Evaluating health visitors' assessment of caries risk for Childsmile Practice

Introduction: Childsmile is the Scottish National Oral Health Programme established in 2005 to address the high levels of dental decay in young children, which is strongly related to deprivation, and causes 28% of elective hospital admissions of Scottish children. Childsmile Practice aims to link families of infants at increased caries risk to practice-based support. 

Methods: In 2006 Health Visitors began assessing infants in order to refer those at increased caries risk to Childsmile Practice.  Caries Risk Assessment (CRA) records received in each quarter of 2007 and 2008 are used in evaluate CRA outcomes, particularly the extent to which at-risk infants were successfully referred. Logistic regression analysis on one quarter's assessments is employed to investigate predictors of referral.

Results:  The number of recorded CRAs remained relatively stable, increasing from 2,313 in the first quarter of 2007 to 2,523 in the last quarter of 2008.  However, referrals to Childsmile rose from 45.6% to 59.3% of all assessments over the same period (fig 1). The proportion of CRAs which did not identify risk fell from 39.2% to 23.4%, while the proportion where parental consent for referral was not given remained relatively stable. Referrals in the first quarter covered represented 31.3% of estimated births occurring in the relevant areas (3,360 per quarter), rising to 44.6% in the last quarter covered.  By that time an estimated 75.6% of new-born babies received a CRA from their Health Visitor.

The logistic regression analysis of first quarter 2008 assessments identified the following predictors of referral: ethnic minority family, smoker household, parent's last dental visit was for pain relief, living in an areas of high deprivation, child aged four weeks plus, first born child, CHP area  (n=1772).

Conclusions:  Referrals increased both numerically and as a proportion of completed assessments, and children from at-risk groups were successfully targeted.