Results Summary
PCORI funded the Pilot Projects to explore how to conduct and use patient-centered outcomes research in ways that can better serve patients and the healthcare community. Learn more.
Background
Decision aids help people choose between two or more health care options based on what is most important to them. Research has shown that decision aids improve patients’ knowledge about their choices but don’t always lead to patients’ making different decisions based on their values.
When it comes to cancer screening, many people believe that more screening is better. Many people don’t know that sometimes screening can have little benefit and can even cause harm (such as unneeded surgery or worry). There has been little research on the use of decision aids to help patients and doctors understand the benefits and harms of cancer screening and make a decision together.
The US Preventive Services Task Force (USPSTF) is an independent panel of experts that reviews the medical evidence about tests and medicines. The USPSTF makes recommendations that patients and doctors can use when making decisions about cancer screenings.
Project Purpose
The research team wanted to understand what sources and types of information about medical evidence patients trust and find most useful. The team wanted to create decision aids that would help people understand the USPSTF recommendations for prostate and breast cancer screenings. The USPSTF recommendations say
- Men do not need a blood test to see if they have prostate cancer.
- Women who are between the ages of 40 and 49 may not need to get a mammogram, a type of x-ray that can reveal signs of breast cancer, to see if they have breast cancer.
The researchers wanted to find out whether the new video decision aid, compared to traditional printed decision aids, made any difference in what people said they would choose to do.
Methods
The study had three parts. The study included men ages 50–74 years who did not have prostate cancer and women ages 40–49 who did not have breast cancer. All participants spoke either English or Spanish. All participants were patients at either an academic or community health center.
In part 1, the research team talked to groups of people to learn how they get health information. The team presented a variety of information about prostate and breast cancer screening and how the USPSTF decides what to recommend. Group discussions included 27 men (19 English speakers and 8 Spanish speakers) and 28 women (18 English speakers and 10 Spanish speakers).
In part 2, the research team presented two or three different videos for discussion by other groups of people. Part 2 had 19 men (12 English speakers and 7 Spanish speakers) and 23 women (13 English speakers and 10 Spanish speakers).
The researchers then developed a video decision aid in English and Spanish based on what they learned in parts 1 and 2. The decision aid included the information that the study participants said helped them understand how and why the USPSTF makes its recommendations about these two cancer screenings.
Part 3 had 27 men (26 English speakers and 1 Spanish speaker) and 35 women (26 English speakers and 9 Spanish speakers). The research team gave participants a survey (called the Decisional Conflict Scale) asking whether they thought they would get cancer screening, what they thought the benefits and harms of screening might be, whether they would discuss screening with their doctors, and how they feel about making decisions when they feel uncertain about what to do. After completing the survey, half of the participants had individual discussions about screening with their doctors and watched the video decision aid. The remaining participants looked at traditional, printed decision aids prepared by health organizations. The researchers then asked participants the survey questions again.
The researchers next gave all of the participants access to the decision aid they hadn’t used. People who had talked to their doctor and watched the video got the paper decision aid, and those who started with the paper decision aid watched the video and talked to their doctor.
The research team then looked at how much the different decision aids changed participant’s thoughts and attitudes about screening for prostate and breast cancer.
Findings
In part 1, the research team found two factors that strongly influenced how both men and women thought about cancer screening: recommendations from their doctor, and powerful stories about the cancer experiences of family and friends. For women, seeing mammography and breast health as part of being a responsible woman, mother, or wife also affected how they thought about screening.
In part 2, most study participants were not familiar with any recommendations about cancer screening, and they did not recognize the name of the USPSTF. Giving people information about how health organizations determine recommendations didn’t seem to make a difference in how participants thought about cancer screening itself.
In part 3, people did not change their survey answers after seeing the printed information, but there were large changes after seeing the video decision aid. Participants who saw the video had less desire to get cancer screenings that are not recommended and more awareness of the potential harms of screening. The participants also felt less desire to talk to their doctors about getting screened. Participants in both groups felt better about their ability to make a screening decision after participating in the study. Finally, the order in which the participants viewed the decision aids did not have an effect on the study results.
Limitations
Across all parts of the study, participants were from one location; results might be different for people in other parts of the country. A small number of people tested the decision aids in part 3. The video format is different than the printed information. Results might be different if the researchers had included more people in part 3 or if all the decision aids were given in video format. Finally, the study only measured changes in what people thought they would do about screening. The study did not track whether the decision aid made any difference in whether people actually got screened for breast or prostate cancer.
Conclusions
The findings suggest that people want good and complete information about how well the screenings work and how much harm people might experience if recommendations are followed.
Many people who saw the video decision aid changed their minds to follow the USPSTF recommendations about cancer screening. Similar decision aids may be useful to help people understand other USPSTF recommendations and to consider them when making decisions. Decision aids specifically designed to help people understand evidence-based recommendations may reduce the number of people who get cancer screenings that offer little benefit to them.
Sharing the Results
The research team presented its work at several meetings and is publishing a research paper.
Professional Abstract
PCORI funded the Pilot Projects to explore how to conduct and use patient-centered outcomes research in ways that can better serve patients and the healthcare community. Learn more.
