What Does the Future of Phase III Clinical Trials Look Like in a Population Health Model?

Population health management strategies can change the way clinical care reaches patients. These approaches take into account the interaction among health outcomes, determinants, and policy interventions to impact the health of both individuals and communities.1,2 By including population-level components like health assessments, health promotion, and outcome management, an effective population health model can ensure specific individuals receive appropriate care while improving the health status of the community. Population health management, therefore, has become an appealing strategy for an increasing number of health systems, insurers, employers, and public policy stakeholders.

The shift toward population health may influence clinical research in several meaningful ways. First, research agendas may be refocused to investigate the effectiveness of population health interventions and multilevel determinants of health status.3,4 Second, the scope of information available for research purpose may be expanded. In particular, population health management often involves leveraging technology and electronic data to track patients and outcomes.5,6 Information accumulated in these databases may also have applications for observational studies, such as patient registries.

However, it is unclear exactly which population health trends will impact clinical trials. The Institute of Medicine (IOM) recently published a report detailing the need for “disruptive innovation” in the clinical trials infrastructure, particularly to reduce the separation between healthcare delivery and research.7 The authors maintain that clinical trials will continue to be essential in the process of developing new therapies or new knowledge of existing therapies, thus calling for a need to increase their efficiency, effectiveness, and generalizability to treatment populations. In this regard, population health management may support this objective by facilitating improved integration between clinical care and trials. Furthermore, population health management may also lead to new trial opportunities due to increased availability of patient information and novel approaches to reaching study participants.

The purpose of the project described in this article was to explore the future of Phase III clinical trials in the context of a population health management framework. We focus specifically on Phase III trials and not the broader topic of clinical research, since initial assumptions regarding the outlook of Phase III trials are less defined than Phase IV trials, registries, or other types of observational studies. That being said, the trends identified here may also affect clinical research more broadly.


The study approach consisted of three phases. The first phase of this study included a non-systematic review of evidence that may be accessed in real-world situations, including but not limited to scholarly publications. This evidence review was undertaken by three trained health services researchers. Initially, PubMed was searched using terminology related to trials and population health. A targeted review was then conducted for select studies that appeared particularly relevant. In addition, a focused web search including key healthcare research websites and general web searches were undertaken to identify other appropriate web-only (not peer-reviewed) publications and references. Findings from the evidence review were summarized and seven distinct trends were identified.

For the second phase of the study, eight healthcare professionals from a variety of backgrounds were identified and asked to participate in an electronic survey. These individuals were selected to provide insight on the future of clinical trials from the perspectives of healthcare professionals, clinical researchers, the pharmaceutical industry, academic institutions, state government, and other commercial interests. The survey consisted of 10 questions designed to assess respondents’ knowledge and opinions regarding potential trends affecting the future of clinical trials. These questions were based on information gleaned from the evidence review. In addition to covering predefined topics, panelists were encouraged to comment on any other trends that may have been overlooked. Survey results were aggregated to present descriptive trends only, as no additional statistical analysis was performed.

For the final phase of the study, an expert panel meeting comprised of the healthcare professionals who participated in the survey was convened. Based on results from the survey and evidence review, a discussion guide was developed to help structure the meeting to be an opportunity for respondents to elaborate on survey responses and share other perspectives on how a population health framework may impact the future of clinical trials.

The expert panel meeting was moderated by two health services researchers over a period of several hours. The panel was held on the first day of the annual Cerner Health Conference in Kansas City, Mo., and panelists were offered complimentary registration in exchange for their participation. No other compensation was provided.


