Immunization Coverage of Children in Care of the Child Welfare System in High-Income Countries: A Systematic Review

The characteristics of included studies are presented in Appendix Table 1, (available online). The studies primarily came from the United Kingdom (n=15),^{,  ,,  ,,,  ,  ,  ,,  ,,  ,,}  the U.S. (n=9),^{,  ,,  ,,  ,,  ,} and Australia (n=6),^{,  ,,  ,,}  with a smaller number from Italy (n=1), and Sweden (n=2).^,  They included 29 peer-reviewed journal publications, three government reports, and one thesis. Study designs included cross-sectional (n=22),^{,  ,  ,  ,  ,  ,,  ,  ,  ,  ,,  ,  ,  ,  ,  ,  ,  ,,  ,} case-control (n=1), prospective and retrospective cohort (n=2),^,  retrospective cohort with pre-/post-intervention assessments (n=3),^{,  ,} mixed methods (n=3),^{,  ,} mixed methods with pre-/post-intervention (n=1), and qualitative (n=1). A variety of data sources were used to determine immunization status of children, including caregiver or self-report (n=5),^{,  ,,  ,} caregiver or self-held record (n=1), child welfare data (n=6),^{,  ,,  ,,}  community health records (n=2),^,  electronic health database (n=5),^{,  ,,  ,} medical records (n=4),^{,  ,,}  electronic immunization databases (n=1), and mixed data sources (n=7).^{,  ,,  ,,  ,} Very few studies (n=7)^{,  ,,  ,,  ,} reported on coverage for specific vaccines, whereas the remainder did not specify or reported more broadly on “age appropriate” or “recommended vaccines.”

Terminology, setting, and time in care varied by study. Included studies used various terms to describe immunization coverage, such as “up-to-date immunizations,” “fully immunized,” or “incomplete immunizations.” In some cases, authors did not define these terms, making it unclear whether they meant that the child had received all age-appropriate immunizations and whether these terms encompassed timeliness of immunization. Authors from different countries used different terminology for children in care (e.g., “looked-after children” is typically used in the United Kingdom, “out-of-home care” is more common in Australia, and “foster care” is common in the U.S. and the United Kingdom). It was not always clear from the article whether these terms captured comparable types of care status, especially as many studies also used specific terms (e.g., foster care and kinship care) to identify subsets of children in care. Some studies specified that they studied children in more than one type of care setting (e.g., foster care, residential care), whereas others did not describe the type of care setting beyond the broad terms of looked-after or out-of-home care. Four studies compared coverage between groups of children in different care settings within the child welfare system.^{,  ,,}  The majority of studies (n=15)^{,  ,,  ,,  ,,  ,,  ,  ,  ,,  ,} did not describe how long the children had been in care before the study period, whereas others included children who had been in care varying lengths of time (n=4).^{,  ,,}  Others studies looked at more defined periods of time in care, including in care ≥3 months (n=1), ≥6 months (n=2),^,  or ≥12 months (n=2).^,  Yet, others looked at children who had recently entered care (n=7),^{,  ,,  ,,  ,} with two of these studies comparing coverage upon entry into care to coverage after children were established in care.^,  Two of the studies assessing barriers/supports to immunization did not study any children in care, instead focusing on nurses or foster caregivers.

Fourteen of the included studies were of low quality (score <0.50),^{,  ,  ,  ,  ,  ,,  ,  ,  ,  ,  ,  ,} 12 were moderate quality (score ≥0.50 to <0.75),^{,  ,  ,  ,  ,,  ,  ,  ,  ,  ,} and seven were high quality (≥0.75).^{,  ,  ,  ,,  ,} The most common reasons for a study being assessed as low quality were lack of a comparison group or the study not being focused specifically on immunization (i.e., it focused more broadly on child health and thus lacked the necessary information specific to immunization, even after follow-up with authors). In order to ensure that recommendations for policy and practice are guided by the best evidence available, the findings of only moderate- and high-quality studies were synthesized and presented.

Appendix Table 2 (available online) presents the study findings from all articles, with moderate- and high-quality studies presented first. Of the 31 studies that measured immunization coverage, 18 were moderate or high quality.^{,  ,  ,  ,  ,  ,  ,  ,  ,  ,  ,  ,  ,  ,  ,,  ,}  Of these, ten had a comparison group^{,,  ,  ,  ,  ,  ,,  ,} of the general population or children not in care. Nine of these found children in the child welfare system to have lower coverage than other children^{,,  ,  ,  ,  ,  ,,}  and one found higher coverage. Of the other eight studies (including all the government reports^,  ) that did not have a comparison group of children in the general population or not in care, the majority (n=5) found ≥80% immunization coverage among children in the child welfare system^{,  ,,  ,}; the remaining three found <80% coverage.^{,  ,} Of these eight studies without a comparison group, three compared immunization coverage between groups of children in care by type of care setting.^{,  ,} One of these found no statistical difference in immunization coverage between those in foster care and parental care and another found no significant difference in immunization coverage between foster care and kinship care. The third study found that children living in the parental home were significantly less likely to be immunized than those in kinship care, foster homes, and residential care, though the difference in coverage was minimal. Studies that reported immunization coverage by age found that older children in care were less likely to have received the recommended immunizations for their age than younger children.^{,  ,,}