Guest editor foreword: real-world evidence and medical affairs

The history of medicine is laden with examples of the utility of real-world evidence for solving the problem of disease. John Snow’s epidemiological observations of water distribution systems in mid-nineteenth century London generated evidence describing the real-world problem accurately enough to provide an effective solution [1]. Examples in contemporary medicine show that skilled observation informed by judicious clinical reasoning remains a vital method in the solving of real-world medical problems [2]. However, as systemic and infrastructural fragmentation have contributed to a decline in quality of this clinical skill, so too have they contributed to quality issues in real-world evidence generation. In this introduction to the special issue, I discuss some key issues and ways to redress the balance towards production of real-world knowledge with greater utility, impact and meaning for patients and clinicians.

As defined in health economics, utility is dependent on perspective [3]. This assumption also applies to real-world evidence for implementation in the pharmaceutical industry and clinical medicine. In this case, pragmatic value of real-world evidence depends on the objectives of stakeholder groups: determination of safety, efficacy, cost-effectiveness and clinical effectiveness by regulators, payors and prescribers. Real-world evidence can be categorised according to its proposed end-user due to the requisite properties of real-world knowledge that can be used for effective decision-making on the relative truth of such claims. Quality of real-world evidence can therefore be seen as the extent to which it is able to facilitate regulatory [4,5], payor [6,7], clinical [8] and patient [10,11] decision-making with the goal of meeting patients’ right to receive the best, most appropriate treatments as quickly as possible [12–14].

In this special issue, readers will see examples of real-world evidence for different stakeholders. Aligned with regulatory decision-making, Jost et al. observed effectiveness and safety outcomes of safinamide in an unselected population during routine care in Germany and Lasala et al. conducted a systematic literature review of time to treatment discontinuation as a surrogate efficacy endpoint in non-small cell lung carcinoma [4,5]. Offerings from Diaz et al. and Hoffman et al. provide real-world evidence suggesting long-term health economic benefits of medial collared, triple tapered primary hip system versus other implants for total hip arthroplasty and steroid-eluting implants after sinus surgery in patients with chronic rhinosinusitis, respectively [6,7]. In line with the real-world evidence needs of prescribers, Ray et al.’s expert consensus on the role of β-blockers in the cardiovascular disease continuum aims to provide real-world evidence that resolves conflicting guidelines in the Asia-Pacific region [8].

In contrast, Maihofner’s prospective longitudinal survey and retrospective analysis of pharmacy data generated real-world evidence using outcomes of importance to patients, such as health-related quality of life (HRQoL) and ability to engage in daily activities [9]. Cotton et al.’s point in time study collected real-world data from the US, UK and Germany to provide evidence on unmet needs in patients with episodic cluster headaches [10], while Pathak et al. observed treatment satisfaction in patients self-reporting physician-diagnosed migraine and currently using erenumab [11]. Volpe et al. provided evidence of treatment adherence in hypertensive patients receiving extemporaneous combination therapy with amlodipine/zofenopril that could be used by healthcare decision-makers [15].

While seemingly straightforward, this delineation of real-world evidence by stakeholder raises further questions. For example, patient registries [16–18] have become an important source of real-world data widely published in the literature. However, the extent to which the knowledge they generate is useful for decision-making varies, with clinical interests defined by local guidelines and practices distinct from those of patients [19–21]. Even within stakeholders, fragmented registries may reduce utility of knowledge generated, especially in cases with lack of consensus on disease classification and clinical management. At a more fundamental level, utility of an observation for real-world application is largely dependent on ecological validity, though validation of outcome measures for real-world knowledge tends to rely on hypothetico-deductive reference standards, which have questionable applicability to real-world contexts [22].

In general, validity refers to the extent to which a scale measures the construct it is said to describe. Ecological validity describes how well a scale reflects a construct in its situated form in a specific real-world setting, whereas construct validity is how well it reflects a theoretical ideal derived using the positivist hypothetico-deductive approach, oriented to generalisation of fundamental principles. In other words, construct validation can be considered an academic approach that aims to distil a construct to its purest, most abstract form. Implementation, however, is inherently practical and aims to apply findings to a real-world problem. Tensions arise when attempting to apply a generic abstract concept to specific concrete scenarios with richness and diversity unlike the setting of construct validity studies, designed to reduce variability and ensure standardisation of knowledge production by control of variables.

Suitability of standardised scales for real-world evidence generation is therefore debateable. As the value of real-world evidence is related to its ability to facilitate stakeholder decision-making, real-world studies must deliver knowledge on the ability of treatments, practices and services to effect clinically meaningful change, often in ratio to their corresponding economic and humanistic burden. Such knowledge may vary depending on how constructs are observed. For example, use of generic HRQoL scales in some populations may cause significance of effects to disappear due to failure to observe disease-specific indicators, with potential impact to regulatory and payer decision-making [23,24]. Use of generic outcomes in real-world evidence generation for cost-effectiveness decision-making may impact QALY calculation substantially enough to affect health technology assessment of treatments in some conditions [25,26]. Perspective is an important consideration in generation of real-world evidence, with use of nonpatient utilities decreasing comparative effectiveness and cost-effectiveness of cataract surgery and intravitreal ranibizumab therapy for neovascular age-related macular degeneration, potentially limiting patient access to life-changing treatments [26]. Likewise, applicability of constructs used in the development of health technology assessment models may impact cost-effectiveness decision-making, with potential to affect treatment selection with wide reaching implications affecting patients, healthcare workers and service delivery [27–29].

There is currently no standardised metric for determining applicability of outcomes used to generate real-world evidence. Use of the pragmatic realist construct as a reference standard, in contrast to the theoretical ideal, provides a means for understanding applicability as the relation between a construct measure and the situated knowledge of those directly observing and interacting with the construct in the real world. This limits findings to the original study setting but can provide sufficient standardisation within a disease and context if real-world studies ensure sufficient representation of end users with epistemic authority as defined by pragmatic realism. Application of such hermeneutic principles centring epistemic authority in those with direct experience of a specific context improves applicability of health technology assessment models [30]. Likewise, development of core outcome sets and design of patient registries have recognised the importance of knowledge user perspective in definition of constructs used to produce real-world evidence [16–18,31–33]. However, adjustment of methodological practices in scale development may be required to avoid marginalisation of some groups [20].

In closing, if studies fail to observe the real-world problem accurately enough to constitute proof of its truth, then they cannot be considered evidence of a claim to it solution. Generators of real-world evidence would do well to align themselves with some of the greatest clinical researchers in history, centring their efforts on observing salient aspects of the problem accurately enough to generate evidence truly capable of solving it.

Transparency

Declaration of financial/other relationships

Dr Ravi Jandhyala is a visiting senior lecturer at the Centre for Pharmaceutical Medicine Research at King’s College London and is responsible for research into real-world evidence approaches and is the founder and CEO of Medialis Ltd, a medical affairs consultancy and contract research organization involved in the design and delivery of real-world evidence in the pharmaceutical industry. The Jandhyala method was developed by Dr Jandhyala but is free of commercial licensing restrictions and while used as part of proprietary methodology, is not a direct means of commercial gain for the author.

Peer reviewers on this manuscript have no relevant financial or other relationships to disclose.

Additional information

Funding

The author received no funding for this work.