Table 2.
Overview of the characteristics and findings from the identified systematic reviews.
| Review | Type | Inclusion criteria | Period | Outcomes | Aims | Study quality | Study characteristics | Findings summary |
|---|---|---|---|---|---|---|---|---|
| Boudreau et al. (47) | NQ | Peer-Mediated; Age range 4–18 years; Design: no design restrictions | NR | SC | Evaluate peer-mediated PRT for facilitating the SC of school-aged children with ASD | Modified/expanded (by authors) framework for appraising the quality of evidence Reichow et al. (50) | N = 5 studies; N = 10 participants (8 after removing overlap); Age: 7–10 years; IQ: 55–85 | None of the studies met the criteria for classification as promising or established EBP for improving SC impairments |
| Bozkus-Genc and Yucesoy-Ozkan (28) | M | Design: Single-case; Age range: 1–13 years | 1979–2012 | No restrictions | Evaluate participant and intervention characteristics, effectiveness and moderators | NA | N = 34 studies; Age: 2 years, 5 months-12 years, 8 months; Settings: multiple (44.1%), clinic (26.4%), school (14.7%), home (8.8%), community (5.8%) | Mean PND: 76.10% (SD = 33.65, range: 0–100); effect sizes > 90% in 38.2% of studies, 70–89% in 33.4% of studies, and <70% in 29.4% of studies; PND scores >70% for all of the dependent variables except play and social skills. 14 studies labeled as highly effective, 11 fairly effective, 10 questionable/ineffective. Mean PNCD: 78.03% (SD = 34.38, range: 0-100); effect sizes > 90% in 41.1% of studies, 70–89% in 26.5% of studies, and <70% in 26.4% of studies; PNCD scores >70% for all of the dependent variables except play skills. 14 studies labeled as highly effective, 9 fairly effective, 10 questionable/ineffective. Mean PEM: 89.34% (SD = 22.18, range: 0-100); effect sizes > 90% in 79.4% of studies, 70–89% in 11.7% of studies, and <70% in 8.8% of studies; PEM scores >70% for all of the dependent variables. 27 studies labeled as highly effective, 4 fairly effective, 3 questionable/ineffective. |
| Cardogan and McCrimmon (48) | NQ | <18 years of age | NR | Study quality | Evaluate adherence of PRT studies to specific research quality standards | Seven specific standards chosen by authors | N = 17 studies | Systematic application of an intervention procedure: five studies utilized a pre–post evaluation, 11 multiple baseline procedure, one did not collect any baseline data; Comparison of intervention approaches: two studies compared intervention approaches; Use of standard intervention protocols: 9 studies explicitly aligned with the PRT manuals; |
| Treatment fidelity: two studies adhered to the recommended fidelity standard prior to the study start but there were variations during the intervention, five studies no reference to the fidelity measures; Use of objective evaluators: 11 studies used objective evaluators, remaining studies did not reference the use of objective evaluators (two did not require it given the design); Inter-rater reliability: 12 studies reported some form of inter-rater reliability; Longitudinal studies: 8 studies collected follow-up data after the post-treatment stage and 9 did not. |
||||||||
| Forbes et al. (49) | NQ | Design: Experimental; Other: at least one communication skill as a dependent variable | 1987–2018 | Communication | Evaluate primary linguistic and verbal behavior outcomes following PRT and how generalized and collateral outcomes were reported | NA | N = 50 studies | The majority of studies aggregated results and/or did not report sufficient detail to determine linguistic forms and/or verbal behavior functions; There was evidence for the generalization of communication skills to untargeted people, settings, materials, and/or activities; Only one study indicated untargeted linguistic forms emerged following PRT and none of the studies described results that indicated improved generalized and collateral verbal behavior functions. |
| Ona et al. (46) | M | Design: RCT; Age range: ≤ 18 years of age | up to August 2017 | SC, SI, RRB | Evaluate social communication, social interaction, and repetitive behavior outcomes in PRT RCTs | GRADE | N = 7 studies; N = 181 participants; Age: 2.4–9.2 years | Communication (subjective report): two studies, SMD 1.12 (95% CI −0.49; 2.73), p = 0.17, GRADE: very low; Expressive language (subjective report): one study, SMD 0.45 (95% CI −0.13; 1.03), p = 0.13, GRADE: low; Expressive language (direct measurement): two studies, SMD 0.48 (95% CI.04; 0.93), p = 0.03, GRADE: low; |
| Receptive language (subjective report): one study, SMD 0.22 (−0.35; 0.79), p = 0.45, GRADE: low; Social Interaction: one study (subjective report): SMD 0.48 (−1.10; 1.06), p = 0.10, SMD 0.46 (−0.12; 1.04), p = 0.12 for CGI-S and SMD 1.12 (0.50; 1.74), p = 0.0004 for CGI-I Repetitive Behaviors (direct assessment): one study, SMD 15.97 (95% CI 11.57 to 20.36) p <0.0001, GRADE: low. |
||||||||
| Verschuur et al. (37) | NQ | Age: no constraints; Design: no constraints | Up to June 2014 | No restrictions | Evaluate: the range of targeted skills; PRT effectiveness for improving children's outcomes; PRT effectiveness for improving parental and staff outcomes and skills; the certainty of evidence; identify limitations and future directions | Quality of evidence (51); Certainty of evidence following classification by Ramdoss et al. (52) into suggestive, preponderant and conclusive | N = 37 studies N = 420 participants; Age: 1–12.7 years | 56.4% of studies had serious methodological limitations; 43.6% of studies showed conclusive or preponderant evidence that PRT increases self-initiations and results in collateral improvements in communication and language, play skills, affect and reductions in maladaptive behavior; The majority of caregivers and staff members were able to implement PRT techniques; Few studies reported on collateral improvements in caregivers' and staff members' behaviors and evidence was qualified as sparse. |
GRADE, Grading of Recommendation, Assessment, Development and Evaluation; M, meta-analysis; NQ, non-quantitative; NR, not reported; PEM, percentage of data points exceeding median; PNCD, percentage of nonoverlapping corrected data; PND, percentage of nonoverlapping data; PRT, Pivotal Response Training; SC, social and communication; SI, social interaction; SMD, standardized mean difference.