important to assess and establish therelevance of the assessment tool (PPQ) for fathers as well.

METHOD

Subjects and Procedure

Mothers {n = 140) and fathers (n = 96) of 144 children and adolescentsreceiving treatment for leukemia in a medically confirmed first remission com-pleted the PPQ.4 Parents were given the PPQ as part of a packet of questionnairesto complete and return in a postage-paid envelope, or at their next oncologyappointment. The patients were 77 (53%) male and 67 (47%) female childrenand adolescents with acute lymphoblastic leukemia (ALL) (n = 121), acutenonlymphocytic leukemia (ANLL) (n = 20), or other childhood leukemias (i.e.,chronic myelogenous leukemia, mixed lineage leukemia; n = 3). The majority(87%) were Caucasian, 9% were African American, 2% Asian, 1% Hispanic,and 1% Asian Indian. Patient age at diagnosis ranged from 1 month to 17.5 years

4The PPQ was developed as part of a research study examining the impact of a combined pharmacologic-psychologic intervention for procedural pain during cancer treatment. The study, Analgesia Protocolfor Procedures in Oncology (APPO), is a prospective evaluation of a family-focused, preventiveintervention added to a pharmacologic intervention protocol. The pharmacologic guidelines arebased on the American Academy of Pediatrics guidelines for treating cancer pain in children (Zeltzeret al., 1990). The APPO program is described in delail elsewhere (Kazak, Blackail, Himelstein,Brophy, & Daller, 1995). Preliminary data on the PPQ from a subset of patients are presented in areport of procedural distress prior to the initiation of APPO (Kazak, Boyer, et al., 1995)

198 Kazak, Penati, Waibel, and Blackall

(M = 5.8, SD = 4.3). The mean time since diagnosis was 8.8 months (SD =10.3), ranging from 15 days to 34 months. Of the 144 patients, 32.6% were 1month from diagnosis, 58.3% were 6 months from diagnosis, and 87.5% werewithin 2 years of diagnosis at the time they completed the PPQ.

Mothers' age ranged from 19.6 to 48.3 years (M = 34.7, SD = 5.7).Fathers' age ranged from 23.5 to 58.2 years (Af = 38.0, SD = 6.7). Eighty-fourpercent (n = 120) were two-parent families and 16% were single-parent units inwhich the parent was either never married (n = 11), separated {n = 5), divorced(n = 6), or widowed (n = 2).

The majority of patients were enrolled in clinical research protocols ap-proved by the hospital's Institutional Review Board (e.g., studies of the Chil-dren's Cancer Group [CCG]). All protocols required the administration of bothLPs and BMAs. Six to seven lumbar punctures (seven for patients with ANLL)were required during induction and consolidation. Following consolidation (i.e.,during maintenance therapy), patients with ALL received one LP every 3 monthsfor 2 to 3 years depending on the child's sex. The protocols required two to threeBMAs during induction, one aspirate upon ending treatment, and additionalBMAs as determined by CCG protocol or when otherwise clinically indicated(e.g., suspicion of disease relapse). For all patients, the medical regimen con-sisted of multiagent chemotherapy. Type and toxicity of treatment varied depend-ing on disease factors.

Development of the PPQ

The PPQ was developed at a pediatric cancer treatment center at a large,urban, university-affiliated children's hospital by a team of health professionalsfrom the oncology service and the hospital's pain team. The team was theResearch Committee of the APPO Workgroup, a larger multidisciplinary groupwho formulated the APPO treatment protocols. The Research Committee metweekly and included four pediatric psychologists, a pediatric oncologist, andthree pediatric nurse practitioners.

Items were drawn from the empirical literature on pediatric pain and proce-dures and from team members' clinical experiences. Group discussions gener-ated the topics to be included. Individual Research Committee members wrotequestions. The questions were discussed by the group to reach agreement oncontent and structure. Major revisions of the questionnaire were reviewed by theAPPO Workgroup.

