- Email: [email protected]
Categorical medicine residents' experiential curriculum
Categorical Medicine Residents’ Experiential
Curriculum
Maxine A. Papadakis, MD, Margaret K. Kagawa, RN, MS, San Francisco, California PURPOSE: To assess the case mix or experiential curriculum of a university-based categorical medicine residency program and compare the residents’ continuity clinic case mix with the outpatients seen by practicing internists. PATIENTS AND METHODS: Descriptive study during the 1991-1992 academic year of 24,218 inpatients and outpatients from the University of California, San Francisco, Department of Medicine’s three core hospitals: the University of California, San Francisco Hospital; the Department of Veterans Affairs Medical Center; and the San Francisco General Hospital. The diagnoses and ages of patients who made office visits to practicing internists from the National Ambulatory Medical Care Survey (NAMCS) were compared with corresponding data from the categorical medicine residents’ continuity clinic patients. RESULTS: Seventy percent of the general medicine inpatients at the three core hospitals had one of the 25 most common principal diagnoses for inpatients. Eighty-seven percent of the patients seen by the residents in their continuity clinic had one of the 20 most common diagnoses seen by practicing internists in the NAMCS. The age distribution was similar in both groups. CONCLUSIONS: A SysM?I&iC assessment of CliniCal training at one university-based hospital program shows that common internal medicine problems represent the case mix of the great majority of patients, both inpatients and outpatients, seen by categorical medicine residents. Residents’ continuity clinic patients are similar to patients seen by practicing internists, The program’s challenge is to ensure that residents have adequate time with general medicine experiences, both in the inpatient setting and in the ambulatory and longitudinal care settings, while balancing and integrating these activities.
From the Department of Medicine, Student Programs, University of California, San Francisco, and the Department of Veterans Affairs Medical Center, San Francisco, California. This study was funded by a Western Region Department of Veterans Affairs Ambulatory Care Education Grant. Reouests for reorints should be addressed to Maxine A. Papadakis. MD, Department of Veterans Affairs Medical Center, 4150 Clement Street lllA1, San Francisco, California 94121. Manuscript submitted November 23, 1993 and accepted in revised form April 28, 1994.
espite the heightened interest in internal mediD cine residency curriculum reform,’ little is known about current categorical medicine residency experiences. As the acute-care university hospitals become a site of increasing admissions for technologic procedures, many university teaching hospitals have become so specialized that they provide an unbalanced educational experience.2 To address the issue of educational balance, Schroeder and coworkers2 have called for the systematic assessment of clinical training of residents by the monitoring of their clinical experiences. Surgical certifying boards place great importance on the volume of selected procedures performed by house officers during their residency. The American Board of Internal Medicine, however, does not place a similar emphasis on case mix in the training of internal medicine residents3 Based on the 1989 National Ambulatory Medical Care Survey (NAMCS),4 data on types of diagnoses and demographic information on patients seen by practicing internists are now available. A comparison of these data to the types of ambulatory patients seen by categorical medicine residents would help identify what should constitute essential components of ambulatory experiences. As part of a Department of Veterans Affairs (VA) Ambulatory Care Education initiative, this study was undertaken to describe the current inpatient and outpatient case mix of categorical medicine residents at a university program consisting of a university hospital (the University of California, San Francisco [UCSF] Hospital or MoffittLong Hospitals), a municipal hospital (San Francisco General Hospital [SFGH]), and a veterans hospital (San Francisco Department of Veterans Affairs Medical Center [SF DVAMC]). The goal of this study was to provide this information to aid educators in planning a curriculum that produces well-rounded internists, proficient in outpatient as well as inpatient care. We asked two primary research questions: what are the principal diagnoses of patients being seen by categorical medicine residents in a three-hospital training program; and how does the case mix of patients seen by residents in continuity clinics compare to what is seen by office-based internists from the NAMCS data.
PATIENTS AND METHODS The study period was the 1991-1992 academic year. Each patient visit is a unit of measurement. UCSF house-staff demographics have been previously reporteds5 January 1995 The American Journal of Medicine*
