Participant recruitment and enrolment
We enrolled participants belonging to three different target populations for COVID-19 vaccination: 1) community-dwelling persons aged 60-85 years, of whom approximately half were expected to have a medical risk condition to be prioritized for COVID-19 vaccination (12), 2) community-dwelling persons aged 18-59 years with a medical risk condition, and 3) community-dwelling persons aged 18-59 years without a medical risk condition. Within each target group, we defined strata based on the primary series vaccine brand. A sample size calculation, assuming a 6-month infection rate among unvaccinated individuals of 0.03 and a vaccination coverage of 85% and accounting for loss to follow-up and uncertainty of vaccine strata sizes resulted in the requirement of ~5,000 participants per stratum to be able to detect a VE of 70% (Supplementary file 1). We anticipated 10 different strata, resulting in a targeted sample size of ~50,000 participants . Participants must be able to understand Dutch, as all study materials are written in Dutch, and were included irrespective of their COVID-19 vaccination status or intention to get vaccinated.
A random sample of the national Dutch Personal Records Database, containing all individuals with a home or postal address in the Netherlands, stratified by age group (18-39, 40-59, 60-85 years) was sent an invitation to participate in the study by regular mail. After sending the initial invitations, the number of unvaccinated persons and persons vaccinated with Spikevax, Vaxzevria and Janssen were relatively low in the VASCO study population. Therefore, two additional random samples from the Personal Records Database were taken, specifically approaching individuals who were not registered as being vaccinated and individuals registered as vaccinated with Spikevax, Vaxzevria or Janssen vaccine in the national COVID-19 vaccination Information and Monitoring System (CIMS) of the Dutch National Institute of Public Health and the Environment (RIVM). Vaccinations are registered in CIMS if the vaccinated individual provided informed consent for registration (13). In total, 770,000 individuals received a personal invitation to participate in the study. In addition, recruitment was done through (social) media. Specific recruitment of persons aged 18-59 years with a medical risk condition due to which they were prioritized for COVID-19 vaccination was done via general practitioners (GPs).
Potential participants could register themselves by entering the study website, answer some screening questions (age, not living in a health care facility), and submit their contact details. Information regarding use of the study website, the study-specific mobile phone application, and login code was sent by e-mail. Between 3 May 2021 and 15 December 2021 (Supplementary file 2, Figure S1), 60,390 persons subscribed on the study website and received a baseline package including study information, an informed consent form and a kit for self-collection of a blood sample. As we did not collect data on recruitment method (personal invitation or social media) until 11 June 2021, the actual response rate to the personal invitations remains unknown. In total, 46,619 (77.2%) participants signed informed consent, of which 45,547 (97.7%) completed the baseline questionnaire and 44,985 (96.5%) returned the baseline fingerprick sample.
Data collection
Participants are followed for 5 years after enrolment. All study procedures are done remotely via a study website and mobile phone application, and self-collection of fingerprick blood samples for SARS-CoV-2 serology. At baseline, participants were asked to complete a baseline questionnaire and to donate a fingerprick sample. Follow-up information is collected using monthly online questionnaires in the first year, and three-monthly online questionnaires in years 2-5. They are asked to complete validated wellbeing questionnaires every three months. In addition to the routine questionnaires, participants are asked to report all COVID-19 vaccinations and positive SARS-CoV-2 tests via the website or app. Each time a participant reports a positive test, follow-up questions are asked about symptoms and disease course. These questions are repeated after one month and supplemented with questions on test positivity and vaccination status of household members. One month after a participant reports a new vaccination, follow-up questions are asked about side effects around the time of vaccination. An overview of the data collected through various questionnaires is presented in Table 1.
In addition, participants are requested to collect fingerprick blood samples at 6, 12, 18, 24 and 30 months and one month after completion of the primary vaccination series (only when at least four weeks have passed since the baseline sample). Serum from fingerprick samples is tested for the concentrations of total immunoglobulin against the receptor binding domain of the SARS-CoV-2 Spike (S1) protein and the SARS-CoV-2 Nucleoprotein (see Supplementary file 3 for more detail).
Where relevant, data from registries on COVID-19 vaccination, SARS-CoV-2 infections, (COVID-19 related) death, and from hospitals and GPs on health status can be linked to the study data. Supplementary file 2, Figure S2 shows examples of data collection schedules of individual participants in VASCO from start until end of the study. From April 2022 onwards, VASCO participants receive SARS-CoV-2 self-tests to be used when having symptoms in order to keep track of SARS-CoV-2 infections, as testing at Public Health Service test centers was scaled down.
