This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Received: 20 October 2018 Revised: 7 February 2019 Accepted: 14 February 2019
DOI: 10.1002/bin.1664
L I T E R A TUR E R E V I EW
Efficacy of functional analysis for informingbehavioral treatment of inappropriate mealtimebehavior: A systematic review and meta‐analysis
Valdeep Saini1 | Joshua Jessel2 | Julia A. Iannaccone2 |
Rooker, Jessel, Kurtz, & Hagopian, 2013). This is done to reduce the effects of any variability in responding during
the initial exposure to the treatment that may not necessarily reflect the terminal therapeutic outcomes. Therefore,
the final statistic was similar to the MBLR calculation with the exception being that the means from the baseline and
treatment were calculated from the last five sessions of each respective phase (MBLR‐5). A treatment that eliminates
IMB would result in the maximum percentage of 100% when using the statistics MBLR and MBLR‐5. If the mean of
the treatment were to equal the mean of baseline (i.e., no effect), the quotient would be 0%. Furthermore, the MBLR
and MBLR‐5 scales are sensitive to possible treatment worsening effects. Any value below 0% indicates a treatment
had a negative impact on IMB.
3 | RESULTS
We identified 11,471 articles during the initial search of large databases and journal archives. Of these, we excluded
11,198 articles because they were duplicates from the various search engines or because the titles, abstract, or key-
words did not meet the initial inclusion criteria. The remaining 273 articles met criteria for a final detailed review. We
removed 249 articles after the detailed review because authors (a) did not display IMB graphically, (b) did not target
IMB, or (c) used an inadequate experimental design. This resulted in the identification of 23 articles total,
representing 82 cases. Of these cases, 37 where designated as treatments for IMB that were informed by a func-
tional analysis, and 45 were treatments that were not informed by a functional analysis. The treatment effect sizes
of the 23 articles were analyzed using the five aforementioned nonparametric statistics.
3.1 | Participant characteristics
Table 1 displays the participant characteristics. Most participants were below the age of 6 years old (92%) and were
male (65%). Just over half the participants did not have a developmental diagnosis or one was not reported (51%),
followed by 30% diagnosed with an intellectual or developmental disability, and 14% diagnosed with ASD. Complex
medical histories were common with 39% of participants diagnosed with gastroesophageal reflux disease and 37%
diagnosed with failure to thrive. Thirty‐three percent of participants did not have additional comorbid medical diag-
noses or they were not reported in the individual studies.
TABLE 1 Participant characteristics
Characteristic n %
Age
0–5 years old 45 92
6–10 years old 4 8
Sex
Male 32 65
Female 17 35
Developmental diagnosis
Developmental/intellectual disability 15 30
Autism spectrum disorder 7 14
Down syndrome 1 2
Dual diagnosis 1 2
None/Not reported 25 51
Medical history
Gastroesophageal reflux disease 19 39
Failure to thrive 18 37
Anatomical abnormalities 5 10
Respiratory disorder/dysfunction 4 8
Gastrointestinal problems 3 6
Cardiac disorder/dysfunction 3 6
Gastrostomy‐tube dependence 3 6
Prematurity 3 6
Enzyme deficiency/excess 2 4
Vision impairments 2 4
Nervous system disorder/dysfunction 1 2
Other 3 6
None/not reported 16 33
236 SAINI ET AL.
3.2 | Functional analysis characteristics
Table 2 displays descriptive information on the functional analyses and treatments for each application. Because mul-
tiple treatments were sometimes developed based on the results of a single functional analysis, there were 25 func-
tional analysis applications (see Data S1 for a list of participants and the corresponding study from which the data
were extracted). Escape was identified as a functional reinforcer in all 25 functional analyses and the only function
in 20 of the 25 applications. Multiple control via attention and escape was identified as the maintaining variables
in the remaining five applications.
