Background: Autism Spectrum Disorder (ASD) is associated with elevated levels of child behavioural problems that limit child daily activities and impose challenges to parenting. Parent training programmes have been shown to be effective in improving the behavioural adjustment of children with ASD, but there has been limited evidence to date of the effectiveness of such programmes in China. This DPhil thesis involved: a) the conduct of a large systematic review of parent training programmes for families of children with a broad range of developmental disabilities in mainland China and b) a real-world evaluation of the effectiveness of a short-term intensive parent training programme, called Beijing Stars and Rain Education Institute for Autism (SREIA), delivered to Chinese caregivers of young children with ASD in the context of routine service provision. Methods and Analysis: The systematic review involved a search of Chinese and English databases, as well as the grey literature. Where possible, multi-level meta-analyses were conducted to synthesise the data evaluating the effectiveness of such programmes in reducing child emotional and behavioural problems, improving parent-child relationships, and diminishing child abuse. Subgroup analyses and meta-regression were carried out to investigate treatment effects for different diagnoses, including ASD, as well as to explore potential moderators with a focus on the intervention and delivery components. Risk of bias was assessed for each study and outcome domain. The SREIA programme was evaluated using a quasi-experimental design involving a mixed-methods approach. Participants were drawn from the SREIA programme waiting list. Data were collected at baseline and immediately post-intervention. The primary outcome was child behavioural problems measured by the Child Behaviour Checklist Externalising Scale. Secondary outcomes included child ASD symptoms, caregiver mental health, caregiver knowledge, parenting styles, and family function. Between-group comparisons were conducted using a difference-in-differences technique with propensity score weighting. A process evaluation was undertaken in parallel to assess participant involvement, programme acceptability, and programme delivery. Qualitative in-depth interviews were conducted with caregivers who were identified based on participant attendance and engagement rates, and focus group discussions were held with all programme practitioners, with the purpose of developing a deeper understanding of the quantitative results and exploring the experiences of stakeholders in participating in and delivering the programme. Data collection was structured around three theoretical concepts - participant involvement, programme acceptability, and programme delivery - as well as stakeholder perceived changes and areas for improvement. Qualitative data were analysed using thematic analysis with a combination of data-driven and theory-driven approaches. Results: The systematic review identified 20 randomised controlled trials and 11 quasi-experimental studies involving 2410 families of children diagnosed with ASD, Attention Deficit/Hyperactivity Disorder, Epilepsy, or Tourette Syndrome. Overall, the programmes showed effectiveness in reducing child emotional and behavioural problems (g = -1.37, 95% CI [-2.03, -0.71]) and improving the parent-child relationship (g = 0.47, 95% CI [0.21, 0.73]). Only one study assessed the impact on child abuse, finding that the programme reduced physical and verbal harsh disciplines (RR = 0.51, 95% CI [0.28, 0.92]). A subset of ASD-focused programmes was found to have a large effect on child emotional and behavioural adjustment (g = -1.47, 95% CI [-2.42, -0.52]; 9 studies, 627 participants). Moderator analyses suggested that improving parental knowledge, self-regulation strategies, and empathy increased programme success, whereas the use of reinforcement appeared to temporarily increase child behavioural problems. In terms of programme delivery methods, the systematic review found that programmes with a longer duration; a combination of group and individual sessions; rapport building; ongoing communication extended beyond sessions; and delivery in hospitals or service agencies, were more effective. However, the risk of bias was generally high across studies and outcome domains. Another feature of the included programmes was their brevity, although parent training programmes for ASD can often be delivered with higher intensity in contexts such as China, where the need for such programmes is greater. Moreover, there was an absence of evaluations of programmes delivered as part of routine services. The findings of the review highlighted the need for an evaluation of a more intensive parent training programme within a Chinese context of routine service delivery. The final sample size for the real-world SREIA evaluation was 111, with 63 caregiver-child dyads in the treatment group and 48 dyads in the waitlist control group. The statistical analyses showed that the programme successfully reduced child externalising behaviours (b = -2·71, 95% CI [-5·23, -0·18]; Cohen's d = -2.80), compared to the waitlist control group. It was also effective in increasing parental knowledge of Autism and child development (b = 2.08, 95% CI [2.07, 2.17] ; d = 2.91), reducing parental mental health symptoms (b = -5.96, 95% CI [-11.74, -0.17] ; d = -2.89), and diminishing over-reactive parenting (b = -0.63, 95% CI [-0.98, -0.27]; d = -5.02). An investigation of version differences found that the version with a smaller group size was more effective in reducing externalising problems (b = -3.77, 95% CI [-6.26, -1.28]; d = -4.26), improving parental mental health (b = -12.52, 95% CI [-19.87, -5.18]; d = -4.79), and increasing parental knowledge (b = 3.03, 95%CI [1.82, 4.24]; d = 7.05), while the other version with a larger group size had a greater treatment effect in reducing parental over-reactivity (b = -0.72, 95% CI [-1.15, 0.28]; d = -4.68). However, the version analysis was underpowered and thus should be interpreted with caution. The programme had high levels of participant attendance, engagement, and satisfaction. Exploratory analyses suggested that initial parent mental health concerns were inversely related to participant engagement, and that higher satisfaction and engagement could potentially increase treatment effects. The qualitative study identified two caregivers from each of the seven caregiver groups based on their levels of attendance and engagement as rated by programme practitioners on a daily basis, resulting in 14 caregivers participating in the in-depth interviews. Two focus group discussions were conducted with four programme practitioners in each group. Results of the qualitative data highlighted the benefits of practice and feedback during and between sessions, flexible mentoring, peer support, and ASD information that is delivered in the initial sessions, as well as the need for more individualised coaching, ASD-related resource sharing, formal mental health support, structured activities for children, brief training for other family members, and organisational resources for practitioners. Conclusion: Overall, this DPhil thesis provides support to the implementation of programmes that combine parent training and support to caregivers of children with ASD, especially in areas where there is a paucity of ASD-related resources and substantial unmet family need. This thesis also identifies the need for better organisational infrastructure to improve practitioner capacity, offer them mental health support, and reduce their workload. Moreover, it demonstrates the feasibility and advantages of conducting rigorous mixed-methods evaluations in real-world conditions, with the purpose of building the evidence base in low-resource settings. This thesis calls for closer partnerships between implementers, researchers, and programme end users; and for more research to investigate the implementation and effectiveness of parent training programmes delivered in low- and middle-income countries.