Table 1 Summary information on included papers

Bearss and colleagues (2015)

Parent Training (PT)

(Parent Education Programme) (PEP)

91

(89)

4.7 (1.1) yrs.

(4.8 (1.2) yrs)

79

79

67(73.6%) had IQ 70 or above, 7 or 7.7% had missing IQ, 16 (17.6) had IQ below 70

(IQ range 67(75.3) had IQ of 70 or above, 13 (14.6) had IQ below 70 while 10 did not have their IQ available)

16–24 wks

(48 wks)

Parent rated ABC-I,

Parent rated HSQ -ASD

Clinician rated CGI-I

VABS

Week 24 PT showed 47.7% decline in Parent ABC-I (from 23.7 to 12.4) compared to 31.8% decrease (23.9 to 16.3%) for PEP (treatment effect, −3.9,95%CI-6.2 to −1.7; P < .001, standard effect size =0.62

PT, HSQ declined 55.0% (4.0 at baseline to 1.8 at 24 weeks) compared to a 34.2% decrease (3.8 to 2.5) for PEP (treatment effect, − 0.7; 95%CI, − 1.1 to − 0.3; P < .001, standardized effect size = 0.45

On CGI -I, 68.5% (61/89) PT much improved, very much improved compared to 39.6% (36/91) in PEP (P < .001); NNT 4(68.5–39.6% = 28.9%;1/28.9 = 3.5 rounded up to 4)

For children who showed improvement at week 24, retention was 90% (55/61) at week 48. Of those who did not achieve good response at week 24, 17/21(81%) returned at 48 weeks, mean scores on ABC-I and HSQ-ASD were lower than baseline but showed upward trend from wk24. Available participants N = 23 PEP maintained benefit to wk48.

On CGI-I, PT, 48/61(79%) remained much improved at 48 weeks; those who did not improve on PT 9/28 (32%) were rated much improved by blind rater. Of the PEP children 16/23 (70%) maintained positive response at 48 weeks.

Iadarola and colleagues 2017

From Bearss and colleagues 2015 Parent training and parent education - Effects on Stress, Strain, and Competence

Same cohort as Bearss and colleagues 2015

Same cohort as Bearss and colleagues 2015

Same cohort as Bearss and colleagues 2015

Same cohort as Bearss and colleagues 2015

24 weeks

PSI

PSOC

CGSQ

Parent rated ABC-I

On the PSI total score, PT showed a 14%reduction, and PEP showed 9.3% reduction.

On the CGSQ global score, PT showed 17.2% reduction, and PEP showed 7.1% reduction. For PSOC total score, PT showed 16.4% increase, and PEP showed 7.4% increase.

On the PSI difficult child factor, PT produced greater reductions than PEP at week 12 and week 24. The reduction in the PSI total score was greater in PT than PEP at week 24, but the difference was not significant. At week 12 and week 24, PT was superior to PEP on the CGSQ global score and Internalized subscale. The CGSQ Objective subscale reached significance at week 24. On the PSOC, parents in the PT group reported greater gains than parents in PEP at week 12 on the satisfaction subscale but not the efficacy subscale or total score. Improvement was significantly greater in PT compared to PEP on the PSOC total score and the efficacy subscale at week 24. The difference on the satisfaction subscale was no longer significant.

Parents in both groups also reported significant decrease in stress (PT: β = − 0.38, p = .009; PEP: β = − 0.39, p = .006) and strain (PT: β = − 0.50, p < .001; PEP: β = − 0.45, p < .001) from week 12 to week 24. change in stress and strain from week 12 to 24 did not significantly differ across groups (z = 1.15 and z = 0.74, respectively). Mixed method analysis didn’t show significant difference between groups between 12 and 24 weeks.

Bradshaw and colleagues 2018

From Bearss and colleagues 2015- parental response to active control treatment

Same cohort as Bearss and colleagues 2015

Same cohort as Bearss and colleagues 2015

Same cohort as Bearss and colleagues 2015

Same cohort as Bearss and colleagues 2015

24 weeks.

CGI-S

Parent rated-HSQ -ASD, ABC-I, ECI

VABS

PSI

CGSQ

PSOC

PHQ

IQ using Stanford Binet

PEP-R and PEP-NR groups were equally matched were similar on child clinical measures and parent self-report measures, including the CGI-S, ABC, HSQ, ECI, VABS, PSI, CGSQ, or PSOC

The only difference between groups was a higher rate of regular educational placement in the PEP-NR group than PEP-R group (60% vs. 36%, p = 0.04). The rate of positive response was not different for children with IQ ≥ 70 compared to those < 70.

