Abstract
BACKGROUNDS: In 2009, the prevention service "Familieambulatoriet" (FA) was established in three pilot hospital areas offering psychosocial support and health monitoring to parents in high risk regarding mental health and substance use, for the purpose of preventing child mental health and developmental problems through preschool years. This study selected new-born health as a preliminary endpoint for evaluation of population effects in three pilot areas, utilizing national statistics for birth cohorts from 2005 to 2013. The aim of the study is to evaluate changes in population new born health incidences associated with the establishment of new supportive and preventive FA-services at three pilot sites from 2009 in contrast to previous years and the remaining country. This quasi-experimental design evaluated changes in populations with new services available not those receiving the services, and controlled for national historical changes, variation between hospital districts, and random variation across the years before or years after the pilot services were introduced. Our hypothesis was to expect reduced frequencies of preterm births, SGA births, low APGAR scores, pediatric transfer, and new born abstinence symptoms in the pilot areas. METHODS: The baseline was established through 4 years preceding 2009, contrasting changes at pilot sites the following 4 years 2009-2013 using the remaining hospital area populations in Norway 2005-2008 and 2009-2013 as contrasts. RESULTS: Related to the introduction of FA services, we found three significant improvements in new born health using mixed effects logistic regression. 1) In the population rate of babies born prematurely with small for gestational age (SGA), using the 10th percentile criteria as the definition; odds ratio (OR) = 0.73 (95% Cl: 0.60 to 0.88). 2) A similar reduction using the 2.5th percentile criteria, although with wider confidence limits; OR = 0.73 (95% Cl: 0.54 to 0.99). 3) A decrease in the frequency of low APGAR scores (0-6) 5 min. after birth; OR = 0.80 (95% Cl: 0.68 to 0.95). Thus, the FA-areas remained significantly lowered on SGA rates or Low APGAR rates across the years after FA establishment, despite considerable variation, in contrast to the baseline years and to the remaining country. No significant effect was found for the outcomes frequency of premature births (unrelated to SGA), SGA among full-term babies, child abstinence symptoms or pediatric transfer of the baby. False negative findings may result from low-rate outcomes or studying the population rather than users. CONCLUSIONS: Population rates suggest that introducing FA services offering support and monitoring in high-risk families may contribute to improving aspects of new born infant health. Intervention components and strategies should be studied more closely using individual data.