Introduction
Healthcare payers face challenges in deciding which health technologies to fund, given a limited budget (Reference Levine, Taylor, Ryan and Sculpher1–Reference Chan, Nam and Evans3). This challenge is magnified given the high volume of new treatments seeking timely regulatory and reimbursement approvals, the rising cost of these technologies, and the growing demand for access to effective and innovative treatments and services (Reference Levine, Taylor, Ryan and Sculpher1;Reference Laba, Jiwani, Crossland and Mitton2;Reference Detiček, Janzic, Locatelli and Kos4;Reference Huang and Gau5). With increased pressure to make decisions based on limited or single-arm trial evidence, payers are cautious about which technologies to fund (Reference Levine, Taylor, Ryan and Sculpher1;Reference Clausen, Mighton and Kiflen6).
Randomized controlled trials (RCTs) are the gold standard for generating scientifically grounded evidence (Reference Pulini, Caetano, Clautiaux, Vergeron, Pitts and Katz7;Reference Roberts and Ferguson8). However, RCTs seldom reflect conditions in the real world because they tend to have strict inclusion and exclusion patient criteria (Reference Chan, Nam and Evans3;Reference Roberts and Ferguson8–Reference Bowrin, Briere, Levy, Millier, Clay and Toumi10), lack external validity/generalizability, use intermediate outcomes, and have short-term follow-up (Reference Levine, Taylor, Ryan and Sculpher1;Reference Chan, Nam and Evans3;Reference Bowrin, Briere, Levy, Millier, Clay and Toumi10). There are also practical and ethical concerns to conducting longer-term studies, as RCTs are often costly to execute, and patients may be unwilling to undergo randomization if the performance of a product is already known (Reference Chan, Nam and Evans3). With these limitations, funding decisions are made with a great deal of uncertainty as decision-makers operate with imperfect information (Reference Chan, Nam and Evans3;Reference Kirwin, Round, Bond and McCabe11;Reference Kovács, Kaló and Daubner-Bendes12).
Real-world evidence (RWE), which is derived from sources outside of RCTs, can be used by payers and health technology assessment (HTA) agencies to inform decisions regarding comparative effectiveness, cost-effectiveness, safety, and the overall value of a healthcare technology (Reference Chan, Nam and Evans3;Reference Makady, de Boer, Hillege, Klungel and Goettsch13–Reference Lau, Jamali and Loebenberg16). Increasingly, RWE is recommended and used to supplement evidence from RCTs to support coverage decisions (Reference Timbie, Kim and Concannon17). For payers, RWE on the clinical validity and utility of a particular technology can be used in pre- and post-market evaluations to inform price negotiations and future funding/reimbursement arrangements with manufacturers (Reference Makady, ten, de Boer, Hillege, Klungel and Goettsch14;Reference Andersson and Kyhlstedt18–Reference Deverka, Douglas and Phillips25).
Although there are potential benefits to incorporating RWE in HTA and payer decision making, few countries have established the policy infrastructure and frameworks needed to realize these benefits (Reference Chan, Nam and Evans3). Many countries are still grappling with how best to include RWE in healthcare decision-making frameworks because the methods of data collection and analysis are not standardized (Reference Clausen, Mighton and Kiflen6;Reference Deverka, Douglas and Phillips25;Reference Cowie, Bozkurt and Butler26). Although HTA agencies and other multistakeholder groups have provided frameworks to facilitate the adoption of RWE in healthcare decision making (Reference Fleurence, Kent and Adamson27), the application of these frameworks by payers is still not well understood. For these reasons, data from RCTs are used as the principal source of evidence to inform pre- and post-market funding decisions and pricing arrangements, despite the growing interest in the use of RWE among payers (Reference Chan, Nam and Evans3;Reference Timbie, Kim and Concannon17). Incorporating RWE into funding and pricing decisions can facilitate the establishment of more innovative funding arrangements and incentivize manufacturers to bring new products to the market (Reference Deverka, Douglas and Phillips25;Reference Gray and Kenney28).
The primary objective of this scoping review is to summarize the existing literature on how RWE has been used by healthcare payers by (i) identifying the types of RWE healthcare payers have considered, (ii) highlighting the barriers that limit the use of RWE in payer decision making, and (iii) exploring how RWE has been factored into payer/manufacturer funding arrangements. This review focused on both medical devices and medicines.
Methods
This scoping review was conducted based on the five-step framework proposed by Arksey and O’Malley (2005), that was later updated by Levac et al. (2010), the Joanna Briggs Institute (2015), and Peters et al. (2020) (Reference Arksey and O’Malley29–Reference Peters, Marnie and Tricco31). Reporting was done according to the Preferred Reporting Items for Systematic Reviews and Meta-analysis for Scoping Reviews (PRISMA-ScR) guidelines (Reference Tricco, Lillie and Zarin32;Reference Page, McKenzie and Bossuyt33). The protocol for this scoping review is available on the Open Science Framework (https://osf.io/8wpnb/).
Search strategy
A literature search was conducted by an information specialist (CC) and included literature published between January 1, 2014, and March 26, 2025. The publication date range was selected as this timeframe aligns with the development of the discourse on RWE in HTA and payer decision-making. The databases searched included PubMed (Medline), OVID Embase, Cochrane Library, and ProQuest Dissertations and Theses Global. The first three databases were chosen based on recommendations from Cochrane (Reference Lefebvre, Glanville, Briscoe, Featherstone, Littlewood, Metzendorf, Higgins, Thomas, Chandler, Cumpston, Li and Page34). The last database was chosen as it included dissertations that may lead to published sources. The search strategy included a combination of keywords (limited to title and abstract fields only) and controlled vocabulary (dependent on database availability) related to the concepts of RWE, payers, and decision making. The complete search strategies for each database are provided in Supplementary File 1. The reference lists of included studies were scanned to identify additional relevant studies. All database search results were imported into Covidence (Veritas Health Innovation) for screening.