Twenty-seven items were selected. Four items were open-ended and 23items were constructed with scores based on a 7-point Likert-type scale, withlower scores indicating higher satisfaction, more involvement, and less child andparent distress. Items were phrased such that parents rated their child's and their

Parent Report of Procedures 199

own distress. Quantitative and qualitative items addressed parents' perceptions oftheir children's and own distress before and during procedures (e.g., spinal tapsand bone marrow aspirations).

Analysis Strategy

1. Due to the large number of correlations conducted, a Bonferroni correc-tion was applied throughout the paper, with the significance level set at/? < .005.Analyses reported at p < .01 or .05 are discussed as trends.

2. Item analyses were performed to examine correlations between individu-al items and the total PPQ score. A conservative approach to eliminating itemswas taken to assure more robust factors. First, for both mothers and fathers,items with low correlation coefficients (r < .30) were eliminated. Second, itemswith low communality scores (<.40) were discarded.

3. Factor analyses were conducted using pairwise deletions, for mothers andfathers separately. The Kaiser-Meyer-Olkin measure of sampling adequacy wasused to assess the appropriateness of factor analysis. Scree plots determined thenumber of factors to be used in the model. The principal components analysiswas used as the method for factor extraction. Varimax rotations were conductedto better identify substantively meaningful factors.

4. Pearson product-moment correlations were conducted to examine therelationships between PPQ factors assessing parents' perceptions of child's dis-tress and parents' own distress.

5. The internal consistency of the PPQ was assessed using Cronbach'scoefficient alphas. Interrater reliability between mothers and fathers was assessedfor PPQ child distress factors.

6. Hierarchical regression analyses were conducted to assess the extent towhich the factor variance was explained by child's age at diagnosis and by timesince diagnosis.

7. To assess concurrent validity, Pearson product-moment correlations be-tween maternal and paternal PPQ factor scores and the Pediatric Oncology Quali-ty of Life Scale (POQOLS; Goodwin, Boggs, & Graham-Pole, 1994) and theParenting Stress Index-Short Form (PSI-S; Abidin, 1990) were conducted.

The POQOLS, a 21-item, 7-point Likert-type scale, measures frequency ofpediatric oncology patients' daily activity over a 2-week period. The POQOLyields a total score and three factor scores: Factor 1 assesses physical functionand role restriction, Factor 2 measures emotional distress, and Factor 3 concernsresponse to current medical treatment. We included the latter two subscales.Lower scores indicate higher quality of life. Internal consistency as measured bycoefficient alphas yielded reliability scores for the three factors ranging from r =.68 to .87. Pearson product-moment correlations also yielded an interrater re-liability score of r = .89.

200 Kazak, Pcnati, Walbd, and Btackal]

The PSI-S, a 36-item Likert-type questionnaire, is widely used in pediatricsettings with three subscales (Parental Distress, Parent-Child Dysfunctional In-teraction, and Difficult Child). Lower scores indicate less stress. Internal con-sistency (Cronbach's coefficient alphas) ranged from r = .80 to .91. Test-retestreliabilities ranged from r = .68 to .85.

8. Validity was also assessed by conducting nonparametric correlationsusing Kendall's tau-Z> between the PPQ factors of child distress and parent andnurse observational ratings of child distress during procedures. Observationalratings of child distress were obtained immediately after the procedure by askingparents and nurses to rate child distress based on a 7-point Likert-type scale, with1 representing no distress and 7 indicating extremely distressed. Only a smallsubset of patients [n= 13-17) were able to reliably rate their own distress on thesame 7-point Likert-type scale. We therefore do not report these data.

9. Finally, the PPQ parent distress factor was correlated, using Kendall'stau-fo, with parents' ratings of their own distress during the procedure. Againparent self-ratings of distress were obtained immediately following the procedureand were based on a 7-point Likert-type scale, with higher scores indicating moredistress.

RESULTS

Factor Analyses

Item-total Pearson product-moment correlations were conducted for moth-ers and fathers separately. The same three questions, for both mothers andfathers, were found to be poorly correlated (r < .30) with the total PPQ score andwere therefore excluded from the factor analyses. One item was removed as ithad a communality score less than .40 for both mothers and fathers. The finalscale consisted of 19 questions (Table I).

