ERIC EJ750845: Treating Bedtime Resistance with the Bedtime Pass: A Systematic Replication and Component Analysis with 3-Year-Olds

2006·0.15 MB·English

by  ERIC

JOURNALOFAPPLIEDBEHAVIORANALYSIS 2006, 39, 423–428 NUMBER4 (WINTER2006)
TREATING BEDTIME RESISTANCE WITH THE BEDTIME PASS:
A SYSTEMATIC REPLICATION AND COMPONENT
ANALYSIS WITH 3-YEAR-OLDS
KURT A. FREEMAN
CHILDDEVELOPMENTANDREHABILITATIONCENTER
OREGONHEALTH&SCIENCEUNIVERSITY
Bedtimeresistance,acommonpediatricproblem,thatwasdisplayedby4unrelated3-year-old
childrenwastreatedwiththebedtimepass(i.e.,provisionofasmallnotecardexchangeablefor
onetripoutofthebedroomafterbedtime)plusextinction.Bedtimeresistancewaseliminatedfor
allparticipants.Further,treatmentdidnotproduceextinctionbursts,asiscommonwhenusing
extinctionproceduresalone.Componentanalysiswith1participantsuggestedthatuseofboth
componentsoftheinterventionproducedthebestoutcomes.Findingsextendtheliteratureon
the treatment of pediatric bedtime resistance as well as the application of behavior analysis to
clinicalpsychology andpediatriccare.
DESCRIPTORS: bedtime resistance, bedtime behavior problems, behavioral treatment,
bedtimepass
_______________________________________________________________________________
Active bedtime resistance (i.e., calling out Paavonen et al., 2002; poor parental satisfac-
from or leaving one’s room after bedtime) is tion, Gelman & King, 2001). Thus, effective
one of the most common difficulties in young intervention is important.
children, seen in approximately20% to 25% of Interventions for pediatric bedtime resistance
children 1 to 5 years of age (see Metzler & typically are behavioral in nature, the most
Mindell, 2004). According to the Diagnostic common being extinction (i.e., planned paren-
and Statistical Manual of Mental Disorders tal ignoring). Although it is effective, parental
(DSM-IV-TR; American Psychiatric Associa- acceptability of extinction is often low (France,
tion,2000),bedtimeresistancemaybeclassified 1994; Friman et al., 1999), partly due to
as dyssomnia not otherwise specified, and temporary increases in bedtime resistance often
bedtime problems can be the fundamental seen early in intervention (i.e., the extinction
component of behavior clusters that are classifi- burst;seeBlum&Friman,2000).Alternatively,
able with other diagnoses (e.g., oppositional the bedtime pass (Friman et al.) involves (a)
defiant disorder, attention deficit hyperactivity a small notecard exchangeable for one trip out
disorder).Leftuntreated,bedtimeproblemscan of the bedroom after being put to bed and (b)
persist for years (Kataria, Swanson, & Treva- extinction. Two male siblings, ages 3 and 10
thon, 1987). Further, pediatric sleep distur- years,weresuccessfullytreatedwiththebedtime
bances are related to various adverse outcomes pass, without observation of extinction bursts
(e.g., disturbed cognitive functioning, Steenari during initial intervention periods. Further, 20
et al., 2003; increased behavior problems, parents rated the intervention as more accept-
able than traditional extinction.
I thank Nancy Farrell and Wendy DeCourcey for Additional research on the bedtime pass is
assistancewithdatacollectionandCynthiaAndersonand
warranted to address two primary issues. First,
Douglas Woodsfor assistance withdesign.
although Friman et al. (1999) suggested that
Correspondenceregardingthisarticleshouldbesentto
KurtA.Freeman,ChildDevelopmentandRehabilitation their results supported 3 years of age as the
Center, Oregon Health & Science University, P.O. Box lowerlimitforeffectiveuseofthebedtimepass,
574, Portland, Oregon 97207 (e-mail: freemaku@ohsu.
