C. Christopher Smith, MD

There is little scientific evidence about professional help‐seeking behavior among resident physicians. Although junior physicians have many sources of information available to them in the course of clinical practiceprint materials, internet resources, curbside consultations, and advice from senior residents and facultywe have little empirical knowledge about when, why, and how physician trainees ask for help.

To study this phenomenon, we examined the use of a medical procedure service (MPS) by resident physicians. The MPS is an inpatient service at a Boston teaching hospital that provides education, supervision, and evaluation of internal medicine residents who perform common bedside procedures; it has been described previously.1 Residents who call the MPS review an online curriculum with self‐assessment quizzes, perform procedures with faculty supervision and feedback, and assess their own performance using online checklists. This program has been available to internal medicine residents since 2002. In a previous study, we found that residents reported greater comfort performing bedside procedures when they used the procedure service, when the operator was a postgraduate year (PGY)2 or PGY3 resident (compared to PGY1 residents), and while placing central venous catheters (CVCs) (compared to thoracenteses).2

The goal of the current study was to examine help‐seeking behavior among resident physicians as they placed CVCs and performed thoracenteses. We interpreted the decision to use the MPS to indicate that the resident successfully sought and received assistance from pulmonary attending physicians or interventional pulmonary fellows. We hypothesized that: (1) residents earlier in their training would choose to use the procedure service due to their relative lack of experience; (2) they would seek consultation when the procedure was performed in high‐risk patients, as indicated by the number of comorbidities, presence of medications that increase the risk of bleeding, and treatment in an intensive care unit; and (3) residents would be less likely to call the MPS for urgent or emergent situations, when timely assistance may be difficult to obtain. To examine the potentially confounding influence of procedures supervised by non‐MPS physicians, we also investigated differences between informally supervised procedures (i.e. by a non‐MPS attending or fellow) and unsupervised procedures (i.e. no attending or fellow supervision) to determine whether any significant differences in their characteristics existed.

Methods

Results

Discussion

To understand professional help‐seeking behavior by internal medicine resident physicians, we studied factors associated with the use of a MPS for performing 2 common bedside procedures. We found that residents used the MPS more often when they performed procedures on patients with more comorbidities and less often during urgent or emergent procedures.

These results are consistent with our hypothesis that residents use formal supervision when caring for high‐risk patients. We had also hypothesized that they would seek the MPS for patients on medications that increase the risk of bleeding, but this was not borne out. One possible explanation is that invasive procedures on anticoagulated patients may be deferred or avoided. Additionally, we did not collect prothrombin times nor platelet count, which may represent better proxies for coagulopathy. Our hypothesis that residents would not seek the MPS for urgent and emergent procedures was confirmed; the time delay between contacting the faculty member and performing the procedure may have inhibited or obviated consultation of the MPS. We hypothesized that interns would use the MPS preferentially; we found instead that level of training did not influence use of the MPS. A resident early in training may struggle with the balance between autonomy and supervision, wanting instead to establish himself as able to solve clinical problems independently and by seeking consultation only as a last resort. Alternatively, interns may be primarily supervised by their residents and may seek expert assistance only for particularly challenging or high‐risk cases. Additionally, as newcomers to the training program, they may not be well acquainted with the role and availability of the service (although periodic announcements were made throughout the year). Our examination of procedures not supervised by the MPS showed that informally supervised and unsupervised procedures are quite similar to each other; the inverse relationship between informal supervision and patient gender is difficult to explain and may be spurious.

To our knowledge, only 1 author has postulated a theoretical foundation for help‐seeking in trainees, depicted in the context of the patient‐resident‐attending triadic relationship.4 The mature help‐seeker, whether patient, resident, or attending, is willing to confront problems, receptive to new information, able to acknowledge dependence on expertise, and able to apply new input with self‐reliance. However, little is known about how this model manifests itself empirically in professional help‐seeking or what the optimal conditions of faculty supervision are. One observational study suggested that faculty who spent more time on hospital floors created environments with higher resident satisfaction scores, higher perceived quality of patient care, and, paradoxically, increased perceptions of autonomy.5 These results are consistent with our previous work showing that residents' comfort with bedside procedures increased with use of the MPS.2 In the related field of consultation medicine, 2 studies6, 7 showed that family practitioners prefer to consult internal medicine subspecialists over general internists. One of these studies7 demonstrated that the primary need was for a consultant with technical (ie, procedural) skills. Our use of MPS faculty who are specifically skilled in performing medical procedures appears to be consistent with this observation that specific technical expertise is valued over general supervision or guidance.

