ePROMs in Clinical Oncology : What is the Evidence ?

Abstract

There is no denying fact that digital therapeutics (DTx) such as electronic patient monitoring outcome measures (ePROMs) will play a decisive role in the future of clinics, especially as efficient tools for addressing some unmet needs in the management of chronic conditions such as cancers. While many conventional pharmacotherapies have undergone multistage stringent verification and safety measurements well over a substantial period of time, barriers to the application and adoption of DTx in clinics have been dramatically low (Kim, H. S. 2020). One of the major barriers to the adoption of DTx into clinical practice remains the unavailability of adequate data quality and robustness of clinical evidence.

The aim of the current thesis was to examine the robustness of scientific evidence for the effectiveness and utility of ePROMs in oncological settings. As per the hierarchy of evidence for the evaluation of health care outcomes randomized control trials (RCTs) are considered as the gold standard since they not only deliver the highest level of evidence but are also limited in all kinds of study bias and control confounding variables. Therefore to realize our aims we took an approach that focused on studying the methodology and design integrity of RCTs using several critical appraisal tools such as our modified Critical Appraisal Skill Programme (mCASP) checklist (primary analysis tool) comprising of 23 items that also incorporated items from additional RCT appraisal tools such Joanna Briggs Institute Critical Appraisal tools for RCT, BMJ pest practice and John Hopkins RCT appraisal checklist. To check the authenticity of mCASP checklist we also employed a battery of some other well-known critical appraisal tools such as the modified Jadad scale (mJadad scale), van Tulden scale and Cochrane Tool Effective Practice and Organisation of Care (EPOC) Risk of Bias Tool that served as a secondary analysis tool.

Using specific keywords, we screened about 4751 RCTs on MEDLINE, and with our inclusion and exclusion criteria, we were finally able to narrow it down to 7 ePROM RCTs from the last 10 years.

Examination of selected RCTs using our primary appraisal tool mCASP checklist showed that >50% of the studies lacked critical components of a RCT such as allocation concealment, blinding, true intention to treat (ITT) analysis, and similarity at baseline. Precisely about 70-85% of studies(n=6-7) scored negatively on the items related to blinding with only a single study recording a score above zero. Similarly, scores of true ITT analysis were 0 to negative for about 85% of studies (N=6) and only a single study was able to score maximum points for ITT analysis. Internal validity total scores as per mCASP checklist showed that more than 50% of studies had total score values below the average score value of 9. Based on grading criteria for mCASP checklist roughly about 60% (n=4) of studies were graded as low quality, about 25% (n=2) as medium quality and only a single study could meet the criteria for a high-quality study.

Although tools for secondary analysis like mJadad Scale and van Tulder scale were not as elaborative as the mCASP checklist, in scrutinizing the internal validity of the RCTs we did find a good correlation between different appraisal tools. Comparative analysis of the four critical appraisal tools categorized n=3 studies as medium-high quality (with mCASP, mJadad scale and van Tulder scale), n=2 studies were categorized as low-quality (mCASP and mJadad scale). Comparative analysis further revealed a trend for a low risk of bias with a higher-quality score for a study.

We conclude that the majority of the RCTs that were examined had a number of serious methodological flaws that could likely compromise the quality of ePROMs RCTs and as such raise a question mark of the internal and external validity of these studies. Apparently, the foundation of empirical evidence advocating the clinical utility of ePROMs for the management of cancer patients could be somewhat shaky. Due to a lack of high-quality evidence, and because of limited generalizability for a major portion of RCTs we recommend a more in-depth scrutiny of the clinical evidence generated by ePROM RCTs in oncological settings and exciting caution while prescribing these DTx to patients.