Communication between members of the cardiac arrest team--a postal survey

Aims: Effective communication enhances team building and is perceived to improve the quality of team performance. A recent publication from the Resuscitation Council (UK) has highlighted this fact and recommended that cardiac arrest team members make contact daily. We wished to identify how often members of this team communicate prior to a cardiopulmonary arrest. Method: A questionnaire on cardiac arrest team composition, leadership, communication and debriefing was distributed nationally to Resuscitation Training Officers (RTOs) and their responses analysed. Results: One hundred and thirty (55%) RTOs replied. Physicians and anaesthetists were the most prominent members of the team. The Medical Senior House Officer is usually nominated as the team leader. Eighty-seven centres (67%) have no communication between team members prior to attending a cardiopulmonary arrest. In 33%, communication occurs but is either informal or fortuitous. The RTOs felt that communication is important to enhance team dynamics and optimise task allocation. Only 7% achieve a formal debrief following a cardiac arrest. Conclusion: Communication between members of the cardiac arrest team before and after a cardiac arrest is poor. Training and development of these skills may improve performance and should be prioritised. Team leadership does not necessarily reflect experience or training.     

How comfortable are emergency physicians with pediatric patients?

We sought to understand how emergency physicians perceive the adequacy of their training in pediatrics. A survey was distributed to emergency physicians regarding residency training, clinical experience, importance of Core Content curricular areas, and the adequacy of their training. The results demonstrated that 84.0% of respondents felt well, completely, or adequately prepared with pediatric cardiopulmonary arrests compared to 96.4% who felt prepared for adult arrests. Trauma resuscitations and care of acutely ill patients revealed differences in preparedness for children versus adults [81.5% vs. 90.1%, respectively (p < 0.001) and 92.2% vs. 97.1%, respectively (p < 0.001)]. Pediatric arrest was the most often cited clinical situation giving trouble to first year attendings (24%) and infants were second (22%). Pediatric disorders were cited as the fourth most important area in training. Certain residency characteristics were associated with an increased sense of preparedness. These results may interest Emergency Medicine educators in planning pediatric curricula and experiences for residents.     

Teaching an integrated approach to complementary, alternative, and mainstream therapies for children: a curriculum evaluation.

BACKGROUND: Increasing numbers of patients seek information about complementary and alternative medicine (CAM) from their primary physicians. We sought to evaluate our 4-year old curriculum integrating mainstream and CAM care for common outpatient pediatric problems within a family medicine residency. DESIGN: Cross-sectional survey. METHODS: Subjects included current (1998) third-year residents and recent graduates from our program and nearby University of Washington-affiliated family medicine residency programs. The survey included items on training experiences, knowledge, attitudes and behavior regarding CAM. RESULTS: Among the 18 respondents from our program and 21 from comparison programs, the average age was 32 years and one-third were male. Over 80% of respondents felt that residencies should provide training in CAM. Substantial numbers of respondents from all programs recommended CAM therapies to patients in the past year. All respondents had recommended special diets and nutritional supplements; more than 50% recommended herbal remedies, acupuncture, meditation or progressive relaxation, massage or home remedies. Respondents from all groups had similar attitudes and knowledge about integrative medicine; those from the intervention program were more likely than comparison respondents to agree that their residency training had prepared them to answer patients' questions about CAM (50% vs. 19%, p = 0.04). CONCLUSIONS: Primary care residents increasingly seek training to answer patients' questions and are already recommending a variety of CAM therapies. Primary care residencies need to develop and evaluate responsible, evidence-based curricula integrating mainstream and CAM therapies.     

The knowledge and perceptions of medical personnel relating to outcome after cardiac arrest

OBJECTIVE: We sought to evaluate the knowledge of probable outcome by medical personnel for in-hospital and out-of-hospital cardiac arrests, and self-reported history of CPR training referrals for family members of cardiac patients. METHODS: One hundred people from each of three population lists were randomly selected at a large, urban school of medicine and affiliated medical center: (1) year III and IV medical students; (2) residents in family medicine, emergency medicine, internal medicine, anesthesia, and surgery; (3) attending physicians in the same departments. A questionnaire was distributed that elicited estimates of in-hospital and out-of-hospital cardiac arrest (IHCA and OHCA, respectively) survival rates, and CPR training referral history. Estimates were compared against published data for accuracy (IHCA: 5-20%; OHCA 1-10%) RESULTS: The overall response rate was 63%. Accurate in-hospital cardiac arrest estimates [% (95% CI)] of survival were provided by 51.1% (36.8-63.4%), 47.3% (35.9-58.7%), and 36.7% (23.2-50.2%) of students, residents, and attending physicians, respectively. Accurate out-of-hospital estimates of survival were provided by 51.1% (36.8-63.4%), 52.1% (40.6-63.5%), and 70.8% (57.9-83.7%), respectively. Most thought that family members of cardiac patients ought to be CPR trained (92.6%). However, few had referred any for training in the past year (16.5%). There was strong support across respondent groups for including death notification information in the ACLS training program, with 80.4% of all respondents in favor. CONCLUSIONS: This study demonstrates that medical experience is not associated with accurate estimates of cardiac arrest survival. Overwhelmingly, medical personnel believe family members should be trained to perform CPR, however, few refer family members for CPR training.     

