Abstract
The financing of postgraduate medical education (PGME) becomes an important topic. PGME is costly, and in most western countries is partly paid by public funding. One of the models that can help to reduce costs is time-variable PGME. Moving to true outcome-based education can lead to more efficient training programs while maintaining educational quality. We analyzed the financial effects of time-variable PGME by identifying the educational activities of PGME programs and comparing the costs and revenues of these activities in gynecology training as an example. This resulted in a revenue–cost balance of PGME activities in gynecology. As gynecology consists of both surgical and non-surgical parts, this specialty is a good starting point for a training cost analysis that can be used for a more general discussion. Shortening PGME programs without losing educational quality appears to be possible with time-variable structures. However, shortening is only safely possible on those areas in which residents have already obtained the desired level of competence. This means that time can be gained at the expense of those educational activities in which residents generate the highest revenues. We therefore conclude that shorter education with the help of time-variable training schemes leads to overall higher costs at the hospital level.
Introduction
Postgraduate medical education (PGME) programs are costly, and in most western countries they are partly paid for by public funding (Frenk et al. Citation2010). Economic restraints call for new financial models for PGME to reduce costs, which has generated interest in new financial models (Asch and Weinstein Citation2014; O’Rourke Citation2014) to lower educational costs while maintaining quality standards (Iglehart Citation2012). One of the models that is frequently discussed is time-variable outcome-based PGME (Harden et al. Citation1999). Competency-based medical education (CBME) can be used to redefine PGME programs in order to better prepare physicians for future practice (Carraccio et al. Citation2016).
Adjusting educational programs according to pre-defined competency-based outcomes has led to a discussion about the time that is needed to become a competent physician (Frenk et al. Citation2010). Result driven, outcome-based education has the potential to offer more flexibility in educational strategy (Harden, Crosby and Davis Citation1999), but also in educational time. Shortening training programs means reducing the time spent on specific, apparently redundant elements of the training program, and it can therefore be an alternative strategy for reducing the cost of PGME. The financial attractiveness of outcome-based PGME sped up a nation-wide implementation of time-variable PGME in the Netherlands. Two years later, however, we see unintended and unforeseen financial effects that are important to consider in the implementation process of CBME and time-variable PGME.
In the Netherlands, residents are employed by the teaching hospitals which they are affiliated with. Residents receive a salary, and teaching hospitals receive a subsidy from the national government for each resident they employ (IBO Universitair Medische Centra Citation2012). In this article, we looked at the financial effects of time-variable outcome based PGME in gynecology. Our objective was to look at how, and to what extent, alterations in the length of a PGME program with gynecology as a relevant example, would impact on the cost of time-variable training.
Methods
To explore the effects of time-variable PGME on the total costs of medical care in teaching hospitals, we unraveled the structure of PGME programs and described the effects for each part of the training. Although PGME programs may differ between specialties and countries, we can distinguish five categories of resident activities that are generally applicable (Van Baalen and Bosman Citation2012):
Direct patient-related learning. Concerns activities with resident–patient contact in the context of clinical service, for example on the patient ward.
Indirect patient-related learning. These are clinical activities that do not involve direct contact between resident and patient, such as handovers.
Non-patient-related course education. These are educational activities that are usually performed outside the clinical arena, such as training courses.
Scientific education. These are activities related to scientific development, such as participating in research.
Activities related to the learning process itself. These are activities like receiving and giving feedback and portfolio building.
Although the specific time required for each of these five categories may differ, direct patient-related learning plays a major part in all PGME programs, followed by indirect patient-related education. In the absence of a resident, direct patient-related activities can only be performed by either a medical specialist (patient care at specialist level) or by other health care workers (patient care at resident level), often at higher costs. During resident training, these activities are performed under decreasing supervision. The other categories have smaller time requirements (Van Baalen and Bosman Citation2012). Room for shortening training time can be found in patient-related activities for which residents have obtained the required level of competence. These activities may, at least in part, be considered to be redundant, as they are not primarily aimed at the learning process but at providing service. In this case, the merits of the activities performed by residents can be seen as saved costs. Of note, activities aimed at retaining the level of competence should remain a part of the learning process.
Costs and revenues of medical education
In general, residents are paid wages and are causing a variety of costs; in return, they provide substantial service to patients, thus generating revenues for their teaching hospitals (Chandra et al. Citation2014). Van Baalen and Bosman (Citation2012) defined three types of costs and revenues connected to PGME:
Fixed costs: The costs that emerge when a resident is available in a teaching hospital, consisting of wages, employers’ costs, and educational costs.
Variable costs: The costs generated by medical specialists for supervising residents, which are based on a variation in time allocation; time allocated to education cannot, or not as efficiently, be used for clinical service.
Variable revenues: Residents perform activities that benefit the clinical process and ease, or even substitute, the workload of other professionals. These can be either specialist or other (non-specialist) health care workers. This saves costs we label as variable revenues.
The total costs are the difference between costs (fixed and variable) and revenues. These costs depend on the specialty, the experience of the resident, and the specific training site.
As the time reduction will differ between individual learners, we made two assumptions. The first assumption is that the total number of hours of training will proportionally decrease with the shortening of the training program, and that this time can be found in activities in which a resident works either at the level of other hospital workers (non-specialist) or at the level of a specialist. Residents can work at both levels at different moments during the course of their training, assuming they have reached the desired level of competence. The second assumption is that the time reduction will not be saved at the expense of teaching time for scientific education, nor at the expense of activities related to the learning process. We made this assumption because we think that shortening the time spent on these activities will adversely affect the quality of education.
