Impact of the COVID-19 nonpharmaceutical interventions on influenza and other respiratory viral infections in New Zealand

The study investigates how stringent nonpharmaceutical interventions (NPIs) implemented in New Zealand (NZ) to eliminate COVID-19 affected influenza and other respiratory viral infections in 2020. NZ introduced progressive border controls starting 2 February 2020 and a Level-4 national lockdown from 25 March to 27 April 2020, including border closures and quarantine, widespread testing, isolation and contact tracing, physical distancing (stay-at-home orders, closure of schools, non-essential businesses and public venues, and travel restrictions), personal hygiene measures, and risk communication. These measures eliminated community COVID-19 transmission during the first wave, achieving 101 consecutive days without community cases. Influenza’s short serial interval, early peak viral shedding, and substantial asymptomatic infection have led to assumptions that NPIs are ineffective for influenza control; however, robust field data are limited. NZ’s stringent NPIs offered a natural experiment to assess their impact on influenza and other respiratory viruses.

WHO pandemic influenza guidance historically does not recommend stringent NPIs (e.g., lockdowns, border closures) once sustained transmission occurs, based largely on historical observations and modeling. Prior reports during the COVID-19 pandemic and earlier epidemics suggest NPIs can suppress influenza and other respiratory viruses, including observations from Australia, Chile, South Africa, and Hong Kong (during SARS 2003 and COVID-19). The study also considers other factors discussed in the literature that may influence influenza circulation: vaccination coverage, climatic effects (e.g., lower stability of influenza virions at higher temperatures), and viral–viral interactions (e.g., interferon-mediated interference between influenza and rhinovirus).

Design: Observational analysis using multiple established national surveillance systems in New Zealand, comparing 2020 (pre-, during, and post-lockdown) to a 2015–2019 reference period. Ethics: SHIVERS (including SARI and ILI), SHIVERS-II, and SHIVERS-III (WellKiwis) cohort studies received approval from NZ Northern A Health and Disability Ethics Committee (NTX/11/11/102). ICD-coded hospitalization and laboratory surveillance were conducted under public health surveillance; ethics approval not required. Surveillance components: - Hospital-based SARI surveillance: Ongoing since 2012 in Central and South Auckland (catchment ~1 million). Research nurses screened daily admissions to enroll patients with SARI (cough and history of fever ≥38 °C within past 10 days) per WHO definition. Respiratory specimens collected (nasopharyngeal/nasal/throat swabs) tested by real-time RT-PCR for influenza, RSV, rhinovirus, PIV1–3, enterovirus, adenovirus, and hMPV. - Sentinel GP-based ILI surveillance: Operating since 1990, usually covering ~10% of NZ population via ~90 sentinel practices. Patients meeting ILI definition (acute respiratory illness with history/measured fever ≥38 °C, cough, onset ≤10 days, consulting GP) were swabbed for influenza and other respiratory viruses. In 2020, patient flow shifted to community COVID-19 testing centers; practice participation was 18–57% of usual during winter; swabbing requirements and pandemic response reduced consultations and detections. - SHIVERS-II (adult) and SHIVERS-III/WellKiwis (infant) cohorts: Prospective cohorts in Wellington; ~1400 adults (20–69 years) and ~80 infants followed in 2020. Weekly surveys (May–September 2020) identified ILI and ARI episodes. 2020 ARI definition: fever/feverishness and/or respiratory symptoms (cough, runny nose, wheezing, sore throat, shortness of breath, anosmia/ageusia) onset ≤10 days. ILI definition (2018–2020): ARI with cough and fever ≥38 °C, onset ≤10 days. Research nurses swabbed ILI/ARI cases for influenza, other respiratory viruses, and SARS-CoV-2 by RT-PCR. - ICD-coded influenza hospitalizations: Extracted from NZ Ministry of Health National Minimum Dataset by discharge date for ICD-10AM-VI codes J09–J11 (principal diagnosis). Patients with length of stay <1 day excluded; repeat admissions included. - Laboratory-based surveillance: Year-round testing via NZ virus laboratory network (National Influenza Centre at ESR and six hospital labs in Auckland [2], Waikato, Wellington, Christchurch, Dunedin). Primarily clinician-ordered diagnostic specimens for inpatients/outpatients (serving ~70% of population), plus specimens from SARI, ILI, SHIVERS-II, and WellKiwis surveillance. Sampling based on clinician judgment (potential selection bias). During COVID-19 response, some labs prioritized SARS-CoV-2 testing. Data management and analyses: Data captured in REDCap 10.0.19; analyses in Stata 16.1. Weekly observed influenza-PCR-confirmed SARI/ILI/ARI incidence corrected for missed swabs using positivity among those tested. Epidemic thresholds and intensity defined using the Moving Epidemic Method (MEM) based on 2015–2019 SARI and ILI data (defining pre-, epidemic, post-epidemic periods; calculating epidemic and intensity thresholds). Cochran-Armitage tests for temporal trends were applied to ICD-coded hospitalizations and virus detections. For lab surveillance and ICD-coded hospitalizations, median weekly values over 2015–2019 were the reference; percent reduction calculated as 1 – [No. of virus (2020)/median no. of virus (2015–2019)]%.