Two reviewers (JZ and DL) screened the titles, abstracts, and relevant full-text articles independently. Before screening, both reviewers pilot-tested the record screening with a random sample of 50 records. Screening commenced when at least 75 percent agreement was reached. Any discrepancies were discussed between reviewers until a consensus was reached. Disagreements were resolved by a third reviewer (LM). Following the title/abstract screening, only relevant articles were selected for full-text review by the reviewers (JZ and DL).
Eligibility criteria
Eligible articles were those that (i) discussed the use of RWE among healthcare payers/decision-makers for health technology reimbursement decisions or (ii) discussed the use of RWE in innovative funding arrangements. Studies could explicitly mention RWE or the use of data obtained outside of randomized-controlled trials. Only articles written in English were included. Conference abstracts were removed from the results of all databases. No restriction was placed on the type of healthcare technology included (e.g., medical device or medicine) because it was anticipated that the literature did not sufficiently address any one type of health technology. A PRISMA flow chart (Reference Tricco, Lillie and Zarin32) was generated to show the outcome of the study selection process (see Supplementary Figure S1).
Types of sources
This scoping review considered qualitative, quantitative, or mixed-methods articles as we sought to obtain a discourse around the use of RWE.
Data charting and synthesis
Data were extracted for all relevant articles by one author, with a second author checking for accuracy. Data were extracted and stored in a customized data abstraction form created in Microsoft Excel. The following variables were extracted: disease area, definition of RWE, reason for using RWE, type of RWE used, strengths and weaknesses of RWE as identified by healthcare payers, whether RWE was used in any funding arrangements, and what type of funding arrangement was used. Data synthesis was informed by the data analysis recommendations proposed by Pollock et al (Reference Pollock, Peters and Khalil35). This included a narrative account of the results of the review and the use of descriptive statistics, such as counts and proportions, to present the results.
Results
The literature search generated 13,655 articles from the respective databases (OVID Embase: 6,887, PubMed: 5,159, ProQuest Dissertations and Theses Global: 832, Cochrane Library: 777). A total of 4,024 duplicates were removed, and a further 9,584 articles were excluded at the title/abstract screening stage because they did not meet the inclusion criteria. Following the full-text review, nineteen of the remaining forty-seven articles were considered relevant for further analysis. The reasons for excluding articles at the full-text screening stage are included in Supplementary Figure S1. Supplementary Tables S1 and S2 present the data extracted from the studies identified in this scoping review.
Overview of included studies
Study countries
The general characteristics of the studies included in this review are summarized in Table 1. The majority of studies were based on payers from North America; five studies each from both the United States and Canada (Reference Laba, Jiwani, Crossland and Mitton2;Reference Chan, Nam and Evans3;Reference Clausen, Mighton and Kiflen6;Reference Roberts and Ferguson8;Reference Lau and Dranitsaris15;Reference Timbie, Kim and Concannon17;Reference Hampson, Towse, Dreitlein, Henshall and Pearson20;Reference Gray and Kenney28;Reference Brixner, Biskupiak and Oderda36;Reference Husereau, Nason, Ahuja, Nikaï, Tsakonas and Jacobs37). Other country-specific studies were on payers from the Netherlands, Germany, Italy, and the Kingdom of Saudi Arabia, respectively (Reference Mohseninejad, van Gils, Uyl-de Groot, Buskens and Feenstra38–Reference Abu-Shraie, Alhammad, Balkhi and Al-Jedai41). There were three multicountry studies (Reference Pulini, Caetano, Clautiaux, Vergeron, Pitts and Katz7;Reference Kovács, Kaló and Daubner-Bendes12;), and two studies were not specific to a country (Reference Kirwin, Round, Bond and McCabe11;Reference Pearson, Dreitlein, Towse, Hampson and Henshall43). All nineteen studies were conducted from the perspectives of payers, with relevance for health technology developers/manufacturers, HTA agencies, patients, and regulatory bodies.
Table 1. General characteristics of the included studies
Box 1. Challenges with the use of real-world evidence
-
▪ Quality of RWE generated (bias and confounding, incomplete or inaccurate data sets)
-
▪ Access to timely and relevant data (legal concerns about data sharing)
-
▪ Cost involved in accessing and generating data
-
▪ Absence of standards for reporting
Study design
Nine studies either surveyed or interviewed payers and other stakeholders to gather their perspectives on the incorporation of RWE into funding and reimbursement arrangements (Reference Clausen, Mighton and Kiflen6;Reference Pulini, Caetano, Clautiaux, Vergeron, Pitts and Katz7;Reference Kovács, Kaló and Daubner-Bendes12;Reference Timbie, Kim and Concannon17;Reference Hampson, Towse, Dreitlein, Henshall and Pearson20;Reference Gray and Kenney28;Reference Brixner, Biskupiak and Oderda36;Reference Husereau, Nason, Ahuja, Nikaï, Tsakonas and Jacobs37;Reference Bharmal, Katsoulis and Chang42). In these studies, all stakeholders were asked to share their views on the perceived benefits, challenges, enabling factors, and application of RWE.