3.3 | Treatment characteristics
Of the 37 treatments informed by a functional analysis, 33 (89%) used escape extinction either alone or as a com-
ponent in a treatment package including supplemental reinforcement procedures (Table 3). Escape extinction took
TABLE 2 Functional analysis and treatment characteristics
Characteristic n %
Functional analysis
Yes 25 36
Escape 20 80
Attention 0 0
Escape and attention 5 20
No 45 64
Treatment procedures
Escape extinction 34 41
Escape extinction and DRA 16 20
Escape extinction and NCR 11 13
Escape extinction and DNRA 7 9
Escape extinction and AE 4 5
Escape extinction and HP 3 4
DRA 3 4
DRA and response cost 2 2
NCR and response cost 1 1
NCR 1 1
Reinforcement thinning
Yes 2 5
No 42 38
N/A 49 60
Generalization
Yes 0 0
No 82 100
Social validity
Yes 7 8
No 75 92
Note. AE refers to attention extinction. D(N)RA refers to differential (negative) reinforcement of alternative behavior. RC
refers to response cost. NCR refers to noncontingent reinforcement. HP refers to high‐probability sequence.
SAINI ET AL. 237
the form of nonremoval of the spoon (Hoch, Babbitt, Coe, Krell, & Hackbert, 1994), physical guidance (Ahearn,
Kerwin, Eicher, Shantz, & Swearingin, 1996), or representation if IMB occurred after acceptance and was associ-
ated with expulsions (Sevin, Gulotta, Sierp, Rosica, & Miller, 2002). Of the five cases where attention (in addition
to escape) was also identified as a reinforcer for IMB, four treatments included attention extinction in the form of
ignoring IMB (80%). Although it should be noted that the majority of studies described ignoring or providing no
consequence for IMB, even in cases where attention was not identified as a reinforcer during the functional
analysis.
Supplemental reinforcement procedures were typically associated with bite acceptance or consumption. There
were 21 of the 37 applications of treatments following a functional analysis that used various forms of functional
and arbitrary supplemental reinforcement. Functional differential negative reinforcement (DNRA) was used in 16%
(six of 37) of applications wherein escape was provided contingent on acceptance or consumption. Noncontingent
TABLE 3 Treatment characteristics
Treatment n %
With pretreatment functional analysis 37
Escape extinction 33 89
Differential reinforcement
Functional 6 16
Nonfunctional 8 22
Noncontingent reinforcement
Functional 0 0
Nonfunctional 7 19
Other 0 0
Without pretreatment functional analysis 45
Procedural escape extinction 42 93
Differential reinforcement
Functional N/A
Nonfunctional 15 33
Noncontingent reinforcement
Functional N/A
Nonfunctional 6 13
Other 6 13
238 SAINI ET AL.
reinforcement (NCR; seven of 37) or differential reinforcement of alternative behavior (DRA; eight of 37) including
nonfunctional reinforcers (e.g., noncontingent access to the feeder's attention throughout the meal; access to pre-
ferred toys contingent on acceptance) was used in the remaining supplemental reinforcement procedures. Addi-
tionally, for 28 of 37 (76%) treatments, the feeder delivered attention in the form of positive praise statements
contingent on acceptance or consumption, regardless of the identified function or co‐occurring treatment
components.
Of the 45 treatments in which the experimenters did not conduct a functional analysis, procedures typical of
escape extinction were arranged in 42 (93%) applications; however, given that a functional analysis was not con-
ducted, it is unclear if these procedures operated as extinction. Procedures consistent with attention extinction were
also common (e.g., ignoring or providing no consequence for IMB); however, the methods of each respective study
were not always clearly described and therefore could not be coded accurately (e.g., some studies clearly stated that
IMB was ignored, whereas others did not specify if an attention contingency was or was not present).
Because functional analyses did not inform these 45 treatments, reinforcers could not be identified as functional,
and any supplemental reinforcement procedures were all considered nonfunctional reinforcement. Supplemental
reinforcement procedures included DRA (15 of 45) and NCR (6 of 45) and were like those that were used for treat-
ments that were preceded by a functional analysis (e.g., contingent use of toys for appropriate behavior, noncontin-
gent attention). There were six applications in which the experimenters included other procedures such as response
cost (e.g., Buckley, Strunck, & Newchok, 2005) or the high‐probability response sequence (e.g., Patel et al., 2006).