Public perceptions and experiences of myocardial infarction, cardiac arrest and CPR in London

INTRODUCTION: The lay public have limited knowledge of the symptoms of myocardial infarction ("heart attack"), and inaccurate perceptions of cardiac arrest survival rates. Levels of CPR training and willingness to intervene in cardiac emergencies are also low. AIMS: To explore public perceptions of myocardial infarction and cardiac arrest; investigate perceptions of cardiac arrest survival rates; assess levels of training and attitudes towards CPR, and explore the types of interventions considered useful for increasing rates of bystander CPR among Greater London residents. METHODS: A quantitative interview survey was conducted with 1011 Greater London residents. Eight focus groups were also conducted to explore a range of issues in greater depth and validate trends that emerged in the initial survey. RESULTS: Chest pain was the most commonly recognised symptom of "heart attack". Around half of the respondents were aware that a myocardial infarction differs from a cardiac arrest, although their ability to explain this difference was limited. The majority overestimated that at least a quarter of cardiac arrest patients in London survive to hospital discharge. Few participants had received CPR training, and most were hesitant about performing the procedure on a stranger. CONCLUSIONS: Awareness and knowledge of CPR, and reactions to cardiac emergencies, reflect relatively low levels of CPR training in London. Publicising cardiac arrest survival figures may be instrumental in prompting members of the public to train in CPR and motivating those who have been trained to intervene in a cardiac emergency.     

Preparation for cancer care: perceptions of newly qualified health care professionals.

The present paper is derived from a larger survey which examined the perceptions of recently qualified health care professionals' experience on evidence-based practice, team working and cancer care. This study reports solely on the findings relating to cancer care. The perceptions of recently qualified professionals in relation to their initial educational input on issues such as confidence, anxiety, communication skills and practice in cancer care as well as adequacy of support, professional supervision and use of reflection were gathered using a cross-sectional postal survey design. A total of 50 graduates from each professional category in nursing, occupational therapy, physiotherapy, and social work were sampled yielding a total sample of 200. Eighty-five questionnaires were returned yielding a response rate of 43%. Twenty-eight (33%) respondents stated that they were currently involved in working with people with cancer. These were as follows: 5 nurses, 8 physiotherapists, 9 occupational therapists and 6 social workers. Despite the low response rate, the findings suggest that health care professionals' educational input and experiences of working with people with cancer were overall positive; for example, in the respondents' confidence, communication skills, decrease in anxiety and application of knowledge gained in classroom to professional practice. Moreover, most respondents learnt about caring for cancer patients through practice rather than classroom teaching. A high percentage (i.e. 64%;18) across all groups felt supported when caring for people with cancer and reported receiving professional supervision as well as being able to actively reflect on their practice. The implications for education and practice were discussed particularly as there have been few studies conducted in relation to the specific needs and collaborative learning of these health care professional groups.     