Two studies surveyed payers (Reference Brixner, Biskupiak and Oderda36;Reference Bharmal, Katsoulis and Chang42). The first study surveyed 221 payers from the United States and had an additional ten payers discuss the survey results (Reference Brixner, Biskupiak and Oderda36). The second study, conducted a survey involving thirty current and former payers from across Europe and the United States, who had experience in the coverage and reimbursement of oncology therapies (Reference Bharmal, Katsoulis and Chang42).
Seven studies interviewed stakeholders involved in pharmaceutical pricing and decision making. Clausen et al. (2020) interviewed thirty Canadian stakeholders, including six from pricing negotiation organizations (Reference Clausen, Mighton and Kiflen6). Kovács et al. interviewed twenty-five decision-makers from twenty-two European countries (Reference Kovács, Kaló and Daubner-Bendes12). Gray and Kenney involved seventeen stakeholders, seven from payer organizations (Reference Gray and Kenney28), whereas Hampson et al interviewed nine, including two US payers (Reference Hampson, Towse, Dreitlein, Henshall and Pearson20). Timbie et al., included twenty-seven stakeholders, four of whom were payers (Reference Timbie, Kim and Concannon17). Husereau et al. held multistakeholder meetings with ninety-one participants, including ten payers (Reference Husereau, Nason, Ahuja, Nikaï, Tsakonas and Jacobs37). Pulini et al. did not report the number of interviewees but identified stakeholders from pharmaceutical and clinical research centers (Reference Pulini, Caetano, Clautiaux, Vergeron, Pitts and Katz7).
Five studies were discussion papers supplemented with literature reviews. (Reference Pulini, Caetano, Clautiaux, Vergeron, Pitts and Katz7;Reference Roberts and Ferguson8;Reference Hampson, Towse, Dreitlein, Henshall and Pearson20;Reference Brandes, Schwarzkopf and Rogowski39–Reference Abu-Shraie, Alhammad, Balkhi and Al-Jedai41). Three studies incorporated a nonsystematic review (Reference Pulini, Caetano, Clautiaux, Vergeron, Pitts and Katz7;Reference Brandes, Schwarzkopf and Rogowski39;Reference Xoxi, Rumi and Kanavos40), one study conducted a systematic literature review (Reference Roberts and Ferguson8;Reference Hampson, Towse, Dreitlein, Henshall and Pearson20), and one study performed a narrative review (Reference Abu-Shraie, Alhammad, Balkhi and Al-Jedai41).
Two studies (Reference Lau and Dranitsaris15;Reference Mohseninejad, van Gils, Uyl-de Groot, Buskens and Feenstra38) adopted predominantly quantitative methods. The first study assessed whether RWE provided by drug manufacturers influenced the funding recommendations of Canada’s Drug Agency for cancer drugs (Reference Lau and Dranitsaris15). This study concluded that the RWE generated by manufacturers did little to spur positive funding recommendations, instead raising concerns about data standards and quality (Reference Lau and Dranitsaris15). The second study estimated an optimal period of no more than two years to collect data from patient registries to inform access with evidence arrangements for the coverage of a colon cancer treatment in the Netherlands (Reference Mohseninejad, van Gils, Uyl-de Groot, Buskens and Feenstra38).
Five studies provided decision-making frameworks or methods on how RWE can be incorporated into payer decision-making and reimbursement schemes (Reference Laba, Jiwani, Crossland and Mitton2;Reference Chan, Nam and Evans3;Reference Kirwin, Round, Bond and McCabe11;Reference Kovács, Kaló and Daubner-Bendes12;Reference Xoxi, Rumi and Kanavos40). Two studies were discussion papers presenting a framework (Reference Kirwin, Round, Bond and McCabe11;Reference Xoxi, Rumi and Kanavos40), one study presented a process for developing a RWE framework (Reference Chan, Nam and Evans3), one was a case study (Reference Laba, Jiwani, Crossland and Mitton2), and another utilized a focus group to validate a decision-making tool (Reference Kovács, Kaló and Daubner-Bendes12).
Clinical setting
Studies focused on several clinical settings where RWE can prove valuable, primarily oncology (n = 6) (Reference Chan, Nam and Evans3;Reference Clausen, Mighton and Kiflen6;Reference Lau and Dranitsaris15;Reference Brixner, Biskupiak and Oderda36;Reference Mohseninejad, van Gils, Uyl-de Groot, Buskens and Feenstra38;Reference Bharmal, Katsoulis and Chang42), pharmaceuticals (n = 5) (Reference Laba, Jiwani, Crossland and Mitton2;Reference Pulini, Caetano, Clautiaux, Vergeron, Pitts and Katz7;Reference Husereau, Nason, Ahuja, Nikaï, Tsakonas and Jacobs37;Reference Abu-Shraie, Alhammad, Balkhi and Al-Jedai41;Reference Pearson, Dreitlein, Towse, Hampson and Henshall43), medical devices (n = 2) (Reference Kovács, Kaló and Daubner-Bendes12;Reference Timbie, Kim and Concannon17), and other new health technologies (n = 2) (Reference Roberts and Ferguson8;Reference Brandes, Schwarzkopf and Rogowski39).