Volume 98
7
TABLE I
angina’
39.7
2 9.8
8.5 * 5.4
9
5
19
2.0 + 1.2 3.4 + 2.4 6.0 i- 7.7
43.0
+ 9.3
SFGH (n = 1,192) LOS (d)
11.3 + 8.4 5.1+ 3.7
16.4 19.2 15.7 17.1 14.2
+ 9.8
43.4
9.2 + 8.0 11.2 zt 8.1 8.4 it 6.2
Age (~1 39.6 * 9.8
3
46.9 + 15.9 43.6 + 10.5 54.7 + 16.0
+ f f + +
9.0 f 10.1
8.9 + 9.3
acquired immuno&fi-
f 15.5
7.7 f 7.9 5.6 + 6.4 6.0 + 3.5
2 2 3 8.2 i 6.6
45.0 + 15.9 76.6 + 7.7 62.3 + 18.8
62.8 + 14.0
2 2 2
7.7 + 9.8 6.8 + 4.7 7.1 + 8.6 6.2 f 6.0
64.3 zt 17.4 66.2 f 11.5 61.7 + 13.3
10.0 + 8.1
56.4 + 22.1
58.1 + 14.2
7.4 i 4.6
2
2
65.5 f 9.7
4 5 21 6.1 + 7.5
2
63.5 54.3 57.3 52.3 38.4 45.6 f 13.9
2.0 + 1.0
4 8 2 2 3 4
41.3 f 17.9
8.4 + 4.8
2
51.5 f 16.7
32.0 + 97.0
2
62.0 + 24.8
8.4 + 4.8
5
48.4 zt 17.5 62.3 + 15.9 71.8 i 9.2
2 2 2
11.0 2 12.5
2.4 + 1.2 6.8 r 5.1 4.9 + 2.8
48.5 + 15.6
49.0 f 24.6 35.0 f 8.2 46.0 + 11.3
2.1 * 1.4
2
2 2 3
45.2 * 19.5
6.5 + 6.7 6.7 ct 4.2 3
65.8 + 12.2
7.7 * 8.8 7.6 * 5.7
5
59.1 i 15.2 42.1 f 11.0
3.0 + 1.7
5 2
47.7 + 19.4
14.3 + 19.2 9.0 i 8.5
9.4 * 16.5
7
65.3 f 8.9 65.0 + 14.9
7.7 + 6.9 8.7 + 10.4
7 5
46.1 + 15.6
4.8 zt 2.3 8.7 * 9.‘0
49.7
57.5 zt 19.8 57.1 ct 18.9
31
2 8 35
29
9 3
Most Common Principal Discharge Diagnoses of 2,096 UCSF Hospital, SF DVAMC, and SFGH Medicine Ward Inpatients UCSF Hospital (n = 264) SF DVAMC (n = 640) % LOS (d) % % LOS (d) Age (~1 Age (~1 10.1 + 11.2 17
AIDS’ Alcohol abuse Alimentary tract Abdominal pain’ Gastritis/gastroenteritis’ Gastrointestinal bleeding’ Stomach and duodenal diseases’ Cardiovascular diseases Arrhythmias/conduction abnormalities’ Cardiac failure Chest pain, R/O myocardial infarction, Hypertension Diabetes mellitus/hypoglycemia Fever Fluid and electrolyte disorders Genitourinary tract Pyelonephritis/urinary tract infection Renal failure Hematologic Anemia Agranulocytosis/neutropenia Hepatitis/cirrhosis Hospice, respite, boarder Poisoning/drug overdose Pulmonary diseases Asthma/chronic obstructive disease Cough Neoplastic and related diseases Pneumonia Respiratory abnormality Other (diagnoses with less than 2% of sample size)
Statistically significant (P ~0.05) difference in the distribution of patients with the same diagnosis among the three hospitals. UCSF= University of California, San Francisco; SF DVAMC = San Francisco Department of Veterans Affairs Medical Center; SFGH = San Francisco General Hospital; LOS = length of stay; AIDS = ciency syndrome; R/O = rule out.
RESIDENTS
TABLE
EXPERIENTIAL
CURRICULUM/PAPADAKIS
AND
KAGAWA
II Most
Common
Diagnoses
of UCSF
Hospital, UCSF
SF DVAMC,
and
Hospital
Continuity
Clinic
SF DVAMCt
ln = :,452) 0
AIDS/HIV-positive Alimentary tract Abdominal pain Peptic ulcer disease Anemia Arthritis and musculoskeletal disorders Back pain Degenerative joint disease/arthritis Hip, knee, shoulder, neck, elbow, ankle pain Cardiovascular diseases Atrial fibrillation Chest pain Congestive heart failure Coronary artery disease Hypertension Depression Endocrine Diabetes mellitus Thyroid disease Follow-up visit Headaches/migraines Health care maintenance/patient education Lipid disorders Obesity Pulmonary diseases Acute bronchitis Asthma Chronic obstructive pulmonary disease Rash Upper respiratory infection Other (diagnoses with less than 2% of sample size)
SFGH
Patients’ SFGH (n = f,9W
(n = 2561)
2
3
3 ; z 3 2
: 2
; 2 2 3
4 1: 17
14 2 4
3: 2
5
19 2
2 2 8 2
5
3 2
2 : 3’ 31
5
;
41
10
‘There is a statistically significant difference in the distribution of the number of patients across the hospitals in all diagnoses. +Data do not total 100% due to rounding. UCSF= University of California, San Francisco; SF DVAMC = San Francisco Department of Veterans Affairs Medical Center; SFGH = San Francisco General Hospital; AIDS = acquired immunodeficiency syndrome; HIV = human immunodeficiency virus.
Description of UCSF Hospital Categorical Medicine Program First-year residents spend approximately 9 months on an inpatient service (7 months on general medicine wards and 2 months on an intensive care rotation, such as the coronary care unit, the intensive care unit, the liver transplant unit, or the cancer unit), 2 months in the emergency room, and 1 month in an outpatient rotation. Second-year residents spend approximately 8 months on an inpatient service, 1 month on an inpatient or outpatient elective, 2 months in the emergency room, and 1 month in an ambulatory care block rotation. Third-year residents spend about 4 months on required inpatient services, 5 months on elective inpatient or outpatient services, 2 months in the emergency room, and 1 month on an ambulatory block rotation. Resident rotations are divided among the three core hospitals (UCSF Hospital, SFGH, and SF DVAMC). Residents have a continuity clinic a half day a week for 3 years at one of the core hospitals. A formal didactic curriculum accompanies both the inpatient and the outpatient rotations.
