Online Social Support Program for Physical and Mental Health of Filipino Migrant Domestic Workers in Hong Kong
Online Social Support Program for Physical and Mental Health of Filipino Migrant Domestic Workers in Hong Kong
The goal of this study is to test the effect of informational and emotional social support (via nutrition education, mental health support, and/or baking classes) through two different communication modalities (online vs. mixed mode / hybrid) on physical and mental health outcomes
Our research questions are the following.
Does an online social support program that provides informational and emotional support improve diet, anthropometry, and mental health?
Does a mixed-mode social support program that provides informational, emotional support through both online and face-to-face mode improve diet, anthropometry, and mental health?
Is a mixed mode social support program more effective in improving outcomes? Does meeting the group members face-to-face change the dynamics of online communication? If yes, what are the mechanisms? Are there differences in the following outcomes by communication modality?
Study design: Randomized controlled trial
Study description
Filipino Migrant Domestic workers (MDWs) working in Hong Kong will be eligible to apply. The investigators will advertise in Newspapers and Facebook commonly read by Filipino MDWs in Hong Kong. The investigators will recruit 400 MDWs over two phases and randomly assign them across three arms, T1) mixed-mode social support program; T2) Online social support program; and C) Control. Online social support program will be delivered through WhatsApp group chat. MDWs will self-fill surveys at Caritas office, but surveys will be assisted by survey enumerators as a group.
Logistics
Due to administrative constraints delivering interventions to 400 participants at the same time, the investigators will deliver interventions over two cohorts.
Cohort 1 (~200 participants)
Cohort 2 (~200 participants)
Note 1: The investigators will enroll participants until the scheduled date for baseline survey or until the investigators recruit target sample size, whichever comes earlier.
Note 2: The investigators will keep the intervention design exactly the same for cohort 1 and cohort 2, and will not analyze data until data collection for both cohort 1 and cohort 2 is complete. Cohort 1 and cohort 2 will be analyzed as one sample.
Statistical analysis model - Estimation of treatment effect
In this section, the investigators outline the basic estimation approach to measuring the effect of the treatments on various outcomes.
Our basic treatment effects specification estimates the following equation:
y= β₀ + β₁T1 + β₂T2+ β₃X+ ε where y is the outcome of interest.
T1 and T2 are dummy variables equal to 1 if the participant was randomly assigned to the T1 and T2 group, respectively, and 0 otherwise. β₁ and β₂ represent the effect of being assigned to the specific treatment arm. X is a vector of baseline outcome and individual's characteristics including age, marital status, years working as migrant domestic workers, level of education, received food allowance. ε is an error term. The primary outcome variable, y, include nutrition knowledge, perceived social support, healthy diet, stress, anxiety, and depression. To compare the difference between T1 and T2, we will use F-test coefficient of equality.
Inclusion Criteria:
Exclusion Criteria:
yehan@ust.hk+85253004009