Too much side effect information 'reduces medication uptake'

Evidence suggests that long lists of side effects on patient information lists do not improve informed consent but instead decrease consumer willingness to take medication.

Informed consent is central to Western medicine, and sharing medication side-effect information with patients is viewed as core to patient autonomy and shared decision making (here). Information about medication side effects can also help prepare patients to manage side effects.

However, these warnings about medication side effects can worsen medication outcomes via the nocebo effect. The nocebo effect is considered the opposite to the placebo effect, where patients experience unpleasant effects through negative expectations (here). Side-effect information can also contribute to non-adherence (here and here), which is one of the biggest barriers to medication effectiveness worldwide (here).

Patients often hear about prescription medication side effects from their doctor, and commonly report (here) that they want their doctor to tell them about all the possible medication side effects. This desire to be told about side effects is echoed in some medical literature, suggesting that the doctor should inform their patient about any potential side effects (here), and government public information campaigns (here) recommend that patients always read their information leaflets before taking medications.

The role of patient information leaflets

Patient information leaflets (PILs), commonly known as consumer medicine information leaflets in Australia, are another main source of medication information for consumers beyond their doctor. PILs are available at pharmacies and in some medication packaging (here), online through the Therapeutic Goods Administration (TGA; here), and are compulsory to share with patients in clinical trials (here). PILs must contain side-effect information, although research has found (here) that this information varies greatly between PILs and is not user-friendly, with poor readability.

Overseas research shows PILs offer extensive side-effect information, which precedes brief information on medication benefits, if benefits are mentioned at all (here). Side effects listed in PILs are not necessarily specific to the medication in question. Indeed, in Australia anyone can report an “adverse event” (ie, undesirable symptom after taking a medication), and these are entered into the TGA Adverse Event Management System (AEMS; here). Health practitioners may refer to AEMS adverse events information in discussions with patients, and these reports can also end up in PILs. This process does not rely on evidence that the undesirable symptom was linked to the medication, and means that non-specific symptoms, such as nausea and fatigue, are commonly included in PILs (here). Consequently, PILs often list a vast number of side effects, with one study (here) showing that an average of 26 side effects across PILs for a single medication was the lowest average number of side effects of 15 commonly prescribed drugs.

Although PILs often go unread for common medications (here), readership is as high as 91% (here) for new and prescription medications (here), so this information can have a significant impact on consumer beliefs about the medication they are taking. Historically, providing consumers with more information is considered beneficial, although research suggests that this is not the case, with one overseas study showing (here) as many as 62% of patients discontinued their medication after reading patient information leaflets.

Information about potential side effects can cause overestimation of the risk of experiencing them, decrease treatment adherence, and even increase side-effect reporting (here). This experience of side effects that are not due to the active ingredients of a medication is known as the nocebo effect (here). Patient information leaflets can also cause fear and worry (here), which may also contribute to a nocebo effect and lack of adherence.

So, side-effect information is important, but it is a delicate balance between giving patients information for informed consent while not worsening outcomes.

What we studied

A study by my research team (here) investigated how consumers respond to being informed about different numbers of side effects. Participants were recruited to take part in a study about a fictitious medication, being either informed about a small number of side effects (one or four); a large number of side effects (26); or had side effect information completely omitted. We then asked participants about their memory of side effect information and their willingness to be involved in future studies in which they may take the (fictitious) medication in question.

Participants were more willing to take the medication when they were told about a small number of side effects, compared with both a large number of side effects and no side effect information at all. We also found that more participants accurately remembered most side effects when they were told a small (rather than large) number of side effects. Interestingly, participants who were not told about any side effects claimed to remember being informed of five side effects on average. Some non-specific side effects, including drowsiness, nausea and fatigue, were more likely to be recalled than others regardless of if they were actually disclosed as side effects to participants.

Our results suggest that there may well be a sweet spot of side effect information in which we can optimise both informed consent and consumer willingness to take medication. The results suggest that medication consumers are wary of side effects even when they are not given any information, but that longer lists (that include side effects lacking evidence of their relation to the specific medication) mean that patients may not remember those of significance (that are evidence-based) as these cannot be distinguished from non-specific symptoms. Therefore, providing medication consumers with a list of evidence-based side effects, resulting in a much briefer list of side effects than is currently provided (here), may be optimal for both their awareness of evidence-based side effects (therefore informed consent) as well as their willingness to take medication.

Conclusion

Overall, the evidence suggests that long lists of side effects do not improve informed consent but instead decrease consumer willingness to take medication. Therefore, side-effect information offered in PILs should be changed to reflect this. Instead of recommending patients seek as much side-effect information as possible (here), we suggest informed consent should be contextualised (here) so that patients are informed about only side effects that are evidence-based and relevant to them specifically, to make side-effect information more useful.

Our research found that particular caution should be taken with disclosing non-specific symptoms that are most easily recalled by medication consumers, such as drowsiness and nausea. Written information requirements in PILs could also be modified such that only side effects shown to be caused by the medication are included, beyond that experienced by a placebo control group.

After all, patients deserve evidence-based side-effect information that allows an accurate risk perception of side effects. Contextualised informed consent may provide a goldilocks zone between informed consent and negative medication expectations.

Jessica Barker is a third year PhD candidate at the University of Sydney, a Certified ESDM Therapist and a Psychologist.

Kate Faase is an Associate Professor at the University of New South Wales Sydney where her main research focus is the nocebo effect.

The statements or opinions expressed in this article reflect the views of the authors and do not necessarily represent the official policy of the AMA, the MJA or InSight+ unless so stated. 

Subscribe to the free InSight+ weekly newsletter here. It is available to all readers, not just registered medical practitioners. 

If you would like to submit an article for consideration, send a Word version to mjainsight-editor@ampco.com.au.