Patient Engagement Surveys in Clinical Trials: Dos, Don’ts and How They Help

Clinical trials study teams are under pressure to be more patient-centric. While many well-intentioned teams have a patient-centric mindset during the study planning phase, they do not always have the correct tools to ensure their plans get carried into the trial itself. Patient satisfaction or engagement surveys can be easily implemented and used to capture the patient perspective beyond initial focus groups, providing important data that can validate engagement measures, help studies course-correct, and inform future decisions when planning similar trials. However, patient engagement surveys are not often used in clinical trials. While sponsors and CROs will be aware of the value of these surveys, they may be uncertain about how to apply the data. There are also important implementation considerations such as ensuring surveys do not increase patient burden. This article will address these concerns and look at how the pragmatic implementation of surveys in today’s clinical trials can provide valuable insight into patient satisfaction levels, motivators, likelihood to follow study requirements and information retention.

Patient engagement surveys in clinical trials

Patient engagement surveys have become standard practice in many of today’s healthcare settings, helping to inform change and incorporate the patient voice by influencing hospital, health system, and individual practices. The pharmaceutical and clinical trials industry, in contrast, has done very little to introduce satisfaction surveys, no doubt in part due to clinical trials being a very different landscape with a very different myriad of circumstances. Trial structure and design is hugely dependent on phase and disease area; for example, patient satisfaction data collected in an oncology trial will likely differ significantly from that in a flu vaccine trial and the data needs to be handled accordingly.

With patient-centricity being well-established as a necessary pillar of clinical research, the FDA, amongst other influencers, is pushing for more patient-centered research design. Their guidance is to involve patient advocacy groups early in the protocol development stages, as well as to conduct focus groups among protocol-specific demographic groups to gather information about perceptions, opinions, and attitudes relating to the disease of study. However, as a trial progresses past the initial stages, there is a noticeable drop-off in the number and frequency of patient interactions. Patients are generally only asked for their opinions on a study design at the very end of the trial, if at all. This provides minimal opportunity to make near real-time changes in the active study, to better support the very patients providing the feedback. Learn how surveys can be used to enhance clinical trial efficiencies at CBI's eCOA/ePRO 2018.

Industry concerns

The limited use of patient engagement surveys in clinical trials may be due to several factors. Firstly, healthcare in general still has much to learn about how to carry out surveys, their pitfalls and how findings can be best applied to ensure needs are understood and change occurs. The collection of unnecessary and idle data from patients is not an advisable practice as it can erode trust that patients’ responses will be considered and adds additional burden on already demanding study commitments.

In the clinical trials industry, as with any industry, there may also be hesitance to collect negative feedback because it can expose weaknesses that might not have otherwise been flagged. Ultimately, this is a risk that needs to be taken. Embracing the feedback is fundamental. To truly walk the walk and talk the talk of patient engagement the feedback must be used to illicit a call-to-action. With any type of feedback request, the notion is that respondents are most motivated to take the time to respond if they’ve had a particularly bad or positive experience, therefore feedback may not always reflect the experience of the broader study population. The greater propensity for patients to share negative insight also means that patient surveys are not necessarily the best way to evaluate a study. Instead sponsors should focus on developing surveys that have a clear call-to-action to make real-time changes in the current study and can potentially improve data quality, retention, decrease time to completion and understand patient satisfaction. Ultimately, the data that is collected should bring value to both the sponsor and the patient.

Unique advantages

Interactive communication empowers patients and provides real-time feedback to sites and sponsors regarding patient retention issues. Patient surveys offer the opportunity to gather feedback on patient happiness and the emotional impact of a trial on individuals. Is it interesting? Do they feel engaged? Is the research team adequately supporting them? It also enables feedback to be collected on the organization of a clinical trial and how things are working, as well as the burden of activities associated with the study; how do tests, transportation, technology, social support, and financial implications affect burden and what might make the patient more likely to drop out? Questions can also be asked surrounding value to the patient; what is the benefit to them and what’s most likely to keep them in the trial? Do they know when their appointments are? When they arrive at the research site is the coordinator ready for them? Does the team do a good job of explaining how they need to prepare? Are visits long, taking up a lot of their time? Often surveys will include a combination of burden and value-type questions. An example of the content flow in a patient satisfaction survey can be seen in Figure 1.