A survey of ventilation strategies during cardiopulmonary resuscitation

BACKGROUND: Many controversies still exist regarding ventilator parameters during cardiopulmonary resuscitation (CPR). This study aimed to investigate the CPR ventilation strategies currently being used among physicians in Chinese tertiary hospitals.METHODS: A survey was conducted among the cardiac arrest team physicians of 500 tertiary hospitals in China in August, 2018. Surveyed data included physician and hospital information, and preferred ventilation strategy during CPR.RESULTS: A total of 438 (88%) hospitals completed the survey, including hospitals from all 31 Chinese mainland provinces. About 41.1% of respondents chose delayed or no ventilation during CPR, with delayed ventilations all starting within 12 minutes. Of all the respondents who provided ventilation, 83.0% chose to strictly follow the 30:2 strategy, while 17.0% chose ventilations concurrently with uninterrupted compressions. Only 38.3% respondents chose to intubate after initiating CPR, while 61.7% chose to intubate immediately when resuscitation began. During bag-valve-mask ventilation, only 51.4% of respondents delivered a frequency of 10 breaths per minute. In terms of ventilator settings, the majority of respondents chose volume control (VC) mode (75.2%), tidal volume of 6-7 mL/kg (72.1%), PEEP of 0-5 cmH₂O (69.9%), and an FiO₂ of 100% (66.9%). However, 62.0% of respondents had mistriggers after setting the ventilator, and 51.8% had high pressure alarms.CONCLUSION: There is a great amount of variability in CPR ventilation strategies among cardiac arrest team physicians in Chinese tertiary hospitals. Guidelines are needed with specific recommendations on ventilation during CPR.     

Induced hypothermia is underused after resuscitation from cardiac arrest: a current practice survey

BACKGROUND: Important recent work has demonstrated that the use of induced hypothermia can improve survival and neurologic recovery after cardiac arrest. We wished to ascertain the extent to which physicians were using this treatment, and what opinions are held by clinicians regarding its use. METHODS: An internet-based survey of physicians was conducted, with physicians chosen at random from published directories of the Society for Academic Emergency Medicine, the American Thoracic Society, and the American Heart Association. Physicians were questioned regarding use of therapeutic hypothermia, methods employed, and/or reasons why they had not incorporated hypothermia into their care of cardiac arrest patients. RESULTS: Completed surveys were collected from 265 physicians, including those practicing emergency medicine (41%), critical care (13%), and cardiology (24%). Respondents were geographically well distributed and the majority (94%) were at post-training level. Most respondents (78%) practiced at either larger referral hospitals or academic medical centers. When asked if they had ever used hypothermia following cardiac arrest, 87% said they had not. Among reasons cited for non-use, 49% felt that there were not enough data, 32% mentioned lack of incorporation of hypothermia into advanced cardiovascular life support (ACLS) protocols, and 28% felt that cooling methods were technically too difficult or too slow. CONCLUSION: Despite compelling data supporting its use, hypothermia has yet to be broadly incorporated into physician practice. This highlights the need for improved awareness and education regarding this treatment option, as well as the need to consider hypothermia protocols for inclusion in future iterations of ACLS.     

Chest compression quality management and return of spontaneous circulation: A matched-pair registry study

Aims

Investigating the effects of any intervention during cardiac arrest remains difficult. The ROSC after cardiac arrest score was introduced to facilitate comparison of rates of return of spontaneous circulation (ROSC) between different ambulance services. To study the influence of chest compression quality management (including training, real-time feedback devices, and debriefing) in comparison with conventional cardiopulmonary resuscitation (CPR), a matched-pair analysis was conducted using data from the German Resuscitation Registry, with the calculated ROSC after cardiac arrest score as the baseline.

Methods and results

Matching for independent ROSC after cardiac arrest score variables yielded 319 matched cases from the study period (January 2007–March 2011). The score predicted a 45% ROSC rate for the matched pairs. The observed ROSC increased significantly with chest compression quality management, to 52% (P = 0.013; 95% CI, 46–57%). No significant differences were seen in the conventional CPR group (47%; 95% CI, 42–53%). The difference between the observed ROSC rates was not statistically significant.

Conclusions

Chest compression quality management leads to significantly higher ROSC rates than those predicted by the prognostic score (ROSC after cardiac arrest score). Matched-pair analysis shows that with conventional CPR, the observed ROSC rate was not significantly different from the predicted rate. Analysis shows a trend toward a higher ROSC rate for chest compression quality management in comparison with conventional CPR. It is unclear whether a single aspect of chest compression quality management or the combination of training, real-time feedback, and debriefing contributed to this result.     

Providing Complementary and Alternative Medicine in primary care: the primary care workers' perspective

