Introduction

The SARS-CoV-2 (COVID-19) pandemic and associated control measures such as national “lockdowns”, involving varying restrictions on leaving the home, work, and socialising, have had a profound impact on daily life and the delivery of healthcare worldwide. In the United Kingdom, a national lockdown was announced on 23 March 2020, with instructions that people should only leave their home for a limited number of “essential” reasons [1]. This was accompanied by the reconfiguration of acute healthcare services to support the anticipated influx of COVID-19 patients, cancellation of most elective activity, and pausing of screening programmes [2]. During lockdown, there was evidence of a change in healthcare-seeking behaviour—for example, in Scotland, the uptake of both emergency and elective hospital-based care dropped substantially over the lockdown period [3]. However, within child health, key routine services including childhood immunisations continued through the UK [4].

It has become increasingly apparent that younger children are at low risk of severe disease due to Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) [5] and may be less susceptible to infection with the virus [6]. Yet the wider impact of the pandemic on children in terms of education, mental and physical health, and safeguarding is not yet fully understood and is likely to be profound [7,8]. One particular area of concern early in the lockdown period was the potential effect on the uptake of routine childhood immunisations [9]. Maintaining high population vaccine coverage is vital for both direct and indirect (via herd immunity) protection against non-COVID-19 infectious diseases.

In July 2020, the World Health Organization (WHO) warned of a potential decline in routine immunisation rates associated with the COVID-19 pandemic, citing a poll from May 2020 in which respondents from 82 countries suggested disruption to immunisation programmes was widespread [10]. Initial reports from England [11], Pakistan [12], South Africa [13], Singapore [14], and the USA [15] were concerning, suggesting a fall in children receiving their scheduled vaccinations in the very early weeks of national lockdowns, though, to the best of our knowledge, the full impact has not yet been assessed. However, the English and US studies relied on surrogate measures of vaccine uptake; number of vaccines delivered/ordered (without a corresponding denominator) [11,15] and the Singaporean study used convenience sampling with multiple assumptions for missing data [14]. Longer-term data were available from the KwaZulu-Natal Province of South Africa [13] and Sindh Province of Pakistan [12], which appeared to show some recovery after an initial fall in uptake; however, in Pakistan, part of the lockdown restrictions involved shutting down of outreach immunisation programme. The overall impact of lockdown on immunisation uptake in higher-income countries, which maintained their routine immunisation programmes, is unclear.

Given the prolonged and repeated periods of lockdown, it is important to evaluate the overall effect on childhood immunisation uptake. The aim of this study was, therefore, to provide a longer perspective than previous studies by describing the pattern of childhood immunisation uptake in Scotland and England before, during, and immediately after the first national lockdown implemented in response to the pandemic (23 March to 31 July), with comparisons to baseline data from 2019, by geographical area and socioeconomic deprivation.

Methods

Results

Data were available for the outcome of 439,754 vaccine invitations in Scotland and 4.08 million for England across all time periods. As many infants or children would be eligible for multiple immunisations during the study, these figures do not represent individual participant numbers. Detailed demographic data are not routinely collected and were therefore not available. For context, S1 Table shows population data from the relevant national records offices for births in Scotland and England by sex, maternal age, and maternal country of birth from 2015 to 2020. It is expected that these data are broadly representative of the children included in this study.

All CIs presented are 95% CIs, and ORs are compared to the baseline year 2019 unless otherwise stated.

Preschool immunisation uptake increased during the lockdown period in Scotland

Across Scotland, the percentage of preschool children receiving their immunisations within 4 weeks of becoming eligible increased during the lockdown period for all 5 immunisations (Fig 1, Table 1, S2 Table). The change in percentage uptake compared to the 2019 baseline ranged from 1.3% for the first dose 6-in-1 (OR 1.28, CI 1.18 to 1.39) to 14.3% for the second dose MMR (OR 1.8, CI 1.74 to 1.87) (Table 1). Across all the immunisations visits, this equated to an additional 7,508 preschool immunisations being delivered in a timely manner over the lockdown period compared to the baseline rates of 2019. Uptake rates dipped immediately before the announcement of a national lockdown in mid-March 2020 (Fig 1), then peaked throughout June before starting to decrease. However, uptake remained significantly higher than 2019 during the postlockdown period (Fig 1, Table 1). Of note, prior to lockdown for both MMR doses, there was already a modest, increase in uptake compared to 2019 (Table 1).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 1. Percentage of children in Scotland immunised within 4 weeks of eligibility.

