Ask the workers first, and you get exercises that fit a dormitory room and the hours they actually have

Myanmar migrant workers in Thailand's seafood plants are injured by repetitive tasks such as shrimp peeling and long hours standing, yet ordinary health programmes never reach them because of language and cultural barriers. This study asked what changes when the workers help design the programme from the start.

A mixed-methods co-design study. Engagement sessions were held with 29 Myanmar workers in three groups by length of service (under 2 years, 10; 2–5 years, 11; over 5 years, 8), all from a single factory, conducted in Myanmar language and scheduled around 12-hour shifts. Five workplace stakeholders (HR managers and line supervisors) were consulted, and four international physical therapists assessed content validity. Consent was taken privately without management present, with 48 hours to decide, and supervisors did not refer participants.

• Every one of the 29 workers reported hand symptoms — numbness, tingling and pain. Twenty-five (86%) had morning hand stiffness that affected life outside work as well. Neck pain 79%, lower back pain 72%, ankle pain 45%.
• 23 of the 29 (79%) bought over-the-counter painkillers at least weekly, and some returned to identical work straight after surgery — this group was already trying to manage its own health, but by risky means.
• 26 of 29 (89.7%) used Facebook daily and chose it because they could read about health without supervisors seeing. They asked for video demonstrations rather than written material, and rejected compulsory workshops on their single day off.
• Barriers to care named by the workers themselves: providers who do not speak Burmese, interpreters in government hospitals who treated them disrespectfully, workplace insurance covering only acute injury rather than chronic conditions, and discrimination.
• The constraints the workers set became the design brief — exercises had to work in a dormitory-sized space, need no equipment, run outside working hours, and never reduce earnings, since all of them send money home.
• The result was a 12-exercise, four-week programme that scored highly on expert content validity (I-CVI 0.95–1.00; S-CVI/Ave 0.94), showing that listening to workers did not cost clinical quality.

The transferable lesson for any health communication work is that the constraints named by the people at the sharp end are not a trade-off against quality — they are the conditions under which a programme gets used at all. Choosing a channel the audience already uses, and can use privately, matters more than choosing the channel the organisation is comfortable with.