OBJECTIVE: The aim is to integrate quantitative and qualitative evidence to understand the effectiveness and experience of advance care planning (ACP) for frail elderly. DESIGN: A mixed-methods systematic review and meta-analysis was conducted. Quality evaluation was conducted using critical appraisal tools from the Joanna Briggs Institute. Data were synthesised and pooled for meta-analysis or meta-aggregation as needed. DATA SOURCES: An electronic search of MEDLINE, CINAHL, Embase, PubMed, PsycINFO, and Cochrane Library databases from January 2003 to April 2022. ELIGIBILITY CRITERIA FOR SELECTING STUDIES: We included experimental and mixed-methods studies. The quantitative component attempts to incorporate a broader study design. The qualitative component aids in comprehending the participant's experience with ACP and its efficacy. DATA EXTRACTION AND SYNTHESIS: Two independent reviewers undertook screening, data extraction and quality assessment. The quantitative and qualitative data were synthesised and integrated using a convergent segregated approach. RESULTS: There were 12 158 articles found, and 17 matched the inclusion criteria. The quality of the quantitative component of most included studies (6/10) was rated as low, and the qualitative component of half included studies (4/8) was rated as moderate. The meta-analysis showed that the intervention of ACP for frail elderly effectively increases readiness, knowledge and process of ACP behaviours. The meta-aggregation showed that the participants hold a positive attitude towards ACP and think it facilitates expressing their preferences for the medical decision. CONCLUSION: ACP is an effective and feasible strategy to facilitate frail elderly to express their healthcare wishes timely and improve their outcomes. This study could provide proof for a better understanding of the subject and help direct future clinical practice. More well-designed randomised controlled trials evaluating the most effective ACP interventions and tools are needed for the frail elderly population.