PurposeStatelessness is the worst possible form of violation of fundamental human rights which can lead to improper health systems management and serious adverse health outcomes in children. To address this, the Thai Cabinet introduced the Health Insurance for People with Citizenship Problem (HIPCP) in 2010. The purpose of this study is to examine the association between insurance affiliations and the health status of stateless children insured with the HIPCP. The presence of pneumonia was selected as a proxy for health status. The comparison groups were Thai children insured with the Universal Coverage Scheme (UCS) which was launched in 2002 and the uninsured children of low-skilled migrants in Thailand.Design/methodology/approachA retrospective study was conducted at four selected district hospitals: Mae Ramat Hospital, Phop Phra Hospital, Tha Song Yang Hospital and Umphang Hospital in Tak Province, located in northwestern Thailand. The study used the medical records of children aged 0-15 years who were admitted to the aforementioned hospitals between January 1, 2013 and December 31, 2017. Multivariate logistic regression model was applied with a binary response variable (ever diagnosed with pneumonia: yes/no). Exposure was three types of insurance status (uninsured, HIPCP and UCS) while covariates were age, sex, domicile and year of hospitalization of children.FindingsOf 7,098 hospitalized children between 2013 and 2017, 1,313 were identified with pneumonia. After controlling for key covariates, multivariate results depicted that the odds of pneumonia was 4 per cent higher in stateless children insured with the HIPCP as compared with uninsured children but non-significant (adjusted odds ratio [AOR] = 1.040, 95 per cent confidence interval [CI] = [0.526, 2.160], p = 0.916). Similarly, the odds of pneumonia was 10 per cent higher in Thai children insured with the UCS as compared with uninsured children but non-significant (AOR = 1.100, 95 per cent CI = [0.594, 2.180], p = 0.767). The children who were hospitalized in 2017 were 26 per cent more likely to have pneumonia as compared with those who were hospitalized in 2013 with statistical significance (AOR = 1.260, 95 per cent CI = [1.000, 1.580], p = 0.050). Results remained robust after performing sensitivity analyses.Social implicationsThis study suggests that health insurance is not associated with the health status of vulnerable children especially in the presence of multiple health interventions for uninsured and/or undocumented children living along the Thai–Myanmar border area. Further experimental studies are warranted to understand the causal relationship between insurance and health outcomes and to overcome the limitations of this observational study.Originality/valueThis study has discovered that age and domicile of children are independently associated with pneumonia. In comparison with the youngest age group (0-1 year), the older age groups presented a significantly lower odds for pneumonia. The children living in Phop Phra, Tha Song Yang and Umphang districts revealed a reduced risk for pneumonia as compared with children living in Mae Ramat district.