Table 2.

Study characteristics.

Citation and methodology	Aim or objectives	Outcome measures	Results	Methodological quality
Hamad et al [49] Pre- or post-test	Investigate the feasibility of an Internet-based “asynchronous” small-scale three module-Web-based learning course presented in a distance-learning medium.	Parent outcome measures: 25-item Web-based knowledge acquisition measure (test) administered prepost intervention.	Internet-based training curriculum could be effective in training parents about methods and procedures related to behavioral interventions. Pretest scores: mean=68.8, SD=15.6, Posttest scores: mean=82.9, SD=4.9. Large effect size (Cohen d=1.21) Paired t-tests: mean prepost test scores statistically significant improvement (P<.001) for all participants combined (n=51).	Kmet rating: strong (82%) NHMRCalevel of evidence: level IV
		Child outcome measures: not specified
Heitzman-Powell et al [50] Pre- or post-test	Evaluate the modified OASIS training intervention for use with parents from a distance.	Parent outcome measures: parent skill assessment in ABAbimplementation	Implementations of ABA skills (41.23% mean increase)	Kmet rating: good (77%) NHMRC level of evidence: level IV
		Parent knowledge assessment (Web-based) on ASDcand ABA principles and procedures	Knowledge assessments (39.15% mean increase)
		Parent satisfaction with training	High levels of importance and significance of Web-based tutorials (mean scale 1-5:4.62 and 4.71 respectively). High levels of importance and significance of telemedicine coaching sessions (mean scale 1-5:4.62 and 4.8 respectively)
		Cost savings (driving miles) Child outcome measures	Mean travel savings per family was 2,263 driving miles using telemedicine if compared with face-to-face coaching. Note: Prepost comparison with no statistical analysis for significance.
		Child outcome measures: not specified
Ingersoll and Berger [43] Ingersoll et al [44] Pickard et al [45] RCTd	Compare parent engagement and effectiveness in self-directed and therapist-assisted versions of a novel telehealth-based parent-mediated intervention for young children with ASD	Parent outcome measures: ImPACT knowledge quiz: 20-item multiple choice quiz taken prepost intervention	Intervention completion was a significant predictor of postintervention knowledge (P=.01) in both groups.	Kmet rating: strong (85%) NHMRC level of evidence: level II
		Videotape parent-child interaction for intervention fidelity using the ImPACT intervention fidelity checklist	Intervention completion (P=.3) and group assignment (P=.45) made significant independent contributions to treatment fidelity. Post intervention fidelity for both groups was significant (P=.004) Statistically significant improvement prepost in parent intervention fidelity in both groups (P<.01, Large effect size: Cohen d=3.21) as well as between groups post intervention (P<.01, Large effect size: Cohen d=0.3). At follow-up statistically significant (P<.001, Large effect size: Cohen d=2.92) prepost in both groups but not between groups.	Kmet rating: strong (85%) NHMRC level of evidence: level II
		Parent sense of competence scale	Statistically significant improvement (P<.01, Large effect size: Cohen d=3.34) prepost intervention in self-efficacy in both groups but not between groups.
		Parent sense of competence scale	Statistically significant improvement (P<.01, Large effect size: Cohen d=1.34) prepost intervention in self-efficacy in both groups but not between groups.
		Family impact questionnaire	Statistically significant improvement (P<.05, Large effect size: Cohen d=1.03) prepost in parent stress in both groups but not between groups post intervention. Statistically significant improvement (P<.05, Large effect size: Cohen d=1.47) prepost in positive perception of the child in both groups as well as between groups post intervention (P<.05, Large effect size: Cohen d=1.16).
		Parent engagement using website analytics	Therapist-assisted group statistical significantly performed better on parent engagement (number of logins and duration on site) and intervention completion when compared with self-directed groups (P<.001 and P<.05 respectively)
		Intervention evaluation survey using 7-point Likert scale measuring treatment appropriateness, website usability, and overall intervention satisfaction.	Participants rated intervention as highly acceptable (mean=6.07, SD=0.79), the website as highly usable (mean=6.36, SD=0.57). Overall satisfaction of intervention was high (mean=6.56, SD=0.71). No statistically significant difference in treatment appropriateness, website usability, and overall intervention satisfaction between groups.
