Covid-19 wave comparison and inequalities analysis
09 Feb 2022
09 Feb 2022
The COVID-19 pandemic has thrown health and care systems across the world into turmoil, creating significant challenges for the provision of safe health and care services. The pandemic has also brought into sharp focus the health inequalities that we face in many of our communities.
More than 152,000 people in the UK have died after contracting the virus, while more than one million survivors are still dealing with the debilitating after effects known as Long Covid, which will continue to heap pressure on our health and care systems.
Much has been learned about the virus so far, with vaccines and treatments now available to reduce the severity of the disease. However, with new variants still evolving, and the likelihood of the Omicron variant leading to significant workforce shortages through isolation and illness, it is critical that we improve in areas where the virus poses the greatest threat in order that its impact is reduced as much as possible.
This report has compared the first two waves of the pandemic, observing which demographic groups were seeing poorer outcomes after being admitted with COVID-19 and highlights several vulnerable groups where further thought and planning is required to improve health inequities and outcomes going forward. The Office for National Statistics estimated that the first wave of COVID-19 started in March 2020 and ended at the end of May 2020. The second wave is estimated to have started at the beginning of September 2020 and ended at the end of April 2021.
In general, crude mortality was lower in the second wave compared to the first wave. Both waves saw more deprived patients making up higher proportions of COVID-19 related hospital activity than their less deprived counterparts and in the second wave almost half of COVID-19 related hospital activity (49.6 per cent) was accounted for by patients in the two most deprived quintiles. It was also clear that patients with several existing health conditions experienced high mortality rates compared to patients with fewer, or no health conditions.
The results are stark and should be the starting point for change with more attention being paid to both the needs of socially disadvantaged groups and how services can be redesigned to address those inequalities. Such inequalities have been a feature of our health systems for many years and are the reason many chronic conditions are now placing pressures on all areas of health and social care. Yet, while the pandemic has brought those inequalities to the fore, it is also widening the gap even further.
Population health management will be a key driver in being able to identify and communicate with disadvantaged communities to help educate, change, and put in place pro-active support and care to reduce this widening gap.
At the beginning of the pandemic, agencies began to work closely together to identify and support those who were most at risk. Data is a vital tool in exposing where these inequalities lie and where extra funding needs to be targeted to make a difference. Bringing agencies together such as health, social care, local authority, voluntary and charitable agencies alongside primary and community care and pooling their data can create the clearest picture of where help is most needed to eradicate inequalities and break down barriers to access healthcare.
COVID-19 spells (identified by a confirmed or suspected COVID-19 diagnosis code in any episode of the spell) in the first and second wave of the pandemic were identified from Hospital Episode Statistics (HES). As a result, the information presented in this analysis is related to COVID-19 patients only where the disease resulted in a hospital admission in England. Wave one included all discharges between March 2020 and May 2020 and wave two included discharges between September 2020 and March 2021 (April 2021 data was not available at the time of analysis).
The cohort of patients seen in each wave of the pandemic was broken down by several demographic variables, comorbidities and by those that received ventilation or were admitted to critical care. Patient outcomes including crude mortality, seven and 28-day emergency readmissions and length of stay (LOS) were analysed across these variables and different geographic regions of England.
The comorbidities included in the analysis were cancer, chronic heart disease, chronic kidney disease, chronic liver disease, chronic neurological conditions, chronic respiratory conditions, diabetes, frailty, obesity, palliative care on admission, stroke, and vitamin D deficiency (Appendix, Table A1).
Summary of differences between wave one and wave two
In general, crude mortality was lower in the second wave compared to the first wave (Figure 1) although both seven and 28-day emergency readmission rates and average length of stay (LOS) increased in the second wave compared to the first.
Figure 1. Crude mortality by region in the first (left) and second (right) wave of the COVID-19 pandemic.