Project Purpose
Decision aids typically yield modest increases in knowledge and little or no change in decisions. New approaches are needed to reduce use of low-value services and help patients make more evidence-informed decisions. The researchers hypothesized that an approach seeking to persuade people to trust and accept counterintuitive, evidence-based recommendations from the US Preventive Services Task Force (USPSTF) about prostate cancer screening for men ages 50-74 and mammography for women in their 40s could be developed and would be more effective in helping people consider these recommendations than traditional, information-based decision aids.
This study was designed to (1) explore consumer experiences and attitudes about credibility and influence of information from a variety of sources, including healthcare providers, evidence-focused organizations, and advocacy groups; (2) develop interventions designed to persuade consumers to trust USPSTF’s evidence-based recommendations against prostate-specific antigen testing for men ages 50-74 and for women ages 40–49 years to make an individual decision about mammography; and (3) assess changes in behavioral intentions among a sample of consumers receiving the interventions as compared with those receiving a more traditional, evidence-focused decision aid.
Study Design
Phase 1: qualitative study with focus groups; Phase 2: formative evaluation of several prototype decision aids using focus groups, with qualitative and quantitative analyses; Phase 3: within-subject crossover trial of persuasive interventions versus more traditional, written interventions.
Participants, Interventions, Settings, and Outcomes
For all phases, eligible participants were men ages 50–74 years without a diagnosis of prostate cancer and women ages 40–49 without a diagnosis of breast cancer who spoke either English or Spanish. Phases 1 and 2: Mailings were sent to potentially eligible persons receiving primary care at an academic health center and a community health center; some participants in the Phase 2 Spanish-language mammography focus group came from outreach to community contacts. Phase 3: Broad-based outreach including: mailings as in Phases 1 and 2; signs posted around the academic medical center; outreach to community contacts; and an ad on Craigslist. Phase 1 had 27 male (19 English-speaking (ES)/eight Spanish-speaking (SS)) and 28 female (18ES/10SS) participants. Phase 2 had 19 male (12ES/7SS) and 23 female (13ES/10SS) participants. Phase 3 had 27 male (26ES/1SS) and 35 female (26ES/9SS) participants. Sessions took place in the offices of community-based organizations.
In Phase 1, researchers presented a variety of information about prostate and breast cancer screening, guidelines, and guideline development processes to focus group participants. Researchers altered the focus group presentations, particularly after the first two, to emphasize information that appeared useful in building trust in the USPSTF recommendations. In Phase 2, the researchers presented each group with two or three different prototype video presentations. In Phase 3, participants saw both a persuasive intervention developed based on Phase 1 and 2 efforts (videos of a scripted slide story/presentation, in English or Spanish) and a print comparator (Summary for Patients from the American College of Physicians for mammography screening in women ages 40–49 and a composite of two decision aids developed by a multidisciplinary prostate cancer work group for the Massachusetts Department of Public Health) in random order.
For the Phase 3 crossover trial, researchers assessed changes in screening intentions, perceptions of the balance of benefits and harms, intention to discuss screening with their clinician, and Decisional Conflict Scale (DCS) responses.
Data Analysis
Phase 1 and 2 focus groups were recorded and transcribed (in English or Spanish). Researchers employed both inductive and deductive approaches for qualitative analysis. For the inductive approach, each transcript was reviewed by two team members to identify categories and themes in the transcripts. The team then met as a group to discuss initial coding and themes and come to consensus about any differences in interpretation. The deductive approach involved evaluating hypotheses related to information researchers presented in each focus group (e.g., that presenting information about risk of bias used in guideline development processes by different groups would influence perceptions of trustworthiness of their guidelines). Phase 3 analyses compared survey question responses before and after seeing each intervention, using nonparametric tests to assess differences in Likert-scaled items and t tests to assess differences in DCS responses.
Findings
Qualitative analysis of the focus groups revealed several factors as influential in determining views about screening. These included clinician recommendations, powerful anecdotes related to cancer experiences of family and friends, and, for mammography, extensive socialization to viewing mammography and breast health as part of being a responsible woman, mother, or wife. Participants had no specific knowledge of any group’s guidelines, and they did not recognize the name of the USPSTF. Discussion of guideline group processes and potential biases related to membership of guideline panels had no appreciable impact.
Phase 2 focus group analyses supported Phase 1 findings. Most participants did not have strong preferences among the prototype interventions, although some expressed that the one framed as a doctor-patient conversation had added credibility.
Phase 3 analyses revealed no appreciable changes in screening preferences after seeing the printed control interventions and substantial changes after seeing the experimental interventions, regardless of order of presentation. The experimental interventions significantly reduced desire for screening and increased perceptions of screening harms. Intentions to discuss screening with clinicians decreased. DCS scores decreased similarly after seeing either intervention.
Limitations
Limitations include the small sample size in Phase 3 and the very different format and content of experimental interventions from control interventions. In addition, all participants were from one geographic area and the study measured changes in behavioral intentions but not long-term effects, particularly of screening utilization.
Conclusions
The experimental interventions changed screening preferences of substantial proportions of viewers to ones consistent with USPSTF recommendations. Findings about how to encourage consumers to consider counterintuitive, evidence-based recommendations may be applicable to a variety of decisions.