(A) All doses 6-in-1 vaccine; (B) MMR vaccine across January to September 2020 in Scotland. Lockdown period = blue shaded area. For January and February, a single mean monthly value is plotted, and from March onwards, weekly uptake is shown (see S2 Table for full data). Dashed horizontal lines (—) indicated the mean uptake in 2019 of the immunisation with the corresponding colour. The increase in uptake during the lockdown period was statistically significant (see Table 1 for details). MMR, measles, mumps, and rubella.

https://doi.org/10.1371/journal.pmed.1003916.g001

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Uptake of preschool immunisations in Scotland by time period with OR compared to baseline of 2019.

https://doi.org/10.1371/journal.pmed.1003916.t001

Variation in uptake of preschool immunisations by geographical area

Baseline data from 2019 showed the percentage uptake of preschool immunisations within 4 weeks of eligibility varied widely by geographical HSCP and immunisation (Fig 2, S1 Fig, S3 Table). In keeping with the rise in mean uptake across Scotland for all vaccines, the percentage of children immunised in most HSCP increased between 2019 and lockdown. However, not all followed this pattern with a minority demonstrating a fall in uptake (Fig 2, S1 Fig, S3 Table). Care must be taken when interpreting percentage results from the island HSCPs (Shetland Islands, Orkney Islands, and Western Isles) given the very small numbers of children involved (S3 Table).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 2. Percentage uptake by Scottish HSCP in 2019 and statistically significant increases between 2019 and “lockdown”.

Left-hand side maps show baseline mean percentage uptake by HSCP in Scotland in 2019, and right-hand side maps show percentage point difference between 2019 and lockdown for areas in which the difference was statistically significant for each vaccine. Grey areas indicated HSCP where there were no statistically significant changes in uptake between 2019 and lockdown. Note that in Grampian, the second MMR dose is delivered at a later age; therefore, data for this area are recorded as “nonapplicable” (further details in S2 File). The full dataset can be found in S3 Table. Base layer map available from https://spatialdata.gov.scot/geonetwork/srv/eng/catalog.search;jsessionid=714BD98E15D22A8116824CA25B30DC02#/metadata/ac5e870f-8fc2-4c21-8b9c-3bd2311a583f, which contains public sector information licenced under the Open Government Licence v3.0. HSCP, Health and Social Care Partnership; LD, lockdown; MMR, measles, mumps, and rubella; Wks, weeks.

https://doi.org/10.1371/journal.pmed.1003916.g002

For individual HSCP, the statistical significance of the change in uptake varied by immunisation (Fig 2, S3 Table). Despite a general trend of improvement for the first 6-in-1 vaccine, we found a significant change for only 8 of the 31 HSCP, mainly centred around the more densely populated, urban central belt of Scotland (Glasgow City, Edinburgh, Stirling and Clackmannanshire, East Dunbartonshire, Falkirk, Fife, South Lanarkshire, South Ayrshire; Fig 2). However, this pattern evolved with the different immunisation visits and with almost all HSCP showing a rise in uptake for both MMR immunisations, with percentage point increases as high as 30% (Angus, 74.1% versus 43.8%, OR 3.38,CI 2.61 to 4.37) (Fig 2, S3 Table).