Our assumptions lead to two extreme scenarios that are based on the Dutch context, in which we aim to shorten a 6-year program by almost 6 months. This context requires a shortening of 0.46 year for a 6-year program. In the first scenario, this time is found in the hours that residents are working at non-specialist level. To achieve this time reduction in the context of our example, we decrease the time spend on non-specialist activities from 100% to 88.7%, which resulted in a shortening of 0.46 year based on a 6-year program. In the second scenario, we aimed to reduce the hours worked on the specialist level. Here, the reduction of 0.46 year can only be achieved by skipping all the hours that residents are working at a specialist’s level and complement these with hours spent working at the non-specialist level. In this second scenario, we have to decrease the activities at the specialist level from 100% to 0% and, in addition, decrease the activities at the non-specialist level from 100% to 98.3%, which together results in a shortening of 0.46 year based on a 6-year program.
Results
presents the average financial effects of a 0.46 year shortening in a gynecology program on the teaching hospitals’ budget.
Table 1. Cost effects per educational activity for gynecology.
The first part of the table shows the costs and revenues of the original program of 6 years. These costs and revenues are based on an average that is reported by Van Baalen and Bosman who evaluated multiple training programs in the Netherlands (Van Baalen and Bosman Citation2012). The second and third parts show the effects of both described scenarios. These two scenarios show a range of options for a shorter educational program without losing educational quality; the reality will lie somewhere in between these extremities.
Discussion and conclusions
Based on the first scenario, we can draw several conclusions. First of all, a shorter training program leads to overall lower cost of employment and training, but also to lower revenues. This leads to a net increase of the costs price per resident per training program of (1.061.711–1.053.240) 8.471 Euros. Based on the second scenario, we can draw the following conclusions. When looking at the effects on revenues generated at the level of medical specialists, the cost increments are (1.129.018–1.053.240) 75.778 Euros per resident. As these scenarios are based on two extreme options in the range of possibilities, in reality the effect will lie between 8.471 and 75.778 Euros.
Implications
Shortening of PGME programs does not automatically lead to lower costs. Although it appears possible to shorten PGME programs with the help of CBME frameworks and a time-variable structure, the time is gained in those educational activities in which residents generate the highest revenues.
Training costs, ranging from costs of supervision and materials to direct fixed costs, are generated in all educational activities. Revenues, on the other hand, are only generated in patient-related activities. In these patient-related activities, the revenues are high when residents are able to work with limited supervision.
Our study has limitations. This study shows the financial effects in the Netherlands and is therefore based on a specific context. Reducing the amount of time spent on activities, especially in which residents work at the level of medical specialists, will have a negative effect on the revenues, which, perhaps counter intuitively, leads to overall higher costs for the teaching hospital in a shorter CBME-based curriculum. While we looked at a specific context, we feel that this effect will also apply to other contexts as well.
Learning to work as an independent physician is an important part of medical training. But this financial analysis shows that residents also have a production value. If the production value of residents diminishes, residents become expensive learners.
Notes on contributors
Tiuri R. van Rossum, MSc, is Lecturer at the School of Health Professions Education (SHE), Maastricht University, Maastricht, The Netherlands.
Fedde Scheele, MD, PhD, is a Professor at the Athena Institute, VU University and VU University Medical Center, Amsterdam, The Netherlands, and practices gynecology at the OLVG Teaching Hospital, Amsterdam, The Netherlands.
Henk E. Sluiter, MD, PhD, is a Manager of postgraduate medical education, Deventer Hospital, Deventer, The Netherlands, and practices internal medicine at the department of internal medicine and nephrology, Deventer Hospital, Deventer, The Netherlands.
Peter J. Bosman, MD, PhD, MBM, is an independent management consultant, Bodegraven, The Netherlands.
Lotte Rijksen, MSc MBA, is a consultant, The Dutch Association of Medical Specialists, The Netherlands.
Ide C. Heyligers, MD, PhD, is a Professor at the School of Health Professions Education (SHE), Maastricht University, Maastricht, The Netherlands, and Orthopaedic Surgeon and Manager of Postgraduate Medical Education at the Zuyderland Medical Center, Heerlen, The Netherlands.
Acknowledgements
We would like to thank Lisette van Hulst for her writing assistance.
Disclosure statement
The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the article.
References
- Asch DA, Weinstein DF. 2014. Innovation in medical education. N Engl J Med. 371:794–795.
- Carraccio C, Englander R, Van Melle E, Ten Cate O, Lockyer J, Chan M-K, Frank JR, Snell LS. 2016. Advancing competency-based medical education: a charter for clinician–educators. Acad Med. 91:645–649.
- Chandra A, Khullar D, Wilensky GR. 2014. The economics of graduate medical education. N Engl J Med. 370:2357–2360.
- Frenk J, Chen L, Bhutta ZA, Cohen J, Crisp N, Evans T, Fineberg H, Garcia P, Ke Y, Kelley P. 2010. Health professionals for a new century: transforming education to strengthen health systems in an interdependent world. The Lancet. 376:1923–1958.
- Harden RM, Crosby JR, Davis MH. 1999. AMEE guide no. 14: outcome-based education. Part 1. An introduction to outcome-based education. Med Teach. 21:7–14.
- IBO Universitair Medische Centra. 2012. Report of the Ministerie van Financiën. The Hague.
- Iglehart JK. 2012. Financing graduate medical education—mounting pressure for reform. N Engl J Med. 366:1562–1563.
- O’Rourke K. 2014. IOM Report calls for overhaul of graduate medical education. JAMA. 312:880–881.
- Van Baalen MR, Bosman PJ. 2012. STZ: Kostprijsonderzoek Opleiding tot Medisch specialist; de resultaten en inzichten van de onderzoeken 2009–2012 gebundeld. Utrecht: Berenschot.