Compared to parents of children in the PEP-NR group, parents of PEP-R children reported significantly greater reductions on the PSI total score, PSI Parent-Child Dysfunctional Interaction subscale, PSI Difficult Child subscale, CGSQ total score, and PHQ-4 total score. Parents of PEP-R also showed greater improvements in the PSOC total score and the PSOC Efficacy domain.

There was no significant difference between the children who responded to PEP and PT on ABC, HSQ, VABS or parent reported scales.

Harden and colleagues 2015

Pivotal Response Treatment Group -PRTG

(Psychoeducation – PEG)

25

(23)

4.1 years (1.2)

(4.1 years) (1.3)

19

17

52.8 - DQ

53.5 - DQ

Final assessment at 12 weeks

SLO

CDI

VABS

MSEL communication subscales

Irrespective of group, children showed significant improvement in total number of utterances across study time points (F(2, 43) =6.12, p = .005) Individuals receiving PRTG showed greater improvement (F(2,43) = 3.53,p = .038), treatment effect was most apparent for imitative (F(2, 43) = 7.67, p = .001) and nonverbally prompted utterances (F(2,43) = 7.06,p = .002 Spontaneous utterances showed a nonsignificant trend towards greater improvement in PRTG.

At week 12, 21/25 of PRTG and none of PEG met fidelity of PRT implementation. Significant treatment effect observed for Vineland Communication scale, with individuals on PRTG (F(2,19) = 3.80, p = .041)

Nonsignificant trend for CDI with children with PRTG saying longer utterances.

Treatment effect was observed for CGI- severity of social and communication symptoms (F2, 42) =6.84, p = .003) and CGI improvement ratings were significantly better in PRTG relative to PEG at weeks 6 and 12 F(1,44) = 15.97, p.001)

Older children with higher IQ had more total utterance. Baseline Mullen visual reception scores were a significant predicator of treatment response.

Gengoux, and colleagues 2019

Pivotal Response Treatment -Package-PRT-P

Delayed treatment group-DTG

23

(20)

49.5 (11.2) months

47.2 (10.0)

21 boys

17 boys

Not available

Final assessment at 24 weeks

SLO- at baseline, week 12, and week 24

SLO, videos, scored using the BOSCC, higher scores indicate greater impairment

The CDI Words and Gestures,

The CDI Words and Sentences,

VABS

MSEL communication subscales

CGI-S

CGI-I

Children participating in the PRT-P showed significantly greater overall improvement between baseline and week 24 in total number of Utterances (F1,41 = 6.07; P = .026) compared with children in the DTG, seen at all time periods, by an increase in nonverbally prompted utterances in PRT-P group (F1,41 = 16.409; P, .001)

Improvement in the PRT-P group was observed on the BOSCC social communication subscale and in the BOSCC total score, across the three time points, F2,39 = 17.597; P, 0.001,

A significant treatment effect was observed for the CDI words produced out of 396 and CDI words produced out of 680 measures, even when controlling for baseline differences.

The treatment effect was also significant on the CGI-S subscale for social communication symptoms (F1,41 = 5.91; P = .019).

Significant group difference was also evident on the CGI-I subscale (F1,41 = 6.86; P < .001).

Although not statistically significant, effect-size calculations suggested a medium-size treatment effect for the Vineland-II expressive v-scale score.

Nefdt and colleagues (2010)

Self-Directed learning Program to provide introductory learning in pivotal response treatment -PRT

Wait list Group-WLG

13

(14)

38.92 months (SD = 14.57)

(38.43(SD = 11.20))

25 males

Not available

No follow up

Fidelity of using PRT procedures; Language opportunities that parents provided;

Child’s verbal utterances;

Parental confidence – measured on a 6-point scale

PSI

Of the 34 dyads who entered the study – 27 (79.4%) completed.

Parents in the TG used motivational procedures of PRT F = 107.02 and p = .000, effect size 4.12

They provided more language opportunities, F = 91.58 and p = .000 and effect size of 2.23

Parent confidence increased, F = 16.37 and p = .001, effect size 1.28

There was a significant difference in child utterances, between training and control group F = 16.23 and p = .001 and effect size .953

Parents found self-directed learning programme easy to understand, useful and informative, changed the way they interacted with their child and reported that their child was trying to communicate with them

Jocelyn and colleagues (1998)

Autism Preschool Program

(Day care centre with childcare worker)

16

(19)

42 + −9.2 months

(43.8 + − 9.0 months)

15

19

58.4 + −27.5 -Leiter IQ

(67.1 + − 27.5)

15 weeks

ABC

EIDP

PDP

TRE-ADD Autism Quiz

Client satisfaction measure

Mothers and CCW of children in the intervention program reported significant increase in understanding of autism on TRE-ADD Autism Quiz (mothers p = 0.02; CCW p = 0.008) compared with mothers and CCWs in the control group.