Preschool immunisation uptake increased across all deprivation levels

Percentage uptake within 4 weeks of becoming eligible rose across all SIMD quintiles, between 2019 and lockdown, for all immunisations (Fig 3). The magnitude of this rise varied by quintile and vaccine (S2 Fig, S4 Table) from 0.3% (SIMD 4, first 6-in-1 dose, OR 1.1, 95% CI 0.9 to 1.3) to 16.2% (SIMD 5, second MMR dose, OR 2.1, 95% CI 1.9 to 2.3). The increase in uptake between 2019 and lockdown was statistically significant for all except first dose 6-in-1 for the least deprived quintiles (4 and 5) (S4 Table). In the postlockdown period, percentage uptake remained significantly higher than the 2019 baseline for all quintiles for each vaccine except for the first 6-in-1 dose, in which only the most deprived quintile retained a significant increase (S2 Fig, S4 Table).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 3. Mean percentage immunised by SIMD quintile in Scotland.

(A) 6-in-1 vaccine; (B) MMR vaccine for Scotland. See S2 Fig and S4 Table for absolute percentage rise compared to 2019 and significance levels. LD, lockdown; MMR, measles, mumps, and rubella; PostLD, postlockdown; PreLD, prelockdown; SIMD, Scottish Index of Multiple Deprivation.

https://doi.org/10.1371/journal.pmed.1003916.g003

In keeping with previous observations [21], children in the least deprived quintiles were more likely to be immunised and this relationship was broadly maintained throughout the study period (Fig 3 and S3). While all quintiles improved uptake between 2019 and lockdown, whether the inequality between most and least deprived increased or decreased varied by vaccine type. For all doses of the 6-in-1 vaccine, there was a tendency to a convergent improvement, i.e., the gap in percentage uptake between the quintiles lessened, while for both MMR doses, there was further divergence in uptake rates between most and least deprived (Fig 3, S4 Fig, S5 Table). The interaction of SIMD quintile and time period was nonsignificant for all 6-in-1 doses; that is to say all SIMD quintiles improved equally for this immunisation (S5 Table). However, for the first MMR dose, the improvement in uptake was statistically greater for SIMD quintiles 3 to 5 compared to SIMD 1, and for the second MMR dose, SIMD 5 showed a significantly larger increase in uptake between 2019 and lockdown (S5 Table). This suggests that for the MMR immunisation, the factors leading to an improvement in uptake had greater positive impact for children living in less deprived areas.

“Catch-up” immunisation rates and comparison with data from England

Thus far, the Scottish data presented have related to children receiving their immunisations as per the recommended schedule (within 4 weeks of the child becoming eligible by age), representing a “gold standard” in which the child is protected as early as possible. It is recognised that some children will receive their immunisations after this point. This “catch-up” effect can be seen in uptakes rate for all Scottish children by the time they reached between 6 and 8 months (6-in-1), 16 months (first MMR), or 3 years 8 months (second MMR) (Fig 4A). Those becoming eligible during lockdown showed minimal change in this longer-term measure of uptake for the 3 doses of the 6-in-1 immunisation, a small increase in uptake of the first MMR, and a more substantial increase in uptake of the second MMR (S6 Table). These data suggest that while lockdown was associated with a beneficial effect on timely uptake of all infant and preschool immunisations, the impact on longer term or final achieved uptake was more variable, possibly reflecting a ceiling effect on maximal uptake, for the earliest immunisations.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 4. Percentage of children up to date at 6/18 months in Scotland and England.

Overall mean percentage of children immunised by approximately 6 months of age (all doses of 6 in1, see y-axis for specific ages) or 16–18 months (first MMR) for Scotland and England. Each bar contains children who became eligible for the immunisation of interest during the time period indicated. (A) Scotland, (B) England *** p-value < 0.001, * p-value <0.05, ns, not significant. p-values calculated using binary logistic regression. LD, lockdown; MMR, measles, mumps, and rubella; PostLD, postlockdown; PreLD, prelockdown.

https://doi.org/10.1371/journal.pmed.1003916.g004

For England, broadly equivalent data were available for children aged 6 months (all doses 6-in-1) and 18 months (first dose MMR) who had become eligible for their immunisations during the time periods of interest (Fig 4B, S7 Table). These data demonstrated a small but statistically significant fall in uptake for all the immunisations in the lockdown periods compared to 2019, ranging from 0.5% (first dose 6-in-1, 95.8% versus 96.3%, OR 0.89, CI 0.86 to 0.91) to 2.1% (third dose 6-in-1, 86.6% versus 88.7, OR 0.82, CI 0.81 to 0.83) (S7 Table). However, much of the fall in uptake took place in the prelockdown period, particularly for the third dose 6-in-1, with a gradual recovery seen over the lockdown period itself (S4 Fig). A general trend towards falling MMR uptake can also be seen to predate the COVID-19 pandemic (S4 Fig).