		49-item 7-point Likert scale quantitative survey administered post intervention examining intervention, appropriateness perceived child social communication gains, burden of the intervention on the family, and frequency of intervention use.	Overall, parent rated intervention favorably with mean scores: 1) Intervention appropriateness 6.59 (SD 0.58), perceived child social communication gains 5.41 (SD 1.24), burden of the intervention on the family 5.72 (SD 1.23), frequency of intervention use 6.36 (SD 0.57) Statistically significant differences between groups (TAevs SDf) for intervention appropriateness (P=.03, Large effect size: Cohen d=0.94) and child social communication gains (P=.05, Large effect size: Cohen d=0.84). No difference in the burden of intervention on the family and frequency of intervention use domains.
		Qualitative interviews— semistructured investigated overall perception of intervention and content, perception of feasibility of intervention, experience of support during intervention, and intervention referral preferences.	Qualitative themes: Positive perception of the appropriateness of intervention. The intervention was easy to learn initially but became more challenging as they progressed. The support of a coach would be essential in the later, more complex sections of the intervention. Parents felt more empowered and better able to interact with their child. Perceptions of barriers included time restrictions and technology failure. Parents suggested the intervention should be made available at the time of ASD diagnosis as it may help empower parents at a stressful time.
		Child outcome measures: language targets	Statistically significant (P<.05, Large effect size: Cohen d=2.26) prepost improvements in language targets in both groups but not between groups post intervention.
		MacArthur communicative development inventories: words and gestures	Statistically significant (P<.01, Large effect size: Cohen d=1.74) prepost improvements in language skills in both groups but not between groups post intervention.
		Vineland adaptive behavior scales, 2ndedition	Statistically significant (P<.05, Large effect size: Cohen d=1.00) prepost improvements in the communication domain in both groups but not between groups post intervention. No statistically significant differences prepost in the social domains in both groups, however, a statistical difference was observed between groups post intervention (P<.05, Large effect size: Cohen d=0.91)
St. Peter et al [46] Quasi-randomized	Compare parental adherence during written or asynchronous video teleconsultation designed to teach parents of children with ASD to implement discrete trial instruction.	Parent outcome measures: Parental adherence between the written (control) and video (experimental) groups Child outcome measures: not specified	Adherence in the video group was significantly higher (P<.001) compared with written instructions.	Kmet rating: good (71%) NHMRC level of evidence: level III-1
Vismara et al [48] Single-subject, multiple- baseline design	To assess if a 12-week videoconferencing and DVD learning module (P-ESDMg) could improve parents’ acquisition of teaching procedures and result in changes in the child’s social communicative behavior [51].	Parent outcome measures: Eight item, 5-point response scale evaluating parental satisfaction (feasibility and appropriateness) with the support and ease of the intervention	All parents reported satisfaction with support and ease of the telehealth learning intervention. Six parents identified DVD’s as more useful teaching aids compared to handouts. All parents agreed they would recommend an approach to other parents of children with ASD with limited access to community services. Significant increases over time from baseline to follow-up (P<.001)	Kmet rating: good (77%) NHMRC level of evidence: level IV
		P-ESDM fidelity tool—5-point Likert rating tool of 13 parent behavior that define the child-centred, responsive interactive style used in PESDM	Significant increases over time from baseline to follow-up (P<.001)
		MBRSh—A 5-point Likert rating scale measuring the parent’s style of interacting to or relating to their child.	Significant increases in parental behavior rating from baseline to follow-up in responsivity (P<.001), affect (P<.001), and achievement orientated behavior (P<.001)
		Child outcome measures: child social communication behavior—10-min videos transcribed and scored for the production of spontaneous and promoted functional verbal utterances and approximations and imitative play actions on objects and gestures.	Significant overall increases from baseline to follow-up in spontaneous functional verbal utterances (P<.001), prompted words over time (P<.001), and spontaneous imitations (P<.001)
		CBRSi[52]—measures engagement and interest in activity as well as joint attention, creativity, and affect demonstrated toward the parent.	Significant increase form baseline to follow-up in child attention (P<.001) and child initiation (P<.001).
