Table 2. The results of Analysis I for individual and group interventions for each outcome.
Individual | Group | ||||||
---|---|---|---|---|---|---|---|
Outcome | p-value | SMD (95%CI) | I2(%) | p-value | SMD (95%CI) | I2(%) | |
Primary outcome | Autism general symptoms | 0.28 | −0.31[−0.63, 0.01] | 21 | N/A | ||
Secondary outcomes | Developmental quotient | 0.02* | 0.36[0.05, 0.66] | 20 | N/A | ||
Developmental quotient (baseline imbalance-adjusted) | 0.07 | 0.31[−0.02, 0.65] | 26 | N/A | |||
Expressive language | 0.18 | 0.13[−0.06, 0.33] | 0 | 0.92 | 0.11[−0.07, 0.30] | N/A | |
Expressive language (baseline imbalance-adjusted) | 0.18 | 0.13[−0.06, 0.33] | 0 | N/A | |||
Receptive language | 0.19 | 0.17[−0.09, 0.42] | 28 | 0.59 | −0.14[−0.65, 0.37] | N/A | |
Reciprocity of social interaction towards others | p<0.001*** | 0.59[0.25, 0.93] | 18 | 0.04* | 0.45[0.02, 0.88] | 18 | |
Adaptive behaviour | 0.60 | −0.05[−0.25, 0.14] | 39 | 0.47 | 0.44[−0.77, 1.65] | N/A | |
Other outcomes | Qualitative impairment in social interaction | 0.61 | −0.15[−0.40, 0.10] | 0 | N/A | ||
Qualitative impairment in communication | 0.85 | −0.03[−0.35, 0.29] | N/A | N/A | |||
Restricted repetitive and stereotyped patterns of behaviour, interests and activities | 0.17 | −0.21[−0.52, 0.09] | 39 | N/A | |||
Initiating joint attention | 0.13 | 0.48[−0.14, 1.10] | 78 | 0.57 | 0.15[−0.38, 0.68] | 15 | |
Responding to joint attention | 0.11 | 0.63[−0.14, 1.39] | 97 | N/A | |||
Imitation | 0.11 | 0.54[−0.25, 1.33] | 62 | N/A | |||
Parental synchrony | p<0.001*** | 0.99[0.70, 1.29] | 0 | N/A | |||
Parenting stress | 0.35 | −0.30[−0.93, 0.32] | 0 | 0.27 | −0.29[−0.81, 0.22] | 0 |
“p-value” indicates value of the test of overall synthesis. SMD indicates standard mean difference of the synthesised effect. 95% CI indicates 95% confidence interval of the SMD of the overall synthesis.
* and *** indicate statistically significant effects (p < 0.05, p < 0.01, and p < 0.001, respectively) in the analysis.
N/A indicates data synthesis could not be performed due to lack of available studies.
“baseline imbalance-adjusted” indicates results of the analysis after excluding the studies with baseline imbalances.