Discussion

We found that early uptake of infant and preschool immunisations (within 4 weeks of a child becoming eligible) rose significantly for the duration of the first lockdown period in Scotland, resulting in thousands more children receiving their immunisations at the scheduled time. Encouragingly, positive results were seen across all levels of deprivation, though some geographical variations were observed across Scotland. Improvement was also seen in longer term uptake of the first and second MMR immunisations (immunisation within 4 months of becoming eligible) in the lockdown period. Findings in England differed, with a small fall in longer-term uptake of immunisations observed for the lockdown period.

The Scottish data in particular challenge the early predictions of an adverse effect of lockdown measures on childhood immunisation uptake and demonstrate the importance of continued surveillance throughout the various phases of pandemic control measures. From these data alone, we are unable to fully determine what changes in both the barriers and motivators to vaccination may have led to this result; however, examining previous research can offer some suggestions. While the concept of vaccine hesitancy is a popular media topic, previous studies have shown that the reasons given by parents for not vaccinating their children are often much more practical. A prepandemic report by the Royal Society for Public Health in the UK found that the major barriers for parents who wanted to immunise their children were timing and availability of appointments and childcare duties [22]. For those who actively chose not to immunise their children, fear of side effects was a key concern and the negative effects of social media are increasing [22]. The lockdown has had a major effect on parental working patterns with almost 9 million UK employees furloughed and millions more working from home, often with the additional tasks of managing childcare as schools and nurseries were closed [23]. While this had made life significantly more challenging for many, more flexible working patterns may have made attending immunisation appointments easier for some.

Jarchow-MacDonald and colleagues [24] from NHS Lothian (which consists of HSCPs Edinburgh, West Lothian, East Lothian, and Midlothian) have suggested that ensuring the accessibility of immunisation centres, either by public transport or by providing mobile services to shielding families, was important in maintaining uptake during the pandemic, as was directly communicating with families with a preappointment phone call and reminder postcards. This gave families an opportunity to discuss the immunisation with a healthcare professional, a strategy that has been showed to be important in addressing parental concerns [25]. In fact, the reminder alone may have been sufficient to encourage parents to attend the first appointment [26]. The clear commitment of the Scottish Government to maintain the immunisation programme was also felt to be an important factor [24].

In England, Bell and colleagues [27] conducted a large-scale online survey of parents of children under the age of 18 months to assess their experiences of accessing immunisations during the early part of the UK lockdown (19 April to 11 May 2020). They highlighted the uncertainty of some families about whether the immunisation service was continuing, particularly among non-white ethnic groups. While most felt it was important that their child was immunised on time, Bell and colleagues reported that many parents felt the risk of catching a vaccine-preventable infectious disease was reduced due to limited mixing with others. This suggests that an enhanced appreciation of the utility of immunisation was not a major motivator for parents to ensure they attended the immunisation appointments. These issues may well be reflected in the fall in immunisation uptake in England during the earlier part of lockdown reported here, and a delay in receiving the first dose 6-in-1 may well led to delays in subsequent doses, meaning that the infants were unable to “catch up” by 6 months of age.

Other factors, which may have had an impact on promoting timely immunisation uptake, could include a reduction in fever, cough, and colds, which may otherwise have caused parents to delay immunisations. Though specific data on this point are lacking, a significant fall in the detection of Rhinovirus in adults was observed in 2020 compared to 2019 [28], suggesting that “normal” childhood respiratory infections are likely to have decreased. More work is required to fully dissect the key factors in improving timely preschool immunisation uptake in Scotland. Clarity and publicity about the continuation of the immunisation programmes, telephone reminders, and the opportunity to discuss with healthcare professions seem likely to have had the most impact. We plan to use the data generated in this study to inform future work investigating these factors, including geographical differences and the introduction of policies such as reminders and publicity campaigns.

