The hybrid intervention delivered over 12 sessions (8 sessions delivered via WhatsApp groups to parents only and 4 sessions delivered in-person to parents and teens together).

Physical abuse and emotional abuse will be measured using 4 items from a reduced version of the ISPCAN Child Abuse Screening Tool-Trial Parent version (ICAST-TP). The ICAST-TP measures parental reports of the incidence of abuse perpetrated against their child over the past month using a frequency score on a scale of 0 to 7, or 8 or more times (e.g., "In the past 4 weeks, how often did you discipline your child by spanking, slapping, or hitting with your hand?").

Positive parenting will be assessed based on parent-report of an adapted version of the Alabama Parenting Questionnaire (6-item APQ subscale; Frick, 1991). The APQ measures the frequency of specific behaviours towards children in the past month on a 7-point Likert-like scale (e.g., "You talk to your child about their friends"; 0 = never; 6 = always). Items are summed to create a frequency score, with higher scores indicating more regular use of positive parenting.

Positive parenting will be assessed based on teen-report of an adapted version of the Alabama Parenting Questionnaire (6-item APQ subscale; Frick, 1991). The APQ measures the frequency of specific behaviours towards children in the past month on a 7-point Likert-like scale (e.g., "Your caregiver talks to you about your friends"; 0 = never; 6 = always). Items are summed to create a frequency score, with higher scores indicating more regular use of positive parenting.

Child behaviour problems will be assessed using parent-report of the Strengths and Difficulties Questionnaire (10-item SDQ; Goodman, 1997). Respondents will be asked to report on the frequency of specific child behaviours (e.g., "Your child is restless, overactive, cannot stay still for long" using a 3-point Likert-like scale (1 = Not true; 3 = Very true). Items are summed, with higher scores indicating more behavioural problems.

Child behaviour problems will be assessed using teen-report of the Strengths and Difficulties Questionnaire (10-item SDQ; Goodman, 1997). Respondents will be asked to report on the frequency of specific child behaviours (e.g., "You are restless, overactive, cannot stay still for long" using a 3-point Likert-like scale (1 = Not true; 3 = Very true). Items are summed, with higher scores indicating more behavioural problems.

Parental depression will be assessed using the Centre for Epidemiologic Studies Depression Scale (10-item CES-D subscale; Radloff, 1977). Caregivers will be asked to respond to items related to how they have felt over the past seven days (e.g., "How often in the past week have you been bothered by things that don't usually bother you?"). Responses are coded on a 4-point Likert-like scale (i.e., 1 = Rarely or none of the time (less than 1 day) to 4 = Most or all of the time (5 to 7 days)) and summed to create a total of parental depression.

Teen internalising behaviours will be assessed using parent-report of the Strengths and Difficulties Questionnaire (SDQ) based subscales for emotional problems (e.g. "I worry a lot") and peer problems (e.g. "I am usually on my own. I generally play alone or keep to myself") (15-item SDQ subscale; Goodman, 1997). Items are summed to create a total measurement of internalising behaviour.

Risk Avoidance will be assessed using parent-report of the Risk Avoidance Planning Scale (5-item RAPS; Cluver et al., 2018). Caregivers will be asked whether they have spoken to their child about risks in the community including topics such as safe sex and community safety. This scale was developed specifically for the PLH for Teens programme in South Africa (Cluver et al., 2018). Caregivers will be asked to respond on a 3-point Likert-like scale as to whether they have spoken to their child about risks in the community (1 = No, I find it too hard to talk about this; 2 = We haven't made plans yet but I would like to talk about it; 3 = We have made a plan together). Topics will include safe sex and community safety.

Risk Avoidance will be assessed using teen-report of the Risk Avoidance Planning Scale (5-item RAPS; Cluver et al., 2018). Teens will be asked to respond on a 3-point Likert-like scale as to whether they have spoken about risks in the community (1 = No, I find it too hard to talk about this; 2= We haven't made plans yet but I would like to talk about it; 3 = We have made a plan together). Topics will include safe sex and community safety.

