Objective To determine if different factors affect children having full, partial or no primary immunisations.
Methods This was a crossgenerational cohort study with linkage to primary care and hospital records conducted in urban and rural settings in Ireland, recruiting in 2001–2003 with 5-year follow-up. A total of 749 children with immunisation information took part.
Results The uptake of reported primary immunisations was 92.8% full, 4.9% partial and 2.3% no primary immunisations. Adjusted relative risk ratios for children receiving no primary immunisations were significant for: having a mother who had ever visited an alternative practitioner 3.69 (1.05 to 12.9), a mother with means tested full general medical services eligibility 8.11 (1.58 to 41.65), a mother who scored <50 for the World Health Organization Quality of Life (WHO-QOL) scale psychological domain 8.82 (1.79 to 43.6) or living in the west of Ireland (rural) 3.64 (1.0 to 13.2). Being born prematurely was associated with partial primary immunisation, adjusted OR 4.63 (1.24 to 17.3).
Conclusions Knowledge of these differences will help target campaigns to increase full uptake of primary immunisations.
Statistics from Altmetric.com
Childhood immunisations are an essential part of child health programmes and their success is reliant on universally high uptake. This ensures protection for those immunised and those too young to receive immunisations or who have contraindications.
Ireland had 69 000 live births in 2008 out of an estimated population of 4.4 million.1 It has a two-tier healthcare system whereby free primary healthcare is obtained via means tested General Medical Services (GMS) eligibility, covering approximately a third of the population including children. The rest of the population pays directly for General Practice (GP) services. Childhood immunisations are voluntary and free of charge to everybody.
WHO immunisation uptake figures for Ireland for three doses of diphtheria, tetanus and polio containing vaccine were 89% in 2004 and 93% in 2008, compared to the UK with 90% in 2004 and 92% in 2008 and USA 96% in 2004 and 95% in 2007.1 Immunisation uptake has slowly increased in Ireland since 2001. Most studies have grouped partially and unimmunised children together, but recent work has found that there may be differences between these groups. The UK's Millennium Cohort2 study found unimmunised children had a higher proportion of mothers who were educated to degree level and were older, in contrast to partially immunised children who lived in disadvantaged wards with larger families and lone or teenage parents. An American study found that under immunised children had younger, unmarried mothers without college education and unimmunised children had married mothers with a college degree and a higher income.3
What is already known on this topic
▶. Most studies on determinants of primary immunisation have grouped partially and unimmunised children together.
▶. More recently the Millennium cohort in Great Britain found unimmunised children had a higher proportion of mothers educated to degree level.
▶. In contrast, partially immunised children were more likely to live in disadvantaged wards with larger families.
What this study adds
▶. This study confirms that being unimmunised is associated with higher educated mothers and also finds an association with previous complementary and alternative medicine (CAM) usage.
▶. There is an association between infant prematurity and partial immunisation, not reported in a general population cohort before, which is counter to immunisation policy.
▶. Different demographic subgroups require more focused and targeted health education strategies to increase their immunisation uptake.
The Lifeways cohort study is one of few prospective datasets with information on demographics including GMS eligibility, lifestyle factors and linkage healthcare usage information for hospitals and GPs. This analysis considers factors affecting uptake of primary immunisations to allow more targeted immunisation campaigns.
The Lifeways cohort study design has been previously described.4 Pregnant women (n=1124) were recruited at their booking hospital appointment between end 2001 to beginning 2003 in Dublin and Galway (Ireland). Women not born in Ireland were excluded for statistical power reasons. Participants completed a questionnaire at recruitment and in 2007. Questions included age, working status, previous pregnancies, self-reported ever use of any complementary or alternative medicine (CAM), education and other sociodemographic information (see table 1) and quality of life tool World Health Organization Quality of Life (WHO-QOL) brief instrument (BREF), which was designed to assess quality of life in a standardised crosscultural manner and comprises 26 items related to health status, social relationships and environment.5 Details on the father and grandparents of the child were requested, including their education.
In 2007 the Health Service Executive (HSE) provided up to date immunisation records of children in the appropriate age cohort and resident in the study areas. Immunisations were recorded manually by general practitioners who forward the form to the HSE. The 2007 family questionnaire also recorded parental recall of immunisations. The information used on the social circumstances of the mother was that collected at recruitment, as this is closest to the time the cohort children would receive their primary immunisations.