Similar to those treatments based on a functional analysis, the use of praise contingent on acceptance or consump-
tion was common, occurring in 41 of 45 treatments (91%).
Table 2 displays various treatment information for each individual application related to reinforcement schedule
thinning, generalization, and social validity. When reinforcement procedures were included as a supplemental
SAINI ET AL. 239
treatment procedure (44 applications), the continuous reinforcement schedules were thinned in only two of those
applications (5%). Tests of generalization across settings were not conducted in any of the 82 applications
(although a few studies did demonstrate durability of the treatment when conducted by caregivers in the same
setting; e.g., Casey, Cooper‐Brown, Wacker, & Rankin, 2006). Experimenters evaluated social validity in only seven
applications (9%).
3.4 | Effect size analyses
Figure 1 displays the results of the treatment effect size analyses across the 23 studies. Since 2001, there were nine
studies in which a functional analysis was conducted before the treatment and 14 studies in which treatment was
initiated without a functional analysis. Large effect sizes were obtained for most studies across the different statis-
tical analyses; however, some variability was present. PEM and MBLR‐5 had the largest effect sizes with all studies
above 80% and 70%, respectively. Slight reductions in effect sizes were obtained from the MBLR analysis with two
studies (no functional analyses) reaching the 40–60% range. The greatest variability was obtained when applying the
PND and PZD statistics. When using PND, three studies (no functional analyses) were below a 50% effect size. More
studies had effect sizes below 50% when applying PZD: seven of 14 without functional analyses and five out of nine
with functional analyses.
The mean effect size for treatments based on a prior functional analysis were 80% (SD = 30), 51% (SD = 32), 88%
(SD = 15), 96% (SD = 8), and 83% (SD = 29) for PND, PZD, MBLR‐5, PEM, and MBLR, respectively. The mean effect
size for treatments not based on a prior functional analysis were 79% (SD = 34), 50% (SD = 34), 91% (SD = 11.91),
97% (SD = 7), and 81% (SD = 21) for PND, PZD, MBLR‐5, PEM, and MBLR, respectively. That is, the mean effect size
was comparable and about equal for all effect size measures regardless of whether the treatment was or was not
based on a functional analysis.
Figure 2 displays results of the five nonparametric analyses for each individual treatment application. Similar to
results obtained at the study level, the greatest variability and lowest effect sizes were obtained when applying
the PND and PZD statistics. Only a minor difference in effect sizes was observed when considering treatment appli-
cations that were and were not based on a functional analysis.
4 | DISCUSSION
We reviewed the extant literature in pediatric feeding disorders to determine the degree to which behavioral treat-
ments for IMB based on a pretreatment functional analysis were more efficacious than those not based on a func-
tional analysis. Across multiple effect size measures, we did not observe markedly greater reductions in IMB for
those treatments in which a functional analysis was conducted. Mean effect sizes for both treatment groups (i.e.,
based on and not based on a functional analysis) were about equal. In many cases, treatments appeared to be based
on the prevailing hypothesis of environmental variables that maintain IMB (i.e., escape) and were efficacious in the
absence of a functional analysis. This result held true across multiple categories of participant characteristics and pro-
cedural variations across studies.