Patient Demographics Ages, length of hospital stay, and discharge diagnoses of the house staffs consecutively treated patients admitted to the UCSF Hospital general medicine service and to the medicine service of the SF DVAMC (which includes the coronary care and medical intensive care units) were obtained from hospital discharge (SF DVAMC) and from billing (LJCSF Hospital) International Classification of Diseases, 9th Reuision (ICD-9) principal diagnostic codes. Demographic data on SFGH patients consecutively admitted to the medical service over 4 random months were obtained from the discharge billing ICD9 codes. Outpatient discharge diagnoses of patients seen consecutively by residents in the UCSF Hospital and SFGH emergency rooms were gathered from emergency room logs. Diagnoses of patients seen consecutively in the continuity clinic over 1 year at UCSF Hospital, SFGH, and SF DVAMC were obtained from the patient encounter forms completed by residents at the end of the patient’s visit and ICD-9 billing codes. The SFGH continuity clinic patient diagnoses January
1995
The American
Journal
of Medicine@
Volume
98
9
RESIDENTS
TABLE
EXPERIENTIAL
CURRICULUM/PAPADAKIS
AND
KAGAWA
Ill Most
Common
Diagnoses
of UCSF
Hospital
and
SFGH
Emergency
Room
Patients
UCSF Hospital % (n = 892)
% (n
2
2 AIDS Alimentary tract Abdominal pain : Gastritis, gastroenteritis Anxiety, hyperventilation’ 2 Cardiovascular diseases 2 Noncardiac/atypical chest pain 3 Rule out myocardial infarction, angina Urinary tract infection/urethritis Infectious diseases Cellulitis’ Pharyngitis/upper respiratory infection Musculoskeletal pain/strain’ Nervous system 2 Migraines* Seizures’ Pulmonary diseases 4 Asthma/chronic obstructive pulmonary disease’ Bronchitis 2 Pneumonia : Skin Substance abuse disorders Alcohol abuse 2 Drug use, overdose Trauma 11 Abrasions, wounds, burns, lacerations’ 4 Fracture, dislocation, cast problem’ 2 Motor vehicle accident, trauma, rape’ 40 Other (diagnoses with less than 2% of sample size) ‘P co.05 when the the number of patients with the same diagnosis at the two sites is compared. UC%=
University of California, San Francisco; SFGH = San Francisco General Hospital; AIDS = acquired immunodeficiency
Analysis Data are reported as mean + SD. Categorical data were tested by the chi-square statistic. Comparisons with a P value of less than 5% (two-tailed) were designated as statistically significant.
RESULTS The most common diagnoses of 24,218 patient visits from the UCSF Department of Medicine’s three January
1995
The American
Journal
of Medicine”
Volume
: 3 2 2 3 3 4 6 2 3
10
55 syndrome.
core hospitals are shown in Tables I, II, and III. Table I compares the distribution of the principal discharge diagnoses, patients’ ages, and length of hospital stay among the inpatients on the general medical wards. Approximately 70% of the inpatients have one of the 25 most common principal diagnoses.The absence of inpatients with cardiac diseaseat the UCSF Hospital is because these patients are cared for by a cardiology team, whereas similar patients at SF DVAMC and SFGH are cared for by the medicine teams. Table II compares the diagnoses of patients seen in the continuity clinics in the three hospitals, and Table III compares the diagnosesof patients seen in the emergency rooms at the UCSF Hospital and SFGH. In almost all comparisons of the number of patients with particular diagnoses among the three hospital sites, there is a statistically significant difference in the distribution of patients with the diagnoses. Table IV compares the frequency of the 20 most common diagnoses seen by practicing internists in the NAMCS with the frequency with which these diagnoses were seen by the residents in their continuity clinic. The 20 most common NAMCS diagnoses constitute 46% of the office visits to practicing internists, while these same diagnoses constitute 87%
represent all patients seen in the general medicine clinic over 1 year. Patient demographic data were not obtained from the UCSF Hospital and SFGH coronary care unit or intensive care unit, from the cancer research unit or the liver transplant unit, or from the 3 elective month’s rotations chosen by the second-year (1 month) and third-year (2 months) residents. Diagnoses and ages of patients who made office visits to practicing internists came from the NAMCS.4 These data were compared with corresponding data of continuity clinic patients seen by categorical medicine residents from all three sites. Since the NAMCS excluded patients seen in the emergency room, only data from continuity clinic patients seen by residents were used in the comparison of diagnoses.
10
SFGH q 1,316)
98
TABLE IV A Comparison of Outpatient Diagnoses Seen in Office Visits to Practicing Internists and to Categorical Medicine Residents NAMCS Practicing internists (n = 78,816) (in thousands) % Essential hypertension 10 Diabetes mellitus 5 Acute upper respiratory infections 4 General medical examination 3 Osteoarthrosis and allied disorders 2 Disorders of lipid metabolism 2 Bronchitis, acute or chronic 2 Chronic ischemic heart disease 2 Chronic sinusitis 2 Other arthropathies 2 General symptoms 2 Cardiac dysrhythmias 2 Asthma 1 Disorders of the back 1 Chronic airway obstruction 1 Acute pharyngitis 1 Symptoms of the respiratory system and chest 1 Heart failure 1 Allergic rhinitis 1 Symptoms of the abdomen and pelvis 1 Total 46 NAMCS = National Ambulatory
Medicine Residents (n = 20,388) % 27 16 1 2 4 3 0 ii 3 1 2 4 s 1 1 4 0 6 87
Medical Care Survey.
TABLE V
Patient Age (y) Under 25 25-44 45-64 65-74 7%
Distribution by Age of Outpatients Seen by NAMCS Practicing Internists and by Categorical Medicine Residents in the UCSF Hospital and SFGH Continuity Clinics. NAMCS Practicing Internists’ Medicine Residents (n = 78,816) (in thousands) (n = 7,262) % of visits % of Visits 4 2 25 26 29 28 26 20 19 16
‘Data do not total 100% due to rounding. NAMCS = National Ambulatory Medical Care Survey; UCSF= University of California, San Francisco; SFGH = San Francisco General Hospital.
of the cases seen by the residents in their continuity clinic. A major reason for that difference in distribution in the two groups is the increased representation of the diagnoses of hypertension and diabetes mellitus in the residents’ patients. Table V shows that the NAMCS and residents’ patients were of similar age.
