International efforts to standardize guidelines for observational research could move the entire field forward
Regulators are increasingly focusing on safety post-approval and are demanding more data from larger and more-representative patient populations. Payers also want more data to demonstrate cost effectiveness. Physicians are driving the need for data showing quality of life and treatment satisfaction, and demonstration of how this can improve therapy adherence to ultimately produce better clinical outcomes.
Observational data—clinical information arising from real-world treatment regimens of patients—is able to meet all of these objectives and, importantly, can be generalized to the broader population. Observational studies can be well designed, but protocols do not have the strict inclusion/exclusion criteria of a Phase III randomized clinical trial (RCT).
There is significant divergence in the design and conduct of observational studies across the biopharma industry, and this is driven by multiple factors. Diverse stakeholders in such disciplines as pharmacovigilance, clinical development, medical affairs, business/therapy management or health economics initiate, fund and implement observational research. There is a further complexity as some pharma companies often rely upon their local marketing affiliates to identify and implement these studies while others are becoming more centralized in their approach.
There is no uniform industry standard underpinning observational research, such as the well understood standards of “Good Clinical Practice” for RCTs, however, many attempts have been made to address this gap. These include the European Union Medicines Agency’s (EMA) pharmacoepidemiology and pharmacovigilance project (ENCePP), the US Agency for Healthcare Research and Quality’s (AHRQ) “User’s Guide to Registries Evaluating Patient Outcomes” , the International Society for Pharmacoepidemiology (ISPE) Guidelines for Good Pharmacoepidemiology Practices  and the “Strengthening the Reporting of Observational Studies in Epidemiology” (STROBE) statement . Moreover, pharma is uncertain whether the results of observational studies will be accepted by authorities, as observational studies have typically been viewed with skepticism.
Most authorities concentrate on RCT data and value observational studies as inferior. For this and other reasons, there is a lack of awareness within the pharmaceutical industry of the potential of observational research. Few companies have the internal experience and organizational structure to guide their organization in strategically undertaking observational research. By some estimates, only half of organizations have standard operating procedures dedicated to observational research design and conduct.
What must change?
To overcome these problems, consensus-building by stakeholders is of major importance: Dialogue between academics, industry, payer and regulatory bodies is needed. Training and education is a precondition to improve the necessary skills concerning the active undertaking of observational studies; also it is necessary to change attitudes and create more visibility with respect to these studies. It’s critically important that the industry be given a backbone in the form of guidelines and best practices for the conduct of observational studies.
Payers would likely be the stakeholder with the most interest in seeing established guidelines for observational studies as they have a very strong need for real-world outcomes data. In June 2010, the International Society for Pharmacoeconomics and Outcomes Research (ISPOR) set up the Prospective Observational Clinical Studies Good Research Practices Task Force to provide good research practices for prospective observational clinical studies that focus on the effectiveness and/or comparative effectiveness of health care interventions. Regulators, too, are showing a keen interest in this topic. The European Medicines Agency (EMA), for example, has published a position paper stressing some level of convergence between regulatory and paying bodies. Stepwise improvement in interaction between the EMA and payers is planned, whilst recognizing distinct roles and responsibilities.
What could these guidelines look like?
At the top line, stakeholders are looking for guidelines on study design. What type of research approach is appropriate to address the needs of payers and regulators? How can bias and confounding be minimized, and how can the data be representative of the broader population? What channels and tools can be used to capture the data and what are the appropriate quality standards? Are there ethical issues and how can these be managed? What are the requirements for safety reporting? How can publication
standards be upheld and what are appropriate levels of remuneration for study participants? Accepted industry guidelines could resolve most or all of these issues.
Guidelines for observational studies would actually represent a true convergence between the clinical and commercial functions in a biopharmaceutical organization. Combining market access with epidemiology will essentially produce a model of “evidence based marketing,” in which each step of the clinical development process includes input from a commercialization perspective. And as we see more efforts focused on patient centricity—as well as more pay-for-performance arrangements and other forms of conditional authorization — having industry-wide, universally accepted guidelines will play a major role in both the development and commercialization of new therapies. PC
4. von Elm E et al; J Clin Epidemiol. 2008 Apr; 61(4):344-9. PMID: 18313558
About the authors
Louise Parmenter (left) is Senior Director, Strategic Planning, at Quintiles. Morna White is Senior Director, Life Sciences Consulting, at the same organization.