Epidemiology of varicella among immigrants and non-immigrants in Quebec, Canada, before and after the introduction of childhood varicella vaccination: a retrospective cohort study.

Background: Many immigrants are susceptible to varicella on arrival to Canada because of different transmission dynamics in their countries of origin and scarcity of vaccination. Universal childhood vaccination programmes decrease varicella incidence rates through herd immunity, but the accumulating number of susceptible adult immigrants could remain at risk for severe varicella. Our aim was to describe the epidemiology of varicella among immigrants and non-immigrants before and after childhood varicella vaccination.

Methods: We did a population-based, retrospective cohort study of all varicella cases in Quebec, Canada, diagnosed between 1996 and 2014 in administrative health databases linked to immigration data. Cases of varicella met diagnostic codes in the International Classification of Diseases, Ninth and Tenth Revision Canadian modifications. Cases with a co-occurring zoster diagnostic code and immigrants from Australia, New Zealand, the USA, and western European countries were excluded. Vaccination periods included pre-vaccination (1996-98), private vaccination (1999-2005), and public vaccination (2006-14). Incidence rate and comparative rate ratios were estimated using census data.

Findings: A total of 231 339 varicella cases diagnosed between Jan 1, 1996, and Dec 31, 2014, were linked to 1 115 696 immigrants who arrived between Jan 1, 1980, and Dec 31, 2014. 1444 herpes zoster cases and 1276 immigrants from Australia, western Europe, New Zealand, and the USA were excluded. Among 228 619 varicella cases, 13 315 (5·8%) occurred in immigrants. In pre-vaccination versus public vaccination periods, varicella incidence declined in immigrants by 87% (95% CI 86·6-87·9; 324·3 cases per 100 000 person-years to 40·9 cases per 100 000 person-years) and in non-immigrants by 93% (92·4-92·7; 484 cases per 100 000 person-years to 36 cases per 100 000 person-years). Mean age at diagnosis increased in both groups (15·1 vs 19·4 years in immigrants and 8·4 vs 12·0 years in non-immigrants). In the public vaccination period, immigrants younger than 50 years had higher varicella rates than non-immigrants, with relative risk ranging from 1·53 (95% CI 1·37-1·72) to 4·64 (3·90-5·53) with the highest risk in adolescents and young adults, and people from Latin America and the Caribbean (age-specific incidence rate ratio [aIRR] pre-vaccination 2·19 and post-vaccination aIRR6·07) and south Asia (aIRR pre-vaccination 3·41 and aIRR post-vaccination 4·46) and in childbearing women (15-40 years; IRR 2·48).

Interpretation: Immigrant adolescents, young adults, and women of childbearing age had higher age-standardised rates of varicella than non-immigrants, with increasing disparities following vaccine introduction. Immigrants younger than 50 years of age would benefit from targeted vaccination upon arrival to host countries.

Funding: The Canadian Institutes of Health Research and The Department of Medicine, Jewish General Hospital, Montreal, QC, Canada.