This document describes the data sources and methodology used for the NSW Health Respiratory Surveillance Reports.
Information related to cases of COVID-19 are stored in the NSW Notifiable Conditions Information Management System (NCIMS), managed by Health Protection NSW. Data reported in the weekly report are sourced from the Notifiable Conditions Records for Epidemiology and Surveillance, NSW Ministry of Health. Data is updated as additional information becomes available and cannot be compared to previous reporting periods.
During the pandemic, a number of different methods have been used to determine the impact of COVID-19 on deaths in a community One way is by counting the number of people for whom a health practitioner considers their deaths was caused by or contributed to be COVID-19. Deaths as identified from the NSW Registry of Births Deaths and Marriages (BDM). If a person dies in NSW, their death must be registered under the Births, Deaths and Marriages Registration Act 1995 (Part 7). NSW Health receives a secure feed from the BDM on a daily basis under the Public Health Act 2010 (Part 129A). Deaths reported to a coroner will be registered with the BDM, however cause of death information may be delayed as it is not recorded until there is a coronial determination. For 2020 through 2021, NSW Health categorised deaths as COVID-19 deaths if they met the definition in Communicable Diseases Network of Australia’s (CDNA) COVID-19 National Guidelines for Public Health Units. As per the CDNA definition, deaths were considered COVID-19 deaths for surveillance purposes if the person died with COVID-19, not necessarily because COVID-19 was the cause of death. NSW Health determined 75% of COVID-19 deaths in 2022 were registered in the BDM in less than four weeks of death.
Another, more comprehensive way to estimate the total impact of the COVID-19 pandemic, and other prolonged and significant health threats, on deaths is to measure changes in the overall number of deaths in a community (regardless of cause), using an indicator called all-cause mortality.
This analysis includes all deaths registered in NSW, including overseas and interstate visitors, sourced from the NSW Registry of Births Deaths and Marriages. We report mortality up to 4 weeks prior to the date of analysis. Not all deaths are registered in this time (in 2021, 93% of deaths were registered within 6 weeks), therefore death rates have been corrected to reflect delays in the most recent weeks. Time series of weekly counts of deaths from January 2017 to 2023 year-to-date were prepared. The date of death was used to count deaths in a week ending Sunday. Death rates were calculated using 2022 NSW Department of Planning and Environment population projections for NSW. The seasonal baseline was estimated by modelling all-cause death rates using seasonally adjusted robust regression. Deaths from 2017 to 2021 were used to fit the model and forecast the seasonal baseline for 2022 and 2023. This may be a slight underestimate due to the effect of lower mortality in winter 2020. The usual variation limits were estimated as 1.96 standard errors above and below the seasonal baseline (95% confidence interval).
This analysis has been undertaken as part the ongoing surveillance of all-cause mortality in NSW. Death rates presented in this report are not directly translatable to analyses in the ABS Provisional Mortality Statistics and Actuaries Institute COVID-19 Working Group reports which make specific comparisons of pre and during pandemic mortality.
NSW Health collects data on the workforce impacts of COVID-19 within Local Health Districts. Healthcare workers may be furloughed if are unable to work due to testing positive to COVID-19, exposure to COVID-19, and/or whilst waiting a negative test result. As healthcare workers can be exposed to COVID-19 within the community when the amount of COVID-19 circulating in the community increases the risk of exposure and transmission also increases leading to increased numbers of healthcare workers being furloughed (absent) from work. This indicator is helpful to assess the level of COVID-19 circulating in the community when community testing decreases. These data also provide an insight into the stress experienced within the healthcare system due to reduced staffing capacity. The data is provided by the People, Culture and Governance Division, NSW Ministry of Health.
The NSW Sewage Surveillance Program tests untreated sewage for fragments of the SARS-CoV-2 virus that causes COVID-19. Gene copy numbers are influenced by many factors including virus shedding by people (which varies individually and over the course of the infection), dilution of virus within sewage – such as during rain, the period over which the sewage sample is collected, and the presence of chemicals and microorganisms in the sewage that affects how well the testing can detect SARS-CoV-2 virus fragments. Gene copy numbers are reported per 1,000 people in the catchment over time. Trends should be interpreted over an extended period and consider these fluctuations in environmental conditions.
Where case notifications are presented alongside sewage detection these represent the number of cases residing within the catchment area.
The NSW Public Health Rapid, Emergency, Disease and Syndromic Surveillance (PHREDSS) system provides daily monitoring of most unplanned presentations to NSW public hospital emergency departments (EDs) and all emergency Triple Zero (000) calls to NSW Ambulance. Emergency hospital presentations and ambulance calls are grouped into related acute illness and injury categories. The number of presentations and calls in each category is monitored over time to quickly identify unusual patterns of illness. Unusual patterns could signify an emerging outbreak of disease or issue of public health importance in the population. PHREDSS is also useful for monitoring the impact of seasonal and known disease outbreaks, such as seasonal influenza or gastroenteritis, on the NSW population. The 88 NSW public hospital EDs used in PHREDSS surveillance account for 95% of all ED activity in NSW public hospitals in 2020-2021, including most major metropolitan public hospitals (99%) and rural public hospitals (89%).
PHREDSS ED diagnosis-based surveillance syndromes include clinician applied provisional diagnoses (ICD9, ICD-10AM or SNOMED-CT codes):
Whole genome sequencing (WGS) is a laboratory procedure that identifies the genetic profile of an organism. WGS can help understand how a virus transmits, responds to vaccination, and the severity of disease it may cause. It can also help to monitor the spread of the virus by identifying specimens that are genomically similar. WGS has been used in NSW since the start of the COVID-19 pandemic to inform epidemiological investigations, and to monitor for and analyse the behaviour of new SARS-CoV-2 variants circulating in the community. WGS is conducted at three NSW reference laboratories.
Community samples are sourced from cases who test via PCR at community pathology services and may not necessarily reflect the distribution in all cases across NSW. NSW continues to monitor results from cases who are admitted from ICU to monitor for increased disease severity and from cases who return from overseas to monitor for new variants introduced into NSW
FluTracking is an online health surveillance system used to detect epidemics of influenza across Australia and New Zealand. Participants complete an online survey each week to provide community level influenza-like illness surveillance, consistent surveillance of influenza activity across all jurisdictions over time, and year to year comparisons of the timing, attack rates and seriousness of influenza in the community. For more information about FluTracking and ways to be involved visit FluTracking.
The NSW sentinel laboratory network comprises of 13 public and private laboratories throughout NSW who provide additional data on positive and negative test results. These data enable the determination of test positivity, that is among the people who have been tested, how many test positive. A high positivity suggests high transmission in the community.