Table 1. Questionnaire data collected during VASCO
Questionnaire
|
Timing
|
Topics
|
Baseline
|
Baseline
|
- sociodemographic factors: age, sex, ethnicity, education, profession,
- health status: comorbidities, medication use, pregnancy, health care consumption, medication use, previous confirmed SARS-CoV-2 infections
- vaccination status: COVID-19, influenza, and/or pneumococcal vaccines
- behavioural responses to COVID-19 measures (e.g. visiting public places, face mask use, contacts, travelling, and physical distancing)
|
Follow-up
|
Monthly in the first year of follow-up, 3-monthly in years 2-5 of follow-up
|
- COVID-19 vaccination
- SARS-CoV-2 testing
- changes in health status
- behavioural responses to COVID-19 measures (e.g. visiting public places, face mask use, contacts, travelling, and physical distancing)
|
Well-being
|
Baseline, 3-monthly
|
- 12-Item Short Form Survey (SF-12)
- Checklist Individual Strength (14-16)
|
COVID-19 vaccination
|
Whenever applicable
|
- Vaccination date, vaccine product
|
Positive SARS-CoV-2 test
|
Whenever applicable
|
- Type of test (either a test by a Public Health Service test centre free-of-charge, a test at a commercial test centre, or a self-test provided to schools and people with limited funds through government and widely available in stores)
- Date of positive test
- Reason for testing
- Symptoms
- Disease course
|
Household members
|
1 month after a positive SARS-CoV-2 test
|
- Vaccination status of household members
- Positive tests of household members
- Age of household members
- Testing of household members
|
Adverse events
|
1 month after a COVID-19 vaccination
|
- Occurrence of injection site or systemic AE
- Occurrence of AE for which contact was sought with a health care professional
|
Follow-up positive SARS-CoV-2 test
|
1 month after a positive SARS-CoV-2 test and from May 2022 during subsequent months for as long as symptoms persist
|
|
Additional data collection can easily be added during the course of the study. Some additional data collection has already been performed. In the period of December 2021 to March 2022, a subset of participants was requested to donate an additional fingerprick blood sample after confirmation of a breakthrough infection. These samples were used to assess boosting characteristics of serum SARS-CoV-2 specific antibody responses (17). In March 2022, participants were asked whether they received pneumococcal or influenza vaccination in the previous Winter season. In January 2023, all female participants <50 years who had reported to be pregnant during the study and were expected to have given birth, were invited to complete a questionnaire regarding neonatal outcomes, including questions on gestational age, birth weight, Apgar score, and hospital and NICU admission. These data are used to study the association of COVID-19 vaccination and SARS-CoV-2 infection on neonatal outcomes. Furthermore, from June 2023 onwards, participants are asked to mail in the test cartridges of positive self-tests to determine the virus variant by sequencing. This enables monitoring of circulating SARS-CoV-2 variants in the Dutch population and estimating VE by specific variant.
Follow-up
After one year of follow-up, 5,502 of the 45,547 participants had dropped out of the study, resulting in an overall attrition rate of 12.1%. The distribution of the non-time-varying characteristics of the (remaining) study population remained largely the same at baseline and 1-year follow-up (Table 2). Attrition analysis (Supplementary file 4) of 32,001 participants showed that attrition during the first year of follow-up was lower among Dutch participants compared to migrants and children of migrants, among participants with higher age, among females, among participants with intermediate education or high education. Higher test intention was associated with a lower attrition with a HR for always testing of 0.40 (95%CI 0.33 – 0.49) to a HR for rarely testing of 0.77 (95%CI 0.61 – 0.98) compared to never testing. Medical risk condition, recruitment method, and having experienced an infection did not affect the attrition rate. Some age group specific differences in attrition rate were detected. For example, attrition was higher among those with medical risk in the 60-85 year age group but lower in the 18-59 year age group (see Supplementary file 4, Table S1). Reasons for dropout were largely unknown (for 92% of dropouts). Reported reasons for dropout included study duration and logistics (e.g. issues with the study app) (4.7%), personal circumstances (2.8%), or death (0.6%).
Cohort characteristics
The main sociodemographic characteristics of VASCO participants are shown in Table 2. The median age is 61 years (95% range 25 – 76). Compared to the general Dutch population, VASCO participants are more often women (i.e. 62.9% vs 50.3% ) (18), highly educated (56.8% vs. 40.0%) (19), and of Dutch origin (89.5% vs. 75.4%) (18). Of the participants between 18 and 60 years, 19.8% have a medical risk condition. In the age group 60-85 years, 38.3% of the participants have a medical risk condition, compared to 51% in the general population (12). The most commonly reported comorbidities in participants with a medical risk condition are cardiovascular disease (n=7,965), lung disease or asthma (n=3,576), and diabetes (n=2,210). Participants reside in all but one of the 352 municipalities of the Netherlands, with more participants residing in more densely populated areas (Fig 1). After one year of follow-up, the COVID-19 vaccination coverage (Supplementary file 2, figure S3) for at least a primary series among VASCO participants was higher compared to the Dutch adult population (97.5% vs 86.0%) and unvaccinated participants are underrepresented in VASCO (2.2% vs. 10.9%) (6). The vaccine product used for primary series vaccination was most often Comirnaty (40.1%) or Vaxzevria (34.4%). The anticipated sample size was (almost) reached for 6 out of the 10 prespecified strata (Supplementary file 1, Table S1). The percentage of participants who experienced a SARS-CoV-2 infection (based on self-report of a positive PCR or antigen (self) test and SARS-CoV-2 serology) increased from 16.6% at baseline to 67.3% after one year of follow-up. Supplementary file 2, Figure S4 shows the incidence of reported positive SARS-CoV-2 tests (either PCR or antigen (self)test) during study follow-up (see www.rivm.nl/vasco/resultaten for the latest update of this graph).