The treatments that were developed based on, or in the absence of, a functional analysis warrant consider-
ation. Our review of treatment results indicated several common components that researchers implemented
regardless of whether a functional analysis was conducted, bringing into question the prescriptive nature of this
practice. A large proportion of studies relied on extinction‐based procedures to decrease IMB (i.e., greater than
80% of treatments based on the results of a functional analysis, and greater than 90% not based on a functional
analysis, arranged escape extinction procedurally). This result, combined with other findings (e.g., Marshall et al.,
2014; Saini et al., in press; Sharp et al., 2010; Silbaugh et al., 2016), emphasizes the role of escape as a reinforcer
for IMB and the efficacy of escape extinction as a treatment component. Studies that have evaluated the efficacy
FIGURE 1 Mean effect sizes for each study. Bars represent standard error measurement (SEM)
240 SAINI ET AL.
of escape extinction relative to other treatment components have repeatedly shown that increased acceptance
and a concomitant reduction in IMB is often due to the escape extinction component (LaRue et al., 2011; Piazza,
Patel, Gulotta, Sevin, & Layer, 2003; Reed et al., 2004). Many studies arranged supplemental reinforcement pro-
cedures with extinction; however, even when other positive reinforcers are included in the treatment package
(functional or nonfunctional), escape extinction has been shown to be the necessary component (Patel, Piazza,
FIGURE 2 Effect sizes of treatments based and not based on a functional analysis. Each circle represents anindividual application. Horizontal lines represent the mean. The asterisk represents an individual applicationbelow zero
SAINI ET AL. 241
Martinez, Volkert, & Santana, 2002), which might explain why escape extinction was procedurally arranged in most
cases reviewed, even those for which a functional analysis was not conducted.
Although extinction is often used when a social function of undesirable behavior is implicated from a functional
Ahearn, W. H., Kerwin, M. E., Eicher, P. S., Shantz, J., & Swearingin, W. (1996). An alternating treatments comparison of two
intensive interventions for food refusal. Journal of Applied Behavior Analysis, 29, 321–332. https://doi.org/10.1901/jaba.1996.29‐321
Allison, D. B., & Gorman, B. S. (1994). “Make things as simple as possible, but no simpler.” A rejoinder to Scruggs and
Mastropieri. Behaviour Research and Therapy, 32, 885–890. https://doi.org/10.1016/0005‐7967(94)90170‐8
*Allison, J., Wilder, D. A., Chong, I., Lugo, A., Pike, J., & Rudy, N. (2012). A comparison of differential reinforcement and non-
contingent reinforcement to treat food selectivity in a child with autism. Journal of Applied Behavior Analysis, 45,
613–617. https://doi.org/10.1901/jaba.2012.45‐613
*Anderson, C. M., & McMillan, K. (2001). Parental use of escape extinction and differential reinforcement to treat food selec-
tivity. Journal of Applied Behavior Analysis, 34, 511–515. https://doi.org/10.1901/jaba.2001.34‐511
*Bachmeyer, M. H., Piazza, C. C., Fredrick, L. D., Reed, G. K., Rivas, K. D., & Kadey, H. J. (2009). Functional analysis and treat-
ment of multiply controlled inappropriate mealtime behavior. Journal of Applied Behavior Analysis, 42, 641–658. https://doi.org/10.1901/jaba.2009.42‐641
*Borrero, C. S., Schlereth, J. G., Rubio, E. K., & Taylor, T. (2013). A comparison of two physical guidance procedures in the