A few studies examined payer perceptions of RWE in oncology funding decisions in the United States and Canada (Reference Chan, Nam and Evans3;Reference Lau and Dranitsaris15;Reference Brixner, Biskupiak and Oderda36). Two focused on the CanREValue project, which is a stakeholder-led initiative aimed at developing a national framework for generating and using RWE in Canadian cancer drug funding decisions (Reference Chan, Nam and Evans3;Reference Clausen, Mighton and Kiflen6). Other studies focused on pharmaceuticals and emphasized integrating RWE into standardized decision-making frameworks to inform reimbursement (Reference Laba, Jiwani, Crossland and Mitton2;Reference Pulini, Caetano, Clautiaux, Vergeron, Pitts and Katz7;Reference Abu-Shraie, Alhammad, Balkhi and Al-Jedai41;Reference Pearson, Dreitlein, Towse, Hampson and Henshall43). Pulini et al. compared how RWE informs market authorization, pricing, and reimbursement in the United States, the United Kingdom, and France. Two other studies proposed decision-making frameworks, including one using multicriteria decision analysis for Canada’s public health system (Reference Laba, Jiwani, Crossland and Mitton2;Reference Pearson, Dreitlein, Towse, Hampson and Henshall43). Abu-Shraie et al. demonstrated how RWE can be linked to reimbursement arrangements like risk-sharing agreements, where payment is based on treatment performance (Reference Abu-Shraie, Alhammad, Balkhi and Al-Jedai41).
Two studies addressed medical devices (Reference Kovács, Kaló and Daubner-Bendes12;Reference Timbie, Kim and Concannon17). One developed a decision tool for guiding reimbursement of late technology adoption to address uncertainty in reimbursement for medical devices based on coverage with evidence development agreements (Reference Kovács, Kaló and Daubner-Bendes12). The other study conducted interviews to explore industry-wide RWE use, highlighting payer’s concerns about RWE rigor despite its value in supplementing trial data (Reference Timbie, Kim and Concannon17). Studies on general health technologies or those that did not have a specific application focused on creating payer decision frameworks and their use in informing reimbursement arrangements (Reference Kirwin, Round, Bond and McCabe11;Reference Kovács, Kaló and Daubner-Bendes12;Reference Timbie, Kim and Concannon17;Reference Hampson, Towse, Dreitlein, Henshall and Pearson20;Reference Brandes, Schwarzkopf and Rogowski39;Reference Xoxi, Rumi and Kanavos40).
Definition of RWE
Twelve of the nineteen studies provided formal definitions for RWE (Reference Chan, Nam and Evans3;Reference Clausen, Mighton and Kiflen6–Reference Roberts and Ferguson8;Reference Kirwin, Round, Bond and McCabe11;Reference Lau and Dranitsaris15;Reference Timbie, Kim and Concannon17;Reference Hampson, Towse, Dreitlein, Henshall and Pearson20;Reference Husereau, Nason, Ahuja, Nikaï, Tsakonas and Jacobs37;Reference Mohseninejad, van Gils, Uyl-de Groot, Buskens and Feenstra38;Reference Bharmal, Katsoulis and Chang42;Reference Pearson, Dreitlein, Towse, Hampson and Henshall43). Three of these studies reported that RWE refers to evidence on the use of medical products, including the potential risks and benefits, based primarily on patient health data and outcomes (Reference Pulini, Caetano, Clautiaux, Vergeron, Pitts and Katz7;Reference Lau and Dranitsaris15;Reference Hampson, Towse, Dreitlein, Henshall and Pearson20). Nine studies highlighted that RWE is generated from the analysis of real-world data (RWD) (Reference Chan, Nam and Evans3;Reference Clausen, Mighton and Kiflen6–Reference Roberts and Ferguson8;Reference Kirwin, Round, Bond and McCabe11;Reference Lau and Dranitsaris15;Reference Hampson, Towse, Dreitlein, Henshall and Pearson20;Reference Husereau, Nason, Ahuja, Nikaï, Tsakonas and Jacobs37;Reference Bharmal, Katsoulis and Chang42). This RWD can be collected either prospectively or retrospectively (Reference Roberts and Ferguson8;Reference Pearson, Dreitlein, Towse, Hampson and Henshall43). Studies added that such data are not derived from RCT settings (Reference Clausen, Mighton and Kiflen6;Reference Pulini, Caetano, Clautiaux, Vergeron, Pitts and Katz7;Reference Lau and Dranitsaris15;Reference Timbie, Kim and Concannon17), and RWD is routinely collected from clinical practice (Reference Kirwin, Round, Bond and McCabe11;Reference Pearson, Dreitlein, Towse, Hampson and Henshall43). Two studies identified the primary sources of RWD as electronic health records (EHRs), claims databases, pragmatic trials, and patient registries (Reference Roberts and Ferguson8;Reference Lau and Dranitsaris15). Smartphones, wearable devices, and survey data were also identified as potential tools for generating RWD (Reference Roberts and Ferguson8).
