Resident Attitudes Regarding the Impact of the 80–Duty-Hours Work Standards
http://www.100md.com
《美国整骨期刊》
The authors report on results from a survey assessing the attitudes of medical residents toward the American Osteopathic Association and the Accreditation Council for Graduate Medical Education duty-hour standards that became effective for all accredited residency programs on July 1, 2003. Data were gathered from 128 residents in four medical specialties: family medicine, general surgery, internal medicine, and obstetrics and gynecology. Participating residents worked at four teaching hospitals with allopathic, osteopathic, or dual-accredited programs. The dominant response of medical residents to duty-hour restrictions is clearly—though not uniformly—positive. Residents tend to agree that there are safety benefits for patients and quality-of-life benefits for themselves. A consistent pattern of positive responses toward the standards among internal medicine residents contrasts with less favorable responses among residents in general surgery programs. Gender differences are noted as well, with women generally more positive about duty-hour restrictions than their male colleagues. Male residents in surgery and obstetrics especially tend to agree that duty-hour restrictions could have negative effects on physician education with regard to their continuity of experience. The most consistent pattern in resident survey responses appears to be by medical specialty, perhaps reflecting variations in the nature of patient care and contact in each specialty.
As of July 1, 2003, common standards concerning duty hours became effective for all postgraduate physician-training programs accredited by the American Osteopathic Association (AOA) and the Accreditation Council for Graduate Medical Education (ACGME). The accreditation bodies were prompted to impose these changes because of: (1) a growing recognition of the deleterious effects of sleep deprivation on the performance of residents, (2) public fears that overly long medical resident duty hours could compromise patient safety and residents' well-being, and (3) the fact that, on average, hospital inpatients tend to be sicker and require more care than in previous decades.1
Facing the possibility that legislation could be enacted or governmental regulatory bodies could step in to set duty-hour standards for the medical profession as a whole,2 a policy of self-policing in the form of common duty-hour standards was seen as the best solution for protecting the interests of patients and medical residents. What ensued, though, was a high level of angst—expressed by medical residents and program directors alike—concerning the impact that the resulting limit of 80 duty-hours per week might have on residency training, patient care, and overall resident well-being. Several questions, including the following, were raised:2,3
Will the restriction in duty hours lead to more limited exposure of residents to different cases and reduced clinician experience of continuity of care
Will the restriction in duty hours really improve patient safety and protect patients against possible medical errors made by fatigued residents
Will the restriction in duty hours lengthen postgraduate training in some disciplines
Will the restriction in duty hours affect the ability of residents to moonlight
Evidence from the few studies that have addressed these questions is far from consistent. In a survey of clinical surgical faculty members, a majority of respondents believed that a reduction in duty hours would compromise surgical education,4–6 whereas a similar study found that only 14% of respondents thought the benefits of a reduction in duty hours would outweigh the costs.5
In a survey of surgical residents, a majority of respondents indicated their expectation that restrictions in duty hours would have a positive effect on the quality of their work, as well as on their personal lives. Although surgical residents thought restrictions on duty hours would benefit them personally, fewer than half (45%) believed that such changes would have a positive impact on patient care.7
After the 1989 enactment of the New York State 405 (Bell) Workforce Regulations, surgical residents in New York state reported a better quality of life for themselves; however, a substantial portion (nearly 60%) stated their belief that these regulations had a negative impact on the continuity of care for surgical patients.8 Among these same respondents, 35% reported that the regulations harmed patient care, whereas 44% indicated there was no change.8
When internal medicine residents in two New York training programs were surveyed, they indicated less fatigue and more personal time as major benefits of the 405 Regulations, but there was no consensus on whether changes brought about through the regulations improved internal medicine practice and clinical supervision.9
Before implementation of the AOA and ACGME 80–duty-hours standard, opinions among physicians and medical educators seemed to be split between those who saw such a proposed change as making a positive contribution to the overall goals of medical education and those who thought it would be deleterious to the educational process.10,11 Given this divergence of opinions, the surveys for the present study were designed to find out how the duty-hour standards were perceived as having an impact on those most affected by them: the residents.
Approximately seven months into the implementation of the standards, we conducted a survey among residents in different medical specialty programs of four institutions located within a 60-mile radius of Detroit, Michigan. The goal of this project was to compare the attitudes toward duty-hour restrictions among residents in different medical specialties. In addition, we sought to identify any demographic characteristics that might predict favorable or unfavorable attitudes toward the duty-hour standards.
Materials and Methods
Four hospital training programs in the Detroit–metropolitan area agreed to participate in the survey. All programs are either (1) accredited by the AOA, (2) accredited by the ACGME, or (3) accredited by the AOA and the ACGME (ie, dual-accredited). To ensure comparability across institutions, only those residency specialties that existed in all four institutions were included: family medicine, general surgery, internal medicine, and obstetrics and gynecology.