While wave two saw a decrease in deaths, possibly due to changes in care pathways and advances in medication and treatment, such as the use of dexamethasone, both waves saw more deprived patients making up higher proportions of COVID-19 activity than their less deprived counterparts. In the second wave almost half of COVID-19 related activity (49.6 per cent) was accounted for by patients in the two most deprived quintiles.
People with several comorbidities also experienced high mortality rates compared to patients with fewer or no comorbidities. In both waves, these were characterised by patients with a record of palliative care on admission, chronic kidney disease, frailty, and stroke. The largest number of deaths were accounted for by patients with a record of coronary heart disease, frailty, or diabetes in both waves.
The second wave of the pandemic was longer and saw a higher number of COVID-19 cases than the first (Figure 2). Additional pressure was added by winter and the emergence of the Alpha variant. These observations may shed light on gaps in current health services and population health.
Figure 2. Number of spells with a confirmed or suspected COVID-19 diagnosis code between 1 March 2020 and 31 March 2021. Grey shading illustrates the periods defined as wave one and wave two for this analysis.
Demographic differences between wave one and wave two
Figure 3. The proportion of COVID-19 spells (y-axis) and crude mortality (size of the point) by demographic features (A) sex; (B) ethnicity; (C) method of admission; (D) source of admission and (E) deprivation quintile.
Figure 4. The proportion of wave one (left) or wave two (right) COVID-19 cases by age group (value at end of each bar) and the crude mortality rates for each age group by wave (red section of each bar).
Of COVID-19 spells in the first wave, 81% had a confirmed COVID-19 diagnosis code and 19% suspected COVID-19. In the second wave 96% of spells had a confirmed COVID-19 diagnosis code and only 4% of spells suspected COVID-19. Crude mortality among patients with confirmed COVID-19, decreased from 30.2% in wave one to 19.8% in wave two while the crude mortality rate for patients with suspected COVID-19 remained fairly consistent in wave one and two (17.9% vs 18.4% respectively).
Figure 5. Proportion of COVID-19 spells where (A) COVID-19 is the primary diagnosis for the spell and (B) where the COVID-19 diagnosis is confirmed or suspected by wave of the pandemic.
Comorbidities
Wave two saw a slightly higher proportion of spells with no comorbidities compared to the first wave and a lower proportion of spells with comorbidities.
Figure 6. The proportion of COVID-19 spells (x-axis) and crude mortality rate (y-axis) for patients in different comorbidity bands in the first and second wave of the pandemic.
Figure 7. The crude mortality rate for patients with each of the included comorbidities (y-axis) and the number of deaths accounted for by patients with each of these conditions in wave one and wave two.
Critical care
The proportion of COVID-19 spells which had a record of critical care was higher in the second wave (9.91%) compared to the first (8.59%) although the crude mortality rate was notably lower in the second wave than in the first wave (19.72% compared with 27.86%, see Figure 8).
Ventilation
We looked at three different ventilation types in this analysis: invasive ventilation, non-invasive ventilation, and continuous positive airway pressure (CPAP). Crude mortality for spells with a record of any of these ventilation types was lower in the second wave compared to the first. There were some interesting differences in the proportion of spells with a record of each of these ventilation types between the waves. While in the first wave 5.40% of spells received CPAP, this increased to 8.88% of spells during the second wave. The proportions of spells where patients received invasive and non-invasive ventilation decreased in the second wave compared to the first.
It is evident from these figures that there have been changes in the care pathways of patients receiving ventilation and improvements in patient care which have resulted in notable improvements in outcomes (Figure 8).
| Figure 8. The proportion of COVID-19 spells with (A) a record of critical care and (B) by ventilation type. The size of the point indicates the crude mortality rate in the first and second wave of the pandemic. |
Region
Differences in both the spread of the virus and susceptibility of the population can influence regional differences in outcomes. Crude mortality rates decreased in wave two compared to wave one across all regions of England (Figure 9).
Figure 9. Crude mortality rate for COVID-19 spells by region in the first and second wave of the pandemic. The size of the point indicates the number of spells relative to other points in this figure.
