Abstract
This study aimed to illustrate and account for immortal time bias in pregnancy observational investigations, using the relationship between late use of antibiotics and risk of preterm birth as an example. We conducted a population-based cohort study including 549,082 deliveries between 2007 and 2017 in Lombardy, Italy. We evaluated the risk of preterm births, low birth weight, small for gestational age, and low Apgar score associated with antibiotic dispensing during the third trimester of pregnancy. Cox proportional hazards models were used to estimate hazard ratios (HRs) and 95% confidence intervals (CI) of the outcomes, considering the use of antibiotics as time-fixed (with biased classification of exposure person-time) and time-varying (with proper classification of exposure person-time) exposure. There were 23,638 (4.3%) premature deliveries. There was no association between time-fixed exposure to antibiotics and preterm delivery (adjusted HR 0.96; 95% CI 0.92 to 1.01) but an increased risk of preterm birth when time-varying exposure to antibiotics was considered (1.27; 1.21 to 1.34). The same trend was found for low birth weight and low Apgar score. Immortal time bias is a common and sneaky trap in observational studies involving exposure in late pregnancy. This bias could be easily avoided with suitable design and analysis.
Highlights
Preterm delivery is estimated to affect 10.6% of all live births around the world, equating to about15 million births each year [1]
The increased risk may be mitigated by treatment with antibiotics, but confidence in this effect is limited given the low certainty of the evidence [10]
All Italian citizens have equal access to healthcare services as part of the National Health Service (NHS); in Lombardy, this is associated with an automated system of databases to collect a variety of information on residents who receive NHS assistance (NHS beneficiaries), diagnoses, procedures performed on inpatients in public or private hospitals, outpatient drugs dispensed in community pharmacies (coded according to the Anatomical Therapeutic Chemical (ATC) codes), and specialist visits and diagnostic examinations reimbursable by the NHS
Summary
Preterm delivery is estimated to affect 10.6% of all live births around the world, equating to about15 million births each year [1]. There is general consensus on the role of intra-amniotic infection as the main cause of preterm births [2,3,4,5,6,7], there is currently no evidence supporting antibiotic prophylaxis for reducing the risk of preterm delivery and other maternal and newborn adverse outcomes due to infection [8,9]. As antibiotics are prescribed mostly as therapy, their use can be thought of as a proxy for the onset of some infection; an increased risk in preterm births among women undergoing treatment with antibiotics is expected. The increased risk may be mitigated by treatment with antibiotics, but confidence in this effect is limited given the low certainty of the evidence [10]. Concern has been expressed that antibiotics may do more harm than good under these circumstances [11,12,13,14,15,16,17]
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