Please rate your level of agreement with the following seven statements:

- I am satisfied with the care I received at the study clinic
(strongly agree, agree, neutral, disagree, strongly disagree)

- Clinic staff was friendly
(strongly agree, agree, neutral, disagree, strongly disagree)

- The clinic staff answered my questions fully
(strongly agree, agree, neutral, disagree, strongly disagree)

- The clinic staff spent adequate time with me during my visits
(strongly agree, agree, neutral, disagree, strongly disagree)

- The study clinic was comfortable
(strongly agree, agree, neutral, disagree, strongly disagree)

- Study visits were easy to schedule and convenient
(strongly agree, agree, neutral, disagree, strongly disagree)

- Wait times for my study visits were reasonable
(strongly agree, agree, neutral, disagree, strongly disagree)

Please select your top three reasons for participating in this study:

- To contribute important information to medical science

- To potentially help other patients with the same condition

- To hopefully receive treatment for this condition

- To benefit from the additional medical attention and testing that the study provided

- Because of the financial incentives offered

- Other: Please explain

Please select your three least favourite components of this study:

- Too many blood draws or other invasive procedures

- Too many study visits

- Too many questions to answer

- I did not like clinic staff

- The financial incentive was not big enough

- Other: Please explain

How did you find out about this study? Please explain.

Figure 1 – Example of questions included in a typical patient satisfaction survey.

There are many reasons why it is important to implement surveys at various points during a study, not least because they offer the opportunity to identify elements of the trial that are causing difficulties for patients and may, if left unattended to, be a reason for them choosing to leave a study or avoid study requirements, which may lead to inaccurate or inefficient data collection. For example, if a patient forgets their appointment, or has trouble following instructions on how to take medication or give themselves an injection, responses to these questions can drive communication. Site staff can send reminders to that patient or provide additional coaching on how to take the medication properly.

If transportation is difficult, is it because a family member must take time off work to take them? Is it expensive? Are there study activities that impact the patient’s ability to transport themselves? In these scenarios the team may consider coordinating with a third party Non-Emergency Medical Transportation (NEMT) provider, or scheduling visits during more convenient times. The key is to keep surveys focused and have a clear goal in mind when both designing and implementing.

On their most basic level, introducing regular patient surveys allows sponsors to make decisions informed by leading, rather than lagging indicators. As it stands, a trial may be shown to be failing when dropout rates are climbing or when, at the end of the study, protocol compliance is low enough to threaten the integrity of data. While it is important to have this information at this stage, it comes at the cost of a wasted trial and enormous expenditure. While data may be gathered that can be applied to future trials, it is not useful for identifying problems that could be addressed in real-time during a trial.

Surveying patients throughout a study can identify points at which participant drop off will be highest. If dropout rates increase or compliance rates drop at a certain point (e.g., certain visits, difficult treatments, transition from one schedule to another), the study team will know to pay special attention at this time and can use the survey data to understand why this is the case. They’ll get a better sense of what to change when they collect feedback from the subject, combining the what (dropout) with the why (patient feedback). In a 500-person study for example, if a protocol ‘hot spot is identified through real-time data from patient surveys of the first 100 patients, the study has the opportunity to mitigate the issue and more carefully manage the patient experience during this phase to avoid the impact for the remaining 400 patients who haven’t hit that ‘hot spot’ yet.

Comparing patient satisfaction results across sites

Patient satisfaction surveys also present a valuable opportunity to compare results across sites. This helps to determine those that are high performing, and those that require further attention. Sponsors rely on sites to provide the optimal patient experience; these patient surveys enable a ‘data-driven’ conversation with sites for the first time. This can prompt sponsors and clinical research associates (CRAs) to further review the practices of these sites to identify any issues that could lead to low patient scores.