Separate principal component factor analyses of mother (n = 124-140) andfather (n = 84—96) were conducted using pairwise deletions and varimax rota-tions on 19 of the original items. Based on a scree plot solution, a five-factorsolution was examined. Varimax rotations produced a five-factor solution thataccounted for 82.2% of the variance for mothers and 78.0% of the variance forfathers. The factors are 1. Parent Satisfaction (6 items); 2. Child Distress:During (5 items); 3. Child Distress: Before (4 items); 4. Parent Distress (2items); and 5. Parent Involvement (2 items).

The factors are the same for mothers and fathers. Means and standarddeviations (M, SD) for the factor scores are as follows. For mothers, Factor 1(11.3, 6.7), Factor 2 (21.9, 8.3), Factor 3 (11.0, 5.6), Factor 4 (8.7, 3.6), Factor5 (3.6, 3.1). For fathers: Factor 1 (11.5, 6.1), Factor 2 (21.6, 8.2), Factor 3

Parent Report of Procedures 201

Table I. Items of Perception of Procedures Questionnaire (PPQ) by Factor"

Factor 1. Parent Satisfaction1. How satisfied are you with the medical services in the division?2. How satisfied are you with each of the following aspects of your child's treatment at The

Children's Hospital of Philadelphia, Division of Oncology?a. Your communication with the treatment teamb. The treatment team's attention to your concerns regarding treatment and side effectsc. The treatment team's attention to providing complete and comprehensive servicesd. The treatment team's ability to minimize pain, fear, and discomfort associated with treat-

mente. Emotional support provided by the treatment team

Factor 2. Child Distress: During3. How much distress does your child experience during spinal taps?4. How much distress does your child experience during bone marrow aspirations?5. How distressed does your child become during routine clinic visits (that is, visits for blood

counts and chemotherapy without spinal taps or bone marrow aspirations)?6. How distressed does your child become immediately prior to a spinal tap?7. How distressed does your child become immediately prior to a bone marrow aspiration?

Factor 3. Child Distress: Before8. How long before a spinal tap do you notice any change in your child's behavior?9. How distressed does your child become in the morning of a visit for a spinal tap?

10. How distressed does your child become in the morning of a visit for a bone marrow aspiration?11. How long before a bone marrow aspiration do you notice any change in your child's behavior?

Factor 4. Parent Distress12. How distressed do you become during spinal taps?13. How distressed do you become during bone marrow aspirations?

Factor 5. Parent Involvement14. How actively are you involved during spinal taps, bone marrow aspirations, and administra-

tion of chemotherapy?15. How actively involved would you like to be?

"All responses are on a 7-point Likert-type scale. Lower scores indicate higher levels of satisfactionand involvement and lower levels of distress.

(11.6, 5.7), Factor 4 (8.1, 3.3), Factor 5 (4.7, 3.5). Eigenvalues and factorloadings are presented in Tables II and III.

Pearson product-moment correlation analyses conducted to assess the rela-tionship between parent perception of child distress and their own distress indi-cated significant correlations. Parent Distress (Factor 4) correlated highly withChild Distress: During (Factor 2), rMo(her,(l 18) = .61, p < .0001; /-FaIhers(76) =.46, p < .0001 and Child Distress: Before (Factor 3), rMotbcrs(\20) = .93, p <.0O0l;rFatheni(76)= .36,/>< .001.

Hierarchical regression analyses were conducted between the PPQ factorsand child's age at diagnosis and time since diagnosis. Data from mothers andfathers indicated a significant negative relationship between child's age andChild Distress: During (Factor 2),'Molhere(2, 116)= - 3 . 5 1 , p < .001; fFathOT(2,75) = -4 .08, p < .0001. In addition, for mothers there were positive relation-ships between time since diagnosis and Factor 4, /(2, 127) = 2.94, p < .004, andFactor 2, t(2, 116) = 2.82, p < .006 (nearly significant).