Evidence Review

Our literature review suggests there is no clear consensus on how the extant clinical trial process can be upgraded to fit within a population health framework. However, we did identify seven trends that may influence the future of clinical trials (see Table 1*):

  • Virtual Trials. To reduce the burden of study visits, virtual trials use web services and/or telemonitoring to carry out most or all of the research. For example, in Pfizer’s REMOTE trial, participants were recruited via the web, screened for eligibility using web-based questionnaires, and entered a run-in phase requiring electronic diaries.8 Despite initial recruitment challenges, a revamped REMOTE 2.0 trial is being planned in Europe.9 Furthermore, with increased use of smart phones and remote health monitoring devices (e.g., Fitbit activity trackers and, more recently, the Apple Watch), virtual trials may be a promising strategy.10
  • Genomics. Although not a novel concept, genomics may have increasing prominence for clinical trials with the shift toward personalized medicine, which requires development of therapies and diagnostic tests targeting specific genetic characteristics.8 Clinical trials that investigate these biomarkers generally employ innovative study designs to identify sufficient patients and capture relevant data.11
  • Contract Research Organizations (CROs). While CROs have been utilized heavily for the last several years to execute complex trials quickly and at lower costs, study sponsors will have to collaborate with CROs in a much more strategic manner.12 Management of the relationships between industry sponsors, CROs, and study sites will likely be increasingly dependent on real-time data and technology.13 By leveraging their expertise and relationships, CROs may be able to execute trials more effectively and at a lower cost than industry sponsors.
  • Globalization. According to one study, approximately one out of every three clinical trials is now being conducted in developing countries.14 The expansion of research in these locations is due to multiple factors, including costs of trials in Western countries and challenges to accruing sufficient sample size.15 Globalization of trials is providing these populations with an opportunity to participate in research and access therapies that would otherwise be unavailable. Additionally, this benefits research efforts by increasing the heterogeneity of study populations, allowing results to be more generalizable.
  • Translational Research. Improving the speed at which healthcare discoveries are applied to meet clinical needs and improve patient care is a priority for the National Institutes of Health.16 According to some estimates, the average time for an innovation in research to reach clinical practice is 17 years.17 Given the length of time between discovery and application, investment is being made in shortening this duration.14
  • Data Analytics. Advanced data modeling is required to effectively leverage the increasing amount of participant-related information that is available due to greater use of technology, especially electronic healthcare records (EHRs). For example, analytics can help researchers better define study designs and outcomes to reduce the occurrence of false negative trials.15 Additionally, some clinical trials may be supplemented or even replaced by data collected from networked groups of patients.8
  • Crowd-Sourcing. The drug development process may be increasingly influenced by patient advocates and clinical trials that are supported by groups of patients, rather than industry sponsors.8 Often referred to as crowd-sourcing, this bottom-up approach may be especially beneficial for advancing the study of rare diseases and orphan drugs that have traditionally struggled with funding and recruitment.18–20 Social media may be key in helping facilitate crowd-sourcing by connecting groups of people with common interests in the clinical trial process.

Survey Findings

Respondents to the survey rated all of the trends identified in the evidence review as being at least slightly important to the future of clinical trials (see Figure 1*). The most highly rated trends were also those most familiar to the panelists: genomics, data analytics, translational research, and CRO partnerships. The average importance rating was lowest for crowd-sourcing and virtual trials.

Respondents indicated that many aspects of the clinical trial process would change in the next five to 10 years (see Figure 2*). In particular, all but one of the eight respondents signified that healthcare provider awareness of clinical research opportunities and participant recruitment would change. Other factors were also rated as likely to change, such as study sponsorship/funding (by six respondents) and institutional review board (IRB) approval process (by five respondents).

When asked what factors would influence the future conduct of clinical trials, respondents rated those related to technology as most significant (see Figure 3*). Specifically, so-called “big data” usage and analytics had the highest average rating, followed by increased use of EHRs and new technology. Government legislation/regulation and healthcare reform were rated least influential.

All eight respondents indicated that privacy or other data security concerns would be barriers to changes in the clinical trial process (see Figure 4*). All but one or two, respectively, also believed that lack of trial funding and regulations on clinical research were inhibiting factors. In contrast, only two rated technology adoption and access to data as barriers. Many also did not indicate the lack of awareness regarding research opportunities, by either patients or healthcare providers, to be a barrier.

