Trends in prenatal alcohol use from 2001 to 2017 in the Northern Territory, Australia: a joinpoint regression analysis.

backgroundPrenatal alcohol use (PAU) is a risk factor for many adverse maternal, fetal, and child health outcomes. In Australia, a range of interventions have been implemented in recent decades to reduce the harms associated with alcohol consumption during pregnancy. While there is evidence of a national decline in the prevalence of PAU, there is limited information on changes in the Northern Territory (NT). This study aims to estimate the trends in PAU among NT Aboriginal and non-Aboriginal women.

methodsWe used de-identified, individual-level linked records from perinatal care, hospital admissions and emergency department (ED) presentations to estimate trends in PAU for all births to NT women between 2001 and 2017. We classified PAU into “No PAU” (a record of no alcohol consumption in pregnancy) and “any PAU” (a record of alcohol consumption in pregnancy). Sub-categories of any PAU included “Early PAU” (a record of alcohol use in early pregnancy only), “Continued PAU” (a record of alcohol use in early and late pregnancy), “Extreme PAU” (a record of hospital admission or ED presentation with an alcohol-related diagnosis during pregnancy), and “Unknown PAU” (insufficient information). We used a joinpoint regression model to estimate the Annual Percentage Change (APC) with 95% confidence intervals (CI).

resultThere were 63,081 births to 39,075 NT mothers (11,723 Aboriginal and 27,352 non-Aboriginal). For all births, from 2001 to 2017, there was a 5% annual decrease in PAU prevalence (APC: -5.2; 95% CI: -6.3, -4.3), with variations in trend by Aboriginal status and PAU category. For births to Aboriginal women, there was no evidence of overall change in annual PAU prevalence (APC: -0.6; 95%CI: -1.8, 0.6), no change in early PAU (APC: 1.1; 95%CI: -0.5, 2.8), a decrease in continued PAU (APC: -4.5; 95%CI: -6.4, -2.7), and an increase in extreme PAU (APC:10.7; 95%CI: 7.7, 15.7). For births to non-Aboriginal women, there was a decline in any PAU prevalence (APC: -10.7; 95%CI: -13.1, -9.1), early PAU (APC: -8.9; 95%CI: -11.4, -7.1) and continued PAU (APC: -13.9; 95%CI: -17.2, -11.84). On average, 17.9% of births had unknown PAU values (23.1% births to Aboriginal and 14.9% births to non-Aboriginal women), with a trend toward improved capture of PAU for births to Aboriginal women only (APC: -1.8; 95%CI: -2.9, -0.9).

conclusionThis study highlights the prevalence of PAU in the NT is decreasing disproportionately among Aboriginal and non-Aboriginal women. The general decline in PAU among NT non-Aboriginal women is consistent with national trends. However, the mixed results for Aboriginal women, and particularly the marked increase in the extreme PAU category, emphasise the need for a more targeted approach to reduce PAU in this population. Additionally, improvements in recording alcohol use at antenatal care visits can enhance data quality and support more effective interventions.