The seasonality of respiratory syncytial virus in Western Australia prior to implementation of SARS‐CoV‐2 non‐pharmaceutical interventions

Abstract Background Respiratory syncytial virus (RSV) seasonality is dependent on the local climate. We assessed the stability of RSV seasonality prior to the SARS‐CoV‐2 pandemic in Western Australia (WA), a state spanning temperate and tropical regions. Method RSV laboratory testing data were collected from January 2012 to December 2019. WA was divided into three regions determined by population density and climate: Metropolitan, Northern and Southern. Season threshold was calculated per region at 1.2% annual cases, with onset the first of ≥2 weeks above this threshold and offset as the last week before ≥2 weeks below. Results The detection rate of RSV in WA was 6.3/10,000. The Northern region had the highest detection rate (15/10,000), more than 2.5 times the Metropolitan region (detection rate ratio 2.7; 95% CI, 2.6–2.9). Test percentage positive was similar in the Metropolitan (8.6%) and Southern (8.7%) regions, with the lowest in the Northern region (8.1%). RSV seasons in the Metropolitan and Southern regions occurred annually, with a single peak and had consistent timing and intensity. The Northern tropical region did not experience a distinct season. Proportion of RSV A to RSV B in the Northern region differed from the Metropolitan region in 5 of the 8 years studied. Conclusions Detection rate of RSV in WA is high, especially in the Northern region, where climate, an expanded at‐risk population and increased testing may have contributed to greater numbers. Before the SARS‐CoV‐2 pandemic, RSV seasonality in WA was consistent in timing and intensity for the Metropolitan and Southern regions.


| BACKGROUND
Respiratory syncytial virus (RSV) is a leading cause of respiratory illness in all ages. RSV is a significant source of childhood morbidity, with an estimated 3 million children under 5 hospitalised annually. 1 Mortality is relatively uncommon in children and is predominantly restricted to low-resource settings. 1 In contrast, mortality in hospitalised older adults is high, with RSV lower respiratory tract infection reaching more than 5%. 2 RSV can be classified into two groups, RSV-A and RSV-B, with the majority group varying between seasons.
RSV seasonality can vary between regions, driven by the prevailing climate. 3 In areas with a temperate climate, RSV epidemics typically occur annually, peaking in the winter months. The associated cold and dry conditions enhance transmission by increasing viral stability on surfaces. 3 Areas with tropical climates have more stable year-round activity, with increases observed in the warm and high humidity rainy season, mediated by enhanced contact-mediated transmission. [3][4][5] Western In 2020, SARS-CoV-2-related non-pharmaceutical interventions were associated with an absent winter RSV season in WA and a subsequent out-of-season surge. 6,7 An improved understanding of RSV seasonality in the region is required to contextualise the significance of this shift. In this report, we assessed the stability of RSV seasonality in WA, by region and viral subtype, between 2012 and 2019.

| METHODS
Laboratory data were collected for all RSV tests from January 2012 to December 2019 from PathWest Laboratory Medicine, the state reference laboratory. This laboratory provides respiratory virus testing for all WA public hospitals as well as some community-based healthcare in regional areas. 8 All clinical samples arriving during the study period requesting respiratory virus testing were tested by RT-PCR using either an in-house assay 9 or GeneXpert Xpress Flu/RSV (Cepheid, USA) for RSV, influenza, human metapneumovirus and parainfluenza viruses 1-3. Until June 2018, patients attending the state tertiary paediatric hospital were also tested by direct immunofluorescence (IMAGEN, ThermoFisher Scientific, Australia) in addition to RT-PCR. A portion of RSV-positive samples had an RSV subtype determined by subtype-specific RT-PCR. 9 Individuals with multiple positive and negative tests were counted once per calendar month.

