Will achieving herd immunity be a road to success to end the COVID-19 pandemic?

Abstract

As the COVID-19 pandemic continues, the availability of several different new vaccines, their varying supply levels, effectiveness, and immunity duration across different ethnic populations, together with natural infection rates, will have an impact on when each country can reach herd immunity (ranging from 15.3% to 77.1%). Here we estimate the population proportions still required to gain immunity (ranging from 0.01% to 48.8%) to reach an overall herd immunity level to stop the exponential virus spread in 32 selected countries.

Dear Editor,

Our previous work estimated the minimum, i.e. ‘critical’, level of population immunity acquired via vaccination or natural infection $\left(P_{\mathit{crit}}\right)$ to stop the spread of Coronavirus Disease 2019 (COVID-19) among 32 selected study populations.¹ Currently, over 1 billion COVID-19 vaccine doses are administered in 208 territories. Early insights from countries with high vaccine uptake offer the hope that mass vaccination can bring an end to the pandemic, though this does not necessarily mean a complete virus eradication, which is likely to persist to become endemic and seasonal in most populations.²

The crucial question of what is the minimal vaccine coverage needed for different countries to achieve SARS-COV-2 herd immunity (i.e. that required to block exponential virus spread in a population) is an important one, when COVID-19 vaccine supplies are limited and unreliable, and different vaccines have different efficacies. With evidence demonstrating natural immunity effectiveness (i.e. immunity acquired after natural SARS-COV-2 infection), we can factor this into the minimum vaccine coverage required for any given population. Much of the COVID-19 vaccine and incidence data can only be estimated from publicly available and various websites, but these can be combined to provide useful estimates of the required herd immunity level - and therefore the COVID-19 vaccine coverage still required - for different countries.

We revisit the calculation of $P_{crit}$ and estimate the current immune proportion, $P_{im}$, with the following formulae:

$$P_{crit}=1-\frac{1}{R_o}$$ (1)

$$P_{im}=P_{v1}\times V{E}_1+P_{v2}\times V{E}_2+P_{cc}\times P_{ni}$$ (2)

where $R_o$ is the basic reproductive number, $P_{v1}$ and $P_{v2}$ are the proportions of the population vaccinated with one and two doses, respectively, $V{E}_1$ and $V{E}_2$ are the overall real-world population effectiveness of the vaccine for one and two doses, respectively, $P_{cc}$ is the proportion of confirmed cases, and $P_{ni}$ is the proportion of the population who have naturally-induced immunity against symptomatic SARS-COV-2 infection. From Eqs. (1) and (2), we define P as the proportion still required to gain immunity for the country to achieve herd immunity: $P=P_{crit}-P_{im}$. A country with P > 0 indicates that its population had achieved herd immunity. All analyses were performed in R (version 4.1.0; R Foundation for Statistical Computing) (Table 1 ).

Table 1.

Characteristics of vaccine deployment in the 32 selected study populations.

Study country	Totalpopulation	R0⁎	Cumulative % of population reported as infected†	First vaccination rollout	Days since first rollout (up to 26/5/2021)	Number of priority groups‡	% of population receiving 1 dose of the vaccine§	%% of population receiving 2 doses of the vaccine
Australia	25,499,881	2.21	0.12	22/02/2021	93	4a	9.5	1.9
Austria	9,006,400	2.31	7.13	27/12/2020	150	3	35.3	15.0
Bahrain	1,701,583	4.36	13.31	25/12/2020	152	4	55.2	43.8
Belgium	11,589,616	2.30	9.08	28/12/2020	149	3	37.4	16.2
Brazil	212,559,409	1.33	7.66	21/01/2021	125	3	22.2	10.0
Canada	37,742,157	1.98	3.64	14/12/2020	163	4b	41.2	4.6
Czech Republic	10,708,982	1.60	15.50	27/12/2020	150	3	38.5	12.3
Denmark	5,792,203	1.32	4.78	27/12/2020	150	3	31.6	21.0
Finland	5,540,718	1.39	1.66	27/12/2020	150	3	32.8	7.9
France	67,564,251	1.86	8.41	27/12/2020	150	5	34.5	15.3
Germany	83,783,945	2.10	4.38	27/12/2020	150	3	35.4	15.6
Greece	10,423,056	1.24	3.79	27/12/2020	150	4	29.2	18.0
Iceland	341,250	1.47	1.92	29/12/2020	148	5	39.5	23.6
Iran	83,992,953	1.45	3.41	9/02/2021	106	3	5.1	0.5
Iraq	40,222,503	1.49	2.94	2/03/2021	85	5	3.2	–9
Israel	8,655,541	3.46	9.70	19/12/2020	158	5	62.5	59.2
Italy	60,461,828	1.95	6.95	27/12/2020	150	4c	34.3	18.1
Japan	126,476,458	1.56	0.58	17/02/2021	98	2	5.1	2.4
Kuwait	4,270,563	1.44	7.10	24/12/2020	153	4	19.3	0.9
Malaysia	32,365,998	1.75	1.65	24/02/2021	91	4	5.6	3.1
Netherlands	17,134,873	1.42	9.70	8/01/2021	138	3	35.5	14.3
Norway	5,421,242	3.87	2.28	27/12/2020	150	3	25.8	16.6
Portugal	10,196,707	1.53	8.30	27/12/2020	150	4	34.8	16.1
Qatar	2,881,060	1.26	7.51	23/12/2020	154	4	46.0	35.5
Singapore	5,850,343	1.47	1.06	8/1/2021	138	4	31.1	27.6
Slovenia	2,078,932	1.18	12.14	27/12/2020	150	3	34.4	17.5
South Korea	51,269,183	2.23	0.27	26/02/2021	89	3	6.4	3.9
Spain	46,754,783	2.36	7.82	27/12/2020	150	3	35.8	18.4
Sweden	10,099,270	1.45	10.57	27/12/2020	150	3	35.5	12.2
Switzerland	8,654,618	1.51	7.99	23/12/2020	154	3	33.3	18.7
United Kingdom	67,886,004	1.66	6.61	8/12/2020	169	4d	51.5	35.4
United States	331,002,647	1.72	10.03	14/12/2020	163	5	50.1	39.8