patternsofobtainedresultssuggestedcautionin
edu).
doi:10.1901/jaba.2006.34-05 this statement. Specifically, the older sibling
423
424 KURT A. FREEMAN
used the pass on 8 of the 13 opportunities, but door of his bedroom (Friman et al., 1999). Use
the 3-year-old used it only twice. Further, of the pass was also recorded during interven-
suppression of bedtime resistance was not tion phases; this was not counted as bedtime
achieved during the initial treatment imple- resistance (Friman et al.). Interobserver agree-
mentation with the younger sibling. Thus, ment, calculated by dividing the frequency of
improvements in the older sibling’s behavior bedtime resistance indicated by the primary
may have facilitated those of the younger child. observer by that of the secondary observer, was
Second, although Friman et al. speculated that 95% for Jim and 100% for Craig, Walter, and
use of the pass might mitigate the likelihood of Greg. A nonconcurrent multiple baseline across
an extinction burst whereas extinction is the participants with an ABAB withdrawal design
active component for behavior change, they (A 5 baseline; B 5 pass plus extinction) was
provided no evidence regarding the necessity of used to investigate treatment effectiveness with
both components. This study attempts a sys- 3 participants (Jim, Craig, and Walter). An
tematic replication of Friman et al. by focusing ACABAB design was used with Greg. The C
on two variations: targeting a group composed phase involved use of the pass while responding
solely of 3-year-olds and providing an initial to bedtime resistance as usual (i.e., pass alone),
attempt at a component analysis. and B involved the combined intervention (i.e.,
pass plus extinction).
METHOD
Procedure
Participants Parents had one meeting with the researcher,
Four typically developing 3-year-old Cauca- during which informed consent was obtained,
sian boys participated. Parents, who responded and data collection and general intervention
to posted flyers about the study, referred all procedures were described (with a written de-
participants. For all, parents were concerned scription provided; copy available from the first
about frequent calling out from or leaving the author). Following the initial meeting, all
room after bedtime. Resistance was reportedly contact between the investigator and parents
problematic due to high frequency, long occurred via telephone and e-mail. Parents
duration, or both. According to parent ratings reported on obtained data daily by either
on the Eyberg Child Behavior Inventory sending an e-mail to or leaving a voice mail
(Eyberg & Pincus, 1999), the children fell into message with the investigator.
the normal or nonclinical range with regards to During baseline, parents were instructed to
the presence of generalized disruptive behavior respond to bedtime behavior problems in their
problems.Allparentsreportedsomeresponseto usual fashion. After completing this phase, the
bedtime resistance (e.g., catering to the child’s investigator contacted parents via the telephone
request,scoldingthechild,allowingthechildto and reviewed the specific details regarding
stay up). Each child slept in his own bedroom. treatment. During intervention, the child was
given a notecard exchangeable for one trip out
Measurement and Research Design of the bedroom for a short (i.e., less than
Each night, the child’s mother measured 3 min), specific activity (e.g., use restroom, one
frequency of calling out from or leaving the more hug). After completing the action, the
bedroom after bedtime; the child’s father child surrendered the pass and was returned to
collected interobserver data every 3rd to 5th bed.TheparentsofJim,Craig,andWalterwere
night.Longandshortcrieswerecountedasone instructed to return the child to his room
instance of calling out; leaving the room was without comment if he came out and to ignore
counted when the child passed through the calling out if either occurred after use of the
TREATING BEDTIME RESISTANCE 425
pass. With Walter, there was an unplanned 0 to 2). When consistent application of
variation in treatment implementation during extinction resumed on Night 52, resistance
the second treatment phase (i.e., Nights 41 decreased (M 5 0.29, range, 0 to 3). The
through 51) such that his parents allowed him planned component analysis conducted with
to use the pass but failed to implement Greg showed that use of the pass alone resulted
extinction. To investigate the effects of the pass in decreased frequency and variability of
with and without extinction, Greg’s parents bedtime resistance (final 3 nights of initial
were initially told to continue to respond to baseline M 5 3.67, range, 2 to 5; final 3 nights
bedtime resistance in their typical fashion. of pass alone M 5 1.67, range, 1 to 2).