BACKGROUND: The use of Complementary and Alternative Medicine (CAM) in primary care is growing, but still not widespread. Little is known about how CAM can/should be integrated into mainstream care. OBJECTIVES: To assess primary care health professionals' perceptions of need and of some ways to integrate CAM in primary care. METHOD: Questionnaire survey of primary health care workers in Northwest London. General Practitioners (GPs) were targeted in a postal survey, other members of the primary care team, such as district and practice nurses, were targeted via colleagues. The questionnaire assessed health care professionals' perspective on complementary medicine, referrals, ways to integrate complementary medicine into primary care and interest in research on CAM. RESULTS: Responses were obtained from 149 GPs (40% response rate after one reminder) and 24 nurses and 32 other primary care team members. One hundred and seventy-one (83%) respondents had previously referred (or influenced referral) for CAM treatments, the main reasons cited were: patients request (68%), conventional treatments failed (58%) and evidence (36%) (more than one reason could be given). Acupuncture and homoeopathy were the therapies for which patients were most frequently referred, followed by manual therapies. There was a significant interest in more training/information on CAM (66%). Only 12 respondents (6%) were against any integration of CAM in mainstream primary care. Most respondents felt that CAM therapies should be provided by doctors (66%) or other health professionals trained in CAM (82%). Twenty-six percent of respondents agreed with provision of CAM by non-state-registered practitioners. It was felt that the integration of CAM could lead to cost savings (70%), particularly in conditions involving pain, but also cost increases (55%) particularly in 'poorly defined conditions'. Fifty-six percent of respondents would consider participating in studies investigating CAM. The greatest interest was in acupuncture (41% of those who expressed an interest in research), homoeopathy (30%) and therapeutic massage/aromatherapy (26%). CONCLUSIONS: There is considerable interest in CAM among primary care professionals, and many are already referring or suggesting referral. Such referrals are driven mainly by patient demand and by dissatisfaction with the results of conventional medicine. Most of our respondents were in favour of integrating at least some types of CAM in mainstream primary care. There is an urgent need to further educate/inform primary care health professionals about CAM.     

Therapeutic hypothermia utilization among physicians after resuscitation from cardiac arrest

OBJECTIVE: We sought to evaluate current physician use of therapeutic hypothermia after cardiac arrest, to ascertain reasons for nonadoption of this treatment, and to determine current cooling techniques employed. DESIGN: Web-based survey. SETTING: International physician cohort in the United States, UK, and Finland. SUBJECTS: Physicians (MD or DO) caring for resuscitated cardiac arrest patients. INTERVENTIONS: An anonymous Web-based survey was distributed to physicians identified through United States-based critical care, cardiology, and emergency medicine directories and critical care networks in the UK and Finland. Recipients were queried regarding use of postresuscitation therapeutic hypothermia. MEASUREMENTS AND MAIN RESULTS: Of the final 13,272 surveys actually distributed to physicians, 2,248 (17%) were completed. Most respondents were attending physicians (82%) at teaching hospitals (76%) who practiced critical care (35%), cardiology (20%), or emergency medicine (22%). Of all replies, 74% of United States respondents and 64% of non-United States respondents had never used therapeutic hypothermia. United States emergency medicine physician adoption of cooling was significantly less than that of United States intensivists (16% vs. 34%, p < .05). The most often cited reasons for nonuse by respondents were "not enough data," "not part of Advanced Cardiac Life Support guidelines," and "too technically difficult to use." Factors associated with increased use included non-United States residence, critical care specialty, and larger hospital size. CONCLUSIONS: Physician utilization of cooling after cardiac arrest remains low. For improved adoption of therapeutic hypothermia, our data suggest that development of better cooling methodology and recent incorporation into resuscitation guidelines may improve use.     

Giving curriculum planners an edge: Using entrance surveys to design family medicine education

ObjectiveTo pilot a survey of family medicine residents entering residency, describing their exposure to family medicine and their perspectives related to their future intentions to practise family medicine, in order to inform curriculum planners; and to test the methodology, feasibility, and utility of delivering a longitudinal survey to multiple residency programs.DesignPilot study using surveys.SettingFive Canadian residency programs.ParticipantsA total of 454 first-year family medicine residents were surveyed.ResultsOverall, 70% of first-year residents surveyed responded (n = 317). Although only 5 residency programs participated, respondents included graduates from each of the medical schools in Canada, as well as international medical graduates. Among respondents, 92% felt positive or strongly positive about their choice to be family physicians. Most (73%) indicated they had strong or very strong exposure to family medicine in medical school, yet more than 40% had no or minimal exposure to key clinical domains of family medicine like palliative care, home care, and care of underserved groups. Similar responses were found about residents’ lack of intention to practise in these domains.ConclusionExposure to clinical domains in family medicine could influence future practice intentions. Surveys at entrance to residency can help medical school and family medicine residency planners consider important learning experiences to include in training.     