Length of stay and readmission outcomes
In all regions besides London, the North East and Yorkshire, median length of stay has increased in the second wave compared to the first. Additionally, both seven and 28-day readmission rates have also increased across all regions (Table 1).
Table 1. Regional differences in median length of stay (LOS), seven-day readmission rates and 28-day readmission rates in the first and second wave.
| NHS England Region | Median LOS | 7-Day Readmission Rate (%) | 28-Day Readmission Rate (%) | |||||
| Wave one | Wave two | Wave one | Wave two | Wave one | Wave two | |||
| South West | 7 | 8 | 5.92 | 7.79 | 11.01 | 13.71 | ||
| South East | 7 | 8 | 6.07 | 7.92 | 11.69 | 13.61 | ||
| North West | 7 | 8 | 5.99 | 7.46 | 11.58 | 13.99 | ||
| North East and Yorkshire | 7 | 7 | 6.43 | 8.26 | 12.33 | 14.61 | ||
| Midlands | 7 | 8 | 6.48 | 8.08 | 12.09 | 14.54 | ||
| London | 7 | 6 | 5.13 | 7.47 | 9.93 | 12.75 | ||
| East of England | 6 | 8 | 6.09 | 7.67 | 11.89 | 13.86 | ||
Age
When looking at median LOS and readmission outcomes by age, the youngest age group (<19 years old) sees a slight reduction in LOS and seven and 28-day readmission rates. The three oldest age groups (65-74; 75-84 and 85+) see increases in median LOS and both seven and 28-day readmission rates (Table 2). This is different to mortality outcomes which have improved for all age groups in the second wave.
Table 2. Differences in median length of stay (LOS), seven-day readmission rates and 28-day readmission rates by age group in the first and second wave of the pandemic.
| Age Group | Median LOS | 7-Day Readmission Rate (%) | 28-Day Readmission Rate (%) | ||||||
| Wave one | Wave two | Wave one | Wave two | Wave one | Wave two | ||||
| <19 | 2 | 1 | 7.40 | 6.16 | 12.29 | 10.93 | |||
| 19-44 | 3 | 3 | 5.65 | 5.69 | 10.33 | 9.55 | |||
| 45-64 | 6 | 6 | 5.21 | 6.83 | 10.05 | 12.10 | |||
| 65-74 | 7 | 9 | 6.36 | 8.48 | 12.35 | 15.26 | |||
| 75-84 | 8 | 11 | 6.62 | 9.28 | 12.54 | 16.73 | |||
| 85+ | 9 | 12 | 5.83 | 8.52 | 11.40 | 15.09 | |||
Comorbidity
Spells with no recorded comorbidities observed the same median length of stay across both waves however spells with comorbidities saw longer length of stays in the second wave compared to the first. All spells, with or without comorbidities, saw increased rates of seven and 28-day readmissions in the second wave, compared to the first. The increase in readmission rates was largest for spells where the patient had 2 or 3+ recorded comorbidities (Table 3).
Table 3. Differences in median length of stay (LOS), seven-day readmission rates and 28-day readmission rates by comorbidity band in the first and second wave of the pandemic.
| Count Of Diagnosis Groups | Median LOS | 7-Day Readmission Rate (%) | 28-Day Readmission Rate (%) | ||||
| Wave one | Wave two | Wave one | Wave two | Wave one | Wave two | ||
| 0 | 4 | 4 | 5.23 | 6.43 | 9.57 | 10.72 | |
| 1 | 7 | 8 | 6.05 | 7.72 | 11.40 | 13.94 | |
| 2 | 8 | 10 | 6.21 | 8.62 | 12.31 | 15.41 | |
| 3+ | 8 | 11 | 6.50 | 8.87 | 12.64 | 16.24 | |
When comparing seven and 28-day readmission rates across the first and second wave by deprivation quintile there are no clear trends (Table 4). In alignment with our wider findings, seven and 28-day readmission rates are generally higher in wave two compared to wave one.