202 Kazak, Penati, Waibd, and BbdtalJ

Table II. Factors and Item Loadings for Mothers' PPQ

Factor

Item

Factor I. Parent Satisfaction (Eigenvalue = 6.6)Team's comprehensive servicesTeam's attention to concernsCommunication with TeamTeam's medical servicesTeam's ability to minimize painTeam's emotional support

Factor 2. Child Distress: During (eigenvalue = 4.28)Immediately prior to spinal tapImmediately prior to bone marrow aspirateDuring routine clinic visitsDuring spinal tapDuring bone marrow aspirate

Factor 3. Child Distress: Before (eigenvalue = 1.98)Amount of time prior to spinal tapAmount of time prior to bone marrow aspirateAmount of distress morning of spinal tapAmount of distress morning of bone marrow aspirate

Factor 4. Parent Distress (eigenvalue = 1.79)During spinal tapDuring bone marrow aspirate

Factor 5. Parent Involvement (eigenvalue = 0.93)Involvement during proceduresDesire to be involved during procedures

Reliability

Reliability of the PPQ measure was assessed through two methods. First,Cronbach's coefficient alphas were calculated for the total measure and for thetotal score of each individual factor. The Cronbach's coefficient alphas for thetotal measure were strong, aMothers( 111) = .89; aFott)CTJ(71) = .86. Cronbach'scoefficient alphas for Factors 1 through 5, respectively, were also strong, for bothparents: aMotb<:rs( 137-120) = .95, .91, .84, .92, .91; aF,thers(96-78) = .91, .90,.84, .94, .82. Interrater reliabilities between mothers and fathers for the childfactors was also high: Factor 2 (Child Distress: During), r{10) = .83, p < .0001and Factor 3 (Child Distress: Before), r(7I) = .77, p < .0001.

Validity

It was expected that mothers' and fathers' ratings of child distress (Factors 2and 3) would correlate significantly with the POQOLS factors assessing emo-tional distress and response to current medical treatment and with the PSI-S

1

.93

.93

.91

.90

.85

.85

.13

.14-.04

.17

.24

-.00-.04-.01-.00

.12

.12

.08

.09

2

.04

.06

.07

.05

.11

.10

.86

.81

.73

.71

.68

.11

.17

.58

.53

.31

.21

-.04-.01

3

.02-.04-.02

.03

.01-.04

.14

.17

.27-.05-.02

.94

.92

.61

.55

.17

.17

.08-.02

4

.04

.02

.06

.00

.21

.14

.17

.27-.02

.55

.56

.10

.06

.08

.22

.86

.87

.07

.12

5

.01

.03

.11

.08

.00

.00

-.03-.03

.06-.07-.02

-.03.13

-.04- 04

.14

.14

.95

.95

Parent Report of Procedures 203

Table III. Factors and Item Loadings for Fathers' PPQ

Factor

Item 1

.93

.93

.87

.77

.71

.80

.04

.08

.19

.03

.11

- .02- .02

.07- .02

.06

.07

.01

.03

2

.00

.00

.05

.07

.20

.09

.87

.86

.55

.87

.86

.00

.05

.42

.35

.17

.29

- .08- .01

3

- .05- . 0 3- .01

.01

.15- .03

.24

.27

.18- .04-.11

.93

.94

.64

.63

.16

.11

- .02- . 06

4

- .11- .02- . 06

.02

.16

.17

.08

.10- .32

.32

.34

.01

.0214

.44

88.85

.13

.15

5

07.11.04

- .01- .18- . 02

- .04- .04- .08- .05- .05

- .0103

- .12- .09

.2218

.90

.89

Factor I. Parent Satisfaction (eigenvalue = 5.4)Team's comprehensive servicesTeam's attention to concernsCommunication with TeamTeam's medical servicesTeam's ability to minimize painTeam's emotional support

Factor 2. Child Distress: During (eigenvalue = 4.00)Immediately prior to spinal tapImmediately prior to bone marrow aspirateDuring routine clinic visitsDuring spinal tapDuring bone marrow aspirate

Factor 3. Child Distress: Before (eigenvalue = 2.25)Amount of time prior to spinal tapAmount of time prior to bone marrow aspirateAmount of distress morning of spinal tapAmount of distress morning of bone marrow aspirate