COMMENTS The definition provided by Ende and DavidofF of a medicine residency curriculum--namely, that the essentials of residency training are “a series of experiences with patients”- is useful. Unlike in a classroom setting, the bulk of the internal medicine curriculum is experiential, and therefore the case mix the residents encounter is a critical element. This study could not identify all elements of the curriculum, such as interaction with other residents, faculty,
conferences, and didactic sessions.However, the experiential content of a categorical medicine training program, composed of patients from a university hospital, a Veterans Affairs hospital, and a municipal hospital, has been described to provide data for a systematic assessmentof the clinical training. Common internal medicine problems comprise the great majority of the case mix of patients, both inpatients and outpatients, seen by categorical medicine residents. It is widely accepted that inpatient training experiences have become less representative of the practice of internal medicine.3,7 However, program changes, such as removing patients admitted to the hospital specifically for procedures such as endoscopic retrograde cholangiopancreatography, cardiac catheterization, or electrophysiologic studies from the general medicine ward teams, can enhance the
January 1995 The American Journal of Medicine@ Volume 98
11
RESIDENTS EXPERIENTIAL CURRICULUM/PAPADAKIS
AND KAGAWA
data This study did not get to the complexity of the patient mix, which, for inpatients, can only be inferred from the patient’s age and length of hospital stay. The accuracy of the listed diagnoses was not validated. Underreporting of certain diagnoses (such as AIDS) for confidentiality or insurance purposes may have occurred. The case mix of patients seen on subspecialty rotations was not reported. Experiences for residents were examined collectively; nonetheless, there may be individual resident variability in patient exposure and experiences. Lastly, this study may not be generalizable to categorical medicine training programs with different structures. This study demonstrates that at one categorical medicine residency program, the hospital-based case mix provides a large number of general internal medicine problems. Diagnoses of patients seen by medical residents in a categorical university hospitalbased educational program can mirror what is seen by practicing internists. “Bread and butter” general medicine can be the rule, not the exception. In terms of a patient population, teaching hospitals and their clinics can provide residents with relevant experience. The program’s challenge entails balancing and integrating these inpatient and outpatient educational experiences.
general medicine experiential curriculum of the ward rotations. Such changes have occurred in this program, where at least 25% of all medical admissions to the UCSF Hospital are for a specialized procedure (personal communication, Department of Medicine Administration office). We do not suggest, however, that these program adaptations can reverse all the recent changes in inpatient demographics, such as increased patient acuity and decreased length of hospital stay. When comparing outpatient house staff experiences with those of practicing internists, the 20 most common principal diagnoses seen by practicing internists comprised nearly 99% of continuity clinic cases seen by the residents. The age distribution of the patients is also similar. The study has examined the continuity clinic case mix. What cannot be assessed from these data is whether the time spent on the outpatient services is appropriate for categorical residents to learn the skills necessary for the practice of internal medicine. Data from graduates of the UCSF Primary Care F’rogram do not support the disproportionate time allocation of medicine residents to the inpatient rotations8 Although primary care residents received shortened training on inpatient rotations compared with the categorical residents, the primary care program graduates felt adequately trained in inpatient medicine, but wished for even greater ambulatory care traming.8 Differences in the distribution of diagnoses were found among the university hospital, the Veterans Affairs hospital, and the municipal hospital. The individual hospital’s educational imbalance was offset by the rotations in the other hospitals, highlighting an advantage of a multiple hospital training program. The limitations of this study include a recognition that, while exposure to a relevant population of patients is necessary, patient experiences alone are not sufficient for optimum training. The full curriculum includes not only caring for patients by the residents, but also how the patients are used for purposes of learning. In addition, although there is similarity between the diagnoses of patients seen by residents and practicing internists, patient acuity, chronicity, and comorbidity may differ, and, therefore, the populations may be different. Only the patients’ primary diagnoses were obtained, as was true of the NAMCS
12
January
1995
The American
Journal
of Medicine@
Volume
ACKNOWLEDGMENT We gratefully acknowledge and Ms. Sylvia Miles.
the technical assistance of Ms. Cynthia Chatterjee
REFERENCES 1. lnui TS, Nolan JP eds. Internal medicine curriculum reform. Ann Intern Med. 1992;116:1041-1115. 2. Schroeder JSA, Showstack JA, Gerber-t B. Residency training in internal medicine: time for a change? Ann Intern Med. 1986;104:554-561. 3. American Board of Internal Medicine. Policies and Procedures for Certification in Internal Medicine. Philadelphia: American Board of Internal Medicine; July 1993;1-16. 4. Woodwell DA. Office visits to internists, 1989. Advance Data from Vital Health Statistics, No. 209. Hyattsville, Maryland: National Center for Health Statistics; 1992. 5. Papadakis MA, Kagawa MK. Influence of ambulatory care rotations on gatekeeping referral patterns of categorical medical residents. Acad Med. 1994;69:299-303. 6. Ende J, Davidoff F. What is a curriculum? Ann intern Med. 1992; 116:1055-1057. 7. Karpf M, Levey GS. Training primary care physicians. Ann Intern Med. 1992;116:514-515. 8. McPhee SJ, Mitchell TF, Schroeder SA, et al. Training in a primary care internal medicine residency program. JAMA. 1987;258:1491-1495.
