Developing competence in biostatistics and research methodology during medical specialty training

Do we equip junior specialists with the skills to conduct research throughout their careers?

The application of an evidence-based approach to patient care underpins modern clinical practice for all medical disciplines. In many specialty training programs, there has been emphasis in recent years on ensuring that trainees develop the skills needed to interpret medical literature and conduct research, as a part of overall training. In turn, it is hoped that these abilities will instil in trainees the capacity and enthusiasm to continue taking an active part in clinical research over their careers. In addition, specialist doctors are ideally placed to direct efforts towards medical education research and research on translation of evidence to practice. In this article, we focus on the need to assist trainees in attaining foundation skills required for participation in research activities, ideally as a stepping stone to leadership in research. This is distinct from looking at tools and strategies (such as journal clubs and critically appraised topic instruments) used to teach critical appraisal of published evidence, although all these concepts are closely linked.1

The case for supporting trainees in learning research skills

It may be no surprise that, when it comes to the knowledge and skills required for junior doctors to undertake clinical research, the evidence suggests that these have not been mastered at the medical student level, and are not likely to be acquired simply by osmosis as training proceeds.2 Several studies highlight the need for innovative learning tools to actively engage junior doctors in biostatistics and research skills learning, and the value of collaborative teaching efforts, combining the expertise of clinicians and biostatisticians.3

Additional learning is necessary, then, to meet the scientific rigor essential for the conduct of high-quality clinical (and other) research by vocational trainee and specialist investigators. Potential waste and challenges to high-quality research arise from weaknesses in study conduct, design and reporting.4 Furthermore, the pressure on doctors in specialty training to fulfil ever-increasing clinical service duties competes with learning in non-clinical areas. Consequently, the imperative to design more efficient and effective teaching strategies, especially for non-core medical expert skills, becomes ever stronger.

Incorporating research competencies into specialty training curricula

Contemporary competency-based curricula now reflect the multifaceted role of medical specialists. Within these, the knowledge and skills needed for basic research literacy are clearly articulated. However, the processes to achieve these and then translate them into research activity are not well defined. The Canadian Medical Education Directive for Specialists (CanMEDS) curriculum framework5 is the most common structure applied to specialty medical training programs in the global setting. Competencies relating to research and biostatistics skills are encapsulated within the Scholar Role. The CanMEDS framework has been adopted by several Australian training bodies, including the Royal Australian and New Zealand College of Ophthalmologists, the Royal Australian and New Zealand College of Radiologists6 (RANZCR, which also trains radiation oncologists) and the Australian Orthopaedic Association. For the Royal Australasian College of Physicians, the relevant learning objectives are encompassed in the Physician Readiness for Expert Practice basic trainee curriculum in Domain 3 (research) and defined within advanced learning curricula for discipline-specific programs.7 Despite these explicit curriculum statements of required knowledge and skills, the teaching of competencies within the Scholar Role or equivalent can be difficult to integrate into typical clinically orientated training programs — a concern expressed by pedagogical experts in the global arena.8 The challenge is magnified by the facts that this skill set can only be fully grasped and refined through application to real projects (ie, doing research), and many clinical supervisors do not feel adequately qualified to teach these subjects.

How might we support attainment of research skills?

Stand-alone practical courses in trial concept development, such as the Australia and Asia Pacific Clinical Oncology Research Development (ACORD) Workshops9 support selected advanced trainee participants in developing research skills. The Royal Australasian College of Surgeons runs a 2-day Critical Literature Evaluation and Research (CLEAR) course10 that is available to trainees and Fellows, and mandatory for trainees in some subspecialties. These types of courses, which typically run externally to other training program elements, may well be helpful for those who attend. However, they are not routinely integrated into training and may not be accessed by the majority, resulting in wide variation in understanding research methods and biostatistics within a specialty stream, and between disciplines. The Faculty of Radiation Oncology, RANZCR, have developed a blended learning approach, integrating activities throughout training for all trainees. Annual centralised workshops, led by biostatisticians and Fellows involved in clinical research, are run in conjunction with the main national trials group meeting. Key examples from the oncology literature (often study outcomes or protocols from that trials group) are used to illustrate specific learning points relating to biostatistics concepts, and to promote small-group discussion. In this model, one goal of alignment with the trials meeting is to increase trainees’ exposure to clinical trial development and conduct, and to provide opportunities for mentorship by senior clinician–scientists for trainees starting their own research.11 Through all stages of the training program, self-directed learning activities are selected from a suite of resources, and recorded in an electronic learning portfolio. Through attending the workshop and taking part in these other learning opportunities, trainees accumulate a mandatory number of research points by the end of their training. In addition, as for some other specialty training programs in Australia, there is a compulsory manuscript submission requirement for their research projects.

Biostatisticians’ input into doctors’ research learning is important because biostatisticians, unlike pure non-medical statisticians, have the necessary skills to bridge the gap between ensuring statistical rigor and placing this in the clinical context. However, in Australia, opportunity for professional biostatistics development is limited, hence the services of experts are in high demand and may be costly. Taking shared responsibility for assisting with junior medical researchers’ education may be constrained by this shortage but is necessary to maintain quality.

Examples of strategies used in Australian and overseas settings to promote research-related learning within medical specialty training programs, and the major benefits and limitations of these, are shown in Box. More than one of these activities may currently be available to trainees on a largely ad hoc basis, depending on the training program. To our knowledge, there are no published recommendations or guidelines on which interventions might best develop the skills that lead to the conduct of good research. The lack of educational literature evaluating the impact of these methods holds back progression in this area. Tools to equip supervisors to facilitate learning, and thus better integrate this throughout specialty training, might reduce reliance on external courses and better contextualise learning. A systematic approach to measuring the research literacy that learners derive from these activities, and determining how this translates into research conduct and leadership, will be valuable. These findings could then inform further development across specialist training programs relating to teaching in the research skills domain, and should increase the chance that high-quality clinician-led research output is maintained into the future. Efforts to develop effective and program-integrated learning opportunities that are accessible by all specialty trainees, regardless of discipline and geographic location, should, in our view, be a priority.