Aim. To describe the trends in anti-infective use during pregnancy between 2010 and 2019 and determine whether they were prescribed according to drug fetal safety international classification systems. Methods. We conducted a population-based, nationwide study using the French national health data system including all pregnancies ended between 2010 and 2019. Anti-infective were considered according to their pharmacological group and potential harmful risk using the Australian and Swedish classification systems. Prevalence rate was estimated annually and by trimester. Average annual percent change (AAPC) and 95% confidence intervals (CI) were calculated using Joinpoint regression. Results. Among 7,571,035 pregnancies, 3,027,031 (40.0%) received ≥ 1 antibacterial. This proportion decreased significantly from 41.5% in 2010 to 36.1% in 2019 (AAPC=-1.7%, [95%CI, -2.5% to -1.0%]). Conversely, use of antiviral agents increased during the 10 years’ study period for anti-HSV agents (AAPC=4.4%, [3.7%-5.2%]), influenza agents (AAPC=25.4%, [6.2%-48.1%]), and for HIV-antiretroviral agents (HIV-ART) (AAPC=1.3%, [0.6%-2.0%]). Use of influenza vaccine increased from 0.2% in 2010 to 4.2% in 2019 (AAPC=49.7%, [95%CI, 39.3% to 60.9%]). Among all pregnancies, 0.9% had been exposed to a potentially harmful anti-infective agent increasing from 0.7% in 2010 to 1.2% in 2019 (AAPC=6.4%, [4.4%-8.5 %]). Conclusion. Based on more than 7 million pregnancies identified from French nationwide data, this study showed that antibacterials are frequently prescribed during pregnancy although their use has decreased over the past ten years. Our results suggest that anti-infective are generally prescribed in accordance with recommendations, with however a potential for improvement in influenza vaccination.

Background Rapid identification of patients with high suspicion of COVID-19 may become a challenge with the co-circulation of multiple respiratory viruses (RVs). We have identified clinical or biological characteristics to help distinguish SARS-CoV-2 from other RVs. Methods We used a prospective cohort including all consecutive patients admitted through the emergency department’s (ED) and presenting respiratory symptoms from November 2019 to April 2020. Patients were tested for RV using multiplex polymerase chain reaction (mPCR) and SARS-CoV-2 RT-PCR. Results 203/508 patients were positive for an RV during the non-SARS-CoV-2 epidemic period (November to February), and 268/596 patients were SARS-CoV-2 positive during the SARS-CoV-2 epidemic (March to April). Younger age, male gender, fever, absence of expectoration and absence of chronic lung disease were statistically associated with SARS-CoV-2 detection. Combining these variables allowed for the distinguishing of SARS-CoV-2 infections with 83, 65, 75 and 76% sensitivity, specificity, PPV and NPV, respectively. Conclusion Patients’ characteristics associated with a positive PCR are common between SARS-CoV-2 and other RVs, but a simple discrimination of strong SARS-CoV-2 suspicion with a limited set of clinical features seems possible. Such scoring could be useful but has to be prospectively evaluated and will not eliminate the need for rapid PCR assays.