Factor 4. Parent Distress (eigenvalue = 2.04)During bone marrow aspirateDuring spinal tap

Factor 5. Involvement (eigenvalue = 1 1 )Involvement during proceduresDesire to be involved during procedures

Difficult Child and the Parent-Child Dysfunctional Interaction subscales. Formothers, strong associations between emotional distress (POQOLS) and PPQFactors 2 and 3, and a weaker association between Factor 3 and the response tocurrent medical treatment were found (Table IV). The PSI-S correlations showeda significant association between Parent-Child Dysfunctional Interaction andFactor 3 (Child Distress: Before) and a statistical trend for the association be-tween Difficult Child and Factor 2 (Child Distress: During). The associations forfathers were less striking, with only one significant correlation, between theemotional distress factor of the POQOLS and Factor 3 (Child Distress: Before).

Validity correlations for parent and nurse ratings were strong. As predicted,Factor 2 (Child Distress: During) showed strong associations with nurse andparent self-report of distress during the procedure. Consistent with prediction,these associations were not found for Factor 3 (Child Distress: Before), suggest-ing that the PPQ discriminates between distress at the time of the procedure andmore general distress driven by anticipation of the procedure.

Parent Distress (Factor 4) was expected to correlate significantly with thePSI-S Parental Distress scale and with self ratings of parent distress. For both

Tab

le I

V.

Cor

rela

tions

Am

ong

PPQ

Fac

tors

and

Sta

ndar

dize

d an

d O

bser

vatio

nal

Mea

sure

s"

Fact

ors*

Mot

hers

(n

=

132-

107)

Fa

ther

s {n

=

95-7

2)

Mea

sure

s 1

2 3

4 5

12

3'

Pedi

atri

c on

colo

gy q

ualit

y of

life

Em

otio

nal

expr

essi

onR

espo

nse

to t

reat

men

tPa

rent

ing

stre

ss i

ndex

-Sho

rt f

orm

Par

ent-

chil

dD

iffic

ult

child

Pare

nt d

istr

ess

Obs

erva

tiona

l di

stre

ss r

atin

gs'"

Nur

se o

f ch

ildPa

rent

of

child

Pare

nt d

istr

ess

—

—

—

.39«

—

—

—

—

.5

9'

"Das

hes

indi

cate

the

cor

rela

tion

was

not

exa

min

ed a

s it

was

not

app

licab

le.

*1.

Pare

nt S

atis

fact

ion;

2.

Chi

ld D

istr

ess:

Dur

ing;

3.

Chi

ld D

istr

ess:

Bef

ore;

4.

Pare

nt D

istr

ess;

5.

Pare

nt I

nvol

vem

ent.

••O

bser

vatio

nal

ratin

gs a

re a

vaila

ble

for

a su

bset

of

the

sam

ple

(n =

62)

."p

< .

05.

'p

< .

01.

fp

< .

005.

«p

<

.001

.

.00

-.1

2

.15

.10

—

.29/

.14

..32*

.25'

.34*

.54*

.43

'.2

4'

.41*

.15

.06

.14

.10

.08

.17

.03

.02 —

-.0

3-.

06

.00

-.0

9

—

.10

.20

.12

.13

—

.22

.10

.01

.20

.45'

.61'

.32/

.17

.19

.06

.14

.12

.10 06 .05

.12

.32/ —

-.0

7-.

03

-.0

2-.

05

—

Parent Report of Procedures 205

mothers and fathers, KendaJl's lau-b correlations were significant for self-ratings:'•mo.h«,(57) = -39, p < .0001, rfaJhcrs(25) = .59, p < .0001 (Table IV). Forfathers, Pearson product-moments correlations indicated significant associationsbetween Factor 4 and the PSI-S parent distress subscale r(8I) = .32, p < .004,whereas this association did not reach statistical significance for mothers, r(126)= .02, p < .82.