During later phases, they were instructed to However, use of both treatment components
use the pass plus extinction. (pass plus extinction) resulted in elimination of
resistance(final3nightsofcombinedtreatment
M 5 0).
RESULTS
Figure 1 shows nightly occurrence of bed-
DISCUSSION
time resistance across experimental phases for
each participant. High and variable rates of The results of this study extend the literature
callingoutorleavingtheroomwereobservedin on the treatment of bedtime resistance in
baseline phases, and reduced rates were ob- general and the use of the bedtime pass in
servedduringintervention.Theseeffectsusually particular. First, the results show that the pass
occurred rapidly. In addition, the reductions in effectively reducedbedtimeresistancein3-year-
bedtime resistance were achieved without being old children who were the only targets of the
accompanied by extinction bursts that are often intervention (as distinct from Friman et al.,
seen with extinction-based procedures. With 1999). These results help to establish that the
Greg, resistance approximated baseline rates intervention is effective with children of this
during initial use of the pass plus extinction; age. Second,these resultsdemonstratedthat the
further, reduction of target behavior occurred bedtime pass reduced bedtime resistance with-
less rapidly when the combined intervention outproducinganextinctionburst,aneffectthat
was used compared to other participants. could heighten treatment acceptability and
However, calling out or leaving the room never thereby improve treatment adherence (France,
occurred more frequently during intervention Henderson, & Hudson, 1996; Rapoff, 1999;
than during baseline. Further, resistance ap- Rickert & Johnson, 1988). This is important
proximated baseline rates on 1 night each for 2 given that inconsistent application of extinction
participants (i.e., Night 19 for Jim and Night encourages persistence of bedtime resistance
64 for Walter), with parental report indicating and decreases responsiveness to future extinc-
that these spikes were accompanied by unusual tion attempts (Pritchard & Appelton, 1988).
events that probably influenced the results (i.e., Third, the results of the component analyses
complaints of illness from Jim, sister calling to indicatedthatboththepassandextinctionwere
Walter). necessary to produce optimal results. Although
Data from 2 participants offer tentative both Walter and Greg used the pass more
support for the benefit of both components of frequently during the pass-alone phases than in
the bedtime pass. Unplanned use of the pass the pass plus extinction phases, elimination of
without extinction by Walter’s parents resulted resistance occurred only when the combined
in bedtime resistance at rates similar to baseline intervention was instituted. With Greg, reduc-
(combined baseline M 5 1.0, range, 0 to 3; tions in bedtime resistance occurred more
unplanned use of pass alone M 5 0.67, range, slowly when the combined intervention was
426 KURT A. FREEMAN
Figure1. Nightlyfrequencyofcombinedcallingoutfromandleavingtheroom.Dataarepresentedtoreflectthe
passageoftime,andnightswithmissingdatapointsrepresentlackofdatacollectionbycareproviders.Asterisksindicate
nightson whichparticipants usedthe bedtime pass.BL 5baseline.
TREATING BEDTIME RESISTANCE 427
implemented. The reason for this is unclear, combined bedtime pass intervention is limited
although the possibility that the order in which because the component analysis was not con-
treatment variations were presented affected ducted with each child. As a related issue, data
data patterns cannot be ruled out. on the component analysis with Greg are
It is important to speculate how the bedtime potentially confounded by the fact that exper-
pass reduces bedtime resistance without pro- imental phases were not of similar lengths.