^⁎We first estimate $R_o$with the exponential growth method¹ using COVID-19 case series from 21st January 2020 to 31st July 2020 (Fig. 1) coupled with estimates of the serial interval² (mean = 4.7 days, standard deviation = 2.9 days). Each country's exponential phase was defined as the period from onset (the first day of a consecutive 3-day period with at least 3 cases) to the peak (maximum cases) of the first wave. The first wave was defined as the period from onset to the day when the number of cases decreased by more than 50% of the maximum up to that day for at least 3 consecutive days or did not exceed the maximum for 7 consecutive days.

^†Information updated on 26/5/2021.

^‡Three priority groups were key workers, clinically vulnerable people and the elderly.2, 3: vaccines available for 2 and 3 of the above priority groups, respectively.4: vaccines available for all of three priority groups plus partial additional availability for various other subgroups or age groups.5: universal availability, when vaccine is available to everyone ≥16 or ≥18 (depends on the lowest age permitted by the vaccine brand currently).

^§Information updated on 26/5/2021, except for Iceland and Malaysia (updated 25/5/2021), Iran and Singapore (updated 24/5/2021), Netherlands (updated 23/5/2021), Iraq (updated 11/5/2021), and Kuwait (updated 18/4/2021).

^aIndigenous people aged 50 or above were eligible as a priority group under the current phase of vaccinations. (Reference: https://www.health.gov.au/initiatives-and-programs/covid-19-vaccines/phase-1b#aboriginal-and-torres-strait-islander-people).

^bIn Canada, some provinces including Saskatchewan, Alberta, New Brunswick and Ontario added pregnancy to the vaccine priority groups. (Reference: https://www.cbc.ca/news/canada/montreal/pregnant-women-not-prioritized-covid-19-vaccine-1.5999304).

^cStudents in the final year of high school in Lazio, Italy were prioritized to receive the COVID-19 vaccine. (Reference: https://www.salutelazio.it/vaccinazione-maturandi).

^dAdults experiencing homelessness in Scotland were one of the eligible priority groups for COVID-19 vaccination.(Reference: https://www.nhsinform.scot/COVID-19-vaccine/invitations-and-appointments/who-will-be-offered-the-coronavirus-vaccine).

^eIraq had no data for 2 doses.References:

1. Wallinga J, Lipsitch M. How generation intervals shape the relationship between growth rates and reproductive numbers. Proc Biol Sci 2007;274(1609):599–604.

2. Nishiura Hiroshi, Linton Natalie M, Akhmetzhanov Andrei R. Serial interval of novel coronavirus (COVID-19) infections. Int J Infect Dis 2020. Doi: 10.1016/j.ijid.2020.02.060.