How can we best design formal procedural supervision programs that allow residents to obtain help when they need it? In addition to fostering mature help‐seeking behavior, help‐giving requires: (1) an environment that encourages help‐seeking; (2) a mechanism to provide assistance when and where it is needed; (3) supervisors with technical expertise; and (4) supervision that supports learning, skill acquisition, and graduated autonomy. It is difficult to devise mechanisms that include all of these elements. For instance, 24‐hour per day faculty coverage may be logistically challenging and expensive. Physicians with technical expertise may not be good teachers despite faculty development on procedural teaching. Obstacles to successful help‐seeking may include differences in residents' and supervisors' perceptions about the need for supervision. For example, a supervisor may be available and willing to assist, but the resident may feel capable of performing independently. When assistance is provided, residents and supervisors may differ in their perceptions of the quality of supervision.8 Ultimately, any educational intervention to increase supervision must confront a cultural norm of self‐sufficiency among many residency programs, in which managing a situation without assistance is equated with competence. To address this issue, our hospital has mandated the use of the MPS for all bedside procedures since 2005 and staffed the program 24 hours a day, in recognition of the potential risk of procedural complications9, 10 among inexperienced trainees.

This study has several limitations. We had a small number of thoracenteses. The study was not designed or powered to examine differences in complication rates among MPS and non‐MPS procedures. Because we represent a single institution, our findings may not be generalizable to other teaching hospitals or nonteaching settings. Our data on procedure characteristics were ascertained through resident self‐reports and, though typically submitted in a timely way, are subject to recall bias. In particular, discrepancies in the reported level of urgency may have affected our results about the time‐dependent nature of help‐seeking. Additionally, our findings about the types of patients about which residents seek consultation are somewhat at odds; use of the modified Deyo criteria to adjust for clinical severity weighs chronic conditions heavily and may translate into complication risk, but the level of urgency may better reflect the acuity of the clinical presentation. We could not distinguish between resident‐supervised procedures and those performed without supervision because of limited data. We also acknowledge the possibility that some non‐MPS faculty (classified for the study as informal supervisors) may serendipitously provide an equal quality of supervision that our MPS faculty did, by being present throughout the procedure and giving structured and valuable feedback.

Nevertheless, our results suggest that many residents do seek formal help appropriately when they perform procedures on the sickest patients, recognizing the risk and technical difficulty associated with bedside procedures in these patients. Our results also point to a greater area of inquiry: how do we optimally address the help‐seeking needs among trainee physicians? How do we create an environment in which help‐seeking is encouraged? How do we overcome the logistical barriers of providing timely assistance to residents, particularly at times of greatest need (urgent or emergent procedures)? How do we confront a longstanding culture in which independence is equated with competence, especially as it relates to procedural skills? A better understanding of how the widespread availability of programs like our MPS would affect the residents' use of supervision in general may guide the design of resident curricula and the development of mechanisms to ensure safe and effective clinical care.

There is little scientific evidence about professional help‐seeking behavior among resident physicians. Although junior physicians have many sources of information available to them in the course of clinical practiceprint materials, internet resources, curbside consultations, and advice from senior residents and facultywe have little empirical knowledge about when, why, and how physician trainees ask for help.

To study this phenomenon, we examined the use of a medical procedure service (MPS) by resident physicians. The MPS is an inpatient service at a Boston teaching hospital that provides education, supervision, and evaluation of internal medicine residents who perform common bedside procedures; it has been described previously.1 Residents who call the MPS review an online curriculum with self‐assessment quizzes, perform procedures with faculty supervision and feedback, and assess their own performance using online checklists. This program has been available to internal medicine residents since 2002. In a previous study, we found that residents reported greater comfort performing bedside procedures when they used the procedure service, when the operator was a postgraduate year (PGY)2 or PGY3 resident (compared to PGY1 residents), and while placing central venous catheters (CVCs) (compared to thoracenteses).2

The goal of the current study was to examine help‐seeking behavior among resident physicians as they placed CVCs and performed thoracenteses. We interpreted the decision to use the MPS to indicate that the resident successfully sought and received assistance from pulmonary attending physicians or interventional pulmonary fellows. We hypothesized that: (1) residents earlier in their training would choose to use the procedure service due to their relative lack of experience; (2) they would seek consultation when the procedure was performed in high‐risk patients, as indicated by the number of comorbidities, presence of medications that increase the risk of bleeding, and treatment in an intensive care unit; and (3) residents would be less likely to call the MPS for urgent or emergent situations, when timely assistance may be difficult to obtain. To examine the potentially confounding influence of procedures supervised by non‐MPS physicians, we also investigated differences between informally supervised procedures (i.e. by a non‐MPS attending or fellow) and unsupervised procedures (i.e. no attending or fellow supervision) to determine whether any significant differences in their characteristics existed.