| Geographical groupings
The residential postcode at the time of sample collection was used to determine the geographical location. Patients with no address listed and out-of-state postcodes were excluded from this analysis. The state was divided into three geographical areas, determined by population density and climate ( Figure 1): Metropolitan region, Northern region and Southern region. Based on 2016 census data, the Metropolitan region has a population of 1.9 million, accounting for 80% of WA's total. 10,11 The Northern region, situated north of the Tropic of Capricorn, has a population of approximately 94,000. 12 The Southern regions, situated mostly below the Tropic of Capricorn, experience a temperate climate, with cooler weather occurring June-August and has a total population of approximately 430,000. 13-15

| Analysis
Season onset and offset threshold was set at 1.2% of the annual total RSV cases for each region and year. 16 Season onset was defined as the first of two consecutive weeks where RSV cases exceeded this threshold. Season offset was defined as the last week above this threshold before declining below this level for more than two consecutive weeks. The intensity was defined as the highest number of cases per week per year. In instances with more than 1 week with equal peak cases, the median of those weeks was used for timing. Statistical comparisons between regions were calculated using the chi-square test. Grubbs test was used to detect outliers in season intensity within regions, as measured by the number of cases in the peak week.
Detection rate was calculated using population data as the denominator, [10][11][12][13][14][15] and detection rate ratios were used to compare detections between regions. A p-value of less than 0.05 was considered significant in all statistical tests.

| RESULTS
Between 2012 and 2019, PathWest tested 144,590 WA individuals for RSV, 12,424 (8.6%) of which were positive (Table 1; Table S1). Age of cases were predominantly children, with 41.5% under 12 months, 28.3% 1-4 years, 19% 19.3 years and 10.9% over 65 years; 52.1% of cases were male. Most cases (68.5%) were detected in the more populous metropolitan region. The mean detection rate of RSV during the study period in WA was 6.3 per 10,000. The metropolitan region's mean annual cases per 10,000 population was 5.5 (range, 4.3 to 7.2).
The mean cases per 10,000 was 8.1 (range, 6.4 to 12.4) in the Southern Region. The highest rates were observed in the Northern region, with a mean of 15 per 10,000 (range, 11.2 to 20.7). Rates were consistently higher each year in the Northern regions and lower in the Metropolitan region, contributing to a detection rate ratio (DRR) of 2.7 (95% confidence interval [CI], 2.6-2.9). Overall, the proportion positive was similar between the Metropolitan region (8.6%) and the Southern region (8.7%). The lowest percentage-positivity was observed in the Northern region (8.1%).

| Seasonality
The Metropolitan region experienced a single annual peak each year, occurring in the winter months ( Season onset and offset were not defined for the Northern region, as data analysis could not determine a seasonal pattern. The northern regions did not experience an annual peak but instead had continued activity throughout the year (Figure 2).   Note: Rate calculated per 10,000 resident population. N = number RSV detected per year. Case and population numbers are supplied in Table S1. Abbreviations: %pos, percentage test positivity; 95% CI, 95% confidence interval; Metropolitan DRR, detection rate ratio compared with Metropolitan region. *Proportion statistically higher than the two other regions (p < 0.05). **Proportion statistically lower than the other two regions (p < 0.05).

| RSV subtypes
No difference was seen in RSV season timing when analysing RSV A and RSV B dominant seasons separately.

| DISCUSSION
This report examined retrospective laboratory data to determine RSV detection rate and seasonality in Western Australia. There were 6.3 cases of RSV per 10,000 population in WA during the study period.
The highest detection rate was observed in the Northern region, Metropolitan region and, to a lesser extent, the Southern region appears to be uncommon, with other temperate regions experiencing more variability in their season timing. 4,16 Perth has minimal year-to- year climate variability, with July being the coldest, wettest month. 20 This predictability, combined with an endemic northern reservoir, may contribute to stable seasonality.
The Northern region did not experience a distinct RSV season but had ongoing RSV activity throughout the year. Previous studies have shown RSV seasonality in tropical regions to be diverse, 4

CONFLICT OF INTEREST
David W. Smith is Director of the Asia-Pacific Alliance for the Control of Influenza, an independent not-for-profit organisation that receives funding from vaccine manufacturers. The position is unpaid and he receives only reimbursement of expenses.

DATA AVAILABILITY STATEMENT
The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.

PEER REVIEW
The peer review history for this article is available at https://publons. com/publon/10.1111/irv.13117.