Development of a Family-Based Prevention Model for Adolescent Mental Health Problems: A Socio-Ecological Approach in the Indonesian Context

Abstract

Background

Adolescent mental health is a growing global concern, particularly in low- and middle-income countries where preventive frameworks remain underdeveloped. In Indonesia, existing adolescent mental health initiatives remain largely school- and health-service oriented, with limited operational integration of families as preventive actors. This study aimed to develop a Family-Based Prevention Model for Adolescent Mental Health Problems using a socio-ecological approach tailored to the Indonesian context.

Methods

Employing an exploratory sequential mixed-methods design, qualitative data from adolescents, parents, educators, health providers, and policy stakeholders were analyzed to construct a conceptual model, followed by quantitative validation through a three-round Delphi process with 14 experts.

Results

The qualitative phase identified five key dimensions of the socio-ecological model: individual, family, organizational, community, and policy, with the family dimension emerging as the most influential factor in adolescent mental health prevention. Within the socio-ecological framework, the family dimension emerges as a central determinant in adolescent mental health, given its position as the closest and most immediate environment to adolescents. The themes identified within the family dimension included family attachment, family supervision, family communication, parenting practices, family relationships, and family support. Quantitative findings achieved full expert consensus across six elements, confirming the model’s coherence and applicability.

Conclusion

The model positions family engagement as the core preventive mechanism, rather than a family-only approach, reinforced by cross-sectoral collaboration and enabling policy environments. This study contributes to the body of knowledge by integrating family processes within the socio-ecological model and offering a culturally grounded framework for adolescent mental health promotion. Further empirical testing is required to evaluate the model’s effectiveness across diverse Indonesian regions and adolescent age groups. The model has the potential to inform national strategies and guide future empirical evaluations aimed at strengthening family-based prevention and intersectoral collaboration in Indonesia.

Introduction

Adolescence marks a dynamic stage of life involving profound developmental transitions that shape mental well-being. Worldwide, the escalating prevalence of mental health disorders among adolescents has emerged as a pressing public health challenge. According to the World Health Organization, mental disorders affect one in seven adolescents aged 10–19 years, accounting for an estimated 15% of the global disease burden in this population.¹ In low-and middle-income countries (LMICs), the prevalence of adolescent mental health problems ranges from 10% to 20%, highlighting their growing recognition as a significant global concern.^2,3 Mental health problems in adolescents play a significant role in the global disability burden.⁴ Most adolescents experiencing mental health problems do not seek or receive appropriate treatment, leading to persistent emotional distress, limited social functioning, and reduced quality of life. They are also at increased risk of stigma, social exclusion, discrimination, educational challenges, risky behaviors, and human rights violations.¹ These worldwide trends highlight the pressing need for prevention-focused approaches that address early adolescence as a crucial stage for intervention. The present study focuses on adolescents aged 12–15 years in the junior high school context, as this developmental stage represents a critical window for early prevention, given that approximately 50% of mental disorders begin by age 14.⁵

The situation in Indonesia mirrors global concerns regarding adolescent mental health. Indonesia, classified by the World Bank as an upper-middle-income country, is included among the low-and middle-income countries (LMICs).⁶ Findings from the Indonesia National Adolescent Mental Health Survey (I-NAMHS) reveal that one in three Indonesian adolescents experience mental health problems (equivalent to 15.5 million individuals) and one in twenty (2.45 million) suffer from diagnosable mental disorders, only 2.6% of those with mental health problems access counseling or mental health services.⁷ These data illustrate the wide gap in mental health service utilization and underscore the urgency of developing comprehensive prevention-oriented approaches. This aligns with Indonesia’s ongoing health system transformation, which emphasizes prevention as the main focus of primary care in the first pillar of reform.⁸ In a setting where formal service utilization remains limited, prevention strategies that mobilize adolescents’ closest social environments, particularly families, are especially important.

Several national initiatives aimed at preventing mental health problems for adolescents have been implemented in Indonesia. Three key ministries have introduced programs addressing adolescent mental health. The Ministry of Education and Culture has youth counseling services in school.⁹ The Ministry of Health provides community-based services for adolescents that include screening and counseling.¹⁰ The Ministry of Women’s Empowerment and Child Protection issued school-based guidelines to address emerging psychosocial problems among students.¹¹ However, the existing national framework does not explicitly include family participation as an integral component of mental health promotion and prevention. This omission is critical because families play an irreplaceable role in shaping adolescents’ psychological development and well-being, functioning as a key protective factor against psychological distress.¹² The absence of family-based approaches creates a gap between policy and the lived realities of adolescents, whose mental health is deeply influenced by their family environment. Thus, a systematic model that integrates family engagement into the socio-ecological framework of prevention is needed to strengthen existing efforts. Such a model should not position the family as the only intervention setting, but rather as a central preventive hub that interacts with schools, health services, community resources, and policy environments.

Research across diverse settings has repeatedly demonstrated that families serve as powerful agents of prevention in adolescent mental health. Berkel et al found that family-based prevention programs in the United States effectively reduced psychological distress and risk behaviours associated with racial discrimination by strengthening parenting skills, particularly through core parenting components such as consistent discipline, monitoring, warmth, and involvement.¹³ Similarly, Stormshak et al highlighted the importance of family management domains, including positive behavioral support, boundary setting, and relationship building, in promoting adolescent psychological well-being.¹⁴ In South Africa, Donenberg et al reported that culturally tailored family programs enhanced adolescent resilience, reduced anxiety symptoms, depression and trauma.¹⁵ These studies collectively demonstrate the universal relevance of family engagement as a cornerstone of adolescent mental health promotion. However, most established family-based prevention evidence originates from high-income or non-Indonesian settings, and its direct transferability to Indonesian adolescents requires contextual adaptation. Family-based prevention strategies may improve one or more indicators of adolescent mental health and behavioral outcomes by actively engaging family members in the intervention process. Overall, the family constitutes a powerful context for delivering preventive interventions.¹⁶ In this study, family-based prevention is conceptualized not as a family-only intervention, but as an approach that places family processes at the center of interaction with schools, health services, communities, and policy systems.