Reason for using RWE
Studies recognized that although payers rely primarily on evidence generated from RCTs, there is growing interest in incorporating RWE in payer decision-making for health technologies, including pharmaceutical and medical devices (Reference Clausen, Mighton and Kiflen6;Reference Roberts and Ferguson8;Reference Bharmal, Katsoulis and Chang42). Four studies explicitly stated that RWE complements RCTs by strengthening the evidence that payers need to make funding and formulary decisions (Reference Clausen, Mighton and Kiflen6;Reference Roberts and Ferguson8;Reference Timbie, Kim and Concannon17;Reference Brixner, Biskupiak and Oderda36). Although RCTs demonstrate the efficacy of a technology, they are costly, focus on relatively short timeframes, and target narrowly defined populations, limiting their generalizability (Reference Clausen, Mighton and Kiflen6;Reference Roberts and Ferguson8;Reference Timbie, Kim and Concannon17;Reference Husereau, Nason, Ahuja, Nikaï, Tsakonas and Jacobs37). In contrast, RWE can address these limitations by using data from routine clinical practice, offering broad patient representation and greater external validity (Reference Chan, Nam and Evans3;Reference Hampson, Towse, Dreitlein, Henshall and Pearson20;Reference Mohseninejad, van Gils, Uyl-de Groot, Buskens and Feenstra38). RWE may also be less costly to produce than evidence from RCTs because data can be collected from existing databases such as clinical registries (Reference Chan, Nam and Evans3;Reference Timbie, Kim and Concannon17). Other studies added that the generation of RWE is not restricted to short study timeframes (Reference Chan, Nam and Evans3;Reference Clausen, Mighton and Kiflen6;Reference Timbie, Kim and Concannon17) and can be used to illustrate the clinical and economic value not captured in trials (Reference Chan, Nam and Evans3;Reference Clausen, Mighton and Kiflen6;Reference Hampson, Towse, Dreitlein, Henshall and Pearson20). Comparative effectiveness, cost analyses, treatment adherence, and patient-reported outcomes were highlighted as critical for payer decision making (Reference Clausen, Mighton and Kiflen6;Reference Roberts and Ferguson8;Reference Timbie, Kim and Concannon17;Reference Brixner, Biskupiak and Oderda36;Reference Husereau, Nason, Ahuja, Nikaï, Tsakonas and Jacobs37). RWE was viewed as valuable when other forms of evidence are not available because of rare indications and small study samples, or when data on key variables are not collected in clinical trials (Reference Kovács, Kaló and Daubner-Bendes12;Reference Timbie, Kim and Concannon17;Reference Husereau, Nason, Ahuja, Nikaï, Tsakonas and Jacobs37).
Barriers limiting the use of RWE in payer decision making
Several studies identified key barriers to the use of RWE in payer decision making (Box 1). The most cited concern was the quality of evidence generated (Reference Pulini, Caetano, Clautiaux, Vergeron, Pitts and Katz7;Reference Roberts and Ferguson8;Reference Timbie, Kim and Concannon17;Reference Husereau, Nason, Ahuja, Nikaï, Tsakonas and Jacobs37;Reference Pearson, Dreitlein, Towse, Hampson and Henshall43), particularly issues of bias and confounding, noted in seven studies (Reference Chan, Nam and Evans3;Reference Clausen, Mighton and Kiflen6;Reference Roberts and Ferguson8;Reference Lau and Dranitsaris15;Reference Hampson, Towse, Dreitlein, Henshall and Pearson20;Reference Mohseninejad, van Gils, Uyl-de Groot, Buskens and Feenstra38;Reference Pearson, Dreitlein, Towse, Hampson and Henshall43). These concerns largely stem from the non-randomized nature of RWE, lack of standardization in study design and analysis, limited transparency, and insufficient expertise to generate and analyze RWE. Reporting, selection, and information biases were also reported in two studies (Reference Lau and Dranitsaris15;Reference Hampson, Towse, Dreitlein, Henshall and Pearson20). As Hampson et al. emphasized, these challenges are evident in observational studies because they are often seen as less robust than RCTs (Reference Hampson, Towse, Dreitlein, Henshall and Pearson20).
Data access was another major barrier, cited in seven studies (Reference Clausen, Mighton and Kiflen6–Reference Roberts and Ferguson8;Reference Timbie, Kim and Concannon17;Reference Hampson, Towse, Dreitlein, Henshall and Pearson20;Reference Husereau, Nason, Ahuja, Nikaï, Tsakonas and Jacobs37;Reference Pearson, Dreitlein, Towse, Hampson and Henshall43). Four studies highlighted that legal and privacy constraints, especially regarding patient-identified data, limited timely data sharing (Reference Pulini, Caetano, Clautiaux, Vergeron, Pitts and Katz7;Reference Roberts and Ferguson8;Reference Hampson, Towse, Dreitlein, Henshall and Pearson20;Reference Husereau, Nason, Ahuja, Nikaï, Tsakonas and Jacobs37). Additionally, it can take 3–7 years to generate prospective RWE and 2–3 years to access it, making it difficult to use up-to-date RWE to inform payer decisions throughout the lifecycle of products (Reference Clausen, Mighton and Kiflen6;Reference Pearson, Dreitlein, Towse, Hampson and Henshall43). High costs of data access and generation, especially for establishing registries, were also described as prohibitive (Reference Lau and Dranitsaris15;Reference Brixner, Biskupiak and Oderda36;Reference Mohseninejad, van Gils, Uyl-de Groot, Buskens and Feenstra38).
Other factors highlighted as limiting the use of RWE by payers were incompleteness and inaccuracy in data sets (Reference Pulini, Caetano, Clautiaux, Vergeron, Pitts and Katz7;Reference Hampson, Towse, Dreitlein, Henshall and Pearson20;Reference Pearson, Dreitlein, Towse, Hampson and Henshall43), due to human error in recording data, omission of data points, and gaps in patient’s medical histories (Reference Lau and Dranitsaris15). Interoperability issues were noted in two studies (Reference Timbie, Kim and Concannon17;Reference Husereau, Nason, Ahuja, Nikaï, Tsakonas and Jacobs37), with inconsistencies in outcome definitions and data formats across platforms making it difficult to integrate data sets (Reference Husereau, Nason, Ahuja, Nikaï, Tsakonas and Jacobs37).
