Consensus development of core competencies in intensive and critical care medicine training in China

Leadership in intensive and critical care medicine in mainland China convened a broad array of clinical experts, credentialing and certifying bodies, and all national critical care societies. We used consensus techniques to develop a set of core competencies for intensive and critical care medicine training, which have been approved by representatives of all national critical care organizations. This is the first time that such a consensus in ICM has been developed in China.

The Delphi technique has been used in the field of healthcare education and training since the 1990s. One of the major advantages that make this method so popular is the ability to allow participants from different geographic regions to share their opinion with each other, and revise them by iteration. In addition, our study used the NG methodology, which has been validated to represent the views of the wider critical care community in developing national research priorities [9, 13]. Moreover, we made efforts to minimize bias, because all NG members in our study remained blinded to the composition of the NG as well as the individual response of any other NG members. Although a methodology involving face-to-face meetings may have led to greater agreement, while working remotely might compromise the level of agreement and may also be more time consuming, this approach may minimize the “tyranny of majority” during the rating process. In the meantime, keeping raters blind to each other during iteration has been well documented in several similar situations [14, 15].

It is noteworthy that there are significant differences in the core competencies developed in this study and those by the CoBaTrICE (Table 3, and Additional file 1: Table S3), highlighting the different insights of the Chinese intensive care community from those of Europe and North America [5–8]. Possible explanations for such differences may vary depending on different themes and merit further investigation. In consideration of heterogeneity of diseases and uneven distribution of healthcare resources all over the nation, we tried to determine a detailed requirement of basic knowledge and skills under certain acute conditions instead of a package of proposal that “manages the care of the critically ill patient with specific acute medical conditions” [6]. A list of 62 specific diseases and conditions were included in the initial questionnaire and yielded 46 items in the final set. Despite the differences in the level of focuses (disease level vs organ system level), our list and the CoBaTrICE syllabus covered almost the same spectrum of acute and critical conditions in the domains of disease management and therapeutic intervention/organ system support. The considerably shorter list of practice procedures might reflect the limitation of training resources in our country. Competencies related to catastrophe management were not included, because some NG members believed that emergency physicians and nurses should be more involved. Despite a similar enthusiasm for professionalism to the CoBaTrICE study [6], an unexpected result is that issues related to end-of-life care such as palliative care and brainstem death testing were rated less important than in Europe and North America. Data from rounds four and five showed that some NG members believed these items should be performed under proper supervision until the end of training. One possible reason is that end-of-life care in China is far underdeveloped, which limited the training sources for these parts [16, 17].

The core competencies generated from our study should be considered minimum requirements of intensive and critical care training which should be used to define a qualified intensivist, whereas those submitted to the NG for initial review but excluded during iterative rating might serve as optional competencies. We used a standardized descriptive term in each statement to clarify at which level the competence should be evaluated. Knowledge and skills can be evaluated by multiple-choice questions (MCQs) and objective structured clinical examination (OSCE) with high reliability [18], while portfolios and faculty rating are more used in attitude and behavior assessment [19]. We understand that some contents need to be revised, expanded, or added, highlighting the need for continuous reviewing and updating in the future. Our ultimate goal is to produce a standardized curriculum and evaluation system for intensive and critical care training, a time-consuming process requiring a more detailed guideline. However, the final set at the current stage should still be considered a starting point—that is, the fundamental standards that may guide future education goals and professional development in intensive and critical care specialty across the whole nation—and should allow curriculum managers to use these competencies as building blocks to develop a curriculum responsive to any special local training needs.

Our study has a few limitations. First, due to limited human resources and time boundaries, membership of the NG was quite small. Moreover, all members were selected from university hospitals, and should not be considered representative of the whole nation. However, almost all of the training bases for resident standardized training programs, which were approved by the Ministry of Health, were university hospitals. We understand that in the future many of the trainees will end up working in smaller or community hospitals, where competency requirements might be different from those in university hospitals. We thus kept reminding NG members to consider resource diversity of the national medical system through each round of rating. Our results show that 19 out of 47 (40.4 %) items in “practical procedures” were selected, whereas 46 of 62 (74.1 %) items in “disease management” were chosen. This may indicate that most competencies requiring advanced training resources were excluded from the final set.

Another limitation is that we did not invite patients or patient families to participate in our survey, which might lead to less attention to the opinions of “consumers”. However, our questionnaire was based on core competencies in the literature, including those generated by the CoBaTrICE [6]. The CoBaTrICE coordinators used a separate survey questionnaire including items about communication and interpersonal skills, and decision-making in addition to medical knowledge and skills, to seek for views from ICU patients and their relatives. Responses from this survey were also integrated with those from medical professionals during their NG rating [6]. However, the NG rating and subsequent iteration process involved only healthcare stakeholders. Therefore, our result of less attention to several ethical issues such as palliative care cannot be explained by an absence of consumers’ opinion during the whole process.

One last limitation is that during the recirculation phase (Phase 3) we did not receive a response from anyone other than NG members, which was unexpected. This may cast doubt on the usefulness of this phase.