treatment of pediatric food refusal. Behavioral Interventions, 28(4), 261–280. https://doi.org/10.1002/bin.1373
*Buckley, S. D., Strunck, P. G., & Newchok, D. K. (2005). A comparison of two multicomponent procedures to increase food
*Casey, S. D., Cooper‐Brown, L. J., Wacker, D. P., & Rankin, B. E. (2006). The use of descriptive analysis to identify and
manipulate schedules of reinforcement in the treatment of food refusal. Journal of Behavioral Education, 15, 39–50.https://doi.org/10.1007/s10864‐005‐9001‐7
*Dawson, J. E., Piazza, C. C., Sevin, B. M., Gulotta, C. S., Lerman, D., & Kelley, M. L. (2003). Use of the high‐probabilityinstructional sequence and escape extinction in a child with food refusal. Journal of Applied Behavior Analysis, 36,
105–108. https://doi.org/10.1901/jaba.2003.36‐105
Didehbani, N., Kelly, K., Austin, L., & Wiechmann, A. (2011). Role of parental stress on pediatric feeding disorders. Children's
Health Care, 40, 85–100. https://doi.org/10.1080/02739615.2011.564557
Drevon, D., Fursa, S. R., & Malcolm, A. L. (2017). Intercoder reliability and validity of WebPlotDigitizer in extracting graphed
Hoch, T. A., Babbitt, R. L., Coe, D. A., Krell, D. M., & Hackbert, L. (1994). Contingency contacting: Combining positive rein-
forcement and escape extinction procedures to treat persistent food refusal. Behavior Modification, 18, 106–128. https://doi.org/10.1177/01454455940181007
Horner, R. H., Carr, E. G., Strain, P. S., Todd, A. W., & Reed, H. K. (2002). Problem behavior interventions for young children
with autism: A research synthesis. Journal of Autism and Developmental Disorders, 32, 423–446. https://doi.org/10.1023/A:1020593922901
Iwata, B. A., Pace, G. M., Cowdery, G. E., & Miltenberger, R. G. (1994). What makes extinction work: An analysis of
procedural form and function. Journal of Applied Behavior Analysis, 27, 131–144. https://doi.org/10.1901/
jaba.1994.27‐131
Jessel, J., Hanley, G. P., & Ghaemmaghami, M. (2016). Interview‐informed synthesized contingency analyses: Thirty replica-
tions and reanalysis. Journal of Applied Behavior Analysis, 49, 576–595. https://doi.org/10.1002/jaba.316
Kahng, S., Iwata, B. A., & Lewin, A. B. (2002). The impact of functional analysis on the treatment of self‐injurious behavior. InS. R. Schroeder, M. L. Oster‐Granite, & T. Thompson (Eds.), Self‐injurious behavior: Gene–brain–behavior relationships (pp.119–131). Washington, DC: American Psychological Association.
*Kahng, S., Tarbox, J., & Wilke, A. E. (2001). Use of a multicomponent treatment for food refusal. Journal of Applied Behavior
Patel, M. R., Piazza, C. C., Martinez, C. J., Volkert, V. M., & Santana, C. M. (2002). An evaluation of two differential reinforce-
ment procedures with escape extinction to treat food refusal. Journal of Applied Behavior Analysis, 35, 363–374. https://doi.org/10.1901/jaba.2002.35‐363
Patel, M. R., Piazza, C. C., Santana, C. M., & Volkert, V. M. (2002). An evaluation of food type and texture in the treatment of
a feeding problem. Journal of Applied Behavior Analysis, 35, 183–186. https://doi.org/10.1901/jaba.2002.35‐183
*Patel, M. R., Reed, G. K., Piazza, C. C., Bachmeyer, M. H., Layer, S. A., & Pabico, R. S. (2006). An evaluation of a high‐probability instructional sequence to increase acceptance of food and decrease inappropriate behavior in children with
pediatric feeding disorders. Research in Developmental Disabilities, 27, 430–442. https://doi.org/10.1016/j.