Methods

Results

Discussion

To understand professional help‐seeking behavior by internal medicine resident physicians, we studied factors associated with the use of a MPS for performing 2 common bedside procedures. We found that residents used the MPS more often when they performed procedures on patients with more comorbidities and less often during urgent or emergent procedures.

These results are consistent with our hypothesis that residents use formal supervision when caring for high‐risk patients. We had also hypothesized that they would seek the MPS for patients on medications that increase the risk of bleeding, but this was not borne out. One possible explanation is that invasive procedures on anticoagulated patients may be deferred or avoided. Additionally, we did not collect prothrombin times nor platelet count, which may represent better proxies for coagulopathy. Our hypothesis that residents would not seek the MPS for urgent and emergent procedures was confirmed; the time delay between contacting the faculty member and performing the procedure may have inhibited or obviated consultation of the MPS. We hypothesized that interns would use the MPS preferentially; we found instead that level of training did not influence use of the MPS. A resident early in training may struggle with the balance between autonomy and supervision, wanting instead to establish himself as able to solve clinical problems independently and by seeking consultation only as a last resort. Alternatively, interns may be primarily supervised by their residents and may seek expert assistance only for particularly challenging or high‐risk cases. Additionally, as newcomers to the training program, they may not be well acquainted with the role and availability of the service (although periodic announcements were made throughout the year). Our examination of procedures not supervised by the MPS showed that informally supervised and unsupervised procedures are quite similar to each other; the inverse relationship between informal supervision and patient gender is difficult to explain and may be spurious.

To our knowledge, only 1 author has postulated a theoretical foundation for help‐seeking in trainees, depicted in the context of the patient‐resident‐attending triadic relationship.4 The mature help‐seeker, whether patient, resident, or attending, is willing to confront problems, receptive to new information, able to acknowledge dependence on expertise, and able to apply new input with self‐reliance. However, little is known about how this model manifests itself empirically in professional help‐seeking or what the optimal conditions of faculty supervision are. One observational study suggested that faculty who spent more time on hospital floors created environments with higher resident satisfaction scores, higher perceived quality of patient care, and, paradoxically, increased perceptions of autonomy.5 These results are consistent with our previous work showing that residents' comfort with bedside procedures increased with use of the MPS.2 In the related field of consultation medicine, 2 studies6, 7 showed that family practitioners prefer to consult internal medicine subspecialists over general internists. One of these studies7 demonstrated that the primary need was for a consultant with technical (ie, procedural) skills. Our use of MPS faculty who are specifically skilled in performing medical procedures appears to be consistent with this observation that specific technical expertise is valued over general supervision or guidance.

How can we best design formal procedural supervision programs that allow residents to obtain help when they need it? In addition to fostering mature help‐seeking behavior, help‐giving requires: (1) an environment that encourages help‐seeking; (2) a mechanism to provide assistance when and where it is needed; (3) supervisors with technical expertise; and (4) supervision that supports learning, skill acquisition, and graduated autonomy. It is difficult to devise mechanisms that include all of these elements. For instance, 24‐hour per day faculty coverage may be logistically challenging and expensive. Physicians with technical expertise may not be good teachers despite faculty development on procedural teaching. Obstacles to successful help‐seeking may include differences in residents' and supervisors' perceptions about the need for supervision. For example, a supervisor may be available and willing to assist, but the resident may feel capable of performing independently. When assistance is provided, residents and supervisors may differ in their perceptions of the quality of supervision.8 Ultimately, any educational intervention to increase supervision must confront a cultural norm of self‐sufficiency among many residency programs, in which managing a situation without assistance is equated with competence. To address this issue, our hospital has mandated the use of the MPS for all bedside procedures since 2005 and staffed the program 24 hours a day, in recognition of the potential risk of procedural complications9, 10 among inexperienced trainees.

This study has several limitations. We had a small number of thoracenteses. The study was not designed or powered to examine differences in complication rates among MPS and non‐MPS procedures. Because we represent a single institution, our findings may not be generalizable to other teaching hospitals or nonteaching settings. Our data on procedure characteristics were ascertained through resident self‐reports and, though typically submitted in a timely way, are subject to recall bias. In particular, discrepancies in the reported level of urgency may have affected our results about the time‐dependent nature of help‐seeking. Additionally, our findings about the types of patients about which residents seek consultation are somewhat at odds; use of the modified Deyo criteria to adjust for clinical severity weighs chronic conditions heavily and may translate into complication risk, but the level of urgency may better reflect the acuity of the clinical presentation. We could not distinguish between resident‐supervised procedures and those performed without supervision because of limited data. We also acknowledge the possibility that some non‐MPS faculty (classified for the study as informal supervisors) may serendipitously provide an equal quality of supervision that our MPS faculty did, by being present throughout the procedure and giving structured and valuable feedback.

Nevertheless, our results suggest that many residents do seek formal help appropriately when they perform procedures on the sickest patients, recognizing the risk and technical difficulty associated with bedside procedures in these patients. Our results also point to a greater area of inquiry: how do we optimally address the help‐seeking needs among trainee physicians? How do we create an environment in which help‐seeking is encouraged? How do we overcome the logistical barriers of providing timely assistance to residents, particularly at times of greatest need (urgent or emergent procedures)? How do we confront a longstanding culture in which independence is equated with competence, especially as it relates to procedural skills? A better understanding of how the widespread availability of programs like our MPS would affect the residents' use of supervision in general may guide the design of resident curricula and the development of mechanisms to ensure safe and effective clinical care.

There is little scientific evidence about professional help‐seeking behavior among resident physicians. Although junior physicians have many sources of information available to them in the course of clinical practiceprint materials, internet resources, curbside consultations, and advice from senior residents and facultywe have little empirical knowledge about when, why, and how physician trainees ask for help.

To study this phenomenon, we examined the use of a medical procedure service (MPS) by resident physicians. The MPS is an inpatient service at a Boston teaching hospital that provides education, supervision, and evaluation of internal medicine residents who perform common bedside procedures; it has been described previously.1 Residents who call the MPS review an online curriculum with self‐assessment quizzes, perform procedures with faculty supervision and feedback, and assess their own performance using online checklists. This program has been available to internal medicine residents since 2002. In a previous study, we found that residents reported greater comfort performing bedside procedures when they used the procedure service, when the operator was a postgraduate year (PGY)2 or PGY3 resident (compared to PGY1 residents), and while placing central venous catheters (CVCs) (compared to thoracenteses).2

The goal of the current study was to examine help‐seeking behavior among resident physicians as they placed CVCs and performed thoracenteses. We interpreted the decision to use the MPS to indicate that the resident successfully sought and received assistance from pulmonary attending physicians or interventional pulmonary fellows. We hypothesized that: (1) residents earlier in their training would choose to use the procedure service due to their relative lack of experience; (2) they would seek consultation when the procedure was performed in high‐risk patients, as indicated by the number of comorbidities, presence of medications that increase the risk of bleeding, and treatment in an intensive care unit; and (3) residents would be less likely to call the MPS for urgent or emergent situations, when timely assistance may be difficult to obtain. To examine the potentially confounding influence of procedures supervised by non‐MPS physicians, we also investigated differences between informally supervised procedures (i.e. by a non‐MPS attending or fellow) and unsupervised procedures (i.e. no attending or fellow supervision) to determine whether any significant differences in their characteristics existed.

Methods

Results

Discussion

To understand professional help‐seeking behavior by internal medicine resident physicians, we studied factors associated with the use of a MPS for performing 2 common bedside procedures. We found that residents used the MPS more often when they performed procedures on patients with more comorbidities and less often during urgent or emergent procedures.

These results are consistent with our hypothesis that residents use formal supervision when caring for high‐risk patients. We had also hypothesized that they would seek the MPS for patients on medications that increase the risk of bleeding, but this was not borne out. One possible explanation is that invasive procedures on anticoagulated patients may be deferred or avoided. Additionally, we did not collect prothrombin times nor platelet count, which may represent better proxies for coagulopathy. Our hypothesis that residents would not seek the MPS for urgent and emergent procedures was confirmed; the time delay between contacting the faculty member and performing the procedure may have inhibited or obviated consultation of the MPS. We hypothesized that interns would use the MPS preferentially; we found instead that level of training did not influence use of the MPS. A resident early in training may struggle with the balance between autonomy and supervision, wanting instead to establish himself as able to solve clinical problems independently and by seeking consultation only as a last resort. Alternatively, interns may be primarily supervised by their residents and may seek expert assistance only for particularly challenging or high‐risk cases. Additionally, as newcomers to the training program, they may not be well acquainted with the role and availability of the service (although periodic announcements were made throughout the year). Our examination of procedures not supervised by the MPS showed that informally supervised and unsupervised procedures are quite similar to each other; the inverse relationship between informal supervision and patient gender is difficult to explain and may be spurious.