Finally, the absence of proper standards of reporting of results raises concerns about potential data mining and undesirable research practices (Reference Roberts and Ferguson8;Reference Timbie, Kim and Concannon17;Reference Hampson, Towse, Dreitlein, Henshall and Pearson20;Reference Husereau, Nason, Ahuja, Nikaï, Tsakonas and Jacobs37). These limitations contribute to payer mistrust of RWE and the continued preference for RCT evidence (Reference Clausen, Mighton and Kiflen6;Reference Husereau, Nason, Ahuja, Nikaï, Tsakonas and Jacobs37).
How is/was RWE used in healthcare payer/manufacturer funding arrangements
Studies emphasized the growing interest among payers in the use of RWE to make coverage decisions (Reference Clausen, Mighton and Kiflen6;Reference Roberts and Ferguson8;Reference Brixner, Biskupiak and Oderda36;Reference Abu-Shraie, Alhammad, Balkhi and Al-Jedai41). Several studies mentioned that for payers, RWE is specifically considered when making initial coverage decisions at the launch of a health technology and when these decisions are being reassessed, typically after a product has had substantial time in the market (Reference Chan, Nam and Evans3;Reference Clausen, Mighton and Kiflen6;Reference Roberts and Ferguson8;Reference Hampson, Towse, Dreitlein, Henshall and Pearson20;Reference Brixner, Biskupiak and Oderda36;Reference Husereau, Nason, Ahuja, Nikaï, Tsakonas and Jacobs37). There was consensus that RWE is used to inform coverage, pricing, and reimbursement negotiations between payers and manufacturers (Reference Chan, Nam and Evans3;Reference Clausen, Mighton and Kiflen6–Reference Roberts and Ferguson8;Reference Kirwin, Round, Bond and McCabe11;Reference Lau and Dranitsaris15;Reference Timbie, Kim and Concannon17;Reference Hampson, Towse, Dreitlein, Henshall and Pearson20;Reference Husereau, Nason, Ahuja, Nikaï, Tsakonas and Jacobs37;Reference Pearson, Dreitlein, Towse, Hampson and Henshall43). Two studies noted that the initial decision may rely on epidemiological data to define target populations and estimate technology costs (Reference Roberts and Ferguson8;Reference Brixner, Biskupiak and Oderda36). Several studies reported that RWE also supports reimbursement schemes that address uncertainties around a product’s real-world performance (Reference Chan, Nam and Evans3;Reference Clausen, Mighton and Kiflen6–Reference Roberts and Ferguson8;Reference Kirwin, Round, Bond and McCabe11;Reference Kovács, Kaló and Daubner-Bendes12;Reference Brixner, Biskupiak and Oderda36).
Of the ten studies focused on reimbursement schemes, most cited risk-sharing agreements (RSAs)/managed entry agreements (MEAs) as the principal type of coverage schemes (Reference Clausen, Mighton and Kiflen6;Reference Pulini, Caetano, Clautiaux, Vergeron, Pitts and Katz7;Reference Kirwin, Round, Bond and McCabe11;Reference Kovács, Kaló and Daubner-Bendes12;Reference Timbie, Kim and Concannon17;Reference Brixner, Biskupiak and Oderda36;Reference Mohseninejad, van Gils, Uyl-de Groot, Buskens and Feenstra38–Reference Abu-Shraie, Alhammad, Balkhi and Al-Jedai41). Three studies pointed out that outcome-based agreements (OBAs), a subset of MEA, were effective at mitigating financial risk by collecting postlaunch clinical data (Reference Gray and Kenney28;Reference Brixner, Biskupiak and Oderda36;Reference Abu-Shraie, Alhammad, Balkhi and Al-Jedai41). Relatedly, coverage with evidence development arrangements that generate RWE under research settings was also noted as desirable MEA (Reference Kirwin, Round, Bond and McCabe11;Reference Kovács, Kaló and Daubner-Bendes12;Reference Brandes, Schwarzkopf and Rogowski39). However, financial-based MEA, which focuses on cost containment, is more commonly used compared with OBA because they are easier to implement (Reference Abu-Shraie, Alhammad, Balkhi and Al-Jedai41).
Overall, three studies agreed that MEA promotes efficiency and sustainability in the health system by guiding resource allocation and defunding costly, low-value technologies (Reference Chan, Nam and Evans3;Reference Clausen, Mighton and Kiflen6;Reference Husereau, Nason, Ahuja, Nikaï, Tsakonas and Jacobs37). Another three studies highlighted the challenges around establishing OBA, (Reference Gray and Kenney28;Reference Brixner, Biskupiak and Oderda36;Reference Abu-Shraie, Alhammad, Balkhi and Al-Jedai41) including monitoring outcomes, patient adherence tracking, contract complexity and costs, lengthy negotiations, and methodological issues such as endpoint selection, identifying target populations, and appropriate sample sizes for analysis (Reference Gray and Kenney28;Reference Brixner, Biskupiak and Oderda36;Reference Abu-Shraie, Alhammad, Balkhi and Al-Jedai41).
In terms of the application of RWE, four studies explained that little is known about its use in payer decision-making and coverage arrangements (especially in oncology) because this information is embedded in confidential contractual agreements between payers and manufacturers that are restricted from public access (Reference Gray and Kenney28;Reference Brixner, Biskupiak and Oderda36;Reference Brandes, Schwarzkopf and Rogowski39;Reference Xoxi, Rumi and Kanavos40). One study added that only 17 percent of the 99 stakeholder organizations included in their study had experience using RWE in OBA (Reference Brixner, Biskupiak and Oderda36).