- No winter influenza epidemic in 2020: Multiple systems showed influenza activity far below seasonal thresholds; no laboratory-confirmed influenza-associated cases detected in key surveillance streams during winter. - Hospital SARI surveillance (catchment ~1 million): From May–September 2020, SARI incidence remained below the seasonal threshold; no influenza-associated SARI identified. - Sentinel GP ILI surveillance: Participation reduced (18–57% of usual). ILI incidence below seasonal thresholds; no influenza-associated ILI detected. Independent HealthStat GP ILI system also showed low ILI rates. - SHIVERS-II (adult) and SHIVERS-III/WellKiwis (infant) cohorts: ILI incidence lower than 2018–2019; ARI incidence was high. No influenza-associated ILI or ARI identified. - ICD-coded influenza hospitalizations (all NZ public hospitals): 291 total Jan 1–Jul 31, 2020: pre-lockdown 238 (81.8%), lockdown 33 (11.3%), post-lockdown 15 (5.2%). Significant downward trend over time (Cochran-Armitage p<0.001). - Laboratory-based surveillance (Jan 1–Sep 27, 2020): 500 influenza virus detections: pre-lockdown 474 (94.8%), lockdown 20 (4.0%), post-lockdown 6 (1.2%). Significant downward temporal trend (Cochran-Armitage p<0.001). - Reductions versus 2015–2019 reference: • Influenza: 67.7% reduction during lockdown; 99.9% reduction post-lockdown. • RSV: 98.0% reduction post-lockdown. • hMPV: 92.2% reduction post-lockdown. • Enterovirus: 82.2% reduction post-lockdown. • Adenovirus: 81.4% reduction post-lockdown. • Parainfluenza virus types 1–3: 80.1% reduction post-lockdown. • Rhinovirus: 74.6% reduction post-lockdown. - Rhinovirus rebound after easing restrictions (Level 1): Proportion of rhinovirus detections increased significantly post-lockdown: 33% (175/529) from 8 Jun–11 Aug 2020 vs 4.8% (10/209) from 13 May–7 Jun 2020 (p<0.0001). SHIVERS-II&III and SARI surveillance also observed increased rhinovirus-associated incidence after easing restrictions.

Findings demonstrate that stringent NPIs in New Zealand, implemented before the typical winter influenza season and maintained through border controls and ongoing hygiene and distancing, coincided with the near-elimination of winter influenza circulation (99.9% reduction) and substantial reductions in other respiratory viruses. This challenges prior assumptions that such NPIs are ineffective for influenza control and suggests that, when timed early and applied rigorously, they can prevent or markedly mitigate influenza epidemics. The results align with observations from other southern hemisphere countries and Hong Kong, supporting broader applicability. Potential contributing factors include increased influenza vaccination (~22% coverage in 2020, with 35% more vaccinations recorded April–June vs 2019), a warmer-than-average winter (nationwide average 9.6 °C, 1.1 °C above the 1981–2010 average) potentially reducing influenza virus stability, and viral–viral interactions (e.g., interferon-mediated interference). Notably, rhinovirus activity increased after easing restrictions, potentially due to household transmission dynamics, children’s hygiene, and rhinovirus being non-enveloped (less susceptible to soap-and-water inactivation), which may explain its distinct pattern compared with other viruses. Policy implications: In settings facing overlapping COVID-19 and influenza seasons, continuation or strengthening of NPIs could substantially reduce influenza and other respiratory viral burdens, easing pressure on health systems. The timing of interventions (before seasonal seeding) and border controls may be critical. The findings support re-evaluating the role of stringent NPIs in pandemic influenza preparedness and in mitigating severe respiratory viral threats. Further work should delineate which NPI components are most effective and sustainable for seasonal control.

This observational study documents an unprecedented reduction in influenza and other respiratory viral infections in New Zealand in 2020 and the complete absence of the usual winter influenza epidemic. The most plausible driver was the implementation of stringent NPIs (border restrictions, isolation and quarantine, social distancing, and behavior changes) in response to COVID-19. These findings inform pandemic influenza preparedness and suggest that well-timed, rigorous NPIs can mitigate or avert influenza and other respiratory virus epidemics. Future research should identify the most effective and sustainable NPI components and assess their broader societal costs and benefits.

- Observational design with multiple simultaneous measures varying by alert level; unable to attribute effects to specific NPI components. - Laboratory prioritization of SARS-CoV-2 testing during the COVID-19 response reduced testing for influenza and other respiratory viruses; clinician-ordered sampling (non-systematic) may introduce selection bias. - Disruptions to sentinel GP-based ILI surveillance (patient flow to COVID-19 testing centers, requirement to swab ILI patients, and lower GP participation) likely reduced consultations, reporting, and sample collection in 2020. - Despite these constraints, hospital-based SARI surveillance and SHIVERS-II/III cohorts operated as usual and consistently showed apparent elimination of influenza circulation.