Medical Sociology of Health Worker Strikes and Care-Seeking Substitution in Kenyan Public Hospitals: A Difference-in-Differences Estimation of Strike Exposure and Private Facility Switching

Health worker strikes in Kenyan public hospitals create abrupt, unanticipated disruptions to outpatient care, yet the question of whether patients substitute toward private facilities during strikes remains empirically unresolved. The design exploits the staggered timing of county-level strikes as a source of quasi-experimental variation, comparing care-seeking behaviour in exposed counties before and after strike onset against contemporaneous trends in unexposed counties. The protocol specifies the construction of a strike exposure calendar from verified industrial-action records, the definition of public and private care-seeking outcomes from standard household survey instruments, and the estimation of two-way fixed-effects models with county and survey-round fixed effects. It further addresses the principal threats to identification, including strike endogeneity, anticipatory behaviour, and compositional change in survey samples. The contribution is methodological: it provides a transparent, replicable template for estimating substitution effects in settings where random assignment is infeasible and administrative data are incomplete. The protocol is intended to support future empirical application and to clarify the conditions under which strike exposure can be credibly interpreted as a causal driver of sectoral switching.