Autistic Symptomatology ABC not significantly different. Parents of all children reported improvement over time p = 0.000, difference between the two groups was not significant.

Developmental outcome – a significant difference was only seen in language score, experimental group language score changed by 5.3 + − 5 months while control group changed by 1.1 + − 4.6 months.

Client satisfaction levels – parents reported higher degrees of satisfaction in the experimental group p = 0.00007 on knowledge, ability to deal with the needs p = 0.009 and how best to meet them p = 0.002 compared to parents on the control group.

Malow and colleagues (2014)

Sleep Education – Individualised program

(Group Education Program)

47

(33)

5.6 (2.6) years

5.9 (2.8) years

39

25

27 (64%) IQ > 70

15 (45%) IQ > 70

No follow up.

actigraphy -change in latency, wake time after sleep onset, total sleep time

CSHQ

FISH

CBC

PSOS

RBS-R

pedsql

Actigraphy: No difference between the two arms – so results combined which showed improvement in sleep latency, combined mean reduction from 58.2 min to 39.6 with treatment (p < 0.0001), in 36 of children, sleep latency was less than 30 min on more than 5 nights per week, there was a modest 2.9% improvement in sleep efficiency and wake time after sleep and total sleep time did not improve.

No difference was noted in the questionnaires on how the training was given but in the complete data set, improvements were noted in all the insomnia related parameters, behaviours related to anxiety and depression, withdrawal, attention, repetitive behaviours, parenting efficacy and satisfaction and paediatric quality of life. Parents reported a high level of satisfaction with the program and educator.

Oosterling and colleagues (2010)

Parent based intervention – The Focus Parent Training

(Care as Usual)

36

(31)

24 months

(33.3 months)

27

25

58 DQ

(58) DQ

12 months (12 months)

MCDI

ADOS

CGI-I

Erikson Child and Parent Scales

On all language measures there was a main effect of time, meaning that the language skills of children in both groups improved with time.

The change in clinical global improvement, as measured with the CGI-I, from baseline to endpoint was not different between the two groups.

Regarding engagement, no intervention effects were found.

The mothers in the experimental group did not show an improvement in parenting skills relative to the mothers in the control group.

Drew and colleagues (2002)

Social pragmatic joint attention focused parent training programme

(Local Service Model)

12

12

21.4(2.7) months

(23.6(3.8) months)

11

8

88.1(11.2) Nonverbal IQ

(66.0(16.5))

12 months

(12 months)

CDI,

Nonverbal IQ- D &E scales of Griffiths Scales of Infant Development.

ADI-R

PSI

Parent training group had marginally higher language comprehension measured by CDI total words though this missed statistical significance. There were no group differences on words produced or gestures produced.

Significantly more children from parent training group moved from nonverbal to having single words or phrase speech (Fisher exact test p < 0.05)

There was no difference in ADI -R or PSI scores.

Pajareya & Nopmaneejumruslers (2011)

Developmental Individual Difference, Relationship based (DIR) /Floortime™ parent training intervention

(Routine Treatment)

16

(16)

56.6 months (SD 10.1)

(51.5(13.9) SD)

15

13

44.0 (12.9) FEDQ

40.7(15.3) FEDQ

3 months

FEAS

CARS

FEDQ

During the study period the intervention group used DIR/Floortime TM at an average of 15.2 h per week SD = 12.4

One family from the intervention group did not complete the study. Analysis including that child showing no improvement, showed that difference in FEAS was significant (F = 4.6, p = .045), change in CARS was F = 1.9, p = .004, and change in FEDQ was F = 6.4, p = .007

Ho and Lin 2020

Training Programme based on the DIR-Developmental Individualised difference Relationship based model

Training Programme based on the developmental milestones.

12

(12)

48.7 (7.4) months

48.3(6.9) months

All boys

All boys

Not available

Only post intervention, no follow up.

FEAS

CPEP-3

VABS

The FEAS scores for the children and caregivers in the intervention group were much higher than for those in the control group at the end of 14 weeks. The results of repeated measures analyses of variance show that significant interactions were evident between the study group and time for the children’s emotion development, F (1, 22) = 7.559, p = .012, η2 = .274, and parenting skills, F (1,22) = 8.447, p = .009, η2 = .297.