98
Curriculum
Maxine A. Papadakis, MD, Margaret K. Kagawa, RN, MS, San Francisco, California PURPOSE: To assess the case mix or experiential curriculum of a university-based categorical medicine residency program and compare the residents’ continuity clinic case mix with the outpatients seen by practicing internists. PATIENTS AND METHODS: Descriptive study during the 1991-1992 academic year of 24,218 inpatients and outpatients from the University of California, San Francisco, Department of Medicine’s three core hospitals: the University of California, San Francisco Hospital; the Department of Veterans Affairs Medical Center; and the San Francisco General Hospital. The diagnoses and ages of patients who made office visits to practicing internists from the National Ambulatory Medical Care Survey (NAMCS) were compared with corresponding data from the categorical medicine residents’ continuity clinic patients. RESULTS: Seventy percent of the general medicine inpatients at the three core hospitals had one of the 25 most common principal diagnoses for inpatients. Eighty-seven percent of the patients seen by the residents in their continuity clinic had one of the 20 most common diagnoses seen by practicing internists in the NAMCS. The age distribution was similar in both groups. CONCLUSIONS: A SysM?I&iC assessment of CliniCal training at one university-based hospital program shows that common internal medicine problems represent the case mix of the great majority of patients, both inpatients and outpatients, seen by categorical medicine residents. Residents’ continuity clinic patients are similar to patients seen by practicing internists, The program’s challenge is to ensure that residents have adequate time with general medicine experiences, both in the inpatient setting and in the ambulatory and longitudinal care settings, while balancing and integrating these activities.
From the Department of Medicine, Student Programs, University of California, San Francisco, and the Department of Veterans Affairs Medical Center, San Francisco, California. This study was funded by a Western Region Department of Veterans Affairs Ambulatory Care Education Grant. Reouests for reorints should be addressed to Maxine A. Papadakis. MD, Department of Veterans Affairs Medical Center, 4150 Clement Street lllA1, San Francisco, California 94121. Manuscript submitted November 23, 1993 and accepted in revised form April 28, 1994.
espite the heightened interest in internal mediD cine residency curriculum reform,’ little is known about current categorical medicine residency experiences. As the acute-care university hospitals become a site of increasing admissions for technologic procedures, many university teaching hospitals have become so specialized that they provide an unbalanced educational experience.2 To address the issue of educational balance, Schroeder and coworkers2 have called for the systematic assessment of clinical training of residents by the monitoring of their clinical experiences. Surgical certifying boards place great importance on the volume of selected procedures performed by house officers during their residency. The American Board of Internal Medicine, however, does not place a similar emphasis on case mix in the training of internal medicine residents3 Based on the 1989 National Ambulatory Medical Care Survey (NAMCS),4 data on types of diagnoses and demographic information on patients seen by practicing internists are now available. A comparison of these data to the types of ambulatory patients seen by categorical medicine residents would help identify what should constitute essential components of ambulatory experiences. As part of a Department of Veterans Affairs (VA) Ambulatory Care Education initiative, this study was undertaken to describe the current inpatient and outpatient case mix of categorical medicine residents at a university program consisting of a university hospital (the University of California, San Francisco [UCSF] Hospital or MoffittLong Hospitals), a municipal hospital (San Francisco General Hospital [SFGH]), and a veterans hospital (San Francisco Department of Veterans Affairs Medical Center [SF DVAMC]). The goal of this study was to provide this information to aid educators in planning a curriculum that produces well-rounded internists, proficient in outpatient as well as inpatient care. We asked two primary research questions: what are the principal diagnoses of patients being seen by categorical medicine residents in a three-hospital training program; and how does the case mix of patients seen by residents in continuity clinics compare to what is seen by office-based internists from the NAMCS data.
PATIENTS AND METHODS The study period was the 1991-1992 academic year. Each patient visit is a unit of measurement. UCSF house-staff demographics have been previously reporteds5 January 1995 The American Journal of Medicine*