		MacArthur communicative development inventories: words and gestures	Significant increases from baseline to follow-up with vocabulary production (P<.001) and vocabulary comprehension (P<.001).
		Vineland adaptive behavior scales, 2ndedition	Significant increase from baseline to follow-up on the adaptive behavior composite (P<.05).
Vismara et al [12] Single-subject, multiple- baseline design	Pilot study of a 12-week telehealth on the Web (videoconferencing and self-guided website) intervention (P-ESDM) and 3-month follow-up to assess: (1) parents’ perception of the intervention as a useful learning platform, (2) parents’ intervention skills and engagement style improvement, (3) website utility to support the intervention, and (4) improvements in the children’s verbal language and joint attention.	Parent outcome measures: Eight item, 5-point response scale evaluating parental satisfaction with the support and ease of the telehealth learning intervention	All parents reported satisfaction with support and ease of the telehealth learning intervention.	Kmet rating: good (77%) NHMRC level of evidence: level IV
		P-ESDM fidelity tool—5-point Likert rating tool of 13 parent behavior that define the child-centred, responsive interactive style used in P-ESDM	Improvement in parent intervention fidelity. Baseline: 0/8 parents meeting criteria for fidelity in tool. Group mean 2.93 (SD 0.6), post intervention: 6/8 parent meeting criteria for fidelity in tool. Group mean 3.69 (SD.51), follow-up: 7/8 parents achieved at least one fidelity score. Group mean 4.15 (SD 0.51)
		Website use	Average number of logins 30 (SD 18, range 9-60); Average viewing time per day 18 min
		MBRS [53]—A 5-point Likert rating scale measuring the parent’s style of interacting to or relating to their child.	Improvement in parent engagement style. Baseline: low-moderate with MBRS total score mean=2.91, SD=0.68, post intervention: mean=3.50, SD=0.44, follow-up (3 months): moderate to high range with MBRS total score mean=3.87, SD=0.42
		Child outcome measures: behavior scoring of videotaped probes—functional verbal utterances and nonverbal joint attention initiations without gestures	Increase in the range of vocalizations at all time points Baseline: mean=2.97, SD=1.93, post interventions: mean=3.60, SD=2.51, follow-up: mean=4.14, SD=2.04 Joint attention initiations remained constant between baseline (mean=1.67, SD=1.07) to post intervention (mean=1.67, SD=1.21) but increased at follow-up (mean=2.16, SD=1.34)
		MacArthur communicative development inventories: words and gestures	Improvements in VPjand comprehension, Baseline: VP mean=111.87, SD=156.03, comprehension mean=224.37, SD=133.25, post intervention: VP mean=163.88, SD=156.03, comprehension mean=284.88, SD=141.53, follow-up: VP mean=213.88, SD=155.08, comprehension mean=314.88, SD= 94.16
Wacker et al [47] Nonconcurrent multiple baseline design	Conduct functional communication training using coaching from trained behavior analysts to parents via telehealth and compare it with completing the same training in-vivo within families’ homes.	Parent outcome measures: Parent overall appropriateness—7-point Likert scale	Parents rated training as acceptable (mean=6.47. Comparable with in-vivo training (mean=6.18)	Kmet rating: good (73%) NHMRC level of evidence: level IV
		Costs: mileage and consultant costs	Costs through telehealth were considerably lower that for in-home behavior therapy
		Child outcome measures: Interobserver agreement on child-targeted problem behavior using interval-by-interval comparisons.	Reduction in child-targeted problem behavior when parents coached via telehealth (mean reduction=93.5%). Comparable with in-vivo training (mean reduction=94.1%).

^aNHMRC: National Health and Medical Research Council. Designation of levels of evidence: I—Evidence obtained from a systematic review of all relevant randomized controlled trials, II— evidence obtained from at least one properly designed randomized controlled trial, III-1 —evidence obtained from well-designed pseudo-randomized controlled trials (alternate allocation or some other method), III-2—evidence obtained from comparative studies with concurrent controls and allocation not randomized (cohort studies), case-control studies, or interrupted times series with a control group, III-3—evidence obtained from comparative studies with historical control, two or more single-arm studies, or interrupted time series without a parallel control group, IV—evidence obtained from case series, either post-test or pre-test and post-test.

^bABA: applied behavior analysis.

^cASD: autism spectrum disorder.

^dRCT: randomized controlled trials.

^eTA: therapist-assisted group.

^fSD: self-directed group.

^gP-ESDM: parent model—early start Denver model.

^hMBRS: maternal behavior rating scale.

ⁱCBRS: child behavior rating scale.

^jVP: vocabulary production.