The estimates of $R_o$ varied by country, ranging from 1.18 to 4.36, resulting in the corresponding $P_{crit}$ estimates ranging from 15.3% to 77.1%. Our lower end estimate of herd immunity threshold was consistent with those of 10 to 20% from a recent study³. The Czech Republic had the highest proportion of reported cases (15.5%) followed by Bahrain (13.3%) and Slovenia (12.1%). Although vaccine rollout was delayed in most Asian countries compared to Europe and North America, COVID-19 vaccines were currently available in our 32 study populations. According to their eligibility for vaccination within the national guidance of each country⁴, they were further classified with different levels of vaccine availability into three subpopulations. In 16 countries, vaccines were available for at least two priority subgroups (key workers, clinically vulnerable, elderly), in 11 countries vaccines were available to all three aforementioned priority groups and extra availability for selected broader subgroups (e.g. indigenous peoples, pregnant women) or age groups (e.g. ≥18, ≥30) in some countries and in 5 countries vaccines were universally available. Countries with universal vaccine availability such as Israel and the United States had higher $P_{im}$ values (62.5% and 50.1%, respectively). Surprisingly, countries in Asia such as Malaysia, Japan, and South Korea, regardless of level of vaccine availability, reported low single-digit $P_{im}$ values (5.6%, 5.1%, and 6.4%, respectively). Of the 32 study countries, 11 had achieved herd immunity, 6 others required P to be between 0.01% and 8.6% to reach the herd protection level, and the rest required proportions ranging from 11.1 to 48.8%. (Table 1 , Fig. 1 , and Supp Fig. 1)

Fig. 1.

Proportion of population already immune ($P_{im}$) (red) and the additional proportion still required to achieve herd immunity ($P$) (blue) in the 32 study populations stratified by vaccine availability for various key priority groups. With the most recent data for the numbers of vaccine doses given and naturally occurring COVID-19 cases, as reported from each country's population on 26^th May 2021,¹ assumed estimates for $V{E}_1$, $V{E}_2$, and $P_{ni}$ to be 70%², 88%³ and 80%^4,5, respectively, $P_{im}$ can be estimated. Percentages to the right of each bar represent the minimum proportion of the total population required to recover from COVID-19 to confer immunity with vaccine availability $\left(P_{crit}\right)$. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.).

References

1 Coronavirus (COVID-19) Vaccinations. Available at https://ourworldindata.org/covid-vaccinations. Accessed April 28, 2021, n.d.

2 Lopez Bernal Jamie, Andrews Nick, Gower Charlotte, Robertson Chris, Stowe Julia, Tessier Elise, et al. Effectiveness of the Pfizer-BioNTech and Oxford-AstraZeneca vaccines on covid-19 related symptoms, hospital admissions, and mortality in older adults in England: test negative case-control study. BMJ 2021;373:n1088.

3 Bernal Jamie Lopez, Andrews Nick, Gower Charlotte, Gallagher Eileen, Simmons Ruth, Thelwall Simon, et al. Effectiveness of COVID-19 vaccines against the B.1.617.2 variant n.d. Doi: 10.1101/2021.05.22.21257658.

4 Boyton Rosemary J, Altmann Daniel M. Risk of SARS-CoV-2 reinfection after natural infection. Lancet 2021:1161–3.

5 World Health Organization. COVID-19 natural immunity: scientific brief, 10 May 2021. World Health Organization; 2021.

Our study suggested that the majority of the study populations had lower proportions that were immune compared to Israel, the exemplar in reducing the infection rate after successful vaccine deployment.⁵ This might be partly attributable to the inequitable distribution of vaccines globally, which may be shaping different government policies on vaccination, but also cultural and socioeconomic barriers leading to vaccine refusal and hesitancy, particularly amongst Asian and African populations. Thus, to improve COVID-19 vaccination coverage and raise the levels of population immunity, sufficient vaccine supplies need to be more reliable,⁶ with improved, culturally sensitive, and appropriate communication to encourage their uptake. This will partly depend on whether we can successfully identify determinants of vaccine hesitancy and refusal amongst various populations.⁷

The exact proportion in any population that is required to achieve herd immunity to stop the spread of the virus will vary, depending on the virus variant circulating, as well as the natural degree of mixing in that population - which also depends on population density and mobility and so on. In addition, the duration of protection conferred by natural and vaccine-induced immunity is not well-established, and different vaccines may confer differing durations and degrees of humoral (B-cell) and cell-mediated (T-cell) immunity.^{8 , 9} It is also not known how long and effective the immunity conferred by mixed vaccine regimens and third dose boosters will be in different populations - including those of different ethnicities. Finally, children are still not routinely vaccinated as most COVID-19 vaccines are not yet licensed for this subgroup, particularly primary school children, which means they will mostly remain a susceptible population where any degree of herd immunity will be uncertain. Therefore, the precise level of population immunity required, as estimated by the equation of herd immunity, to ‘end’ the pandemic in each country and globally is difficult to determine.

From a practical viewpoint, estimates of $P_{crit}$ will be considered to be transient and herd immunity is likely to be a spectrum instead of a specific threshold that determines if and when the entire pandemic is over.¹⁰ The current pandemic might end gradually with an increasing proportion of immune individuals. Also, since all COVID-19 vaccines seem to protect against severe disease and death, and against most viral variants, universal vaccination is still the key message. As this will take time, maintaining social distancing, universal mask-wearing, and improved ventilation indoors to control the virus spread, are all still important as the vaccine coverage in different countries improves.

Declaration of Competing Interest

None