Sponsors may also want to allow the sites they work with to have access to patient-specific survey results. Doing this will help sites to respond quickly to problems as they arise, before escalation is needed.

Pragmatic implementation

Patient surveys are relatively straightforward to implement. Once the content has been created, the system being used is configured with the survey and then made available to patients at the appropriate time. It becomes more complex when there are many differing opinions on the content. While surveys should be scientific, there needs to be a happy medium between survey instrument robustness and getting feedback that can drive meaningful change.

In developing survey questions, study teams should take learnings from consumer strategies. Simple tactics such as keeping surveys short and relevant to a study activity, and writing in plain, conversational language are essential. Patients hear technical jargon all the time during treatment; survey language should be straightforward and to the point. When possible, explanations should also be provided as to why particular data is being collected and how it will be used. Displaying a progress bar is also an easy way to quickly show patients how far they’ve come and how far they have left to go. A sense of progress can help patients complete an entire survey.

The following process may be a useful guideline for sponsors as they develop survey questions.

1. Firstly, lead and main stakeholders should be identified.

2. Information gathering should then be conducted from various teams, including the study team, clinical research sites and vendors.

3. Input can then be summarized and questions drafted. These may be standardized across studies but are generally most valuable when targeted to a specific audience of the specific study. If targeted, the patient population should be kept in mind when it comes to condition, demographics, psychographics, etc. Questions should focus on gathering feedback that maps to the study need and study activities. Keeping the questions specific and measurable will help ensure information is gathered that will inform a clear call-to-action for making real or near real-time changes for improvement.

4. Rounds of reviews should be limited and specific deadlines set for finalizing content.

Study teams will also want to be strategic about when they administer surveys. If feedback is requested too frequently patients may feel overly burdened. It is also important to think about the state of mind the patients will be in when they are providing the feedback. For example, if there is a particularly arduous visit that requires a lot of tests, it would be prudent to make sure a survey is available around the time of that visit so that patients can anchor their experience to the feedback. If surveys are too far removed from key events, patients will start to think about their experience in more general terms.

There are a lot of variables that come into play in motivating patients to provide feedback.  It can easily become an impersonal experience; patients submit responses and all too often this information disappears into a blackhole with no feedback or follow-up communication from the sponsor to show how the information they have provided has been used. There is value in finding ways that patient surveys can give something back to the patient without biasing their input. For instance, after a survey window closes, enabling patients to find out how many other people responded can be a compelling motivator.

Final thought

Regular data gained through patient engagement surveys can give sponsors a leading indicator as to whether their investment is going to fail because of disengaged patients or overly burdensome protocol requirements. Conversely, if patients are satisfied and unexpectedly compliant, that information could help decide when to close recruitment early and reap the cost savings.

More advanced surveys can also serve many purposes, including gathering information about the effectiveness of engagement measures or the patient mindset at study pain points. They can also test the effectiveness of technology and other practices, which considerably de-risks experimentation. Sponsors can use this information to take corrective action and improve study management. Additionally, patient surveys can provide the opportunity to compare feedback across sites to provide positive incentives for good site performance, address disparities across demographics, and make decisions about whether to reuse study sites in future trials.

Sponsors looking to implement a survey strategy should first begin working with teams to create a standardized approach for patient engagement survey creation. It can be useful to look to internal teams or vendors that have done this before for guidance and insights. It is also important for sponsors to work with vendors to understand their ability to support strategies via technology. Sponsors can then start to get buy-in and make surveys part of the study design process. With all parties, it is vital to emphasize the value in creating real-time improvements over long-term analysis while engaging research patients and humanizing today’s complex trials.

Jeff Lee, President, eCOA and Patient Engagement and Mindy Gruba, Product Manager, Bracket Global