The socio-ecological model, initially conceptualized by Bronfenbrenner in the 1970s, offers a comprehensive theoretical foundation for developing such family-centered prevention approaches. The model posits that human health and behavior are shaped by dynamic interactions among multiple levels of influence—individual, interpersonal, organizational, community, and policy.^17,18 Within this study, Bronfenbrenner’s ecological perspective provides the overarching structure for identifying multiple levels of influence on adolescent mental health. This framework has been successfully adapted for various public health domains, including development of school belonging,¹⁹ athlete safety,²⁰ and resilience promotion among marginalized youth.²¹ Michaels et al extended the socio-ecological model specifically to mental health and wellbeing, emphasizing that individual outcomes are influenced not only by family dynamics but also by broader organizational and community structures.²² Consequently, the socio-ecological approach provides a robust theoretical lens for constructing integrative and multi-level preventive interventions that connect adolescent-level factors with family functioning, school systems, community resources, and policy support.

In Indonesia, the socio-ecological model offers a promising structure for reforming adolescent mental health prevention efforts. Although national guidelines acknowledge schools and community health centers as pivotal implementation units, family engagement remains underrepresented. The existing policies, including the Mental Health Act,²³ emphasize promotive and preventive mental health efforts but do not operationalize the role of family within these frameworks. The socio-ecological perspective can address this gap by positioning the family as a central actor interconnected with institutional and policy layers. As Brady et al argue, conceptual models are essential tools for guiding evidence-based program and policy development.²⁴ A family-based socio-ecological model can therefore serve both as a theoretical guide and a practical instrument for designing contextually relevant prevention strategies. This is particularly important in Indonesia, where adolescent help-seeking may be shaped by family hierarchy, parental authority, community norms, religious values, and stigma surrounding mental health problems.

Building on previous frameworks, Kumpfer developed a family-based prevention model for substance abuse and impulse control disorders, encompassing three domains: individual (intrapersonal) factors such as self-control, academic achievement, and self-efficacy; family (interpersonal) factors such as attachment, supervision, and communication; and environmental (social and community) influences.²⁵ Integrating these domains into a socio-ecological approach allows for a multidimensional understanding of adolescent well-being. The Kumpfer model’s emphasis on the family’s mediating role aligns with ecological assumptions that sustained behavioral change depends on reinforcing supportive relationships across contexts. Michaels et al expanded the socio-ecological approach to mental health promotion by linking individual, family, organizational, community, policy and society determinants into a unified framework for psychological wellbeing.²² Kumpfer’s model informs the family-process components of the present study, particularly attachment, supervision and communication, enriched by conceptual elements from Michaels’s framework, including parenting practices, relationships, and support. Michaels’ framework helps translate ecological thinking into mental health promotion by clarifying how family processes are embedded within broader organizational, community, and policy environments. Applying these principles within the Indonesian context offers an opportunity to develop a culturally grounded, family-inclusive prevention model that bridges the current implementation gap.

The existing literature reveals a growing consensus on the necessity of family involvement but also a persistent lack of context-specific models for LMICs, particularly Indonesia. Previous Indonesian prevention guidelines have been segmented across ministries without a unifying framework to ensure coherence or sustainability. Furthermore, cultural norms, socioeconomic disparities, and limited parental literacy regarding mental health pose additional challenges to effective family engagement. In the Indonesian context, these challenges may be intensified by variation in family structure, parent-adolescent communication norms, school-family relationships, community expectations, and stigma related to seeking psychological support. Comparative studies from high-income settings underscore the importance of tailoring interventions to local sociocultural realities.²⁶ However, empirical evidence guiding such contextual adaptation in Indonesia remains scarce. The integration of socio-ecological principles with a family-based model represents an innovative response to this gap by aligning global best practices with national health priorities and by clarifying how family engagement can be operationalized within existing school, primary healthcare, community, and policy structures.

Therefore, the present study aims to develop a family-based prevention model for adolescent mental health problems using a socio-ecological approach contextualized to Indonesia. This model seeks to operationalize family involvement within the five ecological levels—individual, family, organizational, community, and policy—while promoting intersectoral collaboration among schools, healthcare systems, and government agencies. The novelty of this study lies in synthesizing Bronfenbrenner’s ecological systems perspective, Kumpfer’s family-based model and Michaels’ socio-ecological framework into an applied conceptual structure tailored to the Indonesian health transformation agenda. Bronfenbrenner’s framework defines the ecological levels, Kumpfer’s model specifies core family protective processes, and Michaels’ framework links these processes to mental health promotion within broader systems. By emphasizing family as the foundation of adolescent mental health promotion, this research contributes a context-sensitive model that not only addresses existing policy gaps but also provides a scalable pathway for enhancing adolescent wellbeing. Ultimately, this model aspires to inform evidence-based interventions and policy integration efforts aimed at achieving sustainable mental health prevention across Indonesia’s diverse sociocultural landscape. The study focuses on model development and expert consensus validation; therefore, future empirical testing is needed to determine effectiveness across different Indonesian regions, socioeconomic groups, and adolescent developmental stages.

Materials and Methods

Research Design

This study employed an exploratory sequential mixed-methods design, which integrates qualitative and quantitative approaches to comprehensively explore and validate the development of a family-based prevention model for adolescent mental health problems. Mixed-methods research provides a synergistic integration of qualitative exploration and quantitative verification to generate more robust and contextually grounded findings. The exploratory sequential design begins with a qualitative phase to explore the perspectives of participants, followed by a quantitative phase that builds upon the qualitative results to confirm or refine the findings.²⁷ This design was selected because the study aimed first to identify context-specific prevention components from local stakeholders and then to validate the relevance and applicability of these components through multidisciplinary expert consensus.