A researcher or qualified staff member attended an educational or business meeting of each resident group and asked for residents' participation in the survey. Survey questionnaires were distributed, completed, and collected at that time to ensure maximum participation of respondents (Figure).
Data were collected through self-administered questionnaires that asked respondents their opinions on how the restrictions on duty hours may have affected work-related issues. Using Likert-type fixed response scales, residents indicated their level of agreement with 22 statements, with response categories ranging from 1 (strongly disagree) to 5 (strongly agree).
One question (item 23) asked respondents to choose which of two options has a bigger impact on patient care, "continuity" or "providers that are not tired." In addition, there were six questions (items 24 through 29) related to the demographic characteristics of respondents. Completion of the survey instrument by the residents took approximately five minutes.
The 22 items on the survey instrument were designed to probe residents' attitudes concerning the potential areas of impact for the 80–duty-hours standard. Several of these questions could be grouped as addressing distinct areas of concern. For instance, seven questions probed the extent to which residents agreed with the notion that the duty-hour standards would have a negative impact on graduate medical education (items 2 [reversed], 5, 8, 10, 16, 18, and 21). Responses to these seven questions were highly consistent (mean interitem correlation: r=0.51; Cronbach's =0.88) and were subsequently averaged into a single-scale score, indicating the extent to which a respondent experienced or anticipated experiencing negative impacts and restrictions in current and future physician training as a result of the standards. Items included statements addressing such concerns as possible restrictions in resident exposure to different kinds of patients, negative impact on a resident's continuity-of-care experience, reduction in responsibilities for individual patient care, and so forth.
Another three questions were combined into a scale that probed issues of patient safety and protection from medical errors (items 1, 4, and 11) (mean interitem correlation: r=0.75; Cronbach's =0.90); the scale score indicates the extent to which residents experienced or anticipated experiencing benefits from the duty-hour standards in terms of greater patient safety, fewer medical errors, and greater protection of housestaff.
An additional three questions addressed residents' opinions concerning the impact of the duty-hour standards on residents' quality of life (items 6, 13, and 17), which includes less emphasis on the service component and more access to clinical supervision (mean interitem correlation: r=0.46; Cronbach's =0.72). In addition, issues such as the possible impact of the duty-hour standards on resident indebtedness (item 12) and the future length of postgraduate trainings (item 3) were also probed.
In order to compare responses to the survey questions among residents in different medical specialty programs, we used a two-way analysis of variance, in which the attitude scores were the dependent variables. All medical specialty program comparisons were adjusted for gender effects (as sex appeared to be the only demographic characteristic that was frequently associated with different attitudes toward the 80–duty-hours standard). Responses to some of the attitude statements generated different patterns by sex within medical specialty. For such interactions of medical specialty and respondent's sex, separate means for specialty groups among men and women are reported if the interaction effect is statistically significant (P<.05). All analyses were performed using SPSS statistical software (version 11.5 for Windows, SPSS Inc, Chicago, Ill).
Results
A total of 128 out of the 227 residents enrolled in the four medical specialty areas at the four study recruitment sites participated in the survey for a response rate of 56.4%.
Of the 128 participating residents, 43 (34%) were in family medicine, 22 (17%) in general surgery, 32 (25%) in internal medicine, and 31 (24%) in obstetrics/gynecology. Among respondents, 78 residents (61%) indicated that they were married. Seventy-six participating residents (59%) were women. Sixty-four respondents (50%) were first- or second-year residents (another 64 [50%] were in third to seventh years of residency). Seventy participating residents (55%) were trained in colleges of osteopathic medicine; 58 (45%) were trained at allopathic schools.
The data in Table 1 show the descriptive statistics for the attitude and opinion scales used in the subsequent analysis. Because the scale scores are averaged composites of the responses to the individual questions, values at or close to 1 indicate strong disagreement with all of the statements that are part of the scale, whereas values at or close to 5 indicate strong agreement.
It can be noted that the residents who participated in this survey tended to agree most with the notion that the 80–duty-hours standard would improve patient safety and reduce medical errors (mean, 4.03; median, 4.00; it should also be noted that the general closeness of means and medians in Table 1 indicates that all of these variables have fairly symmetric distributions around the mean).
The statement receiving the second highest rate of agreement involved residents' anticipation of a better quality of life as a result of the standards (mean, 3.5).
All other scale means were either slightly below 3, which indicated, on average, a neutral response—or substantially below 3, which indicated overall disagreement. Thus, residents disagreed more strongly with the suggestion that, as a result of the duty-hour standards, patients would be limited in their access to care (mean, 2.09). Residents also tended to disagree with the suggestion that the duty-hour standards would have a negative impact on the quality of physician training.
In sum, although individual opinions and views among residents varied (see the standard deviations and range of scores in Table 1), the dominant response was clearly positive in its overall evaluation of the changes brought about by the 80–duty-hours standard.
A somewhat different picture emerges if residents' responses are analyzed separately by medical specialty and/or sex. Originally, resident responses were also examined by marital status, years in medical residency, and professional affiliation (osteopathic physician versus allopathic physician). None of these additional variables, individually or jointly, appears to affect resident responses, however, so marital status, years of residency, and professional affiliation were subsequently dropped from the analysis.
Participants' responses to survey questions differed by medical specialty and sex, both of which are confounding factors. For example, 76% of all obstetrics/gynecology residents surveyed were women, whereas 39% of the surgery residents surveyed were women. Therefore, all reported response estimates by medical specialty area were adjusted for sex, and all reported response estimates by sex were adjusted for specialty membership (Table 2).
As the analysis of variance results (bottom of Table 2) indicate, medical residents view the impact of the 80–duty-hours standard quite differently based on their chosen medical specialties. In particular, residents in internal medicine are consistently positive about the 80–duty-hours standard. For the negatively worded scales in columns 2 (Negative Impact on Physician Training) and 4 to 6 (Lengthen Residency Period, Increase Student Debt, Reduce Patient Access to Care), internal medicine residents either show the strongest or second strongest level of disagreement, with mean responses ranging from 2.03 to 2.71.
Although not all individual outcome variables show statistically significant differences by medical specialty, an omnibus multivariate test (Hotelling's T-square distribution with the approximate F statistic: 3.38, P<.01) does so.12 Thus, it can be said that the internal medicine residents tend to disagree the most with the view that the 80–duty-hours standard (1) will have a negative impact on physician education, (2) will result in longer residencies and greater student debt, or (3) will limit patient access to care. At the same time, internal medicine residents show the highest level of agreement with statements extolling the patient safety or resident quality-of-life benefits of the duty-hour standards (mean values of 4.46 and 3.74, respectively).
General surgery residents display a quite different pattern. As a group, general surgery residents tend to be least favorably disposed toward the 80–duty-hours standard. More than medical residents in other specialties, surgery residents often agree with the statement that the standards "will have a serious, negative effect on training future physicians" (item 10). Incidentally, men completing a surgery residency hold this belief more strongly (mean, 3.51) than do women in this specialty group (mean, 3.10). Surgery residents are also often inclined to indicate their belief that the standards will prolong residency training (mean, 3.11) and increase student debt loads (mean, 3.39). At the same time, surgery residents are less likely to agree that the standards could have a positive impact in terms of patient safety and resident quality of life (mean, 3.04 and 2.64, respectively).
This pattern of surgery residents differing from other responding groups can be confirmed when mean response scores for all 22 Likert scale survey items are rank-ordered by medical specialty. For internal medicine, family medicine, and obstetrics/gynecology, the highest levels of agreement occur with statements that stipulate positive consequences of the duty-hour standards: they were "necessary to protect housestaff" (item 1), "will benefit patient care and safety" (item 4), and "will improve the quality of life for residents" (item 13). Surgery respondents appear to agree most strongly with statements stipulating problems or negative consequences of the duty-hour standards; the highest means, indicating greater levels of agreement, are noted for "runs contrary to the normal work of a profession" (item 8), "will reduce my clinical exposure" (item 18), and "necessitates high quality sign-out to ensure the follow-up of the previous shift's patient care" (item 22).
The data in Table 2 also show that women in all medical specialties have a consistently more positive view of the 80–duty-hours standard. Women are more likely than men to agree strongly that the standards "will improve patient care and safety" (item 4) (4.03 versus 3.68); they anticipate greater positive impacts on their quality of life (3.46 versus 3.22); and they consistently expect fewer impacts, such as prolonging their training (.84 versus 2.88) or greater indebtedness (2.92 versus 3.10).
The only small exception to this tendency appears within the group of internal medicine residents, in which some women tend to agree slightly more than their male counterparts that the duty-hour standards may negatively affect physician education. However, the single largest difference of opinion between men and women confirms this general gender pattern: among obstetrics/gynecology residents, male residents are far more convinced of the negative impact on physician education than their female counterparts (3.74 versus 2.59).
Finally, the residents were asked which situation they thought would have a bigger impact on the quality of patient care: patients being able to count on continuity of care among their providers in the hospital, or providers not being fatigued (item 23). Although no differences by sex were noted, once again, differences in response were noted by medical specialty. Whereas 64% of internal medicine residents and 53% of family medicine residents indicate their belief that it is more important for patient care that providers are not fatigued, 78% of surgery and 61% of obstetrics/gynecology residents emphasize provider continuity as more important than provider fatigue (P<.01).
Comment
The introduction of standards for restricting duty hours was in response to a deep concern on the part of medical educators and consumers alike regarding assurances that residents are not overextended in their training by long duty hours, which could potentially put residents and their patients at risk. Before the duty-hour standards were introduced, there was much discussion and debate within the medical community about the impact that the then-potential standards might have. Therefore, we decided to examine residents' experiences and attitudes shortly after these standards were implemented.
Six months after the standards went into effect, anecdotal reports indicated that many surgery residents and, to some extent, obstetrics residents were suggesting that continuity of care was being sacrificed. Continuity of care can be interpreted as following a patient from first encounter to some closure point, or as exposure to all facets of a presenting condition through multiple patients. The window of "opportunity to treat" is clearly narrower for surgery and obstetrics patients. These patients cannot be easily rescheduled to fit into residents' schedules, whereas there is a greater opportunity in internal medicine and family medicine for creating continuity in patient encounters.
Medical educators will have to address these concerns, especially in surgery and obstetrics, to ensure that continuity of experiences is still being met for the residents. To ensure optimal patient care, educators and providers will need to closely monitor possible information deficits among multiple residents treating the same patient (when the resident has reached his or her maximum number of duty hours) that could compromise quality of care.
That residents were seemingly not overly concerned about the impact of duty-hour standards on indebtedness was an unexpected finding, given that these standards may restrict residents' abilities to supplement their income through moon-lighting. Even though surgery and obstetrics residents are somewhat more concerned about indebtedness than are family and internal medicine residents, these differences are not substantial (Table 2). This result may be because, under their previous schedules, surgery and obstetric residents were precluded from much additional work anyway. Hence, the 80–duty-hours standard did not necessarily "deprive" them of any moonlighting opportunities. Family medicine residents have historically worked shorter days than have surgical residents. Again, the 80–duty-hours standard may not have an impact on their residencies. Although restricting duty hours would appear to have the greatest likelihood of affecting internal medicine residents' ability to moonlight, this concern was not expressed by them as a group in the present survey.
Overall, family medicine, internal medicine, and, to a lesser extent, obstetrics residents indicate their belief that the 80–duty-hours standard will benefit their personal lives and will protect their well-being. Few of the respondents, including surgery residents, state the belief that the standards will have a negative impact on patients' access to care, and there is broad agreement—except perhaps among surgery residents—that the standards will benefit patient safety and reduce medical errors.
The residents who appear to view the duty-hour standards in the most negative light are male residents in surgery and obstetrics/gynecology. These residents are far more likely than any other group surveyed to anticipate negative consequences for the quality of physician education. The fact, however, that women in the same specialties do not appear to share these fears may introduce doubt to the notion that these opinions reflect systemic problems. Nonetheless, no change in standards can reduce all the problems in a system, and further monitoring is clearly warranted to ensure the overall integrity and quality of residents' educational experiences and patient care.
Footnotes
This manuscript was submitted for publication to JAOA—The Journal of the American Osteopathic Association on July 27, 2004.
Oakwood Southshore Medical Center in Trenton, Mich (Zonia), Genesys Regional Medical Center in Grand Blanc, Mich (LaBaere), Michigan State University in East Lansing (Stommel), and Henry Ford Bi-County Hospital in Warren, Mich (Tomaszewski).
References
1. Philibert I, Friedmann P, Williams WT; ACGME Work Group on Resident Duty Hours. Accreditation Council for Graduate Medical Education. New requirements for resident duty hours. JAMA.2002; 288:1112 –1114.
2. Croasdale M. Medicine limits resident hours before legislation can. Am Med News. July 8/15,2002; 45:14 .
3. Croasdale M. ACGME gives final nod to 80-hour workweek. Am Med News. March 10,2003; 46:11 –12.
4. Winslow ER, Bowman MC, Klingensmith ME. Surgeon workhours in the era of limited resident workhours. J Am Coll Surg.2004; 198:111 –117.
5. Niederee MJ, Knudtson JL, Byrnes MC, Helmer SD, Smith RS. A survey of residents and faculty regarding work hour limitations and surgical training programs. Arch Surg.2003; 138:663 –669; discussion 669–671.
6. Fischer JE. Continuity of care: a casualty of the 80-hour work week. Acad Med.2004; 79:381 –383.
7. 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 –630.
8. Whang EE, Mello MM, Ashley SW, Zinner MJ. Implementing resident work hour limitations: lessons from the New York State experience. Ann Surg.2003; 237:449 –455.
9. Conigliaro J, Frishman WH, Lazar EJ, Croen L. Internal medicine housestaff and attending physician perceptions of the impact of the New York State Section 405 regulations on working conditions and supervision of residents in two training programs. J Gen Intern Med.1993; 8:502 –507.
10. Whitcomb ME. More on resident duty-hours limits. Acad Med. 2004;79:377 –378.
11. Barone JE, Ivy ME. Resident work hours: the five stages of grief. Acad Med. 2004;79:379 –380.
12. Winer BJ, Brown DR, Michels KM. Statistical Principles in Experimental Design. 3rd ed. New York, NY: McGraw-Hill;1991 .(Susan C. Zonia, PhD; Rich)
As of July 1, 2003, common standards concerning duty hours became effective for all postgraduate physician-training programs accredited by the American Osteopathic Association (AOA) and the Accreditation Council for Graduate Medical Education (ACGME). The accreditation bodies were prompted to impose these changes because of: (1) a growing recognition of the deleterious effects of sleep deprivation on the performance of residents, (2) public fears that overly long medical resident duty hours could compromise patient safety and residents' well-being, and (3) the fact that, on average, hospital inpatients tend to be sicker and require more care than in previous decades.1
Facing the possibility that legislation could be enacted or governmental regulatory bodies could step in to set duty-hour standards for the medical profession as a whole,2 a policy of self-policing in the form of common duty-hour standards was seen as the best solution for protecting the interests of patients and medical residents. What ensued, though, was a high level of angst—expressed by medical residents and program directors alike—concerning the impact that the resulting limit of 80 duty-hours per week might have on residency training, patient care, and overall resident well-being. Several questions, including the following, were raised:2,3
Will the restriction in duty hours lead to more limited exposure of residents to different cases and reduced clinician experience of continuity of care
Will the restriction in duty hours really improve patient safety and protect patients against possible medical errors made by fatigued residents
Will the restriction in duty hours lengthen postgraduate training in some disciplines
Will the restriction in duty hours affect the ability of residents to moonlight
Evidence from the few studies that have addressed these questions is far from consistent. In a survey of clinical surgical faculty members, a majority of respondents believed that a reduction in duty hours would compromise surgical education,4–6 whereas a similar study found that only 14% of respondents thought the benefits of a reduction in duty hours would outweigh the costs.5
In a survey of surgical residents, a majority of respondents indicated their expectation that restrictions in duty hours would have a positive effect on the quality of their work, as well as on their personal lives. Although surgical residents thought restrictions on duty hours would benefit them personally, fewer than half (45%) believed that such changes would have a positive impact on patient care.7
After the 1989 enactment of the New York State 405 (Bell) Workforce Regulations, surgical residents in New York state reported a better quality of life for themselves; however, a substantial portion (nearly 60%) stated their belief that these regulations had a negative impact on the continuity of care for surgical patients.8 Among these same respondents, 35% reported that the regulations harmed patient care, whereas 44% indicated there was no change.8
When internal medicine residents in two New York training programs were surveyed, they indicated less fatigue and more personal time as major benefits of the 405 Regulations, but there was no consensus on whether changes brought about through the regulations improved internal medicine practice and clinical supervision.9
Before implementation of the AOA and ACGME 80–duty-hours standard, opinions among physicians and medical educators seemed to be split between those who saw such a proposed change as making a positive contribution to the overall goals of medical education and those who thought it would be deleterious to the educational process.10,11 Given this divergence of opinions, the surveys for the present study were designed to find out how the duty-hour standards were perceived as having an impact on those most affected by them: the residents.
Approximately seven months into the implementation of the standards, we conducted a survey among residents in different medical specialty programs of four institutions located within a 60-mile radius of Detroit, Michigan. The goal of this project was to compare the attitudes toward duty-hour restrictions among residents in different medical specialties. In addition, we sought to identify any demographic characteristics that might predict favorable or unfavorable attitudes toward the duty-hour standards.
Materials and Methods
Four hospital training programs in the Detroit–metropolitan area agreed to participate in the survey. All programs are either (1) accredited by the AOA, (2) accredited by the ACGME, or (3) accredited by the AOA and the ACGME (ie, dual-accredited). To ensure comparability across institutions, only those residency specialties that existed in all four institutions were included: family medicine, general surgery, internal medicine, and obstetrics and gynecology.
A researcher or qualified staff member attended an educational or business meeting of each resident group and asked for residents' participation in the survey. Survey questionnaires were distributed, completed, and collected at that time to ensure maximum participation of respondents (Figure).
Data were collected through self-administered questionnaires that asked respondents their opinions on how the restrictions on duty hours may have affected work-related issues. Using Likert-type fixed response scales, residents indicated their level of agreement with 22 statements, with response categories ranging from 1 (strongly disagree) to 5 (strongly agree).
One question (item 23) asked respondents to choose which of two options has a bigger impact on patient care, "continuity" or "providers that are not tired." In addition, there were six questions (items 24 through 29) related to the demographic characteristics of respondents. Completion of the survey instrument by the residents took approximately five minutes.
The 22 items on the survey instrument were designed to probe residents' attitudes concerning the potential areas of impact for the 80–duty-hours standard. Several of these questions could be grouped as addressing distinct areas of concern. For instance, seven questions probed the extent to which residents agreed with the notion that the duty-hour standards would have a negative impact on graduate medical education (items 2 [reversed], 5, 8, 10, 16, 18, and 21). Responses to these seven questions were highly consistent (mean interitem correlation: r=0.51; Cronbach's =0.88) and were subsequently averaged into a single-scale score, indicating the extent to which a respondent experienced or anticipated experiencing negative impacts and restrictions in current and future physician training as a result of the standards. Items included statements addressing such concerns as possible restrictions in resident exposure to different kinds of patients, negative impact on a resident's continuity-of-care experience, reduction in responsibilities for individual patient care, and so forth.
Another three questions were combined into a scale that probed issues of patient safety and protection from medical errors (items 1, 4, and 11) (mean interitem correlation: r=0.75; Cronbach's =0.90); the scale score indicates the extent to which residents experienced or anticipated experiencing benefits from the duty-hour standards in terms of greater patient safety, fewer medical errors, and greater protection of housestaff.
An additional three questions addressed residents' opinions concerning the impact of the duty-hour standards on residents' quality of life (items 6, 13, and 17), which includes less emphasis on the service component and more access to clinical supervision (mean interitem correlation: r=0.46; Cronbach's =0.72). In addition, issues such as the possible impact of the duty-hour standards on resident indebtedness (item 12) and the future length of postgraduate trainings (item 3) were also probed.
In order to compare responses to the survey questions among residents in different medical specialty programs, we used a two-way analysis of variance, in which the attitude scores were the dependent variables. All medical specialty program comparisons were adjusted for gender effects (as sex appeared to be the only demographic characteristic that was frequently associated with different attitudes toward the 80–duty-hours standard). Responses to some of the attitude statements generated different patterns by sex within medical specialty. For such interactions of medical specialty and respondent's sex, separate means for specialty groups among men and women are reported if the interaction effect is statistically significant (P<.05). All analyses were performed using SPSS statistical software (version 11.5 for Windows, SPSS Inc, Chicago, Ill).
Results
A total of 128 out of the 227 residents enrolled in the four medical specialty areas at the four study recruitment sites participated in the survey for a response rate of 56.4%.
Of the 128 participating residents, 43 (34%) were in family medicine, 22 (17%) in general surgery, 32 (25%) in internal medicine, and 31 (24%) in obstetrics/gynecology. Among respondents, 78 residents (61%) indicated that they were married. Seventy-six participating residents (59%) were women. Sixty-four respondents (50%) were first- or second-year residents (another 64 [50%] were in third to seventh years of residency). Seventy participating residents (55%) were trained in colleges of osteopathic medicine; 58 (45%) were trained at allopathic schools.
The data in Table 1 show the descriptive statistics for the attitude and opinion scales used in the subsequent analysis. Because the scale scores are averaged composites of the responses to the individual questions, values at or close to 1 indicate strong disagreement with all of the statements that are part of the scale, whereas values at or close to 5 indicate strong agreement.
It can be noted that the residents who participated in this survey tended to agree most with the notion that the 80–duty-hours standard would improve patient safety and reduce medical errors (mean, 4.03; median, 4.00; it should also be noted that the general closeness of means and medians in Table 1 indicates that all of these variables have fairly symmetric distributions around the mean).
The statement receiving the second highest rate of agreement involved residents' anticipation of a better quality of life as a result of the standards (mean, 3.5).
All other scale means were either slightly below 3, which indicated, on average, a neutral response—or substantially below 3, which indicated overall disagreement. Thus, residents disagreed more strongly with the suggestion that, as a result of the duty-hour standards, patients would be limited in their access to care (mean, 2.09). Residents also tended to disagree with the suggestion that the duty-hour standards would have a negative impact on the quality of physician training.
In sum, although individual opinions and views among residents varied (see the standard deviations and range of scores in Table 1), the dominant response was clearly positive in its overall evaluation of the changes brought about by the 80–duty-hours standard.
A somewhat different picture emerges if residents' responses are analyzed separately by medical specialty and/or sex. Originally, resident responses were also examined by marital status, years in medical residency, and professional affiliation (osteopathic physician versus allopathic physician). None of these additional variables, individually or jointly, appears to affect resident responses, however, so marital status, years of residency, and professional affiliation were subsequently dropped from the analysis.