Volume 98
7
TABLE I
angina’
39.7
2 9.8
8.5 * 5.4
9
5
19
2.0 + 1.2 3.4 + 2.4 6.0 i- 7.7
43.0
+ 9.3
SFGH (n = 1,192) LOS (d)
11.3 + 8.4 5.1+ 3.7
16.4 19.2 15.7 17.1 14.2
+ 9.8
43.4
9.2 + 8.0 11.2 zt 8.1 8.4 it 6.2
Age (~1 39.6 * 9.8
3
46.9 + 15.9 43.6 + 10.5 54.7 + 16.0
+ f f + +
9.0 f 10.1
8.9 + 9.3
acquired immuno&fi-
f 15.5
7.7 f 7.9 5.6 + 6.4 6.0 + 3.5
2 2 3 8.2 i 6.6
45.0 + 15.9 76.6 + 7.7 62.3 + 18.8
62.8 + 14.0
2 2 2
7.7 + 9.8 6.8 + 4.7 7.1 + 8.6 6.2 f 6.0
64.3 zt 17.4 66.2 f 11.5 61.7 + 13.3
10.0 + 8.1
56.4 + 22.1
58.1 + 14.2
7.4 i 4.6
2
2
65.5 f 9.7
4 5 21 6.1 + 7.5
2
63.5 54.3 57.3 52.3 38.4 45.6 f 13.9
2.0 + 1.0
4 8 2 2 3 4
41.3 f 17.9
8.4 + 4.8
2
51.5 f 16.7
32.0 + 97.0
2
62.0 + 24.8
8.4 + 4.8
5
48.4 zt 17.5 62.3 + 15.9 71.8 i 9.2
2 2 2
11.0 2 12.5
2.4 + 1.2 6.8 r 5.1 4.9 + 2.8
48.5 + 15.6
49.0 f 24.6 35.0 f 8.2 46.0 + 11.3
2.1 * 1.4
2
2 2 3
45.2 * 19.5
6.5 + 6.7 6.7 ct 4.2 3
65.8 + 12.2
7.7 * 8.8 7.6 * 5.7
5
59.1 i 15.2 42.1 f 11.0
3.0 + 1.7
5 2
47.7 + 19.4
14.3 + 19.2 9.0 i 8.5
9.4 * 16.5
7
65.3 f 8.9 65.0 + 14.9
7.7 + 6.9 8.7 + 10.4
7 5
46.1 + 15.6
4.8 zt 2.3 8.7 * 9.‘0
49.7
57.5 zt 19.8 57.1 ct 18.9
31
2 8 35
29
9 3
Most Common Principal Discharge Diagnoses of 2,096 UCSF Hospital, SF DVAMC, and SFGH Medicine Ward Inpatients UCSF Hospital (n = 264) SF DVAMC (n = 640) % LOS (d) % % LOS (d) Age (~1 Age (~1 10.1 + 11.2 17
AIDS’ Alcohol abuse Alimentary tract Abdominal pain’ Gastritis/gastroenteritis’ Gastrointestinal bleeding’ Stomach and duodenal diseases’ Cardiovascular diseases Arrhythmias/conduction abnormalities’ Cardiac failure Chest pain, R/O myocardial infarction, Hypertension Diabetes mellitus/hypoglycemia Fever Fluid and electrolyte disorders Genitourinary tract Pyelonephritis/urinary tract infection Renal failure Hematologic Anemia Agranulocytosis/neutropenia Hepatitis/cirrhosis Hospice, respite, boarder Poisoning/drug overdose Pulmonary diseases Asthma/chronic obstructive disease Cough Neoplastic and related diseases Pneumonia Respiratory abnormality Other (diagnoses with less than 2% of sample size)
Statistically significant (P ~0.05) difference in the distribution of patients with the same diagnosis among the three hospitals. UCSF= University of California, San Francisco; SF DVAMC = San Francisco Department of Veterans Affairs Medical Center; SFGH = San Francisco General Hospital; LOS = length of stay; AIDS = ciency syndrome; R/O = rule out.
RESIDENTS
TABLE
EXPERIENTIAL
CURRICULUM/PAPADAKIS
AND
KAGAWA
II Most
Common
Diagnoses
of UCSF
Hospital, UCSF
SF DVAMC,
and
Hospital
Continuity
Clinic
SF DVAMCt
ln = :,452) 0
AIDS/HIV-positive Alimentary tract Abdominal pain Peptic ulcer disease Anemia Arthritis and musculoskeletal disorders Back pain Degenerative joint disease/arthritis Hip, knee, shoulder, neck, elbow, ankle pain Cardiovascular diseases Atrial fibrillation Chest pain Congestive heart failure Coronary artery disease Hypertension Depression Endocrine Diabetes mellitus Thyroid disease Follow-up visit Headaches/migraines Health care maintenance/patient education Lipid disorders Obesity Pulmonary diseases Acute bronchitis Asthma Chronic obstructive pulmonary disease Rash Upper respiratory infection Other (diagnoses with less than 2% of sample size)
SFGH
Patients’ SFGH (n = f,9W
(n = 2561)
2
3
3 ; z 3 2
: 2
; 2 2 3
4 1: 17
14 2 4
3: 2
5
19 2
2 2 8 2
5
3 2
2 : 3’ 31
5
;
41
10
‘There is a statistically significant difference in the distribution of the number of patients across the hospitals in all diagnoses. +Data do not total 100% due to rounding. UCSF= University of California, San Francisco; SF DVAMC = San Francisco Department of Veterans Affairs Medical Center; SFGH = San Francisco General Hospital; AIDS = acquired immunodeficiency syndrome; HIV = human immunodeficiency virus.
Description of UCSF Hospital Categorical Medicine Program First-year residents spend approximately 9 months on an inpatient service (7 months on general medicine wards and 2 months on an intensive care rotation, such as the coronary care unit, the intensive care unit, the liver transplant unit, or the cancer unit), 2 months in the emergency room, and 1 month in an outpatient rotation. Second-year residents spend approximately 8 months on an inpatient service, 1 month on an inpatient or outpatient elective, 2 months in the emergency room, and 1 month in an ambulatory care block rotation. Third-year residents spend about 4 months on required inpatient services, 5 months on elective inpatient or outpatient services, 2 months in the emergency room, and 1 month on an ambulatory block rotation. Resident rotations are divided among the three core hospitals (UCSF Hospital, SFGH, and SF DVAMC). Residents have a continuity clinic a half day a week for 3 years at one of the core hospitals. A formal didactic curriculum accompanies both the inpatient and the outpatient rotations.
Patient Demographics Ages, length of hospital stay, and discharge diagnoses of the house staffs consecutively treated patients admitted to the UCSF Hospital general medicine service and to the medicine service of the SF DVAMC (which includes the coronary care and medical intensive care units) were obtained from hospital discharge (SF DVAMC) and from billing (LJCSF Hospital) International Classification of Diseases, 9th Reuision (ICD-9) principal diagnostic codes. Demographic data on SFGH patients consecutively admitted to the medical service over 4 random months were obtained from the discharge billing ICD9 codes. Outpatient discharge diagnoses of patients seen consecutively by residents in the UCSF Hospital and SFGH emergency rooms were gathered from emergency room logs. Diagnoses of patients seen consecutively in the continuity clinic over 1 year at UCSF Hospital, SFGH, and SF DVAMC were obtained from the patient encounter forms completed by residents at the end of the patient’s visit and ICD-9 billing codes. The SFGH continuity clinic patient diagnoses January
1995
The American
Journal
of Medicine@
Volume
98
9
RESIDENTS
TABLE
EXPERIENTIAL
CURRICULUM/PAPADAKIS
AND
KAGAWA
Ill Most
Common
Diagnoses
of UCSF
Hospital
and
SFGH
Emergency
Room
Patients
UCSF Hospital % (n = 892)
% (n
2
2 AIDS Alimentary tract Abdominal pain : Gastritis, gastroenteritis Anxiety, hyperventilation’ 2 Cardiovascular diseases 2 Noncardiac/atypical chest pain 3 Rule out myocardial infarction, angina Urinary tract infection/urethritis Infectious diseases Cellulitis’ Pharyngitis/upper respiratory infection Musculoskeletal pain/strain’ Nervous system 2 Migraines* Seizures’ Pulmonary diseases 4 Asthma/chronic obstructive pulmonary disease’ Bronchitis 2 Pneumonia : Skin Substance abuse disorders Alcohol abuse 2 Drug use, overdose Trauma 11 Abrasions, wounds, burns, lacerations’ 4 Fracture, dislocation, cast problem’ 2 Motor vehicle accident, trauma, rape’ 40 Other (diagnoses with less than 2% of sample size) ‘P co.05 when the the number of patients with the same diagnosis at the two sites is compared. UC%=
University of California, San Francisco; SFGH = San Francisco General Hospital; AIDS = acquired immunodeficiency
Analysis Data are reported as mean + SD. Categorical data were tested by the chi-square statistic. Comparisons with a P value of less than 5% (two-tailed) were designated as statistically significant.
RESULTS The most common diagnoses of 24,218 patient visits from the UCSF Department of Medicine’s three January
1995
The American
Journal
of Medicine”
Volume
: 3 2 2 3 3 4 6 2 3
10
55 syndrome.
core hospitals are shown in Tables I, II, and III. Table I compares the distribution of the principal discharge diagnoses, patients’ ages, and length of hospital stay among the inpatients on the general medical wards. Approximately 70% of the inpatients have one of the 25 most common principal diagnoses.The absence of inpatients with cardiac diseaseat the UCSF Hospital is because these patients are cared for by a cardiology team, whereas similar patients at SF DVAMC and SFGH are cared for by the medicine teams. Table II compares the diagnoses of patients seen in the continuity clinics in the three hospitals, and Table III compares the diagnosesof patients seen in the emergency rooms at the UCSF Hospital and SFGH. In almost all comparisons of the number of patients with particular diagnoses among the three hospital sites, there is a statistically significant difference in the distribution of patients with the diagnoses. Table IV compares the frequency of the 20 most common diagnoses seen by practicing internists in the NAMCS with the frequency with which these diagnoses were seen by the residents in their continuity clinic. The 20 most common NAMCS diagnoses constitute 46% of the office visits to practicing internists, while these same diagnoses constitute 87%
represent all patients seen in the general medicine clinic over 1 year. Patient demographic data were not obtained from the UCSF Hospital and SFGH coronary care unit or intensive care unit, from the cancer research unit or the liver transplant unit, or from the 3 elective month’s rotations chosen by the second-year (1 month) and third-year (2 months) residents. Diagnoses and ages of patients who made office visits to practicing internists came from the NAMCS.4 These data were compared with corresponding data of continuity clinic patients seen by categorical medicine residents from all three sites. Since the NAMCS excluded patients seen in the emergency room, only data from continuity clinic patients seen by residents were used in the comparison of diagnoses.
10
SFGH q 1,316)
98
TABLE IV A Comparison of Outpatient Diagnoses Seen in Office Visits to Practicing Internists and to Categorical Medicine Residents NAMCS Practicing internists (n = 78,816) (in thousands) % Essential hypertension 10 Diabetes mellitus 5 Acute upper respiratory infections 4 General medical examination 3 Osteoarthrosis and allied disorders 2 Disorders of lipid metabolism 2 Bronchitis, acute or chronic 2 Chronic ischemic heart disease 2 Chronic sinusitis 2 Other arthropathies 2 General symptoms 2 Cardiac dysrhythmias 2 Asthma 1 Disorders of the back 1 Chronic airway obstruction 1 Acute pharyngitis 1 Symptoms of the respiratory system and chest 1 Heart failure 1 Allergic rhinitis 1 Symptoms of the abdomen and pelvis 1 Total 46 NAMCS = National Ambulatory
Medicine Residents (n = 20,388) % 27 16 1 2 4 3 0 ii 3 1 2 4 s 1 1 4 0 6 87
Medical Care Survey.
TABLE V
Patient Age (y) Under 25 25-44 45-64 65-74 7%
Distribution by Age of Outpatients Seen by NAMCS Practicing Internists and by Categorical Medicine Residents in the UCSF Hospital and SFGH Continuity Clinics. NAMCS Practicing Internists’ Medicine Residents (n = 78,816) (in thousands) (n = 7,262) % of visits % of Visits 4 2 25 26 29 28 26 20 19 16
‘Data do not total 100% due to rounding. NAMCS = National Ambulatory Medical Care Survey; UCSF= University of California, San Francisco; SFGH = San Francisco General Hospital.
of the cases seen by the residents in their continuity clinic. A major reason for that difference in distribution in the two groups is the increased representation of the diagnoses of hypertension and diabetes mellitus in the residents’ patients. Table V shows that the NAMCS and residents’ patients were of similar age.
