We found 105 PROMs and 57 MCID references reported in 39 CDR-CRR. Though these PROMs and MCIDs are not always validated in the study population, a large majority were. Similar trends in both PROM and MCID reporting regarding frequency and public access were observed. 87% of PROMs had a validation reference; 59% validated in the study population and 60% of MCIDs reported were also validated in the study population.
Although 29% of all CDR-CRRs did not report any PROM or HRQoL measurement, when included, multiple PROMs were used for the same drug and/or trial, each with its’ own purpose. It is challenging to choose the right PROM when designing a study, as patients can provide details on a number of domains that are important for evaluation, such as symptom experience, functional status, well-being, quality of life and treatment satisfaction . When designing a study, a number of relevant existing PROM measures could exist or none may seem appropriate. Furthermore, a drug may not be expected to affect all domains equally within a PROM. Often, there is overlap among generic PROMs with questions that could seem redundant to patients, affecting response and attrition rates.
Recently, the International Society for Quality of Life Research (ISOQOL) conducted a literature review and accompanying survey of its members to identify minimum standards for the design and selection of a PROM for use in patient-centered outcomes research and comparative effectiveness research. Evidence for reliability, validity (content, construct and responsiveness), interpretability of scores, quality translation and acceptable patient and investigator burden were the final recommendations for standards and considerations . They emphasized the important of availability of the documentation of the PROM evidence to generate greater acceptance and use of the measure .
As policy concerns become more focused with research uptake and knowledge translation, the issue of interpretability of scores surfaces. MCID were developed, originally promised to define improvement thresholds, helping to interpret results and guide clinical care. However, some remain skeptical about reaching this type of objectivity. We found 60% of MCID references included in the CDR-CRRs were specific to the study population, which reflects the context-specific nature of the MCID. MCID variability depends on baseline scores, study context and the approach used to determine the MCID , time between baseline and follow up , and may vary according to the direction of the score. In patient management, MCID help clinicians interpret and guide clinical practice in a systematic way. It can enhance patient-provider and provider-provider communication . Beaton et al. conclude that future work on MCID should determine whether or not a MCID for improvement is the same as that for deterioration  within the same context. The MCID plays a crucial role in determining any level of change and responsiveness . It is extremely useful in clinical research, where high powered studies will often report statistically significant results due to the low type 2 errors, without measuring clinical significance . Here, reviewers and policy makers need to acknowledge clinical significance and the situation-specific nature of the MCID, and consider this carefully when making final recommendations.
The goal of any research is for it to be used, applied and expanded upon. Public accessibility is important for the timely dissemination of new research to clinicians, patients and the public. The concept of Open Access increases the potential readership of any article to over a billion individuals with Internet access and indirectly speeds up the spread of information . With increased focus on patients and their families through the use of PROMs, it only makes sense to have these findings available to those who contributed, will use and will further discuss the results in the community. With more accessible PROM and MCID data and reports, it can also influence clinical decision making and policy. Future directions might consider evaluating whether or not patients actually access and read research and how much impact more technical validation studies and publications have to patients. Taking this one step further, we might consider alternative knowledge translation products, keeping in mind the lay person audience, and how the research results can be understood and used by those who contributed to the research in the first place.
Patients have the opportunity to contribute to the current CADTH review process through a patient input section. This is almost always narrative accounts, with very little real-world data. This paper adds emphasis to the increasing number of PROMs included in drug studies and reviews, with a focus on public accessibility of data and reports. Patients are a knowledgeable group, with lived experience. With the right education and organization, patients may be able to collect their own PROMs to allow for greater comparisons across clinical trial and real-world data, giving strength to the patient input section of the CADTH drug review. However, first, we must continue to value PROMs and MCID research at large.
Our review is not without limitations. Though we examined 105 PROM measurements, these were found in only 39 CDR-CRR reports and this sample size limited our analyses. CDR-CRRs are also released in ‘batches’ and it’s possible some reports were not yet publicly available to be included. As well, we could not control for the types of diseases or conditions for which CDR-CRR reports were generated. We were unable to identify which fields or diseases lack PROM and MCID studies. Additionally, academic researchers determined the public availability of the references and although double data entry was performed from public locations, access may have been granted due to affiliation. The public availability of research varies greatly around the world and our findings may not be accurate for other jurisdictions.