JOURNAL
OF SURGICAL
RESEARCH
52,97-100
(1992)
Quality Assurance and Morbidity JON
S.
THOMPSON,
M.D.,
and Mortality Conference
AND MARY A. PRIOR,
R.N.
Surgical Service Omaha VAMC, and Department of Surgery, University of Nebraska Medical and Creighton University School of Medicine, Omaha, Nebraska 68198 Presented
at the Annual
Symposium
of the Association
of Veterans
Many surgeons assert that Morbidity and Mortality (M & M) conference in itself assures an effective quality assurance (QA) program. Recent emphasis on QA in other sectors has resulted in other processes for evaluating quality of care. The goals of QA programs are to identify adverse patient care events, relate these to specific physicians and use this information to improve patient care, and for credentialing and privileging physicians. Our aim was to determine the role of surgical M & M conference in a QA program which also includes occurrence screening, wound infection surveillance, and surgical case review. The weekly M & M conference is a discussion of identified complications and deaths submitted voluntarily by surgeons. During a 2year period 5755 procedures were associated with 255 complications and 82 deaths. Only 74% of events identified by occurrence screening, 35% of cases identified by surgical case review, and 54% of wound infections had been submitted to M & M conference. Seventy-four percent of surgical residents and 33% of staff surgeons were present at M & M conference when their complications were discussed. Level of care (I, accepted practice; II, may have managed differently; and III, would have managed differently) was assessed for each complication at M & M conference and by peer review of the medical record for occurrence screening. The assignment of level of care was similar by either process (I = 49,11= 11,111=2,atM&Mvs1=44,11=16,111=2,r = 0.7405, P < 0.005). M & M conference remains an important component of an overall QA program but does not meet all of the goals. There is excellent agreement between level of care assigned at M BZ M conference compared to peer review of the medical record. However, many adverse events identified via other processes are not reported at M & M conference. In our experience, physicians are often not present when their complications are discussed. o 1~92 Academic press, 1~.
Administration
Surgeons, Milwaukee,
Center,
Wisconsin,
May 9-11, 1991
and Mortality (M & M) conference is an important part of surgical heritage. While the conduct and atmosphere of this conference may vary greatly from institution to institution, the dual goal of open peer review and education is standard [l]. Many surgeons feel an effective, ongoing M & M conference improves the physician’s performance and thus, patient outcome [2]. There has been recent emphasis on quality assurance (QA) in other sectors and this has resulted in other processes for promoting and evaluating quality of care [571. During the past decade the use of generic occurrence screening of adverse patient events has been promoted by the American Hospital Association for quality assessment and is now required of hospitals by Peer Review Organizations [8]. Generic screening has been mandated in all military and Veterans Affairs Hospitals [6-81. The main goal of a QA program should be to improve patient care [4, 51. However, other goals of such programs have been to identify all adverse patient care events, relate these to specific physicians, and to use this information for credentialing and privileging of physicians [6-81. Such information has also been used to identify potentially compensable events for insurance purposes [S, 93. Many surgeons assert that the M & M conference itself assures an effective QA program and that perhaps other types of monitoring and evaluation are superfluous. Thus, the aim of the present study was to determine the role and efficacy of surgical M & M conference in a current QA program. MATERIALS
AND
METHODS
We reviewed all adverse patient care events identified on the surgical service and level of care assigned by physician peer review for fiscal years 1989 and 1990 at the Omaha VAMC. During the 2-year period 5755 invasive procedures were associated with 255 (4.4%) complications and 82 (1.4%) deaths. The distribution of procedures among the surgical specialties is shown in Table 1. Those findings reported and discussed at the surgical M & M conference were compared to those identified and evaluated by other elements of our QA program. Practioner quality of care was evaluated by designating one
Surgeons have long recognized the need for peer review of quality of care [l-3]. Mistakes and misadventures are inevitable in medical practice but can promote learning if physicians accept appropriate responsibility and discuss such events with their peers [4]. Morbidity 97
0022-4804/92 $1.50 Copyright 0 1992 by Academic Press, Inc. All rights of reproduction in any form reserved.
98
JOURNAL
TABLE Invasive
Procedures
Performed
Specialty
OF SURGICAL
RESEARCH:
VOL.
52, NO. 2, FEBRUARY
1
1992
TABLE by Surgical Procedures
Identification
Specialty performed
General surgery Urology Orthopedic surgery Otolaryngology Ophthalmology Oral and maxi110 facial surgery Neurosurgery Thoracic surgery Plastic surgery Other
2526 1167 523 497 427 187 152 143 109 24
Total
5755
of three levels of care for that episode [6]. Level I implies that most experienced, competent practitioners would have handled the case similarly in all respects. Level II implies that most experienced, competent practitioners might have handled the case differently in some respect. Level III implies that most experienced, competent practitioners would have handled the case differently in some respect. Evaluation of care includes diagnostic workup, timeliness and appropriateness of diagnosis and treatment, and recognition and management of clinical deterioration. Our current QA program includes occurrence screening, wound infection surveillance, surgical case review, and continuous evaluation of instances of mortality and morbidity. Occurrence screening is conducted by a nurse assigned to the QA coordinator. The following events are identified: return to OR same admission; admission ~3 days after outpatient procedure, death t24 hr after admission and ~48 hr after operation, transfer to ICU t72 hr after operation, readmission 414 days after discharge. Planned reoperation, e,g., multiple staged procedures and scheduled admissions, are not considered further. Each case is evaluated by peer review by a physician from that specialty involved. The medical record is reviewed and a level of care is assigned. Wound infection surveillance is carried out prospectively by an Infection Control Nurse. A wound infection is defined as the presence of purulent drainage from the surgical wound with or without positive culture. Surgical case review includes all invasive procedures performed on the service using designated screening criteria. The weekly M & M conference is a discussion of complications and deaths occurring in all surgical specialties submitted voluntarily by the Chief Resident of each service. Level of care is assigned at this conference by consensus. Statistical comparison between groups were made with x2 and linear correlation analysis with P < 0.05 for significance.