In the present study, the first phase employed qualitative methods to explore the various factors influencing adolescent mental health and to identify the components necessary for constructing a family-based prevention model. The data derived from this stage informed the second, quantitative phase, which utilized the Delphi method to evaluate and achieve expert consensus on the proposed model. The Delphi technique is particularly suitable for studies that require structured communication among experts to reach collective agreement on complex or emerging issues.²⁸ The integration between phases occurred through the transformation of qualitative themes into Delphi questionnaire items, allowing expert assessment of whether the empirically derived components were conceptually coherent, contextually appropriate, and feasible for implementation.

Research Subjects

The study was conducted in two stages, each involving distinct participant groups selected through purposive and theoretical sampling techniques. In the first, qualitative phase, participants were chosen based on their relevance to adolescent mental health and their potential to contribute to the development of a holistic model. Theoretical sampling ensured that data collection continued until no new insights emerged. Saturation was considered achieved when subsequent interviews and FGDs no longer generated new categories or substantially different explanations related to adolescent mental health prevention. The informants included adolescents aged 12–15 years enrolled in junior high schools, their parents, school counselor, healthcare workers responsible for the Adolescent Care Health Services in community health centers, youth cadres, psychologist, psychiatrist, pediatrician, and policy stakeholders from the Health Department, the Child Protection Department, as well as principal and head of community health centers. Adolescents aged 12–15 years were selected because this age range corresponds to the junior high school period in Indonesia, a developmental stage in which adolescents experience rapid psychosocial transition and increasing exposure to school, peer, family, and digital environments. A total of nine adolescents and nine parents participated in the FGDs, whereas 12 participants were involved in the in-depth interviews. Thus, the qualitative phase involved 30 participants in total. Participant Characteristics are presented in Table 1.

Table 1.

Participant Characteristics

Characteristics	f	%
Focus group discussions (FGDs) Participants (n=18)
Adolescents (n=9)
Age
13 years old	5	55.56
14 years old	4	44.44
Gender
Male	5	55.56
Female	4	44.44
Parents (n=9)
Age
35–45 years old	4	44.44
46–55 years old	5	55.56
Gender
Male	1	11.11
Female	8	88.89
Education level
Primary school	1	11.11
Senior high school	2	22.22
Bachelor’s degree	6	66.67
Occupation
Housewife	6	66.67
Civil servant	1	11.11
Small business owner	1	11.11
Self-employed	1	11.11
In-Depth Interview Participants (n=12)
Age
<30 years old	1	8,33
31–40 years old	5	41.67
41–50 years old	3	25.00
51–60 years old	2	16.67
>60 years old	1	8.33
Gender
Male	4	33.33
Female	8	66.67
Participant Role
School counselor	1	8.33
Healthcare workers	3	25.00
Youth cadres	1	8.33
Psychologist	1	8.33
Psychiatrist	1	8.33
Pediatrician	1	8.33
Health policymaker	1	8.33
Child protection officer	1	8.33
School principal	1	8.33
Head of community health center	1	8.33

In the second, quantitative phase, the study involved a panel of experts who assessed the validity and applicability of the developed model. The Delphi panel consisted of 14 experts including a psychologist, psychiatrist, pediatrician, family medicine specialists, healthcare workers responsible for the Adolescent Care Health Services in community health centers (nurses and midwives), school counselor, parent of adolescents aged 12–15 years, school principal, head of community health center, and policy representatives from local government health and child protection agencies. Table 2 shows the characteristics of experts involved in the Delphi study. The number of experts in a Delphi study can vary depending on the complexity and scope of the issue under investigation.²⁹ In this study, expert selection prioritized diversity of expertise and institutional representation to enhance the comprehensiveness of the consensus process. Experts were recruited based on their direct professional, academic, or policy-related experience with adolescent health, mental health, family health, school counseling, primary care, or child protection. This multidisciplinary composition was intended to capture clinical, educational, family, service-delivery, and policy perspectives, although it was not designed to statistically represent all Indonesian regions or stakeholder groups.

Table 2.

Characteristics of Expert in Delphi Study

Characteristics	f	%
Age
31–40 years old	3	21.43
41–50 years old	5	35.71
51–60 years old	6	42.86
Gender
Male	4	28.57
Female	10	71.43
Education level
Senior high school	1	7.14
Diploma or Bachelor’s Degree	3	21.43
Professional Degree	3	21.43
Master’s or Specialist Degree	4	28.57
Doctoral Degree (PhD)	3	21.43
Area of Expertise
Psychologist	1	7.14
Psychiatrist	2	14.29
Pediatrician	1	7.14
Family medicine specialists	1	7.14
Healthcare workers	3	21.43
School counselor	1	7.14
Parent of adolescent	1	7.14
Health policymaker	1	7.14
Child protection officer	1	7.14
School principal	1	7.14
Head of community health center	1	7.14

Research Methods

Qualitative Phase

The qualitative phase adopted a constructivist paradigm, which assumes that knowledge is socially constructed through the interaction between researcher and participants.³⁰ This paradigm is well-suited for exploring lived experiences and generating theoretical insights grounded in participants’ perspectives. A phenomenological approach was employed to capture in-depth meanings of participants’ lived experiences regarding adolescent mental health issues and preventive factors within family and community contexts.³⁰ Data collection was conducted through in-depth interviews and focus group discussions (FGDs). FGDs were held separately for adolescents and parents to facilitate open communication and minimize social desirability bias. FGDs were used for adolescents and parents because group interaction was expected to stimulate discussion of shared experiences, family communication patterns, and perceived prevention needs. Separate groups were used to protect adolescents’ comfort and reduce the influence of parental authority during discussion. Individual interviews were used for professionals and policy stakeholders because their roles, institutional experiences, and decision-making perspectives required more detailed exploration and could involve sensitive service or policy information. Interviews were guided by semi-structured protocols derived from the conceptual framework developed through the literature review. The guides included open-ended questions on adolescent mental health problems, family roles, school and community support, service access, policy context, and potential prevention strategies.