DISCUSSION

In this paper we report data on a new parent report assessment tool, thePerception of Procedures Questionnaire (PPQ), for patient and parent distressduring procedures for the diagnosis and treatment of pediatric leukemia. ThePPQ is unique in providing an easy-to-complete measure of both parent and childdistress during these invasive procedures. The inclusion of fathers in the devel-opment of the PPQ also represents an advancement in the assessment ofprocedure-related distress.

Psychometrically, the PPQ has strong internal consistency. The five factorsare robust and evidence good reliability. The factors tap a spectrum of procedure-related experiences, reflecting the use of research data and clinical experiencethat were instrumental in its design. For example, two child distress factorsassess parent perceptions of child distress before and during the procedure.Similarly, parent experiences distinguish among parent satisfaction, distress, andinvolvement during treatment. The factors are consistent for mothers and fathers,with equivalent psychometric strength for both genders. In addition, interraterreliability for mothers and fathers is high.

Concurrent validity data for the PPQ appears promising, especially for thechild distress factors (Factors 2 and 3). Parent ratings of their child's distress arerelated to the emotional distress scale of the POQOLS. which taps appraisals ofthe child's withdrawal and fear (Goodwin et al., 1994), more so than to theresponse to current medical treatment subscale, which directly assesses acuteside effects of treatment. This finding suggests that parents' perceptions of theirchildren's distress during procedures are influenced more by the affective experi-ence for their children and themselves rather than the more concrete aspects oftreatment side effects. Procedures are very stressful for parents; the data mayhelp to explain the long-term impact of procedural distress on parents (Kazak,Barakat, et al., 1995; Stuber et al., in press).

The PPQ appears to discriminate between child distress in the actual pro-cedural context and more general distress as evidenced by the strong pattern ofcorrelations among parent and nurse ratings and Factor 2 (Child Distress: Dur-ing) and the lack of associations of these variables with Factor 3 (Child Distress:Before). Anticipatory distress appears to represent a distinct, although probably

206 Kazak, Pcnati, Waibel, and Blackall

related, behavior from the acute upset often seen in the procedure room. ThePPQ's ability to discriminate between these two may be useful clinically intargeting and evaluating procedural distress interventions.

The inverse relationship between child's age at diagnosis and PPQ ChildDistress: During supports previous literature documenting the greater difficultyexperienced by younger pediatric oncology patients during procedures (cf.LeBaron & Zeltzer, 1984). The lack of association between the other PPQ factorsand child's age at diagnosis suggests that age is a less striking variable inunderstanding anticipatory distress, or in terms of the distress experienced byparents. Thus, while younger children may require more support during theactual procedure, children and adolescents of all ages may benefit from attentionto distress before procedures. The distress experienced by parents appears rela-tively unaffected by the age of the child. Developmentally, this is an interestingfinding which may be related to the tendency for distressed children to actyounger than their chronological age, thus potentially contributing to parentdistress as they attempt to comfort or assist their child. Time since diagnosis waspositively associated with mother's ratings of child and own distress duringprocedures. It may be that as procedures become less frequent over the course oftreatment they are experienced as more distressing when they do occur.

Further research is needed to establish the validity of the parent factors.Specifically, it will be useful to introduce other measures to help explain moreprecisely what aspects of treatment parent scales target. Unexpectedly, signifi-cant associations between father (but not mother) PPQ scores and the PSI-SParental Distress subscale were found. One possibility is that the tendency offewer fathers to attend procedures may contribute to a different experience forthem than for mothers. That is, mothers may view their role during procedures asunique (e.g., being there specifically to assist their child cope with the proce-dure), whereas fathers' may see their role during procedures as a general exten-sion of their parenting responsibilities and, therefore, reflect more general typesof distress in their ratings than do mothers.

As the PPQ is based on parent report, some caution is warranted in light ofthe potential drawbacks inherent in parent report measures. That is, parent reportis subjective and could reflect systematic biases in the perception of the child'sdistress or parent response styles. These correlational data show interesting rela-tionships among mother, father, and staff perceptions. However, the generallimitations of correlational data apply to the PPQ.

