Article Text
PDF
Abstract
Introduction Academic medical centres (AMCs) have the tripartite mission of performing research to advance healthcare delivery, educating future clinicians and providing healthcare services. This study investigates the criteria associated with being promoted in a Singaporean AMC.
Methods Using a dataset of 255 candidates for promotion at the studied AMC, we employ logistic regression to determine if these factors are associated with the likelihood of promotion. Further, we use interaction effects to test if the relationship between the H-index and likelihood of promotion differs across the academic levels of the candidates.
Results The logistic regression results based on the best of our three tested models suggest that the H-index is positively associated with promotion for those applying to become clinical associate professors (OR=1.43, p=0.01). Moreover, candidates who provide well-developed education portfolios (OR=3.61, p=0.02) and who have held service/leadership roles (OR=6.72, p<0.001) are more likely to be promoted.
Conclusions This study affirms the correlation between promotion and the advancement criteria outlined by the AMC. This is important for transparency and trust between the AMC and its faculty in their applications for promotion and success in an academic career. Further, our study is one of the few empirical studies linking promotion criteria to promotion outcomes.
- analysis
- career development
- clinical leadership
- medical leadership
Data availability statement
No data are available.
This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.
Statistics from Altmetric.com
WHAT IS ALREADY KNOWN ON THIS TOPIC
Criteria used to review the promotion of academic faculty have traditionally included teaching, research and service provision to the university. Considerable differences have been identified regarding the evaluations of excellence in research, teaching and service provision among medical schools.
WHAT THIS STUDY ADDS
This study is one of the few empirical studies linking promotion criteria, as specified by the academic medical centre (AMC), to promotion outcomes. This is important for transparency and trust between the AMC and its faculty in their applications for promotion.
HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY
The findings of this study could help early-career and mid-career faculty in planning their professional progression. Furthermore, the study serves as a guide for the promotion committees of the AMC. Other promotion committees may use our study as a guide in their respective contexts.
Introduction
Academic medical centres (AMCs) have the tripartite mission of performing research to advance healthcare delivery, educating future generations of clinicians and providing healthcare services. The key factor to the advancement of this mission is the clinician faculty of these centres.1 Mullangi et al noted that clinician faculties in medical schools have different jobs from their university peers owing to their commitment to service or patient care; thus, these faculties are ‘without peers in the larger university system’. (2, p. 1) Previous studies have examined how to thrive and succeed in AMCs.3–6 Common investigated themes are career advancement and the associated criteria as well as the output required to be promoted and its accompanying rewards. However, there is a growing need for transparency around academic appointments and the appointment process because of the increased number of medical schools in the United States and proliferation of faculty tracks and criteria.6
The criteria used to review the promotions of academic faculty have traditionally included ‘teaching, research, and public service to the university’. (7, p. 470) Katz found that public service and committee work are ‘important variables in explaining rank’. (7, p. 476) Ence et al found that a higher H-index and M-index correlate with a higher ranking for orthopaedic faculty.8 Grimm et al 9 found that, among other factors, a faculty member’s number of publications and holding an advanced degree are important for the pursuit of an academic career. However, there is a paucity of empirical research on the criteria associated with the likelihood of promotion within AMCs. Aron3 noted the considerable differences in how the demonstrations of excellence in research, teaching and service provision are evaluated in medical schools. Rice et al 10 found that academics adjust their practices according to the evaluation criteria of their institutions. Thus, understanding the relationship between the evaluation criteria and likelihood of promotion helps faculty, particularly early-career and mid-career faculty, in their academic career planning.
To address this gap in the existing literature, we ask, ‘How do such factors as the number of publications, H-index, teaching as demonstrated in an education portfolio, performing service/leadership roles, and holding an advanced degree affect the likelihood of being promoted?’ Using a dataset of 255 candidates for promotion at a Singaporean AMC, we employ logistic regression to determine if these factors are associated with the likelihood of promotion. Further, we use interaction effects to test if the relationship between the H-index and likelihood of promotion differs across the academic levels of the candidates.
The faculty structure of the AMC in question consists of regular and clinical ranks, adjuncts, and visiting appointments. Regular rank appointments are accorded to clinicians expected to make meaningful contributions to the research and/or educational missions of the AMC and consist of tenure and non-tenure tracks. Regular rank tracks comprise research and educational pathways. Clinical rank appointments are accorded to clinicians who focus on clinical/medical care within the AMC, though they may have quantifiable academic or leadership involvement. Both regular and clinical rank appointments have four levels: instructor, assistant professor, associate professor and professor. Adjunct and visiting appointments are accorded to individuals outside the AMC and visiting academics, respectively. Promotion is typically accorded within the same rank (ie, clinical associate professor to clinical professor and regular rank associate professor to professor). However, there may be an occasional conversion from regular rank to clinical rank or vice versa when faculty change their career objectives.