Economic Strengthening will be measured using a Family Financial Coping Scale (5-item FFCS; Lown, 2011). Caregivers will be asked to respond to household demographic items over the past month. These items cover a variety of financial matters including whether caregivers were worried about money, whether they managed to save money, and whether they ran out of money to buy certain items, including food, transport, and mobile data or airtime. Responses are coded on a 4-point Likert scale (0 = Never, 1 = Rarely, 2 = Sometimes, 3= Often). Items are summed, with a higher score reflected a greater number of financial challenges.

Economic Strengthening will be measured using a Family Financial Coping Scale (5-item FFCS; Lown, 2011). Teens will be asked to respond to household demographic items over the past month. These items cover a variety of financial matters including whether caregivers were worried about money, whether they managed to save money, and whether they ran out of money to buy certain items, including food, transport, and mobile data or airtime. Responses are coded on a 4-point Likert scale (0 = Never, 1 = Rarely, 2 = Sometimes, 3= Often). Items are summed, with a higher score reflected a greater number of financial challenges.

Parent Support of Education will be measured using an adapted version of the Parental Support for School Scale (6-item PSE; Ceballo, Maurizi, Suarez, & Aretakis, 2014). Caregivers will report on the frequency of supportive behaviour towards children's learning (e.g. "I support my child's schoolwork in any way that I can" and "I praise my child for working hard at school") using a 5-point Likert-like scale (1 = never; 5 = always). Items are summed to create a frequency score, with higher scores suggesting more parental support and value for school. Items from this survey are present in both the parent survey.

Parent Support of Education will be measured using an adapted version of the Parental Support for School Scale (6-item PSE; Ceballo, Maurizi, Suarez, & Aretakis, 2014). Teens will report on the frequency of supportive behaviour towards children's learning (e.g. "support your schoolwork in any way that he/she can" and "praise you for working hard at school") using a 5-point Likert-like scale (1 = never; 5 = always). Items are summed to create a frequency score, with higher scores suggesting more parental support and value for school. Items from this survey are present in the teen survey.

Attitude toward punishment will be measured using a single-item from the UNICEF Multiple Indicator Cluster Survey (MICS) 5 Child Discipline module - Teen Report (1-item; ). Teens will be asked to report on whether they agree or disagree with the statement: "In order to bring up, raise up, or educate a child properly, the child needs to be physically punished".

Attitude toward punishment will be measured using a single-item from the UNICEF Multiple Indicator Cluster Survey (MICS) 5 Child Discipline module - Teen Report (1-item; ). Caregivers will be asked to report on whether they agree or disagree with the statement: "In order to bring up, raise up, or educate a child properly, the child needs to be physically punished".

Basic demographic information will be collected at baseline and post-test for teens. Demographic information will include; age, gender, literacy, family structure and dynamics, and risk behaviours. During screening family risk will also be assessed using a Vulnerability assessment tool as a proxy assessment of vulnerability to understand the socio-economic needs of respondents (USAID, 2012). There are six questions that will be asked to respondents which are drawn from the topic of child protection, access to education, nutrition status, child's care and support, child social behaviour, and access to health. Each question will either be in a form of Likert or multiple choice question and have a maximum score of 4. Higher total score indicates lower vulnerability.

Facilitator programme adherence will be assessed by examining rates of enrolment.

Facilitator programme adherence will be assessed by examining rates of attendance.

Facilitator programme adherence will be assessed by examining rates of dropout. Dropout rates for enrolled participants will be defined as the percentage of participants who do not attend at least two consecutive sessions across either the in-person or online components of the programme, and do not attend any sessions at a later stage.

Implementation fidelity of the hybrid PLH programme will be measured using self-report checklists by facilitators to understand the extent to which core intervention components are delivered. These checklists will include specific activities for each session in both the in-person and at home components of the programme. Then, in order to produce a basic level of fidelity, a ratio of programme implementation to programme design will be created for the self-report scores. A score of 80% programme fidelity will be considered as "high treatment fidelity".

Intervention cost of programme for scale-up will be assessed using facilitator reports of the time and resource costs spent delivering the programme. This will include training and preparation for delivery.