ridd.2005.05.005
*Peterson, K. M., Volkert, V. M., & Milnes, S. M. (2017). Evaluation of practice trials to increase self‐drinking in a child with a
feeding disorder. Behavior Analysis in Practice, 10, 167–171. https://doi.org/10.1007/s40617‐016‐0147‐7
Piazza, C. C., Fisher, W. W., Brown, K. A., Shore, B. A., Patel, M. R., Katz, R. M., … Blakely‐Smith, A. (2003). Functional analysis
of inappropriate mealtime behaviors. Journal of Applied Behavior Analysis, 36, 187–204. https://doi.org/10.1901/
jaba.2003.36‐187
*Piazza, C. C., Patel, M. R., Gulotta, C. S., Sevin, B. M., & Layer, S. A. (2003). On the relative contributions of positive rein-
forcement and escape extinction in the treatment of food refusal. Journal of Applied Behavior Analysis, 36, 309–324.https://doi.org/10.1901/jaba.2003.36‐309
Piazza, C. C., Patel, M. R., Santana, C. M., Goh, H. L., Delia, M. D., & Lancaster, B. M. (2002). An evaluation of simultaneous
and sequential presentation of preferred and nonpreferred food to treat food selectivity. Journal of Applied Behavior Anal-
Seubert, C., Fryling, M. J., Wallace, M. D., Jiminez, A. R., & Meier, A. E. (2014). Antecedent interventions for pediatric feeding
problems. Journal of Applied Behavior Analysis, 47, 449–453. https://doi.org/10.1002/jaba.117
*Sevin, B. M., Gulotta, C. S., Sierp, B. J., Rosica, L. A., & Miller, L. J. (2002). Analysis of response covariation among multiple
topographies of food refusal. Journal of Applied Behavior Analysis, 35, 65–68. https://doi.org/10.1901/jaba.2002.35‐65
Sham, E., & Smith, T. (2014). Publication bias in studies of an applied behavior‐analytic intervention: An initial analysis. Jour-
nal of Applied Behavior Analysis, 47, 663–678. https://doi.org/10.1002/jaba.146
Sharp, W. G., Berry, R. C., McCracken, C., Nuhu, N. N., Marvel, E., Saulnier, C. A., … Jaquess, D. L. (2013). Feeding problems
and nutrient intake in children with autism spectrum disorders: A meta‐analysis and comprehensive review of the liter-
ature. Journal of Autism and Developmental Disorders, 43, 2159–2173. https://doi.org/10.1007/s10803‐013‐1771‐5
Sharp, W. G., & Jaquess, D. L. (2009). Bite size and texture assessments to prescribe treatment for severe food selectivity in
autism. Behavioral Interventions: Theory & Practice in Residential & Community‐Based Clinical Programs, 24(3), 157–170.https://doi.org/10.1002/bin.282
Sharp, W. G., Jaquess, D. L., Morton, J. F., & Herzinger, C. V. (2010). Pediatric feeding disorders: A quantitative synthesis of
treatment outcomes. Clinical Child and Family Psychology Review, 13, 348–365. https://doi.org/10.1007/s10567‐010‐0079‐7
Sharp, W. G., Volkert, V. M., Scahill, L., McCracken, C. E., & McElhanon, B. (2017). A systematic review and meta‐analysis ofintensive multidisciplinary intervention for pediatric feeding disorders: How standard is the standard of care? The Journal
of Pediatrics, 181, 116–124. https://doi.org/10.1016/j.jpeds.2016.10.002
Shukla, S., & Albin, R. W. (1996). Effects of extinction alone and extinction plus functional communication training on covari-
ation of problem behaviors. Journal of Applied Behavior Analysis, 29, 565–568. https://doi.org/10.1901/jaba.1996.29‐565
Silbaugh, B. C., Penrod, B., Whelan, C. M., Hernandez, D. A., Wingate, H. V., Falcomata, T. S., & Lang, R. (2016). A systematic
synthesis of behavioral interventions for food selectivity of children with autism spectrum disorders. Review Journal of
Autism and Developmental Disorders, 3, 345–357. https://doi.org/10.1007/s40489‐016‐0087‐8
Tiger, J. H., Hanley, G. P., & Bruzek, J. (2008). Functional communication training: A review and practical guide. Behavior Anal-
ysis in Practice, 1, 16–23. https://doi.org/10.1007/BF03391716
Volkert, V. M., & Piazza, C. C. (2012). Pediatric feeding disorders. In P. Sturmey, &M. Hersen (Eds.),Handbook of evidence‐basedpractice in clinical psychology (pp. 323–339). New York: Wiley. https://doi.org/10.1002/9781118156391.ebcp001013
Wagner, F., Rossi, V., Aubry‐Kientz, M., Bonal, D., Dalitz, H., Gliniars, R., … Herault, B. (2014). Pan‐tropical analysis of climate
effects on seasonal tree growth. PLoS ONE, 9, e92337. https://doi.org/10.1371/journal.pone.0092337