The qualitative phase aimed to identify psychosocial determinants of adolescent mental health, risk and protective factors within families, and the multi-level elements influencing preventive strategies. The trustworthiness of the qualitative data was ensured by meeting four criteria: credibility, transferability, dependability, and confirmability.³¹ Triangulation was achieved through the inclusion of diverse participant categories, peer debriefing, and iterative validation of emerging themes. Credibility was strengthened through member checking of key interpretations with selected participants and discussion of emerging themes within the research team. Dependability was supported by maintaining an audit trail of interview guides, field notes, transcripts, coding decisions, and theme development. Confirmability was enhanced through peer debriefing and reflexive discussion among researchers to reduce individual interpretive bias. Transferability was supported by describing participant categories, study setting, and the socio-cultural context of Yogyakarta. To protect confidentiality, transcripts were anonymized by removing names and other identifying information before analysis.

Quantitative Phase

The second phase of the study employed the Delphi method to obtain expert consensus regarding the components of the family-based prevention model. The Delphi technique followed three iterative rounds. In Round 1, the research team designed the initial questionnaire based on the conceptual model derived from the qualitative findings. The questionnaire was developed by converting qualitative themes and subthemes into model elements and indicators, which were then reviewed for clarity, relevance, and conceptual alignment with the socio-ecological framework. This round also involved a brainstorming session with experts to refine item clarity and content validity. In Round 2, the revised questionnaire was distributed to the expert panel, who rated the importance and relevance of each model component. Their responses were analyzed, and the instrument was modified according to aggregated feedback. Items that did not meet the predefined consensus criteria were planned to be revised, merged, removed, or redistributed in the subsequent round based on expert comments. In Round 3, the updated questionnaire was redistributed to the same panel to confirm agreement levels and finalize consensus. Open-ended comments were reviewed after each round to identify reasons for disagreement and to refine item wording without changing the underlying meaning of the qualitative findings.

Each Delphi round was separated by an interval of four weeks to allow experts adequate time for reflection and response. The Delphi process sought to reach a collective judgment about the essential components of the model, balancing individual expert insights with statistical aggregation. The iterative nature of the process enhanced reliability and minimized bias, ensuring that the final model reflected the shared professional consensus of multidisciplinary experts. A three-round Delphi design involving 10–30 experts is generally sufficient to achieve stable consensus in health research contexts.³² Therefore, the inclusion of 14 experts was considered appropriate for a model-development study requiring multidisciplinary judgment rather than statistical power for hypothesis testing. Consensus was summarized using the median, percentage agreement, and interquartile range (IQR) to show both the level and convergence of expert ratings.

Data Collection Procedures

The study was conducted from October 2024 to July 2025 in Yogyakarta, Indonesia. During the qualitative phase, data were collected using semi-structured interview guides that included open-ended questions exploring participants’ perceptions of adolescent mental health and family roles. These guides were informed by theoretical constructs identified in prior literature. Questions were adapted to each participant group while maintaining comparable domains across adolescents, parents, professionals, and policy stakeholders. FGDs were audio-recorded and transcribed verbatim with participants’ consent. Each discussion lasted approximately 60–90 minutes and was conducted in neutral community settings to foster open dialogue. In-depth interviews with other informants lasted between 45 and 60 minutes. Field notes were used to document non-verbal observations and contextual information relevant to interpretation.

In the quantitative phase, the data collection process began with brainstorming among experts to refine the model elements identified qualitatively. Participants rated each model element on a nine-point Likert scale, where 1–3 indicated “not important”, 4–6 represented “important but not a priority”, and 7–9 signified “very important”.³³ This scale allowed for nuanced differentiation of expert opinions. Numerical ratings were complemented by open-ended comment sections where experts could justify their assessments or suggest modifications. After each Delphi round, aggregated feedback was summarized and used to inform the next round, enabling experts to reconsider their ratings in light of the group response while maintaining independent judgment.

Data Analysis

In the qualitative phase, data analysis followed a systematic process of transcription, reduction, coding, categorization, and theme development. Interview transcripts were read multiple times to ensure immersion in the data. Initial codes—representing keywords or short phrases capturing meaningful units—were generated inductively. Codes were then grouped into categories based on conceptual similarity, and overarching themes were constructed to represent core phenomena. Coding was conducted with reference to the five socio-ecological levels—individual, family, organizational, community, and policy—while still allowing inductive themes to emerge from participants’ narratives. The thematic analysis culminated in a conceptual model linking individual, family, organizational, community, and policy factors influencing adolescent mental health, with the family dimension identified as the central domain for the prevention of adolescent mental health problems. Although all five ecological levels informed the model, the family level was analyzed in greater depth because it emerged as the most immediate and consistently emphasized domain across participant groups. The model was iteratively refined through peer review and member checking to ensure accuracy and contextual relevance.

For the quantitative phase, data analysis focused on establishing expert consensus regarding the importance of each model component. Median scores were calculated for each item, as they provide a more robust representation of collective opinion, particularly in the presence of extreme ratings (very high or very low) from a small number of participants. Interquartile ranges were also calculated to describe the convergence of expert ratings across Delphi rounds. Consensus was defined as an IQR of ≤ 2.³⁴ A median value of ≥7 indicated that consensus was achieved for that item.³⁵ The predefined consensus threshold required that at least 70% of experts rated an item as “very important” (score 7–9), while no more than 15% rated it as “not important” (score 1–3). Agreement by ≥70% of experts was considered sufficient to indicate consensus, reflecting a majority view without requiring unanimity.^36,37 Items meeting these criteria were retained in the final model, while those that failed to achieve consensus were revised or eliminated based on expert feedback. When an item did not meet the consensus threshold, the planned procedure was to review experts’ qualitative comments, revise or merge the item when conceptually appropriate, and redistribute it in the next Delphi round.