An important next step is research evaluating the sensitivity of the PPQ toeffects of intervention to determine how parent report can be integrated intointervention plans and outcomes. In addition, treatment for childhood cancer iscomplex and includes many different combinations of procedures over differentphases of the illness and treatment protocols. Observational data have providedinsights into ways in which these factors influence distress. Further research onthe PPQ is needed to determine whether and how parental report measures could

Parent Report of Procedures 207

augment or replace more labor intensive observational approaches and thereforeenhance the practicality of evaluating procedural distress.

In summary, the Perception of Procedures Questionnaire is a brief, easilyadministered parent report measure assessing child and parent distress duringprocedures. Based on these initial reliability and validity data, the psychometricproperties are strong and support the use of the PPQ for evaluation of, andintervention for, procedural distress in the diagnosis and treatment of pediatricleukemia.

REFERENCES

Abidin, R. (1990). Parenting Stress Index Short Form: Manual. Charlottesville, VA: Pediatric Psy-chology Press.

Beyer, J. E., Denyes, M. J., & Villarrucl, A. M. (1992). The creation, validation, and continuingdevelopment of the Ouchen A measure of pain intensity in children. Journal of PediatricNursing, 7, 335-346.

Bloum, R., Corbin, S., Sturges, J., Wolfe, V., Parter, J., & James, L. (1989). The relationshipbetween adult's coping behavior and child coping and distress during BMA/LP procedures: Asequential analysis. Behavior Therapy, 20, 585-601.

Elliott, C. H., Jay, S. M., & Woody, P. (1987). An observation scale for measuring children's distressduring medical procedures. Journal of Pediatric Psychology, 12, 543-551.

Goodwin, D., Boggs, S., & Graham-Pole, J. (1994). Development and validation of the PediatricOncology Quality of Life scale. Psychological Assessment, 6. 321-328.

Kazak, A., Barakat, L., Meeske, K., Christakis, D., Meadows, A., Casey, R., Penati, B , & Stuber,M. (1995). Post traumatic distress, family functioning, and social support in survivors ofchildhood leukemia and their parents. Manuscript under editorial review.

Kazak, A., Boyer, B., Brophy, P., Johnson, K., Scher, C , Covelman, K., & Scott, S. (1995).Parental perceptions of procedure-related distress and family adaptation in childhood leukemia.Children's Health Care. 24, 143-158.

Kazak, A., Blackall, G., Himelstein, B., Brophy, P., & Daller, R. (1995). The process of change inpediatric practice: A family oriented pharmacologic psychologic intervention for proceduralpain. Family Systems Medicine, 13, 173-185.

LeBaron, S., & Zeltzer, L. (1984). Assessment of acute pain and anxiety in children and adolescentsby self-reports, observer reports, and a behavior checklist. Journal of Consulting and ClinicalPsychology, 52. 729-738.

Manne, S., Bakeman, R., Jacobsen, P., GorfinkJe, K., Bernstein, D., & Redd, W. (1992). Adult-child interaction during invasive medical procedures. Health Psychology. II, 94-105.

Mann, S. L., Jacobsen, P. B., & Redd, W. H. (1992). Assessment of acute pediatric pain: Do childself-report, parent ratings, and nurse ratings measure the same phenomenon? Pain, 48. 45-52.

McGrath, P. (1990). Pain in children. New York: Guilford.Rape, R , & Bush, J. (1994). Psychological preparation for pediatric oncology patients undergoing

painful procedures: A methodological critique of the research. Children's Health Care, 23.51-67.

Rodriguez, C , & Boggs, S. (1994). Behavioral Upset in Medical Patients-Revised: Evaluation of aparent report measure of distress for pediatric populations. Journal of Pediatric Psychology 19.319-324.

Stuber. M., Chnslakis, D., Houskamp, B., & Kazak, A. (in press). Post trauma symptoms insurvivors of childhood leukemia and their parents. Psychosomatics.

Zeltzer, L. K., Altman, A., Cohen, D., LeBaron, S., Munuksela, E. L., & Schechter, N. L. (1990).Report of the subcommittee on the management of pain associated with procedures in childrenwith cancer. Pediatrics. 86. 826-831.