Improvements in residency programs may be implemented by the careful consideration of seven elements directly related to residents’ life and education: the curriculum; the proper balance between service and education; residents’ supervision by motivated and credentialed faculty (as well as residents’ willingness to be supervised); frequent evaluation of the resident, the faculty, and the program followed by timely feedback; a pleasant work and educational environment; program sensitivity to issues like culture, gender, and ethnicity; and the recognition of stress in the lives of residents as an important issue.1
In the United States, the Accreditation Council on Graduate Medical Education (ACGME) aims to enhance the quality of patient care by ensuring and improving the quality of graduate medical education. Since July 2003, residency programs in the United States have had to comply with the ACGME's duty hours standards, which limit residents’ duty hours to a maximum of 80 hours a week, including in-house call time. These standards also restrict the number of weekly in-house calls and the number of consecutive duty hours. And residents must be given one day per week free from all clinical and educational activities.2
The work-hour reform is expected to have an impact on multiple aspects of training programs, such as budget allocation, human resources, attainment of educational goals, quality of patient care, faculty and residents’ relationships, and faculty and residents’ well-being. Program directors will have to measure all these aspects as objectively as possible to manage the impact of the reform appropriately and to maximize the quality of patient care and of residents’ training under the new ACGME standards.3
The Brazilian National Committee on Medical Residency (BNCMR), affiliated with the Ministry of Education, is responsible for establishing the minimum standards of residency training through its Specialty Boards, which include members of the respective national specialty associations among others.3
Among the BNCMR's Common Program Requirements (CPRs), established in 1981,3 are items that protect residents against excessive workload, guarantee time for educational activities, and establish standards for faculty supervision. Specifically, residents can be scheduled to a maximum of 60 hours on duty per week (including in-house calls), cannot be scheduled to more than 24 in-house call hours a week, can be scheduled to a maximum of 36 consecutive duty hours, must have one in seven days free from all educational and clinical responsibilities, and must have 30 consecutive days of vacation each year. A minimum of 10% of a resident's duty hours must be allocated to formal educational activities (e.g., lectures, seminars, problem-based sessions, case discussions). Continuous supervision of residents by faculty is required.4
Because the CPRs aim at assuring residents’ well-being, the working hypothesis of this study was that violations of the requirements might affect residents’ perceptions about their general quality of life, their quality of life during duty hours (in residency), and their perceptions about the educational environment. This study compared residents’ perceptions about dimensions of the general quality of life, the quality of life in the residency, and the educational environment between those reporting their program's compliance with and those reporting their program's violation of the CPRs.
The study was performed at the four teaching hospitals (identified in this study as A, B, C, and D) that currently provide residency training in Florianopolis (SC), Brazil, with the approval of the Institutional Ethics Committees. Eighty-eight residents (years one through four) of both genders and of different specialties were invited to participate during the study period, October through December 2003.
After giving written informed consent, participants were requested to complete four instruments.
We asked participants to complete a seven-item questionnaire about their program's compliance with the CPRs outlined above. The questionnaire asked residents’ to identify the amount of on-duty hours on per week (≤ 60, > 60), in-house call hours per week (≤ 24, > 24), consecutive hours on duty (≤ 36, > 36); whether they received one in seven days free from all educational and clinical responsibilities (yes/no), 30 consecutive days on vacation each year (yes/no), the percentage of duty hours their program allocated to educational activities (< 10%, ≥10%); and their feelings of being unsupervised most of the time during patient care activities (yes/no) and estimated percentage of duty hours under direct faculty supervision (0–100%).
Dundee Ready Educational Environment Measure (DREEM).
DREEM is an instrument designed to assess students’ perceptions about the learning environment. Its 50 items are rated on five-point Likert scales and are divided into five subscales: perceptions about learning, teachers, atmosphere of the school, academic self-perceptions, and social self-perceptions. A global 200-point score results from the addition of the subscale scores. Scores are transformed in percentages of the respective scale, greater scores representing increasingly better perceptions.5 For this study, we adapted a Portuguese version6 to our population of residents by replacing “students” by “residents,” and “school” with “institution.”
Quality of School Life Inventory (QSL).
This 50-item instrument is designed to assess the quality of life at school. The items are rated on five-point Likert scales, two global subscales (positive affect and negative affect), and five specific subscales (identity, opportunity, status, achievement, and teachers). A 200-point global score results from the addition of subscale scores. Scores are transformed in percentages of the respective scale, greater scores representing increasingly better perceptions.7 This instrument was translated into Portuguese specifically for this study and was adapted to our population of residents as was the DREEM.
World Health Organization Quality of Life Inventory (WHOQOL).
WHOQOL is designed to assess general quality of life, according to the World Health Organization's definition “the perception of the individual of his position in life in the context of culture and value system in which he lives and in relation to his objectives, expectations, patterns and worries.” The instrument consists of 100 questions grouped in 25 facets and five domains. Domain I (physical) contains facets 1 (pain and discomfort), 2 (energy and fatigue), and 3 (sleep and rest). Domain II (psychological) comprises facets 4 (positive feelings), 5 (thinking, learning, memory, and concentration), 6 (self-esteem), 7 (bodily image and appearance), and 8 (negative feelings). Domain III (level of independence) encompasses facets 9 (mobility), 10 (daily activities), 11 (dependence on medications), and 12 (working capacity). Domain IV (social relationships) comprises facets 13 (personal relationships), 14 (social support), and 15 (sexual activity). Domain V (environment) covers facets 16 (physical safety and security), 17 (home environment), 18 (financial resources), 19 (health and social care availability and quality), 20 (opportunities for acquiring new information and skills), 21 (participation in and new opportunities for recreation/leisure), 22 (physical environment), and 23 (means of transportation). Domain VI contains only facet 24, referring to spirituality, religion, and personal beliefs. In contrast to DREEM and QSL, WHOQOL does not yield a total score. Instead, each facet and domain must be seen as a measure of one specific aspect of quality of life, with greater scores representing increasingly better perceptions. However, one of its facets (facet 25) rates perceptions about the general quality of life of the respondent.8 In this study, a validated Portuguese version was used.9
Because none of the instruments had been applied to medical residents, a reliability analysis using Cronbach's alpha coefficients was applied to each.
The CPRs’ compliance questionnaire yielded dichotomous categorizing variables: weekly duty hours (1 = ≤ 60 hours, 0 = > 60 hours), in-house or on-call weekly duty hours (1 = ≤ 24 hours, 0 = > 24 hours), consecutive duty hours (1 = ≤ 36 hours, 0 = > 36 hours), weekly day-off (1 = yes, 0 = no), 30 consecutive days on vacation per year (1 = yes, 0 = no), supervision most of the time (1 = yes, 0 = no). Comparisons between each category of residents were performed by z tests for proportions. Student's t-tests were used to compare DREEM and SQL subscale and total scores between each category. The same tests were used to compare WHOQOL facets and domains between categories. Sample sizes were defined a priori on the basis of a pilot sample consisting of 25 DREEM scores. It was estimated that at least 13 residents (the maximum sample size across subscales) had to be included in each category to detect a difference between categories equal to 1.2 standard deviation (SD) from the mean, assuming α = .05 and 1−β = .2. Sample size calculations were performed on GB-STAT v. 6.5 (Dynamic Microsystems, Inc., 1997). All other statistical analysis were performed on Statistica for Windows v. 4.5 (StatSoft, Inc., 1993). The level of significance was set to Type I error probability of less than 5%.
In total, 62 residents (70%) completed the four instruments. The respondents were 37 residents (60%) from institution A, 11 residents (18%) from institution B, seven residents (11%) from institution C, and seven residents (11%) from institution D. The distribution of residents according to specialty was as follows: 13 residents (21%) in anesthesiology (from Institution A), 19 residents (31%) in surgery (11 from institution A and eight from institution B), 21 residents (34%) in clinical areas (13 from institution A, three from institution B, and five from institution C), and nine residents (14%) in pediatrics (two from institution C and seven from institution D). Twenty-nine participants (47%) were first-year, 21 (34%) were second-year, nine (14%) were third-year, and four (5%) were fourth-year residents. There were 44 (71%) male residents. Mean (SD) age of residents in the sample was 25.95 (1.44) years.
Overall Cronbach's alpha coefficients of WHOQOL, QSL, and DREEM were .79, .84, and .92, respectively. Table 1 shows the distribution of residents according to the CPRs compliance of their respective programs. No resident reported their program complied with all seven CPRs.
Perceptions about status, identity, teachers, global quality of life in residency, quality of sleep and rest, opportunities to acquire new information, and general quality of life were significantly lower among residents who reported > 60 routine working hours per week than those reporting ≤ 60 routine work hours. Residents who reported working > 36 consecutive hours on duty reported lower perceptions about positive affect, identity, academic achievements, general quality of life in residency, atmosphere of the residency, quality of sleep and rest, and opportunities to acquire new information than did those reporting ≤ 36 consecutive hours on duty. (see Table 2).
Residents who reported not having one day per week free from clinical and educational activities scored lower on their perceptions of positive affect, identity, teaching, achievement, general quality of life in residency, and their educational environment than did those who reported being given one day off a week. Residents who reported they were not allowed 30 consecutive vacation days rated their educational environment, personal relationships, atmosphere at home, opportunities to acquire new information and their general quality of life lower than did those with 30 consecutive vacation days (see Table 3). Finally, residents who reported spending < 10% of their duty hours on formal educational activities reported lower general quality of life in residency, educational environment, positive affect, and opportunities to acquire new information than did those reporting ≥ 10% of duty hours dedicated to formal educational activities (see Table 4).
Fifty-two residents (84%) reported feeling unsupervised during most of their patient care activities (z = 7.39; p = 0). This subset rated faculty lower on QSL (65.25 ± 15.28) than did those not experiencing such a feeling (76.33 ± 11.27) (p = .03). The median (25th; 75th percentiles) of the estimated percentage of duty hours under direct faculty supervision was 60% (30%; 80%).
The present study disclosed violations to Brazil's National Committee on Medical Residency's Common Program Requirements, as most residents reported working more than the maximum recommended number of weekly duty hours, and more than 36 consecutive hours on duty. Additionally, significant proportions of residents reported being on call for more than 24 hours per week and not being allowed to enjoy 30 consecutive days on vacation per year of training. Almost half of the residents reported not having one day-off each week. Residents reporting CPRs violations rated several aspects of general quality of life, quality of life in the residency, and the educational environment lower than did those reporting their programs complied.
A survey of surgery residents about work-hour reforms showed that residents perceived sleep deprivation had negatively affected their work and suggested that a limitation of work hours might improve their quality of life. However, in the same study, a significant percentage of senior residents had negative perceptions of work-hour limitations, particularly with respect to consequences for patient care. Furthermore, residents who had experienced work-hour restrictions were less positive about work-hour limitations than were those who had not yet experienced them.10 It has been argued that decreasing surgery residents’ duty hours may have profound impact on the quality of patient care.3 In contrast to the above-mentioned opinions, some suggest that the educational content of residents’ work is far more important than are the actual hours spent in the hospital.11 Our study's results add to this controversy by showing that residents who reported violations of CPRs’ maximum duty hour limit tend to rate important aspects of their general quality of life, of the quality of life in the residency, and their learning environment lower.
It has been demonstrated that stress among physicians is directly proportional to their workload.12 Also the increase in nights on call significantly increased the stress and dissatisfaction of residents with the faculty and the learning environment.13 In addition to these findings, our study has shown that residents overwhelmed by excessive routine duty hours, excessive number of on-call hours, and/or subjected to long periods on duty tend to have lower perceptions of other aspects, such as their status and identity as residents, their academic achievements, the educational atmosphere of the institution, their teachers, and their opportunities to acquire new information.
Compliance with requirements regarding the allocation of at least 10% of duty time to formal educational activities was positively associated with better total scores and most subscales scores of both QSL and DREEM in addition to improved residents’ perceptions about their opportunity to acquire new information. These findings suggest residents’ perceptions of their quality of life in the residency and of the institutional learning environment may be enhanced by careful curriculum planning and timely content administration.
In our study's sample, only 16% of the residents felt themselves supervised during most patient care activities. The feeling of being unsupervised most of the time was associated with lower ratings for their instructors. Supervision is a key element for the success of residents’ training. In fact, it has been demonstrated that faculty supervision increases residents’ adherence to published guidelines in emergency medicine.14 Supervision has to be tailored to residents’ needs and progressively decreased in order to foster residents’ independence from faculty. Methods to measure residents’ competence at procedures involved in patient care must be used and developed so that learning can be objectively documented by both faculty and residents before direct supervision is decreased.15 Such a strategy may potentially decrease the residents’ perceptions of abandonment and careless teaching that may impair the resident–faculty relationship.
Program directors are responsible for their programs’ compliance with national regulations, but they are equally responsible for monitoring their residents’ well being.2 This study has demonstrated that the instruments used to assess general quality of life (WHOQOL), quality of life in residency (QSL), and the educational environment (DREEM), although never used before to assess resident's perceptions, demonstrated high internal consistency (reliability) and discriminant validity.16
Our study's results suggest that the implementation of Brazil's CPRs in medical residency may be not just a matter of coping with bureaucracy, but may constitute, by itself, a strategy for providing residents better quality of life and greater satisfaction with the educational environment. However, whether improving the quality of life and residents’ satisfaction with their programs is associated with the attainment of the program educational goals has yet to be demonstrated.
1 Nahrwold DI. Toward a better residency. J Laparoendosc Adv Surg Tech A. 1998;8:335–9.
2 Accreditation Council Graduate Medical Education (ACGME). Common Program Requirements 〈www.acgme.org
〉. Accessed in 22 January 2004. Chicago, IL: ACGME, 2004.
3 Lowenstein J. Where have all the giants gone? Reconciling medical education and the traditions of patient care with limitations on resident work hours. Perspect Biol Med. 2003;46:273–82.
5 Roff S, McAleer S, Harden RM, et al. Development and validation of the Dundee Ready Education Environment Measure (DREEM). Med Teach. 1997;19:295–9.
6 Vieira JE, Nunes MPT, Martins MA. Directing student response to early patient contact by questionnaire. Med Educ. 2003;37:1–7.
7 Mok MMC, Flynn M. Establishing longitudinal factorial construct validity of the quality of school life scale for secondary students. J Appl Meas. 2002;3:400–20.
8 The WHOQOL Group. The World Health Organization quality of life assessment (WHOQOL): development and general psychometric properties: Soc Sci Med. 1998;46:1569–85.
9 Fleck MPA, Louzada S, Xavier M, et al. Application of the Portuguese version of the instrument for the assessment of the quality of life of the World Health Organization (WHOQOL). Rev Saude Publica. 1999;33:198–205.
10 Whang EE, Perez A, Ito H, Mello MM, Ashley SW, Zinner MJ. Work hours reform: perceptions and desires of contemporary surgical residents. J Am Coll Surg. 2003;197:624–30.
11 Boex JR, Leahy PJ. Understanding residents’ work: moving beyond counting hours to assessing educational value. Acad Med. 2003;78:939–44.
12 Bergman B, Ahmad F, Stewart DE. Physician health, stress and gender at a university hospital. J Psychosom Res. 2003;54:171–8.
13 Seelig CB. Quantitating qualitative issues in residency training: development and testing of a scaled program evaluation questionnaire. J Gen Intern Med. 1993;8:610–3.
14 Sox CM, Burstin HR, Orav EJ, et al. The effect of supervision of residents on quality of care in five university-affiliated emergency departments. Acad Med. 1998;73:776–82.
15 De Oliveira Filho GR. The construction of learning curves for basic skills on anesthetic procedures: an application for the cumulative sum method. Anesth Analg. 2002;95:411–6.
© 2005 Association of American Medical Colleges
16 Trochim WMK. The Research Methods Knowledge Basis. Cincinnati: Atomic Dog Publishing, 2001.