Participants' responses to survey questions differed by medical specialty and sex, both of which are confounding factors. For example, 76% of all obstetrics/gynecology residents surveyed were women, whereas 39% of the surgery residents surveyed were women. Therefore, all reported response estimates by medical specialty area were adjusted for sex, and all reported response estimates by sex were adjusted for specialty membership (Table 2).
As the analysis of variance results (bottom of Table 2) indicate, medical residents view the impact of the 80–duty-hours standard quite differently based on their chosen medical specialties. In particular, residents in internal medicine are consistently positive about the 80–duty-hours standard. For the negatively worded scales in columns 2 (Negative Impact on Physician Training) and 4 to 6 (Lengthen Residency Period, Increase Student Debt, Reduce Patient Access to Care), internal medicine residents either show the strongest or second strongest level of disagreement, with mean responses ranging from 2.03 to 2.71.
Although not all individual outcome variables show statistically significant differences by medical specialty, an omnibus multivariate test (Hotelling's T-square distribution with the approximate F statistic: 3.38, P<.01) does so.12 Thus, it can be said that the internal medicine residents tend to disagree the most with the view that the 80–duty-hours standard (1) will have a negative impact on physician education, (2) will result in longer residencies and greater student debt, or (3) will limit patient access to care. At the same time, internal medicine residents show the highest level of agreement with statements extolling the patient safety or resident quality-of-life benefits of the duty-hour standards (mean values of 4.46 and 3.74, respectively).
General surgery residents display a quite different pattern. As a group, general surgery residents tend to be least favorably disposed toward the 80–duty-hours standard. More than medical residents in other specialties, surgery residents often agree with the statement that the standards "will have a serious, negative effect on training future physicians" (item 10). Incidentally, men completing a surgery residency hold this belief more strongly (mean, 3.51) than do women in this specialty group (mean, 3.10). Surgery residents are also often inclined to indicate their belief that the standards will prolong residency training (mean, 3.11) and increase student debt loads (mean, 3.39). At the same time, surgery residents are less likely to agree that the standards could have a positive impact in terms of patient safety and resident quality of life (mean, 3.04 and 2.64, respectively).
This pattern of surgery residents differing from other responding groups can be confirmed when mean response scores for all 22 Likert scale survey items are rank-ordered by medical specialty. For internal medicine, family medicine, and obstetrics/gynecology, the highest levels of agreement occur with statements that stipulate positive consequences of the duty-hour standards: they were "necessary to protect housestaff" (item 1), "will benefit patient care and safety" (item 4), and "will improve the quality of life for residents" (item 13). Surgery respondents appear to agree most strongly with statements stipulating problems or negative consequences of the duty-hour standards; the highest means, indicating greater levels of agreement, are noted for "runs contrary to the normal work of a profession" (item 8), "will reduce my clinical exposure" (item 18), and "necessitates high quality sign-out to ensure the follow-up of the previous shift's patient care" (item 22).
The data in Table 2 also show that women in all medical specialties have a consistently more positive view of the 80–duty-hours standard. Women are more likely than men to agree strongly that the standards "will improve patient care and safety" (item 4) (4.03 versus 3.68); they anticipate greater positive impacts on their quality of life (3.46 versus 3.22); and they consistently expect fewer impacts, such as prolonging their training (.84 versus 2.88) or greater indebtedness (2.92 versus 3.10).
The only small exception to this tendency appears within the group of internal medicine residents, in which some women tend to agree slightly more than their male counterparts that the duty-hour standards may negatively affect physician education. However, the single largest difference of opinion between men and women confirms this general gender pattern: among obstetrics/gynecology residents, male residents are far more convinced of the negative impact on physician education than their female counterparts (3.74 versus 2.59).
Finally, the residents were asked which situation they thought would have a bigger impact on the quality of patient care: patients being able to count on continuity of care among their providers in the hospital, or providers not being fatigued (item 23). Although no differences by sex were noted, once again, differences in response were noted by medical specialty. Whereas 64% of internal medicine residents and 53% of family medicine residents indicate their belief that it is more important for patient care that providers are not fatigued, 78% of surgery and 61% of obstetrics/gynecology residents emphasize provider continuity as more important than provider fatigue (P<.01).
Comment
The introduction of standards for restricting duty hours was in response to a deep concern on the part of medical educators and consumers alike regarding assurances that residents are not overextended in their training by long duty hours, which could potentially put residents and their patients at risk. Before the duty-hour standards were introduced, there was much discussion and debate within the medical community about the impact that the then-potential standards might have. Therefore, we decided to examine residents' experiences and attitudes shortly after these standards were implemented.