Table 2. Characteristics of VASCO participants at baseline and after 1 year of follow-up.
|
Baseline
|
Retention at 1 year follow-up
(n = 40,045)
|
Attrition at 1 year follow up
(n = 5,502)
|
|
Total
(n = 45,547)
|
18-59 years (n = 21,679)
|
60-85 years
(n = 23,821)
|
|
|
Sex (%)
|
Male
|
16,881 (37.1)
|
6001 (27.7)
|
10,862 (45.6)
|
14,743 (36.8)
|
2138 (38.9)
|
Female
|
28,640 (62.9)
|
15,655 (72.2)
|
12,957 (54.4)
|
25,285 (63.1)
|
3355 (61.0)
|
Other
|
26 (0.1)
|
23 (0.1)
|
2 (0.0)
|
17 (0.0)
|
9 (0.2)
|
Age (years)
|
18-59
|
21,679 (47.6)
|
21,679 (100.0)
|
0 (0.0)
|
18,220 (45.5)
|
3459 (62.9)
|
60-69
|
18,981 (41.7)
|
0 (0.0)
|
18,981 (79.7)
|
17,501 (43.7)
|
1480 (26.9)
|
70-85
|
4840 (10.6)
|
0 (0.0)
|
4840 (20.3)
|
4301 (10.7)
|
539 (9.8)
|
Missing
|
47 (0.1)
|
0 (0.0)
|
0 (0.0)
|
23 (0.1)
|
24 (0.4)
|
Medical risk conditiona (%)
|
Yes
|
13,440 (29.5)
|
4289 (19.8)
|
9135 (38.3)
|
11,918 (29.8)
|
1522 (27.7)
|
No
|
32,107 (70.5)
|
17,390 (80.2)
|
14,686 (61.7)
|
28,127 (70.2)
|
3980 (72.3)
|
Migrant status (%)
|
Dutch
|
40,785 (89.5)
|
19,097 (88.1)
|
21,647 (90.9)
|
35,984 (89.9)
|
4801 (87.3)
|
Migrant
|
2482 (5.4)
|
1414 (6.5)
|
1064 (4.5)
|
2133 (5.3)
|
349 (6.3)
|
Child of migrant(s)
|
2280 (5.0)
|
1168 (5.4)
|
1110 (4.7)
|
1928 (4.8)
|
352 (6.4)
|
Educational levelb (%)
|
Low
|
6312 (13.9)
|
1479 (6.8)
|
4826 (20.3)
|
5554 (13.9)
|
758 (13.8)
|
Intermediate
|
13,088 (28.7)
|
6726 (31.0)
|
6348 (26.6)
|
11,400 (28.5)
|
1688 (30.7)
|
High
|
25,890 (56.8)
|
13,403 (61.8)
|
12,462 (52.3)
|
22,875 (57.1)
|
3015 (54.8)
|
Other
|
255 (0.6)
|
71 (0.3)
|
183 (0.8)
|
215 (0.5)
|
40 (0.7)
|
Missing
|
2 (0.0)
|
0 (0.0)
|
2 (0.0)
|
1 (0.0)
|
1 (0.0)
|
Vaccination statusc, * (%)
|
Unvaccinated
|
2916 (6.4)
|
2541 (11.7)
|
371 (1.6)
|
871 (2.2)
|
208 (3.8)
|
Partly vaccinated
|
7294 (16.0)
|
4134 (19.1)
|
3150 (13.2)
|
124 (0.3)
|
204 (3.7)
|
Primary vaccination
|
27,352 (60.1)
|
11,254 (51.9)
|
16,068 (67.5)
|
1914 (4.8)
|
2106 (38.3)
|
One booster
|
7971 (17.5)
|
3747 (17.3)
|
4221 (17.7)
|
16,402 (41.0)
|
2366 (43.0)
|
Two boosters
|
13 (0.0)
|
3 (0.0)
|
10 (0.0)
|
15,242 (38.1)
|
483 (8.8)
|
Three boosters
|
1 (0.0)
|
-
|
1 (0.0)
|
5492 (13.7)
|
135 (2.5)
|
Vaccine product first vaccination* (%)
|
Comirnaty (BioNTech/Pfizer)
|
17,035 (37.4)
|
7757 (35.8)
|
9254 (38.8)
|
16,060 (40.1)
|
2384 (43.3)
|
Spikevax (Moderna)
|
5905 (13.0)
|
5529 (25.5)
|
373 (1.6)
|
5193 (13.0)
|
909 (16.5)
|
Vaxzevria (AstraZeneca)
|
15,018 (33.0)
|
1227 (5.7)
|
13,776 (57.8)
|
13,775 (34.4)
|
1261 (22.9)
|
Jcovden (Janssen)
|
4608 (10.1)
|
4591 (21.2)
|
16 (0.1)
|
4092 (10.2)
|
724 (13.2)
|
Other
|
31 (0.1)
|
20 (0.1)
|
11 (0.0)
|
27 (0.1)
|
8 (0.1)
|
Unknown
|
34 (0.1)
|
14 (0.1)
|
20 (0.1)
|
27 (0.1)
|
8 (0.1)
|
Unvaccinated
|
2916 (6.4)
|
2541 (11.7)
|
371 (1.6)
|
871 (2.2)
|
208 (3.8)
|
SARS-CoV-2 infectiond, * (%)
