Customize your JAMA Network experience by selecting one or more topics from the list below.
In 2009, routine prostate cancer screening using a prostate-specific antigen (PSA) test was described as “the controversy that refuses to die.”1(p1351) Unfortunately, 10 years later, the controversy is still alive and thriving.2 Clinical trials have failed to resolve whether or to what degree screening using PSA tests help reduce prostate cancer–specific mortality, and it has long been clear that PSA screening tests increase the diagnosis of low-risk cancers and can lead to patient harm from potentially unnecessary biopsies and cancer treatment. As a result, many expert groups recommend shared decision-making (SDM) and informed patient choice for routine prostate cancer screening.3,4
Patient decision aids (DAs) have been proposed as a crucial tool for supporting SDM. In a new meta-analysis reported in this issue of JAMA Internal Medicine, Riikonen et al5 report the outcomes of DAs for prostate cancer screening relative to usual care, assessing their association with patient knowledge, decisional conflict, screening discussions, decision satisfaction, and screening decisions. The results are disappointing; compared with usual care, these DAs increased patient knowledge to some extent and decreased decisional conflict but had no effect on screening discussions, decision satisfaction, or receiving screening. The authors conclude that “the available evidence does not provide a compelling rationale for clinicians to use existing decision aids to facilitate shared decision making in their discussions with men considering undergoing prostate cancer screening.”
Although we agree that these data do not provide a persuasive case for use of the DAs included in the meta-analysis, it would be premature to conclude on the basis of these data that DAs do not and could not affect prostate cancer screening decisions. One issue that this study5 highlights is that randomized clinical trials of prostate cancer screening DAs vs usual care are heterogeneous in DA content, design, delivery, and outcome measures. For example, the only information that was common to all the DAs was the purpose of prostate cancer screening. There was considerable variation in DA presentation and measurement of patient knowledge, from a 1-page flyer received by mail, with patients’ knowledge assessed up to 3 weeks later, to an in-clinic DA intervention that included a 12-minute video and an 8-minute coaching session, with knowledge assessed immediately after these interventions.5 Furthermore, some DAs recommended that the patient talk with their doctor, while others did not, and still others actively coached patients on how to address barriers to communication just prior to their appointment. There was also heterogeneity in communication method (eg, booklets, leaflets, computer tools, video), how the DAs were implemented (eg, in clinic, by mail), and when and how the primary outcomes were assessed. Outcome measurements also varied among the studies, and the most robust outcome observed was a reduction in decisional conflict, the only outcome for which there was a standard, validated measure. Each of these factors, if designed or implemented suboptimally, could affect the observed outcome of a DA. The present meta-analysis was not able to test whether these factors (eg, DA presentation, timing of outcome measures, standardization and validation of the measures) explain a significant amount of the heterogeneity in DA effects, but such information would be instrumental to the design and implementation of future DA trials.
Consistent with prior studies, Riikonen et al5 found that DAs had no association with whether patients received prostate cancer screening. Almost 20 years ago, Volk and Spann6 noted that while DAs tended to reduce intentions to receive screening (an apparently robust effect7), they frequently had no effect on the number of men who receive screening. There may be many reasons for this, exclusive of the effectiveness of DAs. Ideally, informed patient preferences should align with the care that patients receive, but this ideal frequently does not occur in practice.7 Physician recommendations can be highly persuasive, potentially overriding patients’ informed preferences.8 Moreover, men do not make health decisions in a social vacuum and may decide to receive screening (or not) because of the preferences of a spouse or loved one.9 Recent research has also highlighted that men do not have uniform reactions to prostate cancer screening information; for example, men who are so-called medical maximizers generally prefer an active approach to health care and want to receive PSA screening after being informed of the harms. By contrast, men who are medical minimizers are mostly content to forego PSA screening after learning about the harms.10 Hence, DAs may have a large effect on some populations of patients but very little effect on other populations. If a given individual does not change his decision as a result of a DA, that does not indicate that the DA has failed. The bottom line is that there are many factors that may determine whether men receive screening, including their personal preferences. Some of these factors may be adequately addressed by DAs that are designed or implemented more effectively. Other factors may not be adequately addressed with DAs alone and need to be addressed with both better SDM interventions and changes to clinical practice.
Decision aids face both design and implementation challenges, and unfortunately the study by Riikonen et al5 does not provide much insight into how to design or implement DAs more effectively, which is a large gap in the existing literature. To know the true potential of DAs, we need to systematically test the influence of different types of DA design, implementation strategies, and patients’ individual attributes (such as medical maximizing or minimizing) on SDM outcomes. It seems likely that DAs are helpful for communicating evidence to patients but are often not sufficient for fostering SDM. In designing DAs, we should ask whether we are giving patients the most important information and designing the DA to have a maximal influence on desired outcomes, such as communication or knowledge. In implementing DAs, we must ask whether we are giving the DA to the right people, in the right setting, and at the right moment in the decision-making process. When assessing DAs, we need to ask whether we are measuring key outcomes at appropriate moments using validated and reliable measures. Without such information, the true usefulness of DAs to sustainably facilitate SDM and serve patients’ and clinicians’ goals and needs remains unknown.
Corresponding Author: Laura D. Scherer, PhD, Department of Medicine, Division of Cardiology, University of Colorado, Denver, 13199 E. Montview Blvd, Ste 300, Mailstop F443, Aurora, CO 80045 (email@example.com).
Published Online: June 24, 2019. doi:10.1001/jamainternmed.2019.0753
Identify all potential conflicts of interest that might be relevant to your comment.
Conflicts of interest comprise financial interests, activities, and relationships within the past 3 years including but not limited to employment, affiliation, grants or funding, consultancies, honoraria or payment, speaker's bureaus, stock ownership or options, expert testimony, royalties, donation of medical equipment, or patents planned, pending, or issued.
Err on the side of full disclosure.
If you have no conflicts of interest, check "No potential conflicts of interest" in the box below. The information will be posted with your response.
Not all submitted comments are published. Please see our commenting policy for details.
Scherer LD, Lin G. Decision Aids for Prostate Cancer Screening—The True Potential Remains Unknown. JAMA Intern Med. Published online June 24, 2019. doi:10.1001/jamainternmed.2019.0753
Create a personal account or sign in to: