BACKGROUND AND OBJECTIVE: Medical education relies heavily on workplace learning where trainees are educated through their clinical experience. Few studies have explored trainees’ perceptions of the educational value of these patient care experiences. The aim of this study was to identify pediatric patient characteristics that medical trainees perceive as educationally valuable.
METHODS: Over 2 months, trainees on pediatric inpatient wards ranked the perceived educational value of patients under their care on a 4-point bipolar Likert scale. Three patient characteristics were examined: complex-chronic and noncomplex-chronic preexisting conditions, difficult social circumstances, and rare diseases. Patient-level predictors of cases perceived as educationally valuable (defined as scores ≥3) were examined by using univariate and multivariate analyses.
RESULTS: A total of 325 patients were rated by 51 trainees (clinical medical students [45%], first-year residents [29%], third-year residents/fellows [26%]). Rare diseases had a higher educational value score (adjusted odds ratio 1.76, 95% confidence interval 1.08–2.88, P = .02). Complex-chronic and noncomplex-chronic preexisting conditions and difficult social circumstances did not affect the perceived educational value.
CONCLUSIONS: Trainees attribute the most educational value to caring for patients with rare diseases. Although trainees’ perceptions of learning do not necessarily reflect actual learning, they may influence personal interest and limit learning from an educational experience. Knowledge of trainee perceptions of educational experience therefore can direct medical educators’ approaches to inpatient education.
Research on physician education has shown that learning in medical practice is very much embedded in clinical work. Workplace learning, which includes interacting with other professionals to diagnose, reflecting on clinical experiences, teaching other learners, and feedback, is the primary environment for expertise development.1 Because of the critical role it has to play, there has been a call to expand research on workplace learning in residents to include an exploration of organizational practices, experiences, and learning attitudes.2 In addition to these contributors to the intended and explicit curriculum of workplace learning, one must also consider the important contribution of the so-called “informal and hidden curriculum.”
The hidden curriculum is defined as the unscripted and highly interpersonal forms of teaching between faculty and students (the informal curriculum) and/or a set of influences at the level of organizational structure and culture (the hidden curriculum).3 One manifestation of the informal and hidden curriculum is the potential for the development and/or consolidation of biases and attitudes related to specific patient types.3–5 It has been well documented that physicians experience bias toward specific patient social and demographic traits, with ethnicity being the most well documented.6,7 Studies of bias toward patients with chronic disease,8 physical disabilities,9 and children with cerebral palsy10 have suggested that these biases develop in the early stages of training as medical students express less optimistic views of outcomes for these patients. Physicians may develop cynicism and frustrations because of an inability to “cure” patients with chronic conditions and disabilities who often have many comorbidities, communication difficulties, and economic constraints.11 In turn, the resulting frustrations for medical professionals and trainees may perpetuate unconscious biases and affect the ability to learn by leading to stressful situations: conditions that have been shown to create poor learning experiences.12 Similarly, biases toward patient characteristics may decrease personal interest in the case and thus limit their educational experience.13 To date, little is known about the trainee’s perception of educational experience when exposed to a wide variety of patient types. Specifically, it is unclear if trainees assign greater or lesser value to their learning when caring for patients with certain characteristics.
The objective of this study was to explore the relationship between specific patient characteristics and the perceived educational value of the pediatric inpatient care experience for trainees in the hopes of identifying opportunities to improve inpatient learning. We selected 3 patient criteria to explore. Noncomplex-chronic (NCC) and complex-chronic conditions (CCC) were selected given their time-consuming nature and previously documented negative bias toward these patients.8–10 Therefore, we hypothesized that children with NCCs or CCCs would be perceived as less educationally valuable than children who were previously healthy. The second patient criterion selected was children with difficult social circumstances, which we hypothesized would be perceived as less educationally valuable than children without difficult social circumstances. Bias toward difficult social circumstances has not been independently explored. However, studies have documented that patients requiring time-consuming care diminish perceived learning.12 Last, we selected children with rare diseases and hypothesized that they would be perceived as more educationally valuable than those with common diseases. This patient characteristic was selected given that the culture in medicine commonly rewards the diagnosis of rare diseases.
This was a prospective cross-sectional study of medical trainees at a large academic pediatric hospital in downtown Toronto, Canada, and was conducted for 2 months, from June 20, 2011, to August 19, 2011, on 3 general inpatient wards. The wards were staffed by a general pediatrician, 4 to 5 resident house staff, and 2 to 6 medical students. Each ward carried an ongoing census of ∼20 patients per day with ∼20 new admissions throughout the week.
Included were fellows, third-year residents, first-year residents, and clinical medical students (3rd and 4th year) who cared for the patients during the previous week. Fellows were trained general pediatricians who were enrolled in either an academic general pediatrics or hospital pediatrics fellowship program. Learners from other professional fields (eg, nursing) and more junior (preclinical) medical trainees were excluded.
All patients who spent at least 1 night on the ward for a nonelective admission were eligible for inclusion. Elective admissions were excluded, as they typically do not require diagnostic critical thinking and trainees are often less involved in their management.
Process of Data Collection
After obtaining informed consent, the trainees were given a list of inpatients on the wards during the previous week and a questionnaire. The trainees were asked to rate the perceived educational value of the patients in whose care they were directly involved. The question was phrased “Compared with the average patient, whose care you have been involved with, how would you rate the educational value that you obtained in your involvement in this patient’s care.” A 4-point Likert scale was used for the responses, containing an even number of categories to force raters to commit to one side or another. The patient was rated either “much less,” “a bit less,” “a bit more,” or “much more” educationally valuable than their impression of the average patient. Multiple trainees could rate each patient. However, individual trainees could rate an individual patient only once. The trainees were aware of the overarching aim of the study (to explore predictors of educationally valuable clinical encounters), but were blinded to specific predictors being examined and to the study hypotheses.
Patient demographics were collected from charts. Length of stay data were collected on a randomized group of 40% of the study population.
Complexity and Chronicity of Preexisting Conditions
Patients were categorized into 1 of 3 mutually exclusive categories (previously healthy, preexisting NCC, and preexisting CCC). CCCs were operationally defined by using the framework of Feudtner et al14 of “a medical condition lasting ≥12 months, and involving several different organ systems or one organ system requiring a high level of specialty care and hospitalization” (eg, cystic fibrosis). NCCs were defined as those diagnosed before the current admission and expected to last ≥12 months (eg, asthma) but were not CCCs.15
The attending physician’s admission note and the discharge summary were used to determine preexisting conditions. Any conditions listed in the past medical history on either document were referenced to a previously published diagnostic list, which outlines complex-chronic diseases.16 Conditions not listed on the list were then considered NCCs.
Both NCCs and CCCs were divided into mutually exclusive categories by chronicity of the current presentation. Either the presenting complaint was “new or unrelated to the chronic condition” (eg, an asthmatic patient who presents with a limp), or it was an “exacerbation” of a chronic problem (eg, a child with cystic fibrosis admitted with pneumonia). Patients with multiple presenting complaints were assessed hierarchically where these patients were assigned to the new presentation category.
To further delineate the impact of functional impairment and disability on perceived education value, complex-chronic children with neuromuscular conditions were examined separately from other CCCs. Neuromuscular conditions were defined by using a previously published algorithm.14
Difficult Social Circumstances
The presence of difficult social circumstances was determined by known involvement of child protection services and/or involvement of a social worker during the current admission. This information was obtained from chart abstraction. As in many hospitals, there were no predefined criteria for social work involvement.
Rare diseases were defined by using the prevalence cutoff from the Rare Disease Act in the United States (<1:1500).17 The “most likely diagnosis” was ascertained from the attending physician’s admission note. The prevalence of this diagnosis was determined by using a hierarchical search strategy starting with UpToDate (http://www.uptodate.com/home), then Medscape (http://www.medscape.com), then PubMed (http://www.ncbi.nlm.nih.gov/pubmed/). All searches were conducted between August 24, 2011, and November 11, 2011. If no prevalence data were found (n = 7), consensus between team members was used based on their clinical knowledge and experience (RF, EC).
Informed consent was obtained from all trainees in the study and from all patients (or their parental proxy) who were still in hospital at the time of data collection. The study protocol was approved by The Hospital for Sick Children’s Research Ethics Board.
The perceived educational value scores were dichotomized into mutually exclusive categories, whereby “educationally valuable” was defined categorically as a score of ≥3 on the Likert scale (as scores of 3 and 4 were considered more educationally valuable than the average patient). Mean perceived educational value scores also were analyzed. Univariate analyses were used to explore significant predictors (patient characteristics). Subsequently, a multivariable analysis was performed to test the effect of covariates after adjusting for other covariates in the model. Repeated measurement logistic regression was performed by using a generalized estimating equation approach. Predefined patient characteristics were tested: gender, rare disease, combined NCCs and CCCs, social work involvement, and child protection services involvement. Missing data points were not imputed. The final model was chosen based on the lowest Quasi-likelihood Information Criterion and statistical significance of the variables.