COMMENTS The definition provided by Ende and DavidofF of a medicine residency curriculum--namely, that the essentials of residency training are “a series of experiences with patients”- is useful. Unlike in a classroom setting, the bulk of the internal medicine curriculum is experiential, and therefore the case mix the residents encounter is a critical element. This study could not identify all elements of the curriculum, such as interaction with other residents, faculty,
conferences, and didactic sessions.However, the experiential content of a categorical medicine training program, composed of patients from a university hospital, a Veterans Affairs hospital, and a municipal hospital, has been described to provide data for a systematic assessmentof the clinical training. Common internal medicine problems comprise the great majority of the case mix of patients, both inpatients and outpatients, seen by categorical medicine residents. It is widely accepted that inpatient training experiences have become less representative of the practice of internal medicine.3,7 However, program changes, such as removing patients admitted to the hospital specifically for procedures such as endoscopic retrograde cholangiopancreatography, cardiac catheterization, or electrophysiologic studies from the general medicine ward teams, can enhance the
January 1995 The American Journal of Medicine@ Volume 98
11
RESIDENTS EXPERIENTIAL CURRICULUM/PAPADAKIS
AND KAGAWA
data This study did not get to the complexity of the patient mix, which, for inpatients, can only be inferred from the patient’s age and length of hospital stay. The accuracy of the listed diagnoses was not validated. Underreporting of certain diagnoses (such as AIDS) for confidentiality or insurance purposes may have occurred. The case mix of patients seen on subspecialty rotations was not reported. Experiences for residents were examined collectively; nonetheless, there may be individual resident variability in patient exposure and experiences. Lastly, this study may not be generalizable to categorical medicine training programs with different structures. This study demonstrates that at one categorical medicine residency program, the hospital-based case mix provides a large number of general internal medicine problems. Diagnoses of patients seen by medical residents in a categorical university hospitalbased educational program can mirror what is seen by practicing internists. “Bread and butter” general medicine can be the rule, not the exception. In terms of a patient population, teaching hospitals and their clinics can provide residents with relevant experience. The program’s challenge entails balancing and integrating these inpatient and outpatient educational experiences.
general medicine experiential curriculum of the ward rotations. Such changes have occurred in this program, where at least 25% of all medical admissions to the UCSF Hospital are for a specialized procedure (personal communication, Department of Medicine Administration office). We do not suggest, however, that these program adaptations can reverse all the recent changes in inpatient demographics, such as increased patient acuity and decreased length of hospital stay. When comparing outpatient house staff experiences with those of practicing internists, the 20 most common principal diagnoses seen by practicing internists comprised nearly 99% of continuity clinic cases seen by the residents. The age distribution of the patients is also similar. The study has examined the continuity clinic case mix. What cannot be assessed from these data is whether the time spent on the outpatient services is appropriate for categorical residents to learn the skills necessary for the practice of internal medicine. Data from graduates of the UCSF Primary Care F’rogram do not support the disproportionate time allocation of medicine residents to the inpatient rotations8 Although primary care residents received shortened training on inpatient rotations compared with the categorical residents, the primary care program graduates felt adequately trained in inpatient medicine, but wished for even greater ambulatory care traming.8 Differences in the distribution of diagnoses were found among the university hospital, the Veterans Affairs hospital, and the municipal hospital. The individual hospital’s educational imbalance was offset by the rotations in the other hospitals, highlighting an advantage of a multiple hospital training program. The limitations of this study include a recognition that, while exposure to a relevant population of patients is necessary, patient experiences alone are not sufficient for optimum training. The full curriculum includes not only caring for patients by the residents, but also how the patients are used for purposes of learning. In addition, although there is similarity between the diagnoses of patients seen by residents and practicing internists, patient acuity, chronicity, and comorbidity may differ, and, therefore, the populations may be different. Only the patients’ primary diagnoses were obtained, as was true of the NAMCS
12
January
1995
The American
Journal
of Medicine@
Volume
ACKNOWLEDGMENT We gratefully acknowledge and Ms. Sylvia Miles.
the technical assistance of Ms. Cynthia Chatterjee
REFERENCES 1. lnui TS, Nolan JP eds. Internal medicine curriculum reform. Ann Intern Med. 1992;116:1041-1115. 2. Schroeder JSA, Showstack JA, Gerber-t B. Residency training in internal medicine: time for a change? Ann Intern Med. 1986;104:554-561. 3. American Board of Internal Medicine. Policies and Procedures for Certification in Internal Medicine. Philadelphia: American Board of Internal Medicine; July 1993;1-16. 4. Woodwell DA. Office visits to internists, 1989. Advance Data from Vital Health Statistics, No. 209. Hyattsville, Maryland: National Center for Health Statistics; 1992. 5. Papadakis MA, Kagawa MK. Influence of ambulatory care rotations on gatekeeping referral patterns of categorical medical residents. Acad Med. 1994;69:299-303. 6. Ende J, Davidoff F. What is a curriculum? Ann intern Med. 1992; 116:1055-1057. 7. Karpf M, Levey GS. Training primary care physicians. Ann Intern Med. 1992;116:514-515. 8. McPhee SJ, Mitchell TF, Schroeder SA, et al. Training in a primary care internal medicine residency program. JAMA. 1987;258:1491-1495.
98