This study focused on (1) regular rank non-tenure track appointments and (2) clinical associate professor and professor appointments. Promotion from instructor to assistant professor was not studied, as associate consultants and consultants can obtain appointments as instructors and assistant professors, respectively. The promotion criteria and pathways are summarised in table 1. The promotion process of the AMC is depicted in figure 1.
General promotion process of associate professors and professors in the AMC (adapted from the AMC’S website). AMC, academic medical centre.
Criteria for academic appointment and promotion of faculty (adapted from the AMC’s website)
Methods
Dataset
The dataset comprised 255 candidates nominated for promotion to four senior academic levels of the AMC from January 2018 to June 2021: clinical associate professor, associate professor, clinical professor and professor. We collected data on the promotion decision (dependent variable) as well as on five independent variables: (1) the number of publications, (2) the H-index (defined as the number of papers with a citation number greater than or equal to H11 obtained from Google Scholar), (3) teaching excellence demonstrated through an education portfolio,12 which had been endorsed by the education institute of the AMC, (4) whether or not they had assumed service/leadership roles within the AMC or externally and (5) whether or not they had an advanced degree in addition to a medical degree. These variables are presented in table 2. In line with local legislation on personal data protection, candidates’ names, gender and other demographic data were removed by the department administration before the dataset was provided to the research team.
Definitions of the variables
Tools
R software was used to view and analyse the data.13 Of the 255 candidate nominations, 197 were promoted and 58 were not. Table 3 provides the summary statistics of the dependent and independent variables.
Summary statistics of the variables
Logistic regression model
We used logistic regression to predict the binary outcomes of the dependent variable: promoted or not promoted. Specifically, we fitted and compared three logistic regression models, as described in the results section.
Results
Baseline model with all five independent variables
The pseudo R2 (McFadden) for this model was 11%, indicating that it explained approximately 11% of the variation in the dependent variable. Among the five independent variables, service/leadership was significant at the 95% confidence level. Hence, only service/leadership was significantly associated with being promoted to all four senior academic levels. Our baseline model thus showed that those participants who had assumed service/leadership roles, whether within the AMC or externally, had a 394% higher likelihood of being promoted to a senior academic level on average than those who did not, controlling for all the other independent variables. Within the sample, we were 95% confident that those who had assumed service/leadership roles had a 137%–922% higher likelihood of being promoted to the senior academic level than those who did not, controlling for all the other independent variables.
Model including academic level
As shown in figure 2, the likelihood of being promoted differs by academic level. Therefore, in the second model, we introduced a categorical variable for academic level grouped into professor, clinical professor, associate professor and clinical associate professor. We then re-ran the analysis to examine the likelihood of being promoted by academic level. The pseudo R2 (McFadden) of the second model was more than double that of the first model (25% vs 11%). The second model was also more parsimonious, with an Akaike information criterion (AIC)14 of 223 compared with 256 for the first model. These observations allowed us to conclude that the second model was better than the first. Here, in addition to service/leadership, the H-index and promotion to clinical associate professor were significantly associated with being promoted to all four senior academic levels.
Promotion decision by academic level.
Revised model including academic level
As we detected multicollinearity between the number of publications and H-index, we tested two additional models: one with the number of publications excluded, but with the H-index included and the other with the number of publications included, but with the H-index excluded. Further, we performed a likelihood ratio test on all three models including academic level. The likelihood ratio test found that the additional model with the H-index was the best among these three models. Given these findings, we built a revised model including academic level that excluded the number of publications. Compared with the baseline model, the pseudo R2 of this revised model dropped by one percentage point from 25% to 24%, while the AIC increased marginally to 225 from 223. Service/leadership, the H-index and promotion to clinical associate professor were significantly associated with being promoted at all four senior academic levels.
Model including academic level and interactions among the independent variables
Finally, we explored how academic level interacted with the independent variables, excluding the number of publications. We found that only the H-index and academic level had interaction effects (interaction model hereafter). The boxplot in figure 3 with the H-index and four senior academic levels supports this result. Moreover, previous research has found that the H-index increases as academic rank rises.15 No significant interaction between academic level and the other independent variables was detected.
Boxplot of the H-index by academic level.