Types of RWE used and utility of various forms of RWE
In examining the role of RWE in payer decision-making and contractual arrangements, four studies (Reference Gray and Kenney28;Reference Brixner, Biskupiak and Oderda36;Reference Mohseninejad, van Gils, Uyl-de Groot, Buskens and Feenstra38;Reference Brandes, Schwarzkopf and Rogowski39) compared different RWD sources: registry data, claims data, and EHRs. Registry data reflect clinical practices, capture larger populations than RCTs, and can be used to estimate real-world cost and effectiveness (Reference Mohseninejad, van Gils, Uyl-de Groot, Buskens and Feenstra38;Reference Brandes, Schwarzkopf and Rogowski39). However, registries are often costly to establish (Reference Brandes, Schwarzkopf and Rogowski39), and the evidence is susceptible to bias due to lack of randomization (Reference Mohseninejad, van Gils, Uyl-de Groot, Buskens and Feenstra38). Claims data, typically from insurance schemes, were considered more appropriate for MEA that seeks to mitigate uncertainty around the utilization and cost of new health technologies but are less useful for agreements that require evidence on safety or clinical effectiveness (Reference Gray and Kenney28;Reference Brandes, Schwarzkopf and Rogowski39). It was also argued that claims data are more readily available and less costly than data generated from registries and clinical trials, requiring minimal input from patients and clinicians (Reference Gray and Kenney28;Reference Brixner, Biskupiak and Oderda36;Reference Brandes, Schwarzkopf and Rogowski39). For clinical effectiveness and safety, EHR data were preferred by payers because information on clinical variables is likely to be recorded (Reference Gray and Kenney28;Reference Brandes, Schwarzkopf and Rogowski39).
Framework/tools that incorporate RWE into payer decision-making and reimbursement schemes
Some studies proposed frameworks on how RWE can be incorporated into payer decision-making and reimbursement schemes (Reference Kirwin, Round, Bond and McCabe11;Reference Kovács, Kaló and Daubner-Bendes12;Reference Xoxi, Rumi and Kanavos40). One study developed a lifecycle HTA framework that provides guidelines for payers to evaluate new evidence (including RWE) on health technologies through decision rules that are outlined in conditional market access agreements aimed at sharing risk and mitigating uncertainty in product performance (Reference Kirwin, Round, Bond and McCabe11). Another study also proposed a framework that allows payers and pharmaceutical manufacturers to develop and optimally utilize RWE in funding decisions (Reference Pearson, Dreitlein, Towse, Hampson and Henshall43). This framework is grounded in the principles of transparency, communication, and collaboration among stakeholders, which in turn influence the process of developing RWE. Using a similar approach, a third study proposed a new pathway for value-based managed entry agreements using the existing pricing and reimbursement mechanism (Reference Xoxi, Rumi and Kanavos40). They argued that this pathway can be adapted to the health systems in other countries with special consideration for factors such as the details of the MEA and the process of evidence generation (Reference Xoxi, Rumi and Kanavos40).
Discussion
This review provided a narrative synthesis of how RWE has been used by healthcare payers, the barriers to use, and how RWE is currently used in funding arrangements. It highlighted the types of RWE considered by payers and described how RWE has been factored into funding arrangements. The review revealed payers’ interest in incorporating RWE into coverage decisions and funding arrangements for new health technologies, especially for cancer drugs and other pharmaceutical and medical devices (Reference Clausen, Mighton and Kiflen6;Reference Roberts and Ferguson8;Reference Timbie, Kim and Concannon17).
Payers agree that RWE can supplement evidence from RCTs to build the evidentiary case for reimbursing health technologies, where there is uncertainty around long-term clinical effectiveness, safety, and real-world cost-effectiveness (Reference Clausen, Mighton and Kiflen6;Reference Roberts and Ferguson8;Reference Timbie, Kim and Concannon17;Reference Brixner, Biskupiak and Oderda36). RWE is also preferred when other forms of evidence are not available because of rare disease indications, small study samples, or when data on key variables are not collected in clinical trials. RWE studies are particularly attractive because evidence generated is likely to result in generalizability (Reference Chan, Nam and Evans3;Reference Clausen, Mighton and Kiflen6;Reference Roberts and Ferguson8;Reference Timbie, Kim and Concannon17).
The literature explained that the use of RWE in payer decision-making is critical at two junctures: (i) when making initial funding decisions and (ii) when reassessing initial funding decisions. Funding arrangements are often in the form of managed entry agreements that allow conditional access to new health technologies, as RWE around the cost and effectiveness is collected and evaluated (Reference Brandes, Schwarzkopf and Rogowski39;Reference Xoxi, Rumi and Kanavos40). The goal of these agreements is to manage uncertainty through risk-sharing between payers and technology developers (Reference Brandes, Schwarzkopf and Rogowski39).
Healthcare payers also acknowledged the many barriers that may limit the use of RWE in payer decision making. These barriers stem from the lack of standard scientific methods of data collection, analysis, and reporting of RWE (Reference Clausen, Mighton and Kiflen6;Reference Pulini, Caetano, Clautiaux, Vergeron, Pitts and Katz7;Reference Lau and Dranitsaris15;Reference Timbie, Kim and Concannon17;Reference Hampson, Towse, Dreitlein, Henshall and Pearson20;Reference Husereau, Nason, Ahuja, Nikaï, Tsakonas and Jacobs37). These challenges give rise to concerns of bias and confounding, making RWE less reliable for scientific research and payer decision-making. They also justify the preference for RCT evidence in payer decision-making (Reference Clausen, Mighton and Kiflen6). Legal barriers around data privacy and data access also limit the availability of RWE for decision-makers (Reference Pulini, Caetano, Clautiaux, Vergeron, Pitts and Katz7;Reference Roberts and Ferguson8;Reference Hampson, Towse, Dreitlein, Henshall and Pearson20;Reference Husereau, Nason, Ahuja, Nikaï, Tsakonas and Jacobs37).