Sample size was calculated to prevent overfitting of the regression model using the predictors of interest that were determined a priori.18 It was estimated that 320 cases would be sufficient. Analyses were conducted by using SAS version 9.3 (SAS Institute, Inc, Cary, NC).
All 60 trainees who participated in patient care during the time of the study were given the opportunity to participate, of whom 51 completed the survey (85% response rate). One student (1.5%) declined participation, 6 (10%) were not on service during data collection, and 2 (3%) did not return the survey; 23 (45%) were clinical medical students, 15 (29%) were first-year residents, 7 (14%) were third-year residents, and 6 (12%) were fellows. Seventy percent of the trainees were women. The median age was 26 years (range 22–42). Trainees cared for a mean (SD) of 13.1 (9.2) patients.
A total of 325 patients were included in the study, who were rated by a mean (SD) of 2.05 (1.29) trainees (668 total trainee assessments). The median patient age was 3.1 years (range 0–17.9) and 162 (52%) were boys. The average length of stay was 14.5 days (range 1–155 days). A total of 151 (47%) were previously healthy, 128 (39%) had a CCC, and 46 (14%) had an NCC. Of those with a chronic medical condition (both complex and noncomplex), 79 (45%) were admitted for an exacerbation of a medical issue. Of the patients with a CCC, 49 (15%) had neuromuscular conditions. Child protection services was involved in 20 (6%) of cases and social work was involved in 74 (23%). Ninety-four patients did not have a differential diagnosis written on admission and, therefore, rarity of disease was not determined in these cases. Of the remaining 231 patients, for whom differential diagnosis data were available, 64 (27%) had a rare disease.
Neither patients who were previously healthy nor those with CCCs and NCCs were perceived as more educationally valuable (χ2 = 2.11, 2, P = .35). Mean (SD) perceived educational value for CCCs and NCCs and previously healthy children was 2.58 (0.74), 2.58 (0.70), and 2.58 (0.70), respectively.
The proportion of patients with social work involvement who were considered educationally valuable did not significantly differ from those without involvement (χ2 = 0.11, 1, P = .74).
Similarly, the proportion of patients with child protection services involvement did not significantly differ from those without involvement (χ2 = 7.53, 1, P = .06). The mean (SD) perceived educational value for patients with social work involvement, child protection services involvement, or no social issues was 2.58 (0.87), 2.64 (0.67), and 2.53 (0.73), respectively.
Patients with rare diseases were perceived as significantly more educationally valuable than those with common diseases (χ2 = 4.43, 1, P = .04) (Fig 1). Similarly, patients with a rare diagnosis were perceived as more educationally valuable than those with a common diagnosis (P = .04; mean [SD]: 2.70 [0.68] vs 2.48 [0.73]).
In the multivariable analysis (Table 1), rare disease was still a significant predictor of perceived educational value when adjusted for child protection services involvement (adjusted odds ratio 1.76, 95% confidence interval [CI] 1.08–2.86). There was no significant relationship between length of stay and perceived educational value (r2 = 0.17, P = .09).
This study was undertaken to explore which pediatric patient characteristics medical trainees intrinsically value in the hopes of identifying biases and improving the effectiveness of their workplace education. Of the 3 hypothesized patient characteristics, only rare diseases predicted heightened perceived educational value. Difficult social circumstances, NCCs, and CCCs did not influence the perceived educational value of the clinical experience.
Our findings are concordant with an analogous study addressing the effect of workload on learning.19 Haney and colleagues19 found that patients with higher acuity, defined as those who were unstable, complex, and had diagnostic uncertainty, were correlated with positive perceived educational value, but the authors did not differentiate between the specific patient characteristics (eg, chronic, complex, or rare diseases) that led to high acuity. Our study delineates these 3 characteristics and suggests that trainees may intrinsically value rare diseases as learning opportunities.