Cohort children were eligible for the primary immunisation schedule 5-in-1 (diphtheria, tetanus, pertussis, polio and Haemophilus influenzae type b) and meningitis C given at 2, 4 and 6 months of age. To categorise immunisations: fully immunised had received all six components on three separate occasions, partially immunised were missing at least one of these components on one occasion, but had received at least one, otherwise they were categorised as unimmunised, by 2007 when the immunisation records were collected.
The records were then matched to cohort information using Microsoft Office Access 2003 and analysed to calculate the univariate odds ratios of being partially or unimmunised using SPSS V.15.0 (SPSS, Chicago, Illinois, USA). Multinomial logistic regression was carried out in Stata 9.2 (Stata, College Station, Texas, USA) to calculate the relative risk of being partially or unimmunised as compared with fully immunised.
Of the 1070 singleton babies in the original cohort, 749 (70%) had immunisation records matched. The uptake of reported primary immunisations was 92.8% (n=524) fully immunised, 4.9% (n=23) partially and 2.3% (n=16) unimmunised. Table 1 shows the characteristics of mothers who were matched to their children's immunisation records and those who were not. Those who could not be matched were more likely to be GMS eligible, lone parent, not working or not married. Agreement between HSE records and parental recall of immunisations in 2007 had 93% agreement for primary immunisations.
Partial immunisation was associated with prematurity, adjusting for other factors (table 2). On adjustment, mother being GMS eligible, scoring <50/100 on psychological domain of WHO-QOL BREF and using CAM are significant factors for no immunisations (table 2). No cut-off for WHO-QOL BREF score could be found in the literature, so a level of less than 50 was assigned as low. The final covariate adjusted model (see table 2) did not contain the WHO-QOL brief physical domain due to questionnaire design issues.
This cohort study shows that there are differences between the characteristics of those who are partially immunised compared to the unimmunised. Similar results were found in the Millennium and American studies2 3 implying these results are applicable in different healthcare systems, suggesting these are generalisable findings.
There is an association between children receiving no primary immunisations and mothers who said during pregnancy they had ever used CAM, which has not been noted before in a prospective context. Immunisation information could emphasise that the British Homeopathic Association themselves state that there is no evidence to show that homeopathic medicine can be used instead of immunisations,6 though this group of mothers may be hard to influence.
Receiving no primary immunisations is strongly associated with GMS eligibility. As immunisations are free, this should not be a barrier, but shows that regardless of free access, deprivation factors prevail. An American study7 also showed that uninsured children or those with Medicaid were less likely to be up to date with immunisations than those with private insurance.
There was a worrying association between prematurity and partial immunisations, a finding we are not aware has been seen before in cohort studies of the general population, rather than specialist cohorts of premature babies. The generalisability of this particular observation therefore merits further investigation. Premature infants are more vulnerable to infections for example, pertussis and should receive immunisations according to the normal schedule. Such infants are also more likely to come into contact with healthcare professionals. In this study they saw a GP an average of 15 times over the first 3 years, so they are having contact with healthcare but are not completing immunisations. It is also possible that the HSE records are incomplete due to lack of paper returns, or premature babies immunised in hospital not having those immunisation records forwarded to the HSE.
Galway is a city in the west with predominantly rural catchment population. Dublin, in the east, is urban and, like most major cities, has a more mobile population and more deprivation. Children from Dublin were less likely to be matched to immunisation records, probably because of greater population mobility. Unmatched people were more likely to be deprived (table 1). If more immunisation records could be matched, the deprivation effect may be still more marked.
In summary, there appear to be three different groups identified; children with no records of immunisations or immunisations not up to date who are the poorest and most disadvantaged, premature babies who are vulnerable to infections and children of mothers who use CAM. Each requires different, targeted strategies to increase their immunisation uptake.
The Lifeways Cohort Study Steering Group oversees the study.
Funding The Health Research Board of Ireland has supported all sweeps associated with this study. The work is independent of the funding sources.
Competing interest None.
Ethics approval Ethics approval was granted by the UCD ethics committee for 2007 sweep, and for original recruitment and follow-up stage by the Irish College of General Practitioners ethics committee and respectively, ethics committees of Coombe University Hospital Dublin and University College Hospital Galway.
Provenance and peer review Not commissioned; externally peer reviewed.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.