ducing an extinction burst. One possibility Whether continued use of the pass alone would
involves viewing the program as a form of have resulted in further reductions of resistance
differentialreinforcementofalternativebehavior remains unclear. Second, treatment fidelity was
(DRA; e.g., Vollmer & Iwata, 1992). DRA not systematically evaluated. The fact that
interventions often include both reinforcement Walter’s parents failed to implement extinction
for positive social behavior and extinction of procedures when instructed highlights this as
problematic responses, and the combination has a potentially important issue. Third, explana-
been shown to reduce the probability of tions from parents regarding apparent aberrant
extinctionbursts(e.g.,Bowman,Fisher,Thomp- data (e.g., illness on a given night) were not
son, & Piazza, 1997; Fisher, Kuhn, & Thomp- collected systematically. It is unclear whether
son, 1998). In the pass program, the pass is similar contextual variations existed at other
acommunicativealternativetoresistantbehavior points in the study but were not reported.
and its use results in satisfaction of one request, Fourth, due to miscommunication, Greg’s
whichisapotentiallyreinforcingevent(i.e.,itis parents implemented the initial intervention
differentiallyreinforced).Furthersupportforthe after several nights of missed data collection
DRA hypothesis is found in research showing (i.e., Nights 8 to 13) without obtaining
that reinforcement of positive social behavior additional baseline data. Fifth, this study offers
may not result in elimination of targeted only speculation regarding the link between the
problem behavior if reinforcement is still avail- bedtime pass and other lines of behavior-
ablefortheproblembehavior(e.g.,Piazzaetal., analytic research (e.g., DRA, manding).
1999; Shirley, Iwata, Kahng, Mazaleski, & These limitations notwithstanding, the pos-
Lerman, 1997). A similar result was observed itive results here (and in Friman et al., 1999)
during the current component analysis.Another warrant additional research. In addition to the
plausible explanation of how the pass produces suggestion above about mechanism, future
its effects involves the concept of manding research should address issues such as pro-
(Skinner, 1957). Previous research has shown motion of parental adherence to treatment
that treatment of problem behavior with differ- procedures, child and family variables that
entialreinforcementandextinctionusingsignals affect intervention success, and technology
(e.g., tone, picture) denoting reinforcement of transfer to applied settings (e.g., prescribed by
mands reduced problem behavior without pro- pediatricians). Further, experimental designs
ducing extinction bursts (e.g., Fisher et al., that offer empirical analysis of speculations
1998). In the pass program, it is possible that regarding why the intervention works are
the pass served as a stimulus that was discrim- needed to solidify its theoretical basis. More
inative for reinforcement of at least one mand systematic and comprehensive investigation of
(e.g.,requestforatriptothebathroom).Clearly the relative importance of both treatment
these possibilities are speculative and are offered components(i.e.,passandextinction)isneeded
here to guide future research. to further establish the external validity of the
Theseresultsshouldbeinterpretedinlightof intervention. Finally, analysis of level of
several limitations. First, external validity of the adherence to, and application of, the interven-
428 KURT A. FREEMAN
tion is warranted, given other research that has Gelman, V. S., & King, N. J. (2001). Wellbeing of
mothers with children exhibiting sleep disturbance.
shown that differential reinforcement proce-
AustralianJournalofPsychology, 53,18–22.
dures may be effective despite less than optimal Kataria, S.,Swanson,M.S.,& Trevathon,G.E.(1987).
implementation (Vollmer, Roane, Ringdahl, & Persistence of sleep disturbances in preschool chil-
dren.Behavioral Pediatrics,110, 642–646.
Marcus, 1999). Given the extent of bedtime
Metzler, L. J., & Mindell, J. A. (2004). Nonpharmaco-
resistance exhibited by children in the United logic treatments for pediatric sleeplessness. Pediatric
States (see Metzler & Mindell, 2004), addi- ClinicsofNorthAmerica, 51,135–151.
Paavonen, E. J., Almqvist, F., Tamminen, T., Moilanen,
tional research on the pass program may not
I., Piha, J., & Ra¨sa¨nen, E., et al. (2002). Poor sleep
only add to extant research on bedtime
and psychiatric symptoms at school: An epidemio-
problems but also help to extend behavior logicalstudy.EuropeanChild&AdolescentPsychiatry,
analysis into mainstream clinical psychology 17,10–17.
