August 18, 2020

expert reaction to analysis of reports on unintended pregnancies with combined use of certain antibiotics and hormonal contraceptives

A study, published in BMJ Evidence Based Medicine*, has analysed reports of unintended pregnancies occurring when certain antibiotics are used with hormonal contraceptives.

Jane Bass, Royal Pharmaceutical Society spokesperson on women’s health and senior pharmacist, women’s services, at Guy’s and St Thomas’ Hospital London said:

“In their review of a large dataset from the UK, the researchers assume that all the women they reviewed were taking hormonal contraception at the time of their unplanned pregnancy although this information was not part of the data they collected or reported.

“Unfortunately the researchers were unable to pinpoint in which women antibiotic usage is likely to cause contraceptive failure and so they recommend that all women prescribed antibiotics should avoid sex or take alternative precautions for a period of time during and after the antibiotic course if they do not want to get pregnant. This general advice for all types of antibiotics was current for many years in the UK but is no longer standard practice, although some guidance for certain antibiotics remains.

“We cannot tell from the paper whether the number of unintended pregnancies has increased since the previous advice to avoid sex or take alternative precautions while taking antibiotics was changed as the numbers of pregnancies the researchers reported in women taking antibiotics was only 46 over approximately 50 years, with no trends given over time.

“If numbers have not gone up in recent years, this could mean that the change in advice has not had any impact on unintended pregnancy rates and therefore there could be very little impact on the current rate by changing the advice back. Although numbers are low, the consequences of an unintended pregnancy for the individual woman are significant.

“Women using hormonal contraceptives (whether oral, patches or vaginal rings) should be aware that “real life” usage means that more women will get pregnant each year than would be expected if the contraceptive was used “perfectly”. The study was unable to take into account which type of contraception women were using when the unintended pregnancy was reported, or even whether they were using hormonal contraception at all as this was only assumed from the data they collected.

“Women who suffer from vomiting or diarrhoea for more than 24 hours while taking a hormonal contraceptive, whether caused by antibiotics, other medicines or an illness, should be aware that the contraceptive may not be as effective as normally and they should follow “missed pills” advice.”

Prof Stephen Evans, Professor of Pharmacoepidemiology at the London School of Hygiene & Tropical Medicine, said: 

“This study is based on spontaneous reports of suspected adverse reactions to drugs. These reports can almost never be taken as anything other than suggestions for causal effects, at best. As the authors point out, the information on antibiotics (and other drugs known to affect the same metabolic pathways) have been noted before.

“Every packet of these relevant contraceptive pills has a patient leaflet in the UK and they contain warnings about possible reduced effectiveness when also taking some other drugs, including antibiotics. It seems likely that the patient leaflets are not read, and it is possible that some prescribers do not mention the problem, or that patients forget they were told.

“The bias that the authors note in their limitations that relates to the possibility that the prescribing information contains warnings about the drugs and hence that the reporting of an unintended pregnancy takes place with antibiotics at a higher rate because of this information is not dealt with by the finding of lower rates with the “control” drugs. These “control” drugs are not warned against in the same way as with antibiotics.

“Nevertheless, there are good reasons why antibiotics could reduce efficacy of some hormonal contraceptives, and the reminder may be helpful. The problem is that these data neither provide good evidence that the disproportionate reporting is a causal effect, nor do they indicate the magnitude of the potential problem.

“With 46 unintended pregnancies reported over 55 years, even with under-reporting, it may not be a serious public-health problem. It would be good to have a proper observational study with a very much larger number of unintended pregnancies to elucidate the issue further. Overall this link is not novel and, as advice is already available in clinical practice, should be reported on with care as creating scares about oral contraceptives never has good outcomes.”

* ‘Analysis of reports of unintended pregnancies associated with the combined use of non-enzyme- inducing antibiotics and hormonal contraceptives’ by Jeffrey K Aronson and  Robin E Ferner was published in BMJ Evidence Based Medicine on Tuesday 18 August.

DOI: doi:10.1136/bmjebm-2020-111363

Declared interests

Prof Stephen Evans: “No conflicts of interest in relation to this.  I am funded (1 day/week) by LSHTM.  They get funding from various companies, including Astra Zeneca and GSK but I am not funded by them, I have no involvement in obtaining funding from them and I am not an investigator for any grants obtained from them.”

None others received.