Six months after the standards went into effect, anecdotal reports indicated that many surgery residents and, to some extent, obstetrics residents were suggesting that continuity of care was being sacrificed. Continuity of care can be interpreted as following a patient from first encounter to some closure point, or as exposure to all facets of a presenting condition through multiple patients. The window of "opportunity to treat" is clearly narrower for surgery and obstetrics patients. These patients cannot be easily rescheduled to fit into residents' schedules, whereas there is a greater opportunity in internal medicine and family medicine for creating continuity in patient encounters.
Medical educators will have to address these concerns, especially in surgery and obstetrics, to ensure that continuity of experiences is still being met for the residents. To ensure optimal patient care, educators and providers will need to closely monitor possible information deficits among multiple residents treating the same patient (when the resident has reached his or her maximum number of duty hours) that could compromise quality of care.
That residents were seemingly not overly concerned about the impact of duty-hour standards on indebtedness was an unexpected finding, given that these standards may restrict residents' abilities to supplement their income through moon-lighting. Even though surgery and obstetrics residents are somewhat more concerned about indebtedness than are family and internal medicine residents, these differences are not substantial (Table 2). This result may be because, under their previous schedules, surgery and obstetric residents were precluded from much additional work anyway. Hence, the 80–duty-hours standard did not necessarily "deprive" them of any moonlighting opportunities. Family medicine residents have historically worked shorter days than have surgical residents. Again, the 80–duty-hours standard may not have an impact on their residencies. Although restricting duty hours would appear to have the greatest likelihood of affecting internal medicine residents' ability to moonlight, this concern was not expressed by them as a group in the present survey.
Overall, family medicine, internal medicine, and, to a lesser extent, obstetrics residents indicate their belief that the 80–duty-hours standard will benefit their personal lives and will protect their well-being. Few of the respondents, including surgery residents, state the belief that the standards will have a negative impact on patients' access to care, and there is broad agreement—except perhaps among surgery residents—that the standards will benefit patient safety and reduce medical errors.
The residents who appear to view the duty-hour standards in the most negative light are male residents in surgery and obstetrics/gynecology. These residents are far more likely than any other group surveyed to anticipate negative consequences for the quality of physician education. The fact, however, that women in the same specialties do not appear to share these fears may introduce doubt to the notion that these opinions reflect systemic problems. Nonetheless, no change in standards can reduce all the problems in a system, and further monitoring is clearly warranted to ensure the overall integrity and quality of residents' educational experiences and patient care.
Footnotes
This manuscript was submitted for publication to JAOA—The Journal of the American Osteopathic Association on July 27, 2004.
Oakwood Southshore Medical Center in Trenton, Mich (Zonia), Genesys Regional Medical Center in Grand Blanc, Mich (LaBaere), Michigan State University in East Lansing (Stommel), and Henry Ford Bi-County Hospital in Warren, Mich (Tomaszewski).
References
1. Philibert I, Friedmann P, Williams WT; ACGME Work Group on Resident Duty Hours. Accreditation Council for Graduate Medical Education. New requirements for resident duty hours. JAMA.2002; 288:1112 –1114.
2. Croasdale M. Medicine limits resident hours before legislation can. Am Med News. July 8/15,2002; 45:14 .
3. Croasdale M. ACGME gives final nod to 80-hour workweek. Am Med News. March 10,2003; 46:11 –12.
4. Winslow ER, Bowman MC, Klingensmith ME. Surgeon workhours in the era of limited resident workhours. J Am Coll Surg.2004; 198:111 –117.
5. Niederee MJ, Knudtson JL, Byrnes MC, Helmer SD, Smith RS. A survey of residents and faculty regarding work hour limitations and surgical training programs. Arch Surg.2003; 138:663 –669; discussion 669–671.
6. Fischer JE. Continuity of care: a casualty of the 80-hour work week. Acad Med.2004; 79:381 –383.
7. 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 –630.
8. Whang EE, Mello MM, Ashley SW, Zinner MJ. Implementing resident work hour limitations: lessons from the New York State experience. Ann Surg.2003; 237:449 –455.
9. Conigliaro J, Frishman WH, Lazar EJ, Croen L. Internal medicine housestaff and attending physician perceptions of the impact of the New York State Section 405 regulations on working conditions and supervision of residents in two training programs. J Gen Intern Med.1993; 8:502 –507.
10. Whitcomb ME. More on resident duty-hours limits. Acad Med. 2004;79:377 –378.
11. Barone JE, Ivy ME. Resident work hours: the five stages of grief. Acad Med. 2004;79:379 –380.
12. Winer BJ, Brown DR, Michels KM. Statistical Principles in Experimental Design. 3rd ed. New York, NY: McGraw-Hill;1991 .(Susan C. Zonia, PhD; Rich)