It is important to acknowledge the limitations of the data presented, many of which arise from opportunistically using routinely collected data rather than that obtained from a specific study design. The SIRS electronic system is well established and captures data on the entire child population in Scotland. However, the aggregate surveillance data derived from the system that we could access lacked detailed information on several potentially important factors, not least of which was ethnicity, which is known to affect both immunisation uptake and attitudes towards immunisation. [29]. As shown in S1 Table, the proportion of mothers born outside the UK is relatively low in Scotland (16.3% to 17.7% between 2015 and 2020) and is almost half that of England (28.4% to 30.2%). It may not be appropriate to extrapolate these data to countries with a significantly different ethnic makeup, and it is plausible that some of the difference seen between the Scottish and English data could be due to these factors.

In using the mean percentage of the entire year 2019 as our baseline comparator, we potentially run the risk of confusing normal seasonal variation in immunisation uptake with the impact of lockdown measures. Ideally, direct weekly or monthly comparisons would be made between 2019 and 2020, though these data are not available. The enhanced detail and frequency of collection of immunisation uptake data and rapid dissemination (through the COVID dashboard) was in response to the pandemic and therefore only began in March 2020. However, quarterly trends published for previous years including 2019 do not show major difference in uptake throughout the year and in fact show a gradual decline in uptake year on year since 2015 [30]. In addition, it is possible that 2020 uptake rates have been underestimated, due to a lag in data entry into SIRS, which would particularly affect the “catch-up” rates (Fig 4A).

Therefore, caution must be taken not to overinterpret the results presented here or extrapolate to significantly different populations with varying baseline immunisation uptakes rates and less robust immunisation programmes, the organisation of which may have been adversely affected by the pandemic. Nevertheless, this study has efficiently and quickly produced useful and valid results, which have the potential to aid the development of future research and guide policy. The positive findings regarding immunisation uptake despite the extreme circumstances of the COVID-19 pandemic and associated lockdowns (in Scotland at least) is an important message to send to support public and professional confidence in the preschool immunisation programme and help normalise timely immunisation uptake for both parents and health services. Improving public and professional confidence is particularly vital given the current importance of promoting vaccination against SARS-CoV-2. Despite these encouraging data, it is not possible to ascertain from the numbers alone, which are the key contributing factors to improving uptake. This is a key avenue of future research as lessons learnt can then be taken forward to optimise future immunisation programmes, both within the pandemic setting and beyond. In addition, we have not yet assessed the potential impact of improved early immunisation uptake on rates of vaccine preventable diseases. Any observed changes in vaccine-preventable diseases (for example, pertussis in young babies) would be heavily confounded by the changes in behaviour such as social distancing and isolation for symptoms such as cough. However, of the 5 vaccinations, there did appear to be an increase in overall coverage for the second MMR; therefore, monitoring trends in measles cases as we move fully out of COVID-related restrictions may be of interest, though we would not be able to infer causation.

The ongoing COVID-19 pandemic continues to stretch health services and adversely affect all areas of life, with children disproportionately bearing the burden of the indirect consequences of pandemic control measures such as the closure of schools and limited social contact. However, opportunities have also been created in terms of enhanced surveillance of health programmes. In this study, we have used such data to investigate the effect of the pandemic on infant and preschool immunisation uptake. We have demonstrated that a robust child immunisation service can continue to deliver high and even increasing uptake rates. Families will respond despite the many difficulties they face, to ensure that children continue to be protected again vaccine-preventable diseases. The challenge now is to use and expand on this knowledge to promote future vaccination programmes, including those targeting SARS-CoV-2.

Supporting information

Acknowledgments

We thank Public Health Scotland for making the data publicly available and acknowledge the support of the HDR UK BREATHE Hub and EAVE II collaborators.