Leveraging the advantages of electronically captured patient-reported data in early phase trials
As clinical trial planning begins, PROMETRIKA Clinical Trial Managers (CTMs) and Project Managers (PMs) are often asked by Sponsors to assess the pros and cons of using paper documentation (e.g., diaries) vs. electronic data entry for patient-reported data. Our broad experience with different phases of clinical trials helps us advise Sponsors on this decision.
Paper may be considered the gold-standard method of collecting certain kinds of patient-reported data, such as diary data, because it easily satisfies the FDA regulatory requirement of the Investigator to control, access, and maintain source documentation.1 But the use of handwritten information comes with disadvantages. Diaries, for example, may have one or more ALCOA-C issues: is the data attributable, legible, contemporaneous, original, accurate, and complete? Additionally, handwriting may include extraneous marks and the diary may accidentally include protected health information (PHI).
The use of paper diaries always raises a concern over whether the data are contemporaneous with the events. Was the diary completed at the time stipulated by the protocol or did the patient (or caregiver) fill in data by recall, hours, days, or even weeks after the event? Because paper diaries have no time stamp, the Sponsor cannot know if the data are affected by a lag-time between event and entry. This lag-time may even result in missing data; i.e., the patient cannot recall some details.
Sponsors often wonder whether using a paper document is more cost effective. The answer is – maybe. By using paper, the Sponsor does not have to provide devices (e.g., electronic pads) or pay a vendor for use of a validated platform. But there are hidden costs associated with paper documents. Depending on the size of a trial, each validated copy of a paper instrument carries a standalone fee; these can add up quickly. And if unvalidated patient reported outcomes (PRO) copies are used, the data may be invalid and not fit for regulatory analysis purposes. Another cost of using paper may be paying the site or another party to transcribe the data from the paper document to the electronic database. And if the handwritten data are difficult to read, there is added time/cost associated with data cleaning and analysis. Missing entries, i.e., what a patient may not recall when completing a diary too late, can lead to missing data crucial to analyzing a study endpoint. Finally, the time between issuing of a diary to the patient and collecting it can be weeks-long, so data completion issues may not be detected for a long period of time.
What are the advantages and challenges of electronic capture of patient-reported data?
What are the considerations for use of an electronic device to capture patient-reported data? An issue that might prove to be a hindrance to the use of electronic capture is access to a stable internet connection via cellular signal or Wi-Fi. A weak or fluctuating signal may result in missing or late-recall data. Common complaints regarding the use of electronic PROs are that the questionnaires are not triggering correctly or data is being collected but not transmitted. In situations where human error has been identified as the cause, additional site training has improved ePRO acceptance. In some targeted populations, electronic literacy may not be strong and paper documents may work better. But given that over 70% of the world’s population has access to smart technology, sponsors should try to take advantage of the benefits of ePRO.2
In the past 10 to 15 years, the use of ePRO has gained traction. The speed at which technology is advancing is demonstrated by the introduction of artificial intelligence in our daily lives as well as by the development of technologies currently available for use in clinical trials. Companies like Medidata are championing the way forward with advancing technologies that would allow for more efficient and timely methods of collecting critical data points. Examples of these technologies are iPads, mobile apps, wearable devices, telehealth apps, and smartwatches. Use of e-diaries, e-questionnaires, etc., can ensure that patient-reported data are captured contemporaneously, correctly, and completely due to automated time stamping and data checks. Patients can be prompted in real time to correct entry errors or complete missed fields. Additionally, the accidental release of PHI is removed by use of an electronic form. Finally, in many locations around the world today, patient populations are using smart devices in their daily lives; thus, incorporating a document into the patient’s phone or providing an electronic pad should not create significant delays in learning to use the document.3 These features of electronic capture save time by eliminating the need to transcribe data and providing near-real-time data to data managers and Sponsors.
When, why, and how could both methods be used in a trial?
Sponsors may wonder if both methods are acceptable in a given trial. When designing a trial, one mode should be primary but it is crucial to have a backup plan in case of emergencies. Both electronic and paper copies of the instruments must be IRB approved and, in some cases, licensed. The number of instances where paper is used due to emergencies (e.g., power outages; internet interruptions) should be low; sites should be discouraged from using paper otherwise. Experience has shown that, when sites have access to the paper backup copies, they revert to utilizing them more often than just for emergencies. Reasons cited for this can be: they perceive technology to be glitchy and unreliable, so it is faster and easier for them to use paper rather than take the moment to troubleshoot, or a particular trial participant struggles with electronic modalities. Regulatory bodies strongly discourage mixing modalities but assert that mixing modality is better than losing data.
If mixed mode data collection is planned to minimize the loss of data, the statistical plan needs to specify whether the primary analysis will include data from both modes, and if so, what analyses will be performed to evaluate whether the treatment effect differs by mode. If the amount of data from paper is minimal (e.g., less than 10%), mixing of modes may not have an impact on the overall result, but analyses are recommended to verify whether that is the case. These analyses may include descriptive summaries of the data by mode, and sensitivity analyses using only the ePRO data, both as observed and with imputation for missing data.4
Final thoughts
Where does this leave Sponsors? The U.S. FDA has expressed that use of electronic data capture in clinical trials source data is preferred over paper.5 Per NIH, electronic data capture has increased data collection capabilities and options in clinical trials.6 Electronic data capture is more accurate, timely, creates less of an administrative burden, and can be more cost effective. Advancing artificial intelligence capabilities and integrations can enhance trial outcome. Along with many other advantages, it can assist in developing various algorithms that help study teams analyze the vast amounts of data, help match participants to clinical trials, improve recruitment, and allow participants to be more engaged in the trial.
Sponsors must consider embracing technological leaps to ensure staying in step with the industry and the regulatory environment. Complete analytical planning and strong site and staff training will ensure successful use of electronic modes of patient data capture.
PROMETRIKA’s clinical and management teams are aware that Sponsors whose only previous experience has been with paper patient-reported documents are looking for experienced guidance when considering the use of electronic documents. Early-phase trials in a new program are usually a good time to pilot the use of electronic documents to test the feasibility with a certain patient population. Use in early phases allows Sponsors to hone the processes and creates data that is more likely to be integrate-able with later-phase studies.