The integration of findings from both phases was achieved through interpretive synthesis, where qualitative insights provided contextual depth and theoretical grounding, while quantitative consensus validated the structure and content of the developed model. Qualitative themes were transformed into Delphi items, and expert ratings were then used to confirm whether the proposed elements were relevant, important, and applicable within adolescent mental health prevention practice. This mixed-methods integration ensured that the final family-based prevention model was both empirically supported and contextually appropriate for implementation within Indonesia’s cultural framework. However, the Delphi process primarily validated the family-based prevention components derived from the qualitative phase; therefore, broader ecological elements require further empirical and stakeholder validation in future implementation studies. The research flow is illustrated in Figure 1.

Figure 1.

Research Flow of the Exploratory Sequential Mixed-Methods Design.

Results

Qualitative Phase Results

The qualitative analysis identified five interrelated socio-ecological dimensions influencing adolescent mental health prevention: individual, family, organizational, community, and policy. Across these dimensions, the family dimension emerged as the most immediate and consistently emphasized domain because participants repeatedly described family interactions as central to adolescents’ emotional security, help-seeking, supervision, and daily coping. Six themes within the family dimension were generated from the thematic analysis process that collectively informed the conceptual framework of the family-based prevention model for adolescent mental health problems. The six themes are: 1) family attachment; 2) family supervision; 3) family communication; 4) parenting practices; 5) family relationships and 6) family support, as shown in Table 3. Other ecological dimensions were incorporated into the model as contextual systems that interact with and support family functioning, drawing on elements from Kumpfer’s model and Michaels’s framework, including organizational aspects (cross-sectoral and interorganizational collaboration, local partnerships, service availability, and service accessibility), community resources (social environment and peer norms), and policy structures (family functioning, policy decisions, and regulations).

Table 3.

Themes of Family Dimension

Themes	Categories
1. Family attachment	1.1 Child comfort focus 1.2 Relationships among family members 1.3 Emotional bonds 1.4 Parental care 1.5 Family meaning
2. Family supervision	2.1 Compliance 2.2 Family commitment 2.3 Parental control 2.4 Family attention 2.5 Family protection 2.6 Parental demands
3. Family communication	3.1 Family discussion 3.2 Family openness 3.3 Family conflict 3.4 Parental response
4. Parenting practices	4.1 Parenting style 4.2 Family routines 4.3 Parental cooperation 4.4 Parental readiness 4.5 Family involvement 4.6 Paternal role 4.7 Parenting management practices
5. Family relationships	5.1 Family togetherness 5.2 Family roles
6. Family support	6.1 Emotional support 6.2 Instrumental support 6.3 Parental knowledge 6.4 Family time

Theme 1: Family Attachment

Family attachment is a key factor in fostering adolescents’ sense of security and emotional support. Adolescents with strong family attachment tend to have better psychological resilience. Thus, promoting close and harmonious family relationships may serve as an important protective factor against adolescent mental health problems. In this study, attachment was expressed through everyday family togetherness, visible parental affection, and adolescents’ perception that parents were emotionally available. This theme reflects the individual–family interface within the socio-ecological model, where adolescents’ sense of security is shaped by their closest interpersonal environment.

Eating together and bonding with my family, my father and my mother. (Adolescent, male, 13 years old)

Parental affection needs to be shown more clearly, so that they know we care for them. We love them. (Adolescent’s mother, 40 years old)

Theme 2: Family Supervision

Family supervision plays an important role in shaping and maintaining adolescent mental health. Adaptive supervision is needed, whereby parents remain present as guides without exerting excessive control, while adjusting their approach to the adolescent’s needs and circumstances. Participants emphasized supervision as a balanced process involving trust, commitment, monitoring, and guidance rather than authoritarian control. This finding also indicates that family supervision is influenced by broader contexts, including school expectations and health-worker perspectives on early detection of adolescent problems.

The point is to give trust to the child. I trust you, I hold on to your word. So, there has to be commitment and mutual support in whatever activities our child does. (Adolescent’s mother, 37 years old)

In my opinion, parents have the greatest control, because they know the child’s true character. (Adolescent’s mother, 46 years old)

So, it is the parents at home who must monitor [the child]. (School counselor, female, 52 years old)

Theme 3: Family Communication

Open family communication enables parents to identify changes in adolescents’ emotions and behavior at an early stage. Daily conversations and discussions about the adolescent’s social life can help parents develop a more comprehensive understanding of their child’s condition. Such communication patterns also make adolescents feel more comfortable expressing themselves, particularly when the family environment fosters open dialogue and healthy interaction. Communication was also interpreted as a preventive mechanism that links family and organizational systems, because open communication at home may support earlier recognition of problems that can then be connected to school counselors or health services when needed.

Good communication means always talking things through, whenever there is a problem, we look for solutions together. (Adolescent’s mother, 37 years old)

Communication within the family is very important. Even though we may all be busy and have little time to meet, we always make time to discuss things and talk. (Adolescent’s mother, 36 years old)

Through communication, there is sharing, if input is needed, then input is given, so the child does not make decisions alone. (Adolescent’s mother, 45 years old)

Theme 4: Parenting Practices

Supportive and non-authoritarian parenting practices help adolescents navigate this transitional stage of development. Adolescents value an empathetic approach that is free from both physical and verbal violence. Appropriate parenting practices can foster self-confidence and support healthy emotional regulation among adolescents. Participants described effective parenting as a balanced approach that avoids both excessive control and excessive permissiveness. This theme also reflects the cultural relevance of parental authority in Indonesia, where parents are expected to guide adolescents while maintaining warmth, empathy, and respect for adolescents’ developmental autonomy.

As parents, we do not force our children to always be perfect, and we try to make them understand. (Adolescent’s father, 46 years old)

This also cannot be separated from the role of parents. What must be prepared is the parents’ positive role in managing their children. (Pediatrician, male, 72 years old)

There is authoritarian parenting and there is permissive parenting. It turns out that the best one is in the middle, not too authoritarian, but not too permissive either, so that a warm environment can be created. (Psychiatrist, female, 37 years old)

Theme 5: Family Relationships

Warm and supportive family relationships provide a nurturing environment for adolescents’ emotional development. Parents who demonstrate commitment to their children’s activities can help strengthen adolescents’ self-confidence. Adolescents who feel heard and understood within the family are better able to cope with life’s pressures and develop a healthy sense of identity. In contrast, distant family relationships, particularly those marked by limited communication and frequent conflict, may increase the risk of mental health problems. This theme highlights the relational quality of the family environment, including shared time, emotional closeness, and adolescents’ perception of being listened to. Within the socio-ecological model, family relationships operate as a proximal protective system that can buffer stress arising from peer, school, and community environments.