Debriefing in the intensive care unit: a feedback tool to facilitate bedside teaching

OBJECTIVE: To develop an assessment tool for bedside teaching in the intensive care unit (ICU) that provides feedback to residents about their performance compared with clinical best practices. METHOD: We reviewed the literature on the assessment of resident clinical performance in critical care medicine and summarized the strengths and weaknesses of these assessments. Using debriefing after simulation as a model, we created five checklists for different situations encountered in the ICU--areas that encompass different Accreditation Council for Graduate Medical Education core competencies. Checklists were designed to incorporate clinical best practices as defined by the literature and institutional practices as defined by the critical care professionals working in our ICUs. Checklists were used at the beginning of the rotation to explicitly define our expectations to residents and were used during the rotation after a clinical encounter by the resident and supervising physician to review a resident's performance and to provide feedback to the resident on the accuracy of the resident's self-assessment of his or her performance. RESULTS: Five "best practice" checklists were developed: central catheter placement, consultation, family discussions, resuscitation of hemorrhagic shock, and resuscitation of septic shock. On average, residents completed 2.6 checklists per rotation. Use of the cards was fairly evenly distributed, with the exception of resuscitation of hemorrhagic shock, which occurs less frequently than the other encounters in the medical ICU. Those who used more debriefing cards had higher fellow and faculty evaluations. Residents felt that debriefing cards were a useful learning tool in the ICU. CONCLUSIONS: Debriefing sessions using checklists can be successfully implemented in ICU rotations. Checklists can be used to assess both resident performance and consistency of practice with respect to published standards of care in critical care medicine.     

A model for a statewide critical incident stress (CIS) debriefing program for emergency services personnel

PURPOSE: Emergency services personnel are highly vulnerable to acute and cumulative critical incident stress (CIS) that can manifest as anger, guilt, depression, and impaired decision-making, and, in certain instances, job loss. Interventions designed to identify such distress and restore psychological functioning becomes imperative. METHODS: A statewide debriefing team was formed in 1988 through a collaborative effort between an academic department of emergency medicine and a social work department of a teaching hospital, and a metropolitan area fire department and ambulance service. Using an existing CIS debriefing model, 84 prescreened, mental health professionals and emergency services personnel were provided with 16 hours of training and were grouped into regional teams. Debriefing requests are received through a central number answered by a communicator in a 24-hour communications center located within the emergency department. Debriefings are conducted 48-72 hours after the event for specific types of incidents. Follow-up telephone calls are made by the debriefing team leader two to three weeks following a debriefing. The teams rely on donations to pay for travel and meals. RESULTS: One hundred sixty-eight debriefings were conducted during the first four years. Rural agencies accounted for 116 (69%) requests. During this period, 1,514 individuals were debriefed: 744 (49%) firefighters, 460 (30%) EMTs, and 310 (21%) police officers, dispatchers, and other responders. Deaths of children, extraordinary events, and incidents involving victims known to the responders (35%, 14%, and 14% respectively) were the most common reasons for requesting debriefings. Feedback was received from 48 (28%) of the agencies that requested the debriefing. All of those who responded felt that the debriefing had a beneficial effect on its personnel. Specific individuals identified by agency representatives as having the greatest difficulty were observed to be returned to their pre-incident state. CONCLUSIONS: CIS debriefings are judged as beneficial. A statewide response team is an effective way to provide these services at no cost to agencies.     

Perceived need for emergency medicine training in Pakistan: A survey of medical education leadership

Objective: To assess the perception of leaders of the academic medical institutions regarding the need for specialty training in emergency medicine. Methods: A cross‐sectional survey was conducted in all medical colleges of Pakistan in September 2005. Our sample included all academic leaders of recognized medical colleges in Pakistan. A questionnaire was designed and sent (mailed and faxed) to vice chancellors, deans, principals or medical directors of the institutions. Reminders were sent through faxes and emails wherever available, followed by phone calls if responses were not available after several attempts. Results: At the time of study, there were 39 medical colleges recognized by Pakistan Medical and Dental Council. Of these, responses were received from 26 teaching institutions in the country. A majority of the respondents (85%) were not satisfied with the care provided in the ED of their primary teaching hospital, and three‐fourth (74%) thought that doctors specialized in other disciplines, like internal medicine and family medicine, cannot adequately manage all emergencies. When asked if Pakistan should have a separate residency training programme in emergency medicine, 96% responded in affirmative, and many (85%) thought that they will start a residency programme in emergency medicine if it was approved as a separate specialty. Conclusion: This survey shows significant support for a separate local training programme for emergency medicine in the country.     

