Background

An important part of CDER’s work is communicating the most up-to-date safety information for medicines to help ensure that health care providers (HCPs) and their patients can make the most informed treatment decisions. One way that CDER does this is by including a boxed warning (BW) in labeling for certain prescription medicines to highlight important information for HCPs about serious risks associated with the medicine. BWs are ordinarily required in a medicine’s label when there are serious risks that can be mitigated by appropriate use of the medicine or in cases where there are serious risks that should be considered in assessing the benefits and risks of the medicine for the individual patient. While BWs are an important risk management tool, little is known about their intended – and potentially unintended—impacts on HCP prescribing or patients’ use of medicines that have a BW.

The Study

CDER and outside researchers conducted a qualitative study to explore how HCPs consider a BW when they are making treatment decisions for their patients, and how HCPs communicate with patients about BW information.¹

In order to explore potential contextual differences, CDER researchers selected two different treatment scenarios: (1) an oral antiviral medicine to treat hepatitis C virus (HCV), which can be a serious, life-threatening condition; and (2) an estrogen vaginal insert to treat vulvovaginal atrophy (VVA), which is a symptomatic, but less serious condition in post-menopausal females. The labeling for each medicine contained a BW, but the BW for these medicines varied in text length and number of identified risks. The BW in the HCV scenario was shorter and addressed only one risk, while the BW in the VVA scenario was longer and addressed multiple risks.

The research team conducted 52 in-depth interviews with HCPs who treated patients in one of the two different treatment scenarios (n=26 in the HCV scenario and n=26 in the VVA scenario). The participants included a mix of primary care providers and medical specialists. The interviews lasted up to 60 minutes each. Participants were first asked questions about what they think about when making treatment decisions for the scenario and their perspectives on benefits and risks of the specific medicine. They were then asked to read through the medicine’s BW and asked follow-up questions to explore whether the BW raised new thoughts or questions regarding the medicine’s benefits or risks. Finally, participants were asked their general thoughts about BWs in medicine labeling and how they discuss medicine risk information with patients.

Interview Findings

Perceptions of the Boxed Warning Across Treatment Scenarios

In general, perceptions of BWs among HCPs varied, with some HCPs finding them more useful than others. For instance, HCPs who had more experience treating patients with the conditions were well-aware of the risk information presented in the labeling and found the BW to be less useful.

Perceptions of the BW were also impacted by the treatment scenario. For example, several HCPs (particularly specialists) in the VVA scenario expressed their belief that the BW generally overstates the risks of the specific medicine. In contrast, HCPs in the HCV scenario expressed a more favorable view, describing the information in the BW as appropriate and important.

Factors That Influence Prescriber Decision-Making

HCPs in both treatment scenarios described BWs (in general) as only one of several factors that influence their treatment decision-making and described prioritizing individual patient factors when making prescribing decisions. Insurance coverage of the medicine was the most referenced factor in both scenarios. Other factors included a patient’s socioeconomic status, health status (e.g., patient’s personal or family history of certain cancers or other coexisting diseases or conditions), how likely the medicine is to improve the condition, and patient behaviors that could impact their medicine adherence.

Communication with Patients and Actions to Reduce Risk

HCPs across both treatment scenarios described discussing risk information and individual patient’s risk factors (e.g., personal and family health history) with their patients. In the VVA scenario, HCPs discussed explaining to their patients that any estrogen has risks and that this VVA treatment, which is inserted directly into the vagina, contains a lower dose of estrogen than estrogen medicines that are taken orally, which they believed could potentially confer lower risk for the side effects presented in the BW. HCPs also mentioned instructing patients to contact them if they begin experiencing any unusual side effects or if their symptoms are not improving.

In the case of the HCV scenario, HCPs emphasized the importance of highlighting to patients the medicine’s high effectiveness and short treatment duration, and they framed the side effects as minimal and manageable. Some HCPs discussed telling patients the potential serious dangers of failing to adhere to treatment, such as developing cirrhosis or liver cancer. HCPs who specialized in this treatment area emphasized that they create a plan to check in with patients to monitor adherence and assess medicine risk.

Discussion

Interviews with HCPs in this study revealed that a BW in labeling for a prescription medicine is only one of several factors that can influence providers’ treatment decisions, and that the potential influence of the BW depends on the context. Factors such as the seriousness of the condition and patient’s ability to access and adhere to the medicine can affect how the BW will influence decision-making. Across both scenarios, HCPs also described considering the benefits and risks for each patient in the context of the BW information and how they would communicate risk information to patients. They also considered how patients may interpret the BW information.

Perceptions of and reactions to BWs may vary across treatment contexts. For instance, some participants involved in the VVA scenario had a more negative reaction to the BW and offered suggestions on specific changes (e.g., to add disclaimer language regarding dosing considerations). In contrast, in the HCV scenario the BW was viewed more positively, and most participants did not have changes to suggest.

Most HCPs in each treatment scenario described being familiar with the medicine’s BW and that they were generally comfortable prescribing the medicine in appropriate situations. They also felt generally prepared to address patients' concerns about risks. A limitation of the study is that findings about HCPs’ familiarity and comfort discussing risk may not be generalizable to different treatment scenarios, particularly in situations where a risk issue is newly identified or highly uncertain.

Conclusions

BWs in labeling for prescription medicines bring HCPs' attention to serious medicine risks, as intended, yet this information is only one of several factors that influence HCP’s treatment decisions. The potential role that BW information plays in informing treatment decision-making often depends on the context (e.g., condition severity). Thus, across both treatment scenarios, HCPs in the current study expressed important differences in their perceptions of the risk information provided in the BWs. This research provides novel insights into how BWs fit within the context of HCP treatment decision-making and their communication with patients. These findings also underscore the nuanced complexity and context-specific nature of individual patient benefit–risk decisions.

¹ Ingersoll, RN, Bui, ET, Coleman, B, Zhou, EH, Eggers, S, 2024, Prescriber Perceptions of Boxed Warnings: A Qualitative Study, Pharmacoepidemiol Drug Saf, 33(3):e5766. doi:10.1002/pds.5766