Expert Panel Discussion

Overall, when the survey respondents were later gathered for a panel discussion, they agreed that population health will have some effect on clinical trials; however, opinions were mixed regarding the extent of impact. Although it was acknowledged that population health tools, such as outreach programs and patient dashboards, may help target specific populations, panelists indicated the population health model focuses largely on prevention and chronic conditions. Thus, it may have limited applicability for interventional trials or studies of rare diseases (although access to patient data may be used to target orphan diseases).

Findings from the evidence review were used to initiate discussion among panel members, and key findings are listed below. The group believed that several types of research partnerships will increase and that collaboration across sites, sponsors, and CROs will become even more essential. In addition, state, federal, and other third-party payers may have a more central role. Specifically, partnerships with these entities would facilitate better integration of data from multiple sources that can be leveraged for clinical trials.

Changing referral patterns may become a barrier for clinical trials. Given the emergence of accountable care organizations (ACOs) and the dynamic landscape of healthcare delivery, providers may be reluctant to share data and/or refer patients outside their own network. Thus, clinical trials may be impacted as providers become less inclined to refer patients for studies outside their network and are motivated to keep trial referrals within their organizations.

The panelists indicated new clinical trial technology may only be appropriate in specific contexts. For example, trends such as virtual trials or telehealth may be challenging for many clinical studies, especially those involving clinical or laboratory measures. Instead, these technical advances may be most beneficial in the collection of patient-reported outcomes, where participants are equipped with new methods of communicating their experiences and opinions.

Leveraging technology for certain therapeutic areas and study designs holds some promise. Specifically, virtual trials may be more successful for monitoring more common and predictable conditions (e.g., diabetes), as opposed to specialty trials in areas such as oncology, which can be more complex and require more oversight. Panelists were also optimistic that technological trends, such as telehealth, could benefit rural and Medicaid populations that may otherwise lack access to clinical sites.

The panelist believed that prospective study participants will eventually be more engaged in clinical research. The concept of crowd-sourcing for clinical trials was found to be intriguing, especially as a way to gauge participation interest. Specifically, social media may also increase public awareness of, and attraction to, clinical trials, especially in terms of online patient communities that connect visitors for education and support.

The shift toward patient-centered care has implications for clinical trials. Panelists suggested that trials should also focus on individuals’ health goals and autonomy. The increased use of technology across the general population is also improving how participants can access information and make more informed healthcare decisions.

Although big data could benefit clinical trials, there are obstacles worth noting. As ACOs become more prevalent, more complex and voluminous data will be available for research purposes. However, fully leveraging this information requires a workforce capable of managing large datasets and performing sophisticated analyses. Furthermore, the integration of multiple data sources presents other challenges, such as those to data governance and quality control efforts.


The findings of our multifaceted exploration suggest changes associated with the population health framework will impact clinical trials in the next decade, and that there is an opportunity to leverage these concepts toward the evolution of the clinical trial process. However, instead of widespread change, our findings suggest clinical trials will advance in a piecemeal fashion, with new approaches implemented only under the right circumstances; that is, for specific study outcomes, therapeutic areas, and target populations.

It is very likely that the shift toward population health will result in greater use of technology, an abundance of data on research participants, and novel approaches to executing clinical trials. While some of these trends are already in practice, further advancement requires an improved ability to leverage big data, sufficient numbers of skilled personnel, and a reimagined approach to ensuring participant protections and regulatory requirements. In this regard, an increase in collaborative partnerships across various stakeholders is likely to occur as researchers consider how to apply concepts from population health management toward innovations in the clinical trial process.