|
Yes
|
7568 (16.6)
|
4523 (20.9)
|
3041 (12.8)
|
26,954 (67.3)
|
2470 (44.9)
|
No
|
37,979 (83.4)
|
17,156 (79.1)
|
20,780 (87.2)
|
13,091 (32.7)
|
3032 (55.1)
|
Recruitment method
|
Personal invitation
|
24,661 (54.1)
|
9341 (43.1)
|
15,299 (64.2)
|
21,870 (54.6)
|
2791 (50.7)
|
Via social media
|
4621 (10.1)
|
4148 (19.1)
|
469 (2.0)
|
3939 (9.8)
|
682 (12.4)
|
Via family, a friend, colleague or acquaintance
|
2682 (5.9)
|
1690 (7.8)
|
990 (4.2)
|
2310 (5.8)
|
372 (6.8)
|
Other
|
321 (0.7)
|
237 (1.1)
|
84 (0.4)
|
264 (0.7)
|
57 (1.0)
|
Missing
|
13,262 (29.1)
|
6263 (28.9)
|
6979 (29.3)
|
11,662 (29.1)
|
1600 (29.1)
|
Test intention*
|
Never
|
616 (1.4)
|
227 (1.0)
|
386 (1.6)
|
405 (1.0)
|
151 (2.7)
|
Rarely
|
598 (1.3)
|
290 (1.3)
|
308 (1.3)
|
1115 (2.8)
|
216 (3.9)
|
Sometimes
|
1138 (2.5)
|
609 (2.8)
|
529 (2.2)
|
2271 (5.7)
|
370 (6.7)
|
Regularly
|
672 (1.5)
|
464 (2.1)
|
206 (0.9)
|
1229 (3.1)
|
183 (3.3)
|
Often
|
6138 (13.5)
|
3811 (17.6)
|
2322 (9.7)
|
8098 (20.2)
|
1091 (19.8)
|
Always
|
35,806 (78.6)
|
16,055 (74.1)
|
19,715 (82.8)
|
26,693 (66.7)
|
3450 (62.7)
|
Missing
|
579 (1.3)
|
223 (1.0)
|
355 (1.5)
|
234 (0.6)
|
41 (0.7)
|
a Medical risk condition: one or more of following conditions: diabetes mellitus, lung disease or asthma, asplenia, cardiovascular disease, immune deficiency, cancer (currently untreated, currently treated, untreated), liver disease, neurological disease, renal disease, organ or bone marrow transplantation. Four most frequent conditions are presented here.
b Educational level was classified as low (no education or primary education), intermediate (secondary school or vocational training), or high (bachelor’s degree, university).
c Unvaccinated (no vaccination received), primary vaccination series received (one dose of Jcovden [Janssen] 28 + days ago, or two doses of Vaxzevria [AstraZeneca], Comirnaty [BioNTech/Pfizer], or Spikevax [Moderna] 14 + days ago), primary vaccination series and one booster received (primary vaccination series + one additional dose 7 + days ago), primary vaccination series and two boosters received (primary vaccination series + two additional doses 7 + days ago), or primary vaccination series and three boosters received (primary vaccination series + three additional doses 7 + days ago)
*Vaccination status, vaccine product, infection status and test intention are time varying variables. The first three columns present the variable status at baseline, in the last two columns the status of the variable at time of dropout or 1 year of follow-up (whichever came first) is presented. Vaccination status was based on data from the questionnaires linked with data from CIMS (see Supplementary file 5).
d Infection status was defined using self-reported test-confirmed infections and serology data (see Supplementary file 4).