To our knowledge, only 1 author has postulated a theoretical foundation for help‐seeking in trainees, depicted in the context of the patient‐resident‐attending triadic relationship.4 The mature help‐seeker, whether patient, resident, or attending, is willing to confront problems, receptive to new information, able to acknowledge dependence on expertise, and able to apply new input with self‐reliance. However, little is known about how this model manifests itself empirically in professional help‐seeking or what the optimal conditions of faculty supervision are. One observational study suggested that faculty who spent more time on hospital floors created environments with higher resident satisfaction scores, higher perceived quality of patient care, and, paradoxically, increased perceptions of autonomy.5 These results are consistent with our previous work showing that residents' comfort with bedside procedures increased with use of the MPS.2 In the related field of consultation medicine, 2 studies6, 7 showed that family practitioners prefer to consult internal medicine subspecialists over general internists. One of these studies7 demonstrated that the primary need was for a consultant with technical (ie, procedural) skills. Our use of MPS faculty who are specifically skilled in performing medical procedures appears to be consistent with this observation that specific technical expertise is valued over general supervision or guidance.

How can we best design formal procedural supervision programs that allow residents to obtain help when they need it? In addition to fostering mature help‐seeking behavior, help‐giving requires: (1) an environment that encourages help‐seeking; (2) a mechanism to provide assistance when and where it is needed; (3) supervisors with technical expertise; and (4) supervision that supports learning, skill acquisition, and graduated autonomy. It is difficult to devise mechanisms that include all of these elements. For instance, 24‐hour per day faculty coverage may be logistically challenging and expensive. Physicians with technical expertise may not be good teachers despite faculty development on procedural teaching. Obstacles to successful help‐seeking may include differences in residents' and supervisors' perceptions about the need for supervision. For example, a supervisor may be available and willing to assist, but the resident may feel capable of performing independently. When assistance is provided, residents and supervisors may differ in their perceptions of the quality of supervision.8 Ultimately, any educational intervention to increase supervision must confront a cultural norm of self‐sufficiency among many residency programs, in which managing a situation without assistance is equated with competence. To address this issue, our hospital has mandated the use of the MPS for all bedside procedures since 2005 and staffed the program 24 hours a day, in recognition of the potential risk of procedural complications9, 10 among inexperienced trainees.

This study has several limitations. We had a small number of thoracenteses. The study was not designed or powered to examine differences in complication rates among MPS and non‐MPS procedures. Because we represent a single institution, our findings may not be generalizable to other teaching hospitals or nonteaching settings. Our data on procedure characteristics were ascertained through resident self‐reports and, though typically submitted in a timely way, are subject to recall bias. In particular, discrepancies in the reported level of urgency may have affected our results about the time‐dependent nature of help‐seeking. Additionally, our findings about the types of patients about which residents seek consultation are somewhat at odds; use of the modified Deyo criteria to adjust for clinical severity weighs chronic conditions heavily and may translate into complication risk, but the level of urgency may better reflect the acuity of the clinical presentation. We could not distinguish between resident‐supervised procedures and those performed without supervision because of limited data. We also acknowledge the possibility that some non‐MPS faculty (classified for the study as informal supervisors) may serendipitously provide an equal quality of supervision that our MPS faculty did, by being present throughout the procedure and giving structured and valuable feedback.

Nevertheless, our results suggest that many residents do seek formal help appropriately when they perform procedures on the sickest patients, recognizing the risk and technical difficulty associated with bedside procedures in these patients. Our results also point to a greater area of inquiry: how do we optimally address the help‐seeking needs among trainee physicians? How do we create an environment in which help‐seeking is encouraged? How do we overcome the logistical barriers of providing timely assistance to residents, particularly at times of greatest need (urgent or emergent procedures)? How do we confront a longstanding culture in which independence is equated with competence, especially as it relates to procedural skills? A better understanding of how the widespread availability of programs like our MPS would affect the residents' use of supervision in general may guide the design of resident curricula and the development of mechanisms to ensure safe and effective clinical care.