Using this interaction model, we found the highest pseudo R2 of 30% for the clinical associate professor level. This model was also more parsimonious, with the AIC reduced to 215 from 225 in the revised second model. In addition to service/leadership, teaching excellence was significant. In contrast to the revised second model, the interaction of the H-index and promotion to clinical associate professor was also significant; however, the interactions of the H-index with the other independent variables were not. Hence, the interaction model provided further insights, showing that when interactions were introduced, the performance of the model improved, with the H-index and clinical associate professor level demonstrating interaction effects.
Table 4 compares the results of the three models. The interaction model was the best, as it explained 30% of the promotion decisions, which was the highest proportion among the three models. It was also the most parsimonious with the lowest AIC value. This model showed that the H-index had a significant interaction effect with promotion to the clinical associate professor level. These faculty candidates were predicted to have a 43% higher likelihood of being promoted than candidates at the other levels on average, controlling for all the other independent variables. The interaction of the H-index and clinical associate professor had an odds ratio OR of 1.43 (95% CI 1.10 to 1.88, p=0.01). Alone, the clinical associate professor variable was not significant, and neither were the interactions of the H-index with the professor, clinical professor and associate professor levels.
Comparison of three logistic regression models in detecting the associations between the variables
Candidates who had a well-developed education portfolio, as endorsed by the AMC, had a 261% higher likelihood of being promoted than those who did not (OR 3.61, 95% CI 1.25 to 10.45, p=0.02). Candidates who had assumed service or leadership roles within or outside of the AMC had a 572% higher likelihood of being promoted on average (OR 6.72, 95% CI 2.73 to 16.54, p<0.001).
Discussion
In this study, we empirically examined the association between the studied AMC’s tripartite mission-related criteria and promotion outcomes. We found that the H-index is positively associated with the promotion of those applying to become clinical associate professors. Teaching excellence, demonstrated through a well-developed education portfolio, and holding service or leadership roles within or outside the AMC were positively associated with promotion to all four senior academic levels studied. Holding an advanced degree was not significant for being promoted.
The interaction of the H-index with promotion to associate professor, clinical professor and professor levels was not significant. This implies that other criteria not included in our study were more significant in the promotion reviews for these three academic levels. Data that were not quantifiable, such as reference letters, were required for the promotion decisions to associate professor, clinical professor and professor, but were not required in the case of clinical associate professor. Further research, including qualitative studies, is required to explore the association between such unquantifiable criteria and promotion decisions. At the time of the study, requirements between clinical and regular rank were not differentiated. The emphasis on grant-funding for regular ranks has increased in recent years, and data could be collected in future research.
Limitations
Of the five promotion criteria outlined in table 1, hospital grade was not studied, as it is relevant to the junior faculty levels of instructor and assistant professor, as explained in the introduction section. Reputation is not quantifiable and needs to be studied through qualitative research in future studies. Grant funding was not studied, as it is not a significant contributor to advancement within clinical ranks.
The variables were locked into binary data for teaching excellence and service/leadership, resulting in a loss of granularity in the findings. The study could not explore further granularity in these two independent variables without identifying the candidates, and this would not pass institutional review. Increased details in a future qualitative study would help readers better contextualise our results.
As the data were obtained from the clinical sciences division of the AMC, the recommended model may not demonstrate similar results for candidates from the basic sciences division. If used as a decision support tool, a constant input of fresh review decisions is thus required to best reflect the promotion committees’ decision-making process.
Conclusion
This study affirms the correlation between promotion outcomes and the promotion criteria outlined by the AMC. This is important for transparency and trust between the AMC and its faculty in their applications for promotion and success in their academic careers. Further, our study is one of the few empirical studies linking promotion criteria to promotion outcomes. In addition to helping early-career and mid-career faculty in planning their professional progression, it serves as a guide for the promotion committees of the AMC. Other promotion committees may use our study as a guide in their respective contexts.
Data availability statement
No data are available.
Ethics statements
Patient consent for publication
Footnotes
Contributors MMY developed the research idea, performed the analysis and wrote the manuscript. S-HL reviewed and edited the manuscript. AK verified the methods used and critically reviewed the manuscript. AWT provided guidance and support throughout the study. MMY accepts full responsibility for the work and/or the conduct of the study, had access to the data, and controlled the decision to publish.
Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.
Competing interests None declared.
Provenance and peer review Not commissioned; externally peer reviewed.
Author note Within the healthcare system of Singapore, associate consultants and consultants are specialists who have completed residency/fellowship or basic and advanced specialist training and fulfil several conditions outlined in the Singapore Ministry of Health Holdings website, https://www.physician.mohh.com.sg/medicine/medical-service-career-path.