Table 4. Differences in seven and 28-day readmission rates by deprivation quintile in the first and second wave of the pandemic.
| Deprivation Quintile | 7-Day Readmission Rate (%) | 28-Day Readmission Rate (%) | ||
| Wave one | Wave two | Wave one | Wave two | |
| Quintile 1 | 6.21 | 7.86 | 12.32 | 14.12 |
| Quintile 2 | 5.81 | 7.62 | 10.94 | 13.51 |
| Quintile 3 | 5.96 | 7.90 | 11.30 | 13.98 |
| Quintile 4 | 5.88 | 8.04 | 11.10 | 14.20 |
| Quintile 5 | 6.01 | 7.76 | 11.17 | 13.72 |
| Unknown | 4.92 | 7.35 | 11.77 | 12.20 |
When COVID-19 first started to spread it seemed to be indiscriminate in its infection. Constant research and learning throughout the pandemic have revealed trends that can help to focus where help is needed most.
Our analysis has revealed definite trends between the first and second waves, with a reduction in deaths likely due to improvements in care pathways and treatments. However, wave two saw longer hospital stays and higher readmission rates among older age groups and patients with comorbidities. This could be the result of more patients, having survived acute illness, requiring more clinical support, such as rehabilitation or step-down treatment. In addition, wave two did not see the complete lockdown of NHS services that wave one saw, so it is likely that more patients were coming forward for care, who may not have sought treatment during wave one. Patients who were admitted to hospital for reasons unrelated to COVID-19 are also likely to have had an increased risk of nosocomial infections, due to compromised immunity.
Overall, it is men who are more likely to be affected than women, yet only by a small margin, which reduced from the first to the second wave. However, it is socio-economic inequalities and those who suffer with several comorbidities that create the greatest challenge for health systems, that is having the greatest impact on these systems across the world.
Our analysis also shows that geographic variation is likely to be influenced by demographic differences. Social inequalities have been exposed in terms of patient groups who have seen the worst outcomes during the pandemic. Both waves saw more deprived patients making up higher proportions of COVID-19 spells, than their less deprived counterparts. During the second wave nearly 50 per cent of spells were accounted for by patients in the two most deprived quintiles.
Patients with several comorbidities had a higher risk of death during both waves and the largest number of deaths were accounted for by patients with a record of coronary heart disease, frailty, or diabetes.
Using the data to highlight these variances is crucial to being able to tackle the areas that are causing the greatest pressures on the NHS. By providing earlier support and intervention for specific groups of patients and putting in place the range of services required to help promote healthier lifestyles and tackle the underlying cause of health inequalities, it may be possible to reduce the most severe effects of the pandemic. This will take a co-ordinated approach from a number of agencies and sectors in the long-term, to avoid further burden on the healthcare system and economy and prevent ongoing disparities in outcomes. Population health management will be a key driver in being able to identify and communicate with disadvantaged communities to help educate, change, and put in place pro-active support and care to reduce this widening gap.
Telstra Health supports healthcare systems with an end-to-end population health management solution suite; providing the technology, services and experts needed to maintain, improve and proactively manage a population’s health and wellbeing.
Table A1. Diagnosis codes included to identify each comorbidity
| Condition | ICD-10 Codes included |
| Chronic respiratory diseases | J40-44, J84 |
| Diabetes | E10-14 |
| Obesity | E66 |
| Cancer | C code or D00-09 |
| Chronic kidney disease | N18, Z49 |
| Chronic heart disease | I10-15, I25, I42-I43, I50 |
| Stroke | I61, I63-64 |
| Vitamin D Deficiency | E559 |
| Chronic neurological conditions | G20, G212-219, G122, G35 |
| Chronic liver disease | K70, K72, K73, B18 |
| Palliative on admission | Z515 in the first episode OR Treatment function code: 315 in the first episode. |
Get the latest insights and updates from Telstra Health UK.
View now