Piazza,C.C.,Bowman,L.G.,Contrucci,S.A.,Delia,M.
and primary pediatric medical care.
D., Adelinis, J. D., & Goh, H. L. (1999). An
evaluation ofthe properties ofattention asreinforce-
ment for destructive and appropriate behavior.
REFERENCES
JournalofAppliedBehavior Analysis, 32,437–449.
Pritchard, A., & Appelton, P. (1988). Management of
AmericanPsychiatricAssociation.(2000). Diagnosticand
sleep problems in pre-school children. Early Child-
statistical manual of mental disorders (4th ed., text
hoodDevelopment andCare,34,227–240.
rev.).Washington, DC:Author.
Rapoff, M. A. (1999). Adherence to pediatric medical
Blum,N.,&Friman,P.C.(2000).Behavioralpediatrics:
regimens.NewYork: KluwerAcademic/Plenum.
The confluence of applied behavior analysis and
Rickert, V. I., & Johnson, C. M. (1988). Reducing
pediatric medicine. In J. Carr & J. Austin (Eds.),
nocturnal awakening and crying episodes in infants
Handbookofappliedbehavioranalysis(pp.161–185).
andyoungchildren:Acomparisonbetweenscheduled
Reno, NV:Context Press.
awakenings and systematic ignoring. Pediatrics, 81,
Bowman, L. G., Fisher, W. W., Thompson, R. H., &
203–212.
Piazza, C. C. (1997). On the relation of mands and
Shirley,M.J.,Iwata,B.A.,Kahng,S.,Mazaleski,J.L.,&
the function of destructive behavior. Journal of
Lerman,D.C.(1997).Doesfunctionalcommunica-
AppliedBehavior Analysis, 30,251–265.
tion training compete with ongoing contingencies of
Eyberg, S., & Pincus, D. (1999). Eyberg Child Behavior
reinforcement? An analysis during response acquisi-
InventoryandSutter-Eyberg StudentBehaviorInvento-
tion and maintenance. Journal of Applied Behavior
ry: Professional manual. Lutz, FL: Psychological
Analysis, 30,93–104.
Assessment Resources, Inc.
Skinner, B. F. (1957). Verbal behavior. Cambridge, MA:
Fisher,W.W.,Kuhn,D.E.,&Thompson,R.H.(1998). PrenticeHall.
Establishing discriminative control of responding Steenari,J.R.,Vountela,V.,Paavonen,E.J.,Carlson,S.,
using functional and alternative reinforcers during Fja¨llberg, M., & Aronen, E. T. (2003). Working
functionalcommunicationtraining.JournalofApplied memory and sleep in 6- to 13-year-old school-
BehaviorAnalysis,31, 543–560. children. Journal of the American Academy of Child
France, K. G. (1994). Handling parents’ concerns andAdolescent Psychiatry,42,85–92.
regarding behavioural treatment of infant sleep Vollmer, T. R., & Iwata, B. A. (1992). Differential
disturbance. Behavioural Change,11, 101–109. reinforcement as treatment for behavior disorders:
France, K. G., Henderson, J. M. T., & Hudson, S. M. Procedural and functional variations. Research in
(1996).Fact,act,andtact:Athree-stageapproachto Developmental Disabilities,13, 393–417.
treating the sleep problems of infants and young Vollmer,T.R.,Roane,H.S.,Ringdahl,J.E.,&Marcus,
children. Child and Adolescent Psychiatric Clinics of B. A. (1999). Evaluating treatment challenges with
NorthAmerica,5, 581–599. differential reinforcement of alternative behavior.
Friman,P.C.,Hoff,K.E.,Schnoes,C.,Freeman,K.A., JournalofAppliedBehavior Analysis, 32,9–23.
Woods, D. W., & Blum, N. (1999). The bedtime
pass:An approach to bedtimecrying andleaving the Received March 14,2005
room. Archives of Pediatric and Adolescent Medicine, Final acceptance August2,2006
153,1027–1029. Action Editor,Pat Friman