Throughout the COVID-19 pandemic, Public Health Scotland (PHS) provided accurate, timely and relevant information that has played a key part in enabling people to make informed decisions during this challenging time.
As part of our continuous review of the information we share on COVID-19, we are changing how we present information on the virus, including the statistics on COVID-19 hospital admissions.
In this blog post, the PHS COVID-19 Data Teams explain why the change is necessary and how the information will be presented going forward to more accurately reflect the prevalence of COVID-19, as we move from the emergency response to living with the virus.
As part of our response to COVID-19, Public Health Scotland (PHS) has provided robust estimates of rates of COVID-19 among the Scottish population and have presented these in our COVID-19 Dashboard and COVID-19 Weekly Statistical Report.
These outputs have been reliable resources of COVID-19 data for both the public and policy makers and were developed when there was widespread SARS-CoV-21 testing in the community, which was conducted to reduce the spread of the virus.
As we now move to living with the virus, the emphasis has shifted from containing the virus to managing the impact, including limiting severe disease and managing demands on the health system. As a result, we have moved away from community testing to targeting testing to specific groups.
Because of this, testing data have become increasingly difficult to interpret and do not reflect the burden of COVID-19 in the community.
Reporting on COVID-19 prevalence in the community
PHS offers trustworthy data and it is therefore important that our practices are continually reviewed to ensure the data we provide continues to be accurate. As a result, several current measures have been replaced with others that are more representative.
Currently, the best data available to support our understanding of community population prevalence is the ONS COVID-19 Infection Survey (external website). The ONS data lags the COVID-19 test data (PCR and LFD cases) by approximately two weeks and so PHS will continue to report PCR and LFD case data as an early indicator of future trends.
Therefore, from 22 September 2022, several measures will no longer be shown on the COVID-19 dashboard including; positive cases by episode of infection/test type, positive tests and cases by local neighbourhood. Instead, the ONS COVID Infection Survey information, which is currently the most robust measure of community COVID-19 prevalence in Scotland, is now included within our weekly report as the primary measure of prevalence.
Measuring those admitted to hospital ‘because of’ COVID-19
Alongside COVID-19 prevalence within the community, it is important to monitor COVID-19 hospital admissions, as it is a measure of severe disease and captures pressures facing NHS hospitals.
PHS currently publishes weekly hospital admissions ‘with’ COVID-19, defined as those testing positive by PCR or LFD test 14 days prior to or during their hospital stay. This is irrespective of their symptoms and whether they were admitted because of, or coincidentally with, their SARS-CoV-21 infection. It includes those admitted for both emergency and elective admissions to all specialties as well as those who acquired the infection in hospital.
The current measure was devised early in the pandemic to understand the severity of COVID-19 but since the emergence of the Omicron variant and in light of high vaccine coverage, as well as changes to testing practices, we are aware that increasing proportions of individuals are not admitted to hospital because of the virus and that their infection is in fact coincidental.
Therefore, a more robust indicator of those more likely to be admitted ‘because of’ COVID-19 is the Scottish Morbidity Records (SMR01 2) which captures clinical discharge diagnoses. Due to a three-month lag in these data becoming available, it is not suitable for real-time monitoring of COVID-19 related admissions. PHS has therefore been working to derive an alternative measure to allow real-time reporting of hospital admissions ‘because of’ COVID-19 to enable monitoring of the impact on hospitals.
As previously published in the Changes to the severity of COVID-19 and impact on hospitals in Scotland report in June 2022, we’ve found that individuals who had acquired their infection in the community and were subsequently admitted to medical specialties as an emergency will give a better indication of admissions ‘because of’ COVID-19. In addition, recent changes in SARS-CoV-21 testing practices in healthcare settings (including a move to symptomatic testing) have strengthened the need to restrict COVID-19 hospital admissions to community acquired infections identified by PCR tests.
The impact of this change can be seen within Figure 1 below, which shows the number of COVID-19 admissions using the current definition alongside our new analyses. The proposed new definition tracks the current ‘with’ COVID definition well until January 2022 when the Omicron variant emerged, and also positive LFD tests were introduced to the positive case definition.
The new measure is a more robust real-time indicator for COVID-19 hospital admissions ‘because of’ COVID-19.
Figure 1: Trend of COVID-19 hospital admissions in Scotland, old and new methodology, January 2021 to September 2022
These changes reflect our efforts to continuously review our COVID-19 outputs to ensure they continue to provide the most accurate, timely and useful information available.
1 Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is a viral respiratory disease strain of coronavirus that causes COVID-19
2 The SMR01 dataset comprises episode-based patient records relating to all inpatients and day cases discharged from non-obstetric, non-psychiatric specialties
3 Community-acquired hospital admissions are defined as those testing positive 14 days before or 2 days after admission. Under the new definition, COVID-19 hospital admissions will only include those with a positive PCR test from emergency admissions to medical specialties, excluding surgical and mental health specialties and emergency admissions for injuries.