BackgroundWe aimed to use New Zealand’s Anzac Day to test the public health effect of secondary public holidays; and to use weekly hospitalisation counts to identify which dates were more health suitable for a potential new public holiday.MethodsWe conducted a retrospective population cohort study of hospital admissions in New Zealand between 23 April and 27 May 1988–2018. We compared acute and arranged hospitalisation and mortality rates in holiday (Anzac Day Monday to Friday) and non-holiday (Anzac Day Saturday or Sunday) years, for mid-week holidays and long weekends; and measured total weekly average acute and arranged hospitalisation counts.ResultsHospitalisation rates were lower in holiday years than non-holiday years (rate ratio (RR) 0.96, 95% CI 0.95 to 0.96, p<0.001), with fewer arranged admissions (RR 0.93, 95% CI 0.93 to 0.94, p<0.001), but no significant difference for acute admissions (RR 1.00, 95% CI 0.99 to 1.00, p=0.087). Holiday year acute admission rates were lower than non-holiday years for children aged 0–4 years, but higher for adults aged 15–44 years. Holidays reduced arranged admissions most in children and areas of higher socioeconomic deprivation. There was no significant difference for mortality. Average weekly acute hospitalisations were higher than arranged admissions between 10 June and 7 October.ConclusionSecondary holidays do not reduce deaths; or acute hospital admissions except in the holiday week; and may delay elective treatment. However, if New Zealand is to add a new public holiday, it would have least detrimental health effect scheduled between 10 June and 7 October.