However, no significant interactions were evident between the study group and time for the children’s developmental abilities

Rogers and colleagues (2019)

Early Start Denver Model -ESDM

Treatment as Usual in three sites

T1 55

T2 51

T3 47

T4 44

(T1 63

T2 52

T3 43

T4 36)

T1 20.58(3.37)

T2 24.33(3.18)

T3 36.47(3.24)

T4 48.53(3.05)

T1 20.70(3.21)

T2 24.13(3.48)

T3 36.58(3.55)

T4 48.75(3.77)

41

51

66.98

64.52

24 months (27 months after enrolment)

Language composite age-equivalent score from Expressive Language and Receptive Language age equivalents of the MSEL at each time point.

DQ

VAB

ADOS

When all three sites were taken together, there was a significant change in favour of ESDM but when sites were analysed separately, in sites 1 and 2 there was a significant effect of treatment on the trajectory of language with the ESDM group increasing more than the community group over time. For site 3, although the ESDM group increased, it was less over time than the community group, the group difference was nonsignificant.

ADOS and DQ did not differ across groups, no difference in Adaptive behaviour age equivalents.

Sofronoff & Farbotko (2002)

Parent management training aimed to improve parental self-efficacy in management of problem behaviours

(Wait List Control Group)

69

(20)

8 years, 3 months

(8 years, 3 months)

Not available

-Not available

6 weeks (3 months)

Parental self-efficacy;

ECBI

Parental assessment of a child’s behaviour problems

Compared with the control group, parents in both intervention groups reported fewer problem behaviours and increased self-efficacy following the interventions, at both 4 weeks and 3 months follow-up. The results also showed a difference in self-efficacy between mothers and fathers, with mothers reporting a significantly greater increase in self-efficacy following intervention than fathers. There was no significant difference between the workshop format and the individual sessions.

Tonge and colleagues (2014)

Parent education and counselling -PEAC;

Parent education and behavioural management-PEBM

(Business at Usual)

70

(35)

10 years

(10 years)

55

32

PEAC group DQ- 48.71

PEBM group −64.74

DQ-63.31

20 weeks (6 months)

VABS

DBC

PEP

RDLS-III

There was a significant improvement in the communication skills of the children whose parents received PEBM compared to the ‘business-as-usual’ control group, but only for the children who had more communication delay. PEBM group performed better than PEAC on VABS – daily living domain, on VABS- socialisation both groups performed better than control group

Keen and colleagues 2010

Professionally supported parent focussed intervention

Self-directed video-based intervention

17

(22)

36.38(7.54) months

35.71 (6.92) months

15 boys

16 boys

53.06(9.06)

62.86 (6.53)

3 months

SIB-R

CSBS-DP

MSEL

Parent Measures self-reports

PSI

PSOC

Both intervention type and parent gender had a significant influence on child-related stress.

Fathers experienced higher levels of child-related stress than mothers, the professionally supported intervention reduced child-related stress relative to the self-directed intervention for both mothers and fathers.

For self-efficacy there is an interaction for intervention group by baseline score. Parents low in self-efficacy at baseline demonstrated relatively higher levels of self-efficacy if they received the professionally supported intervention than if they received the self-directed intervention.

Tellegen and Sanders 2014

Primary Care Stepping Stones Triple P (PCSSTP).

Care as Usual

35

29

5.66 (2.18) years

5.69 (2.2) years

29 boys

26 boys

Not available

6 months

ECBI

PS

DASS 21

PSS

Observation of parent child interaction coded according to family observation schedule.

PPC

RQI

In the short term, there were significant decreases in the intervention group on laxness, F(1, 34) = 26.91, p < .001; verbosity, F(1,34) = 22.81, p < .001; and over reactivity, F(1, 34) = 30.99, p < .001. Parenting confidence found a significant multivariate interaction effect, F (2, 61) = 12.63, p < .001, with significant univariate effects on both subscales.

A significant multivariate interaction effect on parental adjustment was found, F (4, 59) = 4.56, p < .003. Follow-up tests revealed a significant decrease for the intervention group on DASS–21 stress, F(1, 34) = 21.38, p < .001, and the PSS, F(1,34) = 24.21, p < .001,

There was a significant multivariate interaction effect for child behaviour problems, F(2, 61) = 3.33, p = .042, with effects on both scales.

In the long term, there was maintenance of improvement of child behaviour problems. Dysfunctional parenting styles -scores were still significantly better than preintervention.

Differential improvement in parental confidence was significantly maintained.

Reduction in stress was significantly maintained.

There was no significant change in observed child parent behaviours.