Rare diseases may be perceived as educationally valuable for a number of reasons. Rare diseases are commonly touted in medicine as valuable “cases,” exemplified by their emphasis in the formal and informal/hidden curriculum. This may specifically be the case in large tertiary care centers, which have extensive access to rare disease presentations and subspecialists, creating an institutional culture that is more focused on rare diseases than a community-based facility. The informal/hidden curriculum promotes rare diseases in subtle ways, such as praise for solving “diagnostic dilemmas” and the emphasis of these cases as the subject of interesting case rounds or case reports in medical journals.5,20 Similarly, the content of high-stakes medical qualifying examinations often includes a dominant representation of rare conditions.21 Rare diseases also make the diagnostic process more challenging and thus increase trainee engagement and learning.19
Although our study found that trainees perceive patients with rare diseases to be more educationally valuable, it is important to note that trainees’ perceptions of learning do not necessarily reflect actual learning. The concern, however, is that if trainees do not perceive an experience to be educationally valuable and are not intrinsically motivated by the case, they might be less engaged and therefore potentially less likely to gain important learning from these experiences. This concern is well supported by the self-determination theory, where one’s learning is most effective when intrinsically motivated or due to their own personal interest.22 Intrinsic motivation has been shown to improve learning and performance in medical students.23
Because trainees did not perceive common and chronic diseases as enhancing educational value, medical educators are now challenged to highlight the importance and value of these experiences. One solution is encouraging training opportunities where more common and chronic diseases are seen, such as in community hospitals and ambulatory settings.24 In these settings, the primary focus may be on management of a chronic disease rather than on the acute presentation of a disease. Ambulatory clinics provide adequate time to address the issues of a chronic disease, whereas in the acute setting time constraints may not allow learners to address these issues. A number of residency programs have begun to incorporate a satellite campus or rotations at smaller community hospitals to address this concern.
Not only can medical educators work to enhance the perceived educational value of chronic and common diseases, but they also can address the hidden curriculum they personally perpetuate. Physician educators may take it on themselves to improve their cultural and social competency, as many medical school and residency programs provide specific training in this area.25 Through this improved education and recognizing their own bias, one hopes that this would reduce the hidden curriculum reinforcing bias toward rare diseases.26 An additional strategy that supervisors might use is to provide positive feedback to trainees for diagnostic restraint instead of encouraging the use of multiple costly and/or invasive tests in search of the medical “zebra.”20
This study had a number of important limitations. First, the study was completed on an inpatient ward; trainees’ perception of the intrinsic educational value of the patient characteristics may vary depending on the workplace learning setting. Specifically, the inpatient setting focuses on acute treatment, allowing providers to defer most comprehensive care to alternative settings. Similarly, patients were assessed at multiple time points throughout their admission. Therefore, we were limited to using admission diagnosis for determination of rare disease categorization, as we had no way of knowing the diagnosis being entertained at the time of assessment. It may be that assessing a patient earlier in the admission may lead to higher perceived educational value, as there still may be diagnostic uncertainty. Additionally, one-third of the patients did not have an admitting differential listed and were thus excluded from the analysis.
Furthermore, data collection was limited in 2 ways. First, data were collected at various time points in the academic year. Therefore, residents may have been in their first or last months of their respective academic year. Second, social work involvement was defined as any documented connection with social work, whether this was simply a taxi chit or a more complex application for funding. We did not distinguish between the extent of assistance provided, within the relatively small sample (23% of patients in the study) with social work involvement as there was not enough power to perform this analysis.
Lastly, no qualitative data were collected to assess the rationale behind the educational value scores or relationship between medical competencies and patient characteristics. This is an area in which future research should be conducted.
Inpatient medical educators should acknowledge the potential bias toward rare disease and use this intrinsic motivation to improve skills that can be used in common disease management and diagnosis. Although the unconscious biases against noncomplex-chronic and complex-chronic diseases and difficult social circumstances issues do not appear to diminish the perceived educational value of pediatric inpatient care, replication of these findings in broader care settings is required. Inpatient educators may want to reflect on their own patient biases to ensure they do not promote an institutional culture that promotes specific patient characteristics over others. All patients are and should be a learning opportunity for trainees.
We acknowledge the support of research staff Julie DeGroot in conducting this study.
This work was presented in part at the Pediatric Academic Society Conference; April 28–May 1, 2012; Boston, MA.
FINANCIAL DISCLOSURE: The authors have indicated they have no financial relationships relevant to this article to disclose.
FUNDING: All phases of this study were supported by a Pediatric Consultant Education Grant, The Hospital for Sick Children.
POTENTIAL CONFLICT OF INTEREST: The authors have indicated they have no potential conflicts of interest to disclose.
- Chuang AW,
- Nuthalapaty FS,
- Casey PM,
- Kaczmarczyk JM,
- Cullimore AJ,
- Dalrymple JL
- White A,
- Chanoff D
- Ten Cate T,
- Kusurkar R,
- Williams G
- Feudtner C,
- Hays RM,
- Haynes G,
- Geyer JR,
- Neff JM,
- Koepsell TD
- Cohen E,
- Berry JG,
- Camacho X,
- Anderson G,
- Wodchis W,
- Guttmann A
- 17.↵Rare Diseases Act, Pub. L. No. 107-280 (2002)
- Babyak MA
- Copyright © 2015 by the American Academy of Pediatrics