To address these challenges, payers have called for greater collaboration with manufacturers in the generation and use of RWE (Reference Pearson, Dreitlein, Towse, Hampson and Henshall43). Standardizing methods of data collection and analysis are desirable to ensure that RWE is trustworthy (Reference Hampson, Towse, Dreitlein, Henshall and Pearson20). Researchers have also proposed decision frameworks that utilize RWE in payer coverage decision-making throughout the lifecycle of new health technologies (Reference Kirwin, Round, Bond and McCabe11;Reference Pearson, Dreitlein, Towse, Hampson and Henshall43). These frameworks provide a structured approach to generating and assessing RWE in payer decision-making (Reference Kirwin, Round, Bond and McCabe11;Reference Pearson, Dreitlein, Towse, Hampson and Henshall43). Although the literature provides several recommendations on conducting and reporting RWE (Reference Thokala, Devlin and Marsh44–Reference Johnson, Crown, Martin, Dormuth and Siebert47), researchers have reported a lack of consensus on these guidelines (Reference Hampson, Towse, Dreitlein, Henshall and Pearson20). This highlights the need to consolidate and standardize RWE research and reporting practices to better serve various stakeholders, including payers.
An important finding of this scoping review is that published examples of how RWE is incorporated into payer decision-making and funding arrangements are not readily available. This is because contractual agreements between payers and health technology developers are often confidential and not publicly available (Reference Gray and Kenney28;Reference Brixner, Biskupiak and Oderda36;Reference Brandes, Schwarzkopf and Rogowski39;Reference Xoxi, Rumi and Kanavos40). As a result, stakeholders such as payers, manufacturers, and researchers have limited information on how best to incorporate RWE in decision-making and coverage arrangements. Options for disseminating successful cases that do not violate contractual agreements should be explored. This scoping review highlighted the usefulness of certain data sources for generating RWE, including registries, insurance claims, and EHRs. However, the included studies cautioned that payers should be mindful of the advantages and limitations of each source of data.
The results of this scoping review are consistent with a previous US-based review that examined the use of RWE in payer decision-making, with a specific focus on next-generation sequencing tests (Reference Deverka, Douglas and Phillips25). The review highlighted the value of RWE in outcomes-based contracts, the importance of data sharing, integrating clinical and genomic data, and the need for regulatory and practical solutions to support the use of RWE (Reference Deverka, Douglas and Phillips25). However, the narrow focus on a single country and clinical intervention limits its scope, whereas this current review adopts a much broader focus and is not restricted to a particular country or health intervention.
There are a few limitations of this study. First, a risk-of-bias assessment to evaluate the quality of the included studies in this scoping review was not conducted. A risk-of-bias assessment is not typically conducted for scoping reviews, so the authors cannot make a definitive statement on the quality of the included articles. Second, many of the included studies were qualitative with small sample sizes and specific to a region. The individual study findings may not be generalizable, but together they are valuable in summarizing the use of RWE among healthcare payers. Third, we did not review the gray literature, including reports and websites of agencies known to be associated with payer decision-making and coverage decisions. Although these may be useful sources of information, they were beyond the scope of this review. Fourth, this review did not differentiate between RWE for medical devices and medicines, as few studies on medical devices were identified in the scoping review to support such a distinction. It is acknowledged that RWE may be handled differently for each type of health technology, and so this can be addressed in future research when more relevant studies become available.
Conclusions
This scoping review highlighted the growing interest in RWE among payers to inform funding and reimbursement decisions for health technologies, particularly for cancer drugs and other pharmaceuticals. RWE provides evidence on long-term clinical effectiveness and safety, real-world cost-effectiveness, and budget impact as well as complements evidence generated from RCTs. Although RWE is more generalizable than evidence generated from RCTs, it is often subject to bias and confounding due to poor study designs and methods, which limit its use in payer decision-making. Managed entry agreements are used by payers and manufacturers to allow the entry of health technologies on the market while allowing for the generation and collection of evidence on a technology’s performance in a clinical setting to inform future reimbursement decisions. They also help to mitigate the risks borne by payers and manufacturers when the clinical and cost-effectiveness of a new technology remains uncertain. Examples of how these agreements function are in contractual arrangements between payers and manufacturers, which are seldom available to the public. Access to these agreements can help researchers better understand how RWE informs reimbursement decisions for new health technologies.
Supplementary material
The supplementary material for this article can be found at http://doi.org/10.1017/S0266462325100445.
Author contribution
All authors contributed to the conceptualization and development of the final manuscript. L.M., W.W.L.W., C.C., and K.K.W.C developed the study protocol. L.M., C.C., and K.K.W.C developed and executed the search strategy. D.L. and J.Z conducted the title/abstract review, full-text review, and data extraction; L.M. provided support and resolved discrepancies. D.L. prepared the final manuscript. L.M. and W.W.LW. provided additions and editorial feedback to the final manuscript. All authors approved the final manuscript.
Funding statement
This work was supported by the Ontario Institute for Cancer Research (OICR) (grant no. P.CT.927).
Competing interests
The authors have no conflicts of interest to disclose.
Open access