  1. Kindig DA. 2007. Understanding population health terminology. Milbank Q 85(1):139–61.
  2. Kindig D, Stoddart G. 2003. What is population health? Am J Public Health 93(3):380–3.
  3. Milat AJ, King L, Newson R, et al. 2014. Increasing the scale and adoption of population health interventions: experiences and perspectives of policy makers, practitioners, and researchers. Health Res Policy Syst 12:18.
  4. Hawe P, Di Ruggiero E, Cohen E. 2012. Frequently asked questions about population health intervention research. Can J Public Health 103(6):e468– 71.
  5. Diamond CC, Mostashari F, Shirky C. 2009. Collecting and sharing data for population health: a new paradigm. Health Aff (Millwood) 28(2):454–66.
  6. Friedman DJ. 2006. Assessing the potential of national strategies for electronic health records for population health monitoring and research. Vital Health Stat 2 (143):1–83.
  7. Califf RM, Filerman GL, Murray RK, Rosenblatt M. 2012. The clinical trials enterprise in the United States: a call for disruptive innovation. (Accessed October 1, 2014) www.iom.edu/~/media/Files/Perspectives-Files/2012/Discussion-Papers/Drug%20Forum-Call%20 for%20CTE.pdf
  8. Gossen R. 2013. Partnerships 2013: the great inflection of clinical trials. (Accessed October 1, 2014) http://rebarinteractive.com/partnerships-clinical-trials-recap/
  9. Taylor N. 2012. Pfizer director defends virtual trial after recruitment struggle. (Accessed May 6, 2015) www.outsourcing-pharma.com/Clinical-Development/Pfizer-director-defends-virtual-trial-after-recruitment-struggle
  10. Taylor NP. 2013. Pfizer: ‘Bring your own device’ is coming to clinical trials. (Accessed October 1, 2014) www.fiercebiotechit.com/story/pfizer-sees-smartphones- taking-center-stage-clinical-trials/2013-12-02
  11. Kelloff GJ, Sigman CC. 2012. Cancer biomarkers: selecting the right drug for the right patient. Nat Rev Drug Disc 11(3):201–14. 1
  12. Halloran L. 2014. The clinical research crystal ball: looking at the future. (Accessed October 1, 2014) www.clinicalleader.com/doc/the-clinical-research-crystal-ball-looking-at-the-future-0001
  13. Tufts Center for the Study of Drug Development. 2014. Tufts CSDD Outlook 2014. (Accessed May 6, 2015) http://csdd.tufts.edu/files/uploads/Outlook-2014.pdf
  14. Azeka E, Fregni F, Otavio Costa Auler J. 2011. The past, present and future of clinical research. Clinics (Sao Paulo) 66(6):931–2.
  15. Write R. 2014. The future of clinical trials in an outsourced model. (Accessed October 1, 2014) www.clinicalleader.com/doc/the-future-of-clinical-trials-in-an-outsourced-model-0001
  16. About NCATS. 2014. (Accessed October 1, 2014) www.ncats.nih.gov/about/about.html
  17. Morris ZS, Wooding S, Grant J. 2011. The answer is 17 years, what is the question: understanding time lags in translational research. J R Soc Med 104(12):510–20.
  18. Dragojlovic N, Lynd LD. 2014. Crowdfunding drug development: the state of play in oncology and rare diseases. Drug Disc Today 19(11):1775–80.
  19. Byrnes JE, Ranganathan J, Walker BL, Faulkes Z. 2014. To crowdfund research, scientists must build an audience for their work. PloS One 9(12):e110329.
  20. Perlstein EO. 2013. Anatomy of the Crowd4Discovery crowdfunding campaign. SpringerPlus 2:560.

Kathleen M. Aguilar, MPH, (kat.aguilar@outlook.com) was a scientist with the Cerner Corporation’s research consulting office in Culver City, Calif., at the time this article was accepted for publication.

Erika Shah, MS, (Erika.Shah@Cerner.com) is a scientist with Cerner.

Kelly J. Ko, PhD, (kko@westhealth.org) was a scientist with Cerner at the time this article was accepted for publication.

[DOI: 10.14524/CR-15-0032]

*To see all figures and/or tables published originally in this article, please visit the full-issue PDF of the April 2016 Clinical Researcher.