To lead or not to lead? Prospective controlled study of emergency nurses' provision of advanced life support team leadership

Background and objectives: In many emergency departments advanced life support (ALS) trained nurses do not assume a lead role in advanced resuscitation. This study investigated whether emergency nurses with previous ALS training provided good team leadership in a simulated cardiac arrest situation. Methods: A prospective study was conducted at five emergency departments and one nurses'' association meeting. All participants went through the same scenario. Details recorded included baseline blood pressure and pulse rate, time in post, time of ALS training, and subjective stress score (1 = hardly stressed; 10 = extremely stressed). Scoring took into account scenario understanding, rhythm recognition, time to defibrillation, appropriateness of interventions, and theoretical knowledge. Results: Of 57 participants, 20 were ALS trained nurses, 19 were ALS trained emergency senior house officers (SHOs), and 18 were emergency SHOs without formal ALS training. The overall mean score for doctors without ALS training was 69.5%, compared with 72.3% for ALS trained doctors and 73.7% for ALS trained nurses. Nurses found the experience less stressful (subjective stress score 5.78/10) compared with doctors without ALS training (6.5/10). The mean time taken to defibrillate from the appearance of a shockable rhythm on the monitor by the nurses and those SHOs without ALS training was 42 and 40.8 seconds, respectively. Conclusion: ALS trained nurses performed as well as ALS trained and non ALS trained emergency SHOs in a simulated cardiac arrest situation and had greater awareness of the potentially reversible causes of cardiac arrest. Thus if a senior or middle grade doctor is not available to lead the resuscitation team, it may be appropriate for experienced nursing staff with ALS training to act as ALS team leaders rather than SHOs.     

The effectiveness of high fidelity simulation on medical–surgical registered nurses' ability to recognise and respond to clinical emergencies

Background

There is a paucity of evidence regarding the efficacy in preparing medical–surgical nurses to respond to patients with acutely deteriorating conditions.

Study aim

The aim of this study was to evaluate registered nurses' ability to respond to the deteriorating patient in clinical practise following training using immersive simulation and use of a high fidelity simulator.

Methods

This study was a follow-up survey of medical–surgical graduate nurses following immersive high fidelity simulation training. Thirty eight registered nurses practising in medical–surgical areas completed the simulation as part of university graduate study. A follow-up survey of the graduate medical–surgical registered nurses conducted three months following completion of a high fidelity simulation-based learning experience. Outcomes consisted of the number of times skills were used in practise and the usefulness of simulation in preparing for actual emergency events.

Results

Participants reported a total of 164 clinical patient emergencies in the follow-up time period including: 46% cardiac, 32% respiratory, 10% neurological, 7% cardiac arrest and 5% related to electrolyte disturbances. The ability to respond in a systematic way, handover to the emergency team and airway management were identified as the skills most improved during patient emergencies following simulation. The most useful aspects of the simulation experience identified were scenario debriefing and assertiveness training. Participants with less years of clinical experience were more likely to report practising the team leader role and debriefing as the most useful aspects of simulation.

Conclusions

The skills practised in simulation were highly relevant to participants practise in medical–surgical areas. Non-technical skills, including assertiveness skills should be considered in future emergency training courses for nurses.     

Altering residency curriculum in response to a changing practice environment: use of the Mayo internal medicine residency alumni survey

To elicit the opinions of practicing internists who had graduated from a single internal medicine residency program about the adequacy of their training and its relevance to their medical practice, we mailed a survey to 1,342 physicians who had spent at least 1 year in the Mayo internal medicine residency training program. Of this group, 703 alumni (52%) responded to the survey, 532 of whom were currently practicing internal medicine. Our detailed analysis was based on responses from these 532 and, for some aspects of evaluation, on the 121 general internists who had completed residency training after 1970. Of the respondents, 42% spent more than 80% of their time in general medicine, and 53% had at least some subspecialty practice; 55% were involved in teaching, 20% in some research, and 37% in various administrative duties. In 27%, all patient-care activities involved primary care, an increase from 18% in a 1979 survey and 9% in 1972. Of those who were subspecialists, 67% spent more than half their time in subspecialty practice. Of those who were trained after 1970, 90% were board certified. Most respondents thought that their training in the internal medicine subspecialties was adequate, that additional procedure training was needed in joint aspiration, line placement, and flexible sigmoidoscopy, and that many allied medical areas were important to their practice and necessitated additional training. Although virtually all respondents assessed their inpatient training as adequate, only 42% were fully satisfied with their outpatient training. Alumni surveys can be useful in restructuring a residency program to meet the needs of the trainees.