Families can become closer by spending time together, chatting, or on weekends going out for a meal or traveling out of town. That is enough. (Adolescent, female, 13 years old)

I think it is about spending more time together and talking more often with parents about problems. (Adolescent, female, 14 years old, girl)

A good relationship between children and parents, a warm relationship, is the main determining factor, far more than the other factors. (Psychiatrist, female, 37 years old)

Theme 6: Family Support

Family support can be expressed in both material and emotional forms. Material support, such as meeting the adolescent’s needs, can help them feel cared for and motivated. It also serves as a tangible expression of appreciation for the adolescent’s interests. Such support may enhance motivation and, in turn, contribute to better mental health outcomes. Family support also plays a vital role in strengthening adolescents’ psychological resilience. This includes emotional presence, appreciation, and sincere encouragement, all of which are important in preventing adolescent mental health problems. Participants emphasized that emotional support should accompany financial or instrumental support, indicating that adolescents need not only resources but also encouragement, recognition, and psychological presence from family members. This finding is consistent with the family-based orientation of the model, while also suggesting that schools and health services should help parents strengthen supportive communication and early response capacity.

Like being supported, being encouraged. (Adolescent, male, 13 years old)

Parents can support their children’s activities by providing the equipment they need. (Adolescent, male, 14 years old, boy)

Families need to provide emotional support, not only financial support. (Psychologist, female, 39, years old)

The synthesis of these themes produced a comprehensive conceptual framework for the Family-Based Prevention Model for Adolescent Mental Health Problems as presented in Figure 2. The model positions the family as the central preventive mechanism within a wider socio-ecological system, rather than as an isolated intervention unit. In the model, individual adolescent characteristics interact with family processes, while organizational aspects (eg, schools, health services), community resources, and policy structures provide enabling conditions for family-based prevention.

Figure 2.

Conceptual framework of the Family-Based Prevention Model for Adolescent Mental Health Problems.

Quantitative Phase Results

The quantitative validation using the Delphi method involved 14 experts representing diverse disciplines, including a psychiatrist, pediatrician, family medicine specialist, psychologist, health professionals responsible for adolescent health programs, school counselors, and policymakers representing schools, healthcare services, health authorities, and child protection services. The three-round Delphi process successfully achieved expert consensus on the components of the developed model. The Delphi phase focused on confirming the relevance, importance, and applicability of the model components generated from the qualitative phase, particularly the family-based prevention elements embedded within the socio-ecological framework.

The first round involved brainstorming sessions that reviewed the qualitative framework and confirmed the relevance of the main aspects. Experts provided input on operational definitions and measurement indicators, emphasizing the need for contextual alignment with Indonesia’s socio-cultural and healthcare systems. Experts also emphasized that family roles should be connected with school-based counseling, primary healthcare services, and local child protection systems to support feasibility of implementation. The brainstorming process reaffirmed the appropriateness of integrating family roles to address adolescent mental health comprehensively. The results of this round showed that no elements were added or removed. However, several wording refinements were made to improve clarity of operational definitions and measurement indicators before the next Delphi round.

In the second round, experts assessed model components using a nine-point Likert scale. The results showed that, for all elements, more than 70% of the experts assigned a score of ≥7, none assigned a score of ≤3, and the median score was ≥7. The interquartile range also indicated acceptable convergence of expert ratings across the assessed elements. These findings indicate that all elements achieved consensus.

In Round 3, all experts reassessed the results of Round 2. The findings showed consistent expert ratings, as more than 70% of the experts assigned a score of ≥7, no experts assigned a score of ≤3, and the median score was ≥7. The stability of ratings between Rounds 2 and 3 suggested that the panel had reached a shared judgment on the importance of the model elements. This indicates that consensus was reached and that all model elements achieved expert agreement. Table 4 presents the summary of consensus outcomes across the Delphi rounds.

Table 4.

Result of the Delphi Rounds

Element	Round 1	Round 2					Round 3
		Median	Score ≥7 (%)	Score ≤3 (%)	IQR	Consensus	Median	Score ≥7 (%)	Score ≤3 (%)	IQR	Consensus
Family attachment	Brainstorming; no scoring	8	100.00	0.00	1	Achieved	9	100.00	0.00	1	Achieved
Family supervision	Brainstorming; no scoring	8	92.86	0.00	0.75	Achieved	8	92.86	0.00	1	Achieved
Family communication	Brainstorming; no scoring	9	100.00	0.00	1	Achieved	9	100.00	0.00	1	Achieved
Parenting practices	Brainstorming; no scoring	8	85.71	0.00	0.75	Achieved	9	92.86	0.00	1.75	Achieved
Family relationships	Brainstorming; no scoring	8	100.00	0.00	0.75	Achieved	9	100.00	0.00	1	Achieved
Family support	Brainstorming; no scoring	9	100.00	0.00	1	Achieved	9	100.00	0.00	2	Achieved

The statistical results confirmed strong convergence of expert ratings, with overall consensus exceeding 85%. Experts agreed that the model accurately represented the multilevel dynamics influencing adolescent mental health and emphasized the importance of family engagement as a central mechanism within the socio-ecological framework. These findings affirm the model’s conceptual coherence and practical relevance, positioning it as a culturally grounded, evidence-informed framework for future implementation and evaluation within adolescent mental health programs in Indonesia. Nevertheless, the Delphi findings should be interpreted as conceptual and expert-consensus validation rather than evidence of intervention effectiveness. Further empirical evaluation is required to test whether implementation of the model improves adolescent mental health outcomes in practice.

Integration of Qualitative and Delphi Findings

Integration of the two phases showed that the qualitative findings provided the substantive foundation of the model, while the Delphi process confirmed the acceptability and relevance of the proposed elements from multidisciplinary expert perspectives. The six family themes identified in the qualitative phase were translated into Delphi items and received expert consensus, indicating convergence between stakeholder experiences and professional judgment. The integration also clarified that the family functions as a central preventive hub within a broader socio-ecological system, supported by schools, primary healthcare services, community structures, and policy mechanisms. However, because the Delphi process primarily validated the family-based components, future studies should further validate the organizational, community, and policy dimensions through broader stakeholder consultation and implementation research.

Discussion

Family as the Core of Adolescent Mental Health Prevention

The findings of this study underscore the pivotal role of family processes in shaping adolescent mental health outcomes within the socio-ecological framework. Consistent with prior research, family dynamics particularly parental communication and emotional support, emerge as primary mediators influencing adolescent resilience and psychological well-being.^38,39 Enhanced family communication has been significantly associated with lower levels of depression, anxiety, and stress among adolescents.⁴⁰ Adolescents who experience open communication and emotional warmth from parents are more likely to develop adaptive coping strategies, better emotion regulation, and stronger psychological support, thereby reducing their vulnerability to mental health problems.⁴¹ This aligns with the present study’s qualitative findings, where parental attachment and communication were repeatedly cited as critical determinants of mental well-being. Importantly, the present findings do not suggest that prevention should rely solely on families; rather, they indicate that family processes function as the most proximal mechanism through which broader school, community, health-service, and policy supports can influence adolescents. Family involvement plays a critical role in adolescent well-being, emphasizing the importance of strengthening parent-adolescent communication and cooperation in addressing mental health challenges.⁴²

Evidence-based family interventions provide a foundation for operationalizing these findings. The Research and Action for Teens (RAFT) study has demonstrated substantial efficacy in addressing mental health problems by integrating family, school, and community systems.⁴³ The Puentes Program, The Strengthening Families Program (SFP 10–14) and the Strong African American Families (SAAF) program have been shown to improve parenting skills and strengthen adolescent resilience against risk behaviors, emphasizing the enhancement of family functioning as a preventive tool.¹⁶ These findings corroborate the multidimensional approach identified in this study, where family-based interventions operate across emotional, behavioral, and relational domains to promote psychological health. However, most of these intervention models were developed outside Indonesia; therefore, their relevance depends on adaptation to local parenting norms, family hierarchy, school-family relationships, and community-based support systems.

Adapting these approaches to Indonesia’s cultural and religious context necessitates careful cultural tailoring to ensure acceptability and efficacy. The integration of communal values, religious practices, and traditional family hierarchies into prevention frameworks can increase program relevance and engagement.^44,45 In collectivistic societies like Indonesia, family obligation values and shared responsibility are deeply embedded cultural constructs that can be leveraged to strengthen preventive programs. Integrating culture-specific components such as culturally adapted parent training may enhance program resonance and uptake.⁴⁶ Moreover, the adaptation process should consider linguistic diversity and local norms regarding authority and communication to ensure inclusivity and accessibility across Indonesia’s heterogeneous populations. In the present study, culturally relevant family processes were reflected in adolescents’ need for visible parental affection, balanced supervision, shared family time, and emotional support. These findings suggest that Indonesian family-based prevention should respect parental authority while promoting warmer, more dialogical, and less punitive communication patterns.

Socio-Ecological Model as a Framework for Multi-Level Intervention

The socio-ecological model provides a comprehensive structure for understanding how interconnected systems influence adolescent mental health. Originating from Bronfenbrenner’s ecological systems theory, the model has evolved to integrate complex interrelations between individuals, families, organizations, communities, and policy structures.^17,18 The current study’s five-level model (individual, family, organizational, community, and policy) reflects this multidimensional perspective, demonstrating how the interaction among these layers contributes to the success of preventive strategies. The individual level includes adolescents’ self-control, academic achievement, self-efficacy, adaptability, stress management, coping skills; the family level includes attachment, supervision, communication, parenting practices, relationships, and support; the organizational level includes cross-sectoral and interorganizational collaboration, local partnerships, service availability, service accessibility; the community level includes the social environment, peer norms; and the policy level includes family functioning, policy decisions, and regulations that enable prevention. These elements were informed by Kumpfer’s model and Michaels’s framework, which served as the conceptual foundation for developing the present model.

The study’s results corroborate prior evidence that multilevel interventions yield superior outcomes in promoting mental well-being. School-, family-, and community-based interventions grounded in a socio-ecological framework have demonstrated measurable effectiveness in enhancing resilience, strengthening psychosocial competencies, and reducing internalizing and externalizing problems among young people facing adversities.⁴⁷ Another study highlights that adolescent mental health is influenced by multi-level socio-ecological processes, wherein supportive relationships across family, peer, and school contexts are key determinants of distinct mental health profiles.⁴⁸ These findings illustrate that ecological interventions are most effective when they integrate both micro-level (individual and family) and macro-level (organizational and policy) strategies. Accordingly, the proposed model should be interpreted as a family-based socio-ecological model, not a family-only model. Families are positioned as the central mediating context, while schools, health centers, community structures, and policy actors provide reinforcement, referral pathways, and enabling environments.

Family-level support emerged as the key factor in sustaining adolescent mental health prevention efforts. As the social context closest to adolescents, the family has the most direct and enduring influence on their emotional well-being, behavioral development, and psychosocial adjustment. This central position makes the family a crucial protective factor in preventing adolescent mental health problems. In addition, the effectiveness of family involvement may be enhanced through policy-level support. Institutional regulations, including school policies that formalize parental participation in adolescent mental health programs, can strengthen and sustain family engagement over time. These findings emphasize the importance of aligning family-based interventions with policy support to ensure a more comprehensive and sustainable prevention strategy. For example, schools can institutionalize parent engagement sessions, community health centers can provide family-oriented screening and counseling, and local governments can support referral coordination between education, health, and child protection sectors.

Policy and Implementation Implications

The translation of family-based prevention into national mental health policy presents both challenges and opportunities. Despite Indonesia’s policy landscape supporting preventive health through frameworks such as the National Health Transformation Agenda,⁸ family engagement remains underrepresented in mental health governance. Consistent with global observations, the predominant focus on individual treatment limits systemic approaches that integrate family as a preventive resource.^16,49 Cultural stigma surrounding mental illness and limited cross-sectoral coordination further inhibit the scaling of family-centered interventions. Nonetheless, the growing recognition of mental health as an integral component of universal health coverage provides a strategic entry point for embedding preventive, family-based approaches within public health systems.⁵⁰ This integration should explicitly define the roles of families, schools, community health centers, community organizations, and local policy actors so that family engagement becomes an operational component of prevention rather than a general recommendation.

Governance models that promote intersectoral collaboration have demonstrated success in advancing preventive mental health programs. Participatory governance, wherein community members and families actively contribute to program design and implementation, has also proven effective in enhancing ownership and long-term sustainability.⁵¹ This participatory approach aligns with Indonesia’s decentralized governance structure, enabling local health offices and schools to contextualize family-based mental health initiatives within community needs. At the school level, implementation may include parent education sessions, routine communication between parents and school counselors, and early identification of emotional or behavioral changes. At the primary healthcare level, community health centers may integrate family counseling, adolescent screening, psychoeducation, and referral pathways into existing adolescent health services. At the community level, youth cadres, religious leaders, and community organizations may help reduce stigma and promote supportive norms for help-seeking.

Capacity building among health and education professionals is a further determinant of successful implementation. Training providers to identify and engage families in prevention activities can improve early detection and response to adolescent mental health risks. Empirical evidence indicates that structured training and mentorship for educators, healthcare workers, and community leaders significantly enhance the reach and quality of preventive programs.⁵² Furthermore, resource allocation and funding mechanisms that prioritize family-centered interventions can ensure program continuity.⁵³ Policy reform that institutionalizes mental health promotion within broader health system strengthening agendas would provide a structural foundation for such integration.⁵⁴ Based on the model, future implementation could begin with a localized training manual for parents, teachers, school counselors, and primary healthcare workers, followed by pilot testing in selected schools and community health centers. Implementation indicators may include parent participation, family communication practices, referral uptake, adolescent help-seeking, mental health literacy, and changes in adolescent psychosocial outcomes.

The findings affirm that integrating family-based prevention within a socio-ecological framework provides a robust and contextually adaptive pathway for improving adolescent mental health in Indonesia. The alignment of family-level processes with organizational, community, and policy systems offers a foundation for sustainable, culturally grounded interventions capable of addressing both the immediate and structural determinants of adolescent mental well-being. Nevertheless, the present study provides conceptual and expert-consensus validation rather than evidence of intervention effectiveness. Future research should test the model through pilot implementation, longitudinal evaluation, mediation analysis of family processes, and randomized or quasi-experimental designs across diverse Indonesian regions and adolescent age groups.

Conclusion

This study developed a Family-Based Prevention Model for Adolescent Mental Health Problems grounded in a socio-ecological framework contextualized to Indonesia. The findings highlight five interrelated dimensions (individual, family, organizational, community, and policy) that collectively influence adolescent mental health. The model underscores the central role of the family in preventing adolescent mental health problems through family attachment, family supervision, family communication, parenting practices, family relationships, and family support, supported by institutional and policy systems that enable sustainable prevention. Thus, the model should be understood as a family-based socio-ecological model, rather than a family-only model, because family processes are positioned as the central preventive mechanism within broader school, healthcare, community, and policy systems. The Delphi consensus confirmed the validity and relevance of these dimensions, achieving full expert agreement on the proposed model components.

The study’s implications extend to both theory and practice. Theoretically, it advances the socio-ecological model by integrating family processes as a mediating mechanism that links individual and systemic factors affecting mental health. Practically, it provides a framework for developing culturally responsive interventions aligned with Indonesia’s preventive health transformation agenda. Implementation of this model could enhance collaboration between schools, health services, and communities to foster adolescent mental well-being. However, the present study provides conceptual model development and expert-consensus validation, not evidence of intervention effectiveness. Therefore, the preventive impact of the model should be tested empirically before broader implementation. Future research should focus on piloting and evaluating this model in diverse regions, examining its effectiveness in improving resilience, reducing risk behaviors, and strengthening family engagement across socio-cultural contexts. Future studies should also include adolescents across the full 10–19-year age range, assess regional and socioeconomic variation, develop localized training manuals, and evaluate implementation indicators such as parent participation, family communication, referral uptake, adolescent help-seeking, and psychosocial outcomes.

Funding Statement

This publication charge is funded by Unpad through the Indonesian Endowment Fund for Education (LPDP) on behalf of the Indonesian Ministry of Higher Education, Science and Technology and managed under the EQUITY Program (Contract No. 4303/ B3/DT.03.08/2025 and 3927/UN6. RKT/HK.07.00/2025).

Declaration of Use AI in Scientific Writing

The authors declare that ChatGPT 5.5 was used solely to assist in improving the English language quality of this manuscript, including grammar, clarity, and sentence structure.

Data Sharing Statement

The datasets generated and analyzed during the current study are available from the corresponding author on reasonable request, subject to ethical and confidentiality considerations.

Ethical Approval

This study was conducted in accordance with the ethical principles of the Declaration of Helsinki. Ethical approval was obtained from the Ethics Committee of the Faculty of Medicine, Universitas Padjadjaran, Indonesia, with reference number 1031/UN6.KEP/EC/2024.

Consent to Participate

Written informed consent was obtained from all individual participants included in the study. For adolescents, written informed consent was obtained from their parents, and assent was obtained from the participants.

Disclosure

The authors declare that there are no conflicts of interest in the writing or publishing of this article.