2
of Adverse
Method of detection
Events identified
Occurrence screening Wound infection surveillance Surgical case review
Events Events reported at M&M conference
95
70 (74%)
69 14
32 (46%) 5 (35%)
RESULTS
Six hundred and two adverse events were initially detected by occurrence screening but in only 95 (16%) instances were problems in care identified by peer review. Of these only 70 (74%) had been submitted to M & M conference (Table 2). The most frequent finding was unplanned return to the operating room during the same admission (Table 3). Significantly fewer of these events were reported at M & M conference than other identified events (51% vs 18%, P < 0.05). All deaths identified by occurrence screening had been reported at M & M conference. Sixty-nine wound infections were identified during this period but 37 (54%) were identified via wound surveillance by the nurse and not initially the physician at M & M conference. Ninety-nine procedures failed initial screening during surgical case review. Forty-nine (49%) procedures failed criteria for justification or appropriateness and the remainder were discrepancies in either the preoperative and postoperative diagnosis or the postoperative and pathologic diagnosis. Fourteen adverse events (primarily error in diagnosis) were identified by subsequent peer review but only 5
TABLE Adverse
Events
Identified
3 by Occurrence M&M Reported
Screening
conference Not reported
Total
Return to OR same admission Admission
OF SURGICAL
RESEARCH
52,97-100
(1992)
Quality Assurance and Morbidity JON
S.
THOMPSON,
M.D.,
and Mortality Conference
AND MARY A. PRIOR,
R.N.
Surgical Service Omaha VAMC, and Department of Surgery, University of Nebraska Medical and Creighton University School of Medicine, Omaha, Nebraska 68198 Presented
at the Annual
Symposium
of the Association
of Veterans
Many surgeons assert that Morbidity and Mortality (M & M) conference in itself assures an effective quality assurance (QA) program. Recent emphasis on QA in other sectors has resulted in other processes for evaluating quality of care. The goals of QA programs are to identify adverse patient care events, relate these to specific physicians and use this information to improve patient care, and for credentialing and privileging physicians. Our aim was to determine the role of surgical M & M conference in a QA program which also includes occurrence screening, wound infection surveillance, and surgical case review. The weekly M & M conference is a discussion of identified complications and deaths submitted voluntarily by surgeons. During a 2year period 5755 procedures were associated with 255 complications and 82 deaths. Only 74% of events identified by occurrence screening, 35% of cases identified by surgical case review, and 54% of wound infections had been submitted to M & M conference. Seventy-four percent of surgical residents and 33% of staff surgeons were present at M & M conference when their complications were discussed. Level of care (I, accepted practice; II, may have managed differently; and III, would have managed differently) was assessed for each complication at M & M conference and by peer review of the medical record for occurrence screening. The assignment of level of care was similar by either process (I = 49,11= 11,111=2,atM&Mvs1=44,11=16,111=2,r = 0.7405, P < 0.005). M & M conference remains an important component of an overall QA program but does not meet all of the goals. There is excellent agreement between level of care assigned at M BZ M conference compared to peer review of the medical record. However, many adverse events identified via other processes are not reported at M & M conference. In our experience, physicians are often not present when their complications are discussed. o 1~92 Academic press, 1~.
Administration
Surgeons, Milwaukee,
Center,
Wisconsin,
May 9-11, 1991
and Mortality (M & M) conference is an important part of surgical heritage. While the conduct and atmosphere of this conference may vary greatly from institution to institution, the dual goal of open peer review and education is standard [l]. Many surgeons feel an effective, ongoing M & M conference improves the physician’s performance and thus, patient outcome [2]. There has been recent emphasis on quality assurance (QA) in other sectors and this has resulted in other processes for promoting and evaluating quality of care [571. During the past decade the use of generic occurrence screening of adverse patient events has been promoted by the American Hospital Association for quality assessment and is now required of hospitals by Peer Review Organizations [8]. Generic screening has been mandated in all military and Veterans Affairs Hospitals [6-81. The main goal of a QA program should be to improve patient care [4, 51. However, other goals of such programs have been to identify all adverse patient care events, relate these to specific physicians, and to use this information for credentialing and privileging of physicians [6-81. Such information has also been used to identify potentially compensable events for insurance purposes [S, 93. Many surgeons assert that the M & M conference itself assures an effective QA program and that perhaps other types of monitoring and evaluation are superfluous. Thus, the aim of the present study was to determine the role and efficacy of surgical M & M conference in a current QA program. MATERIALS
AND
METHODS
We reviewed all adverse patient care events identified on the surgical service and level of care assigned by physician peer review for fiscal years 1989 and 1990 at the Omaha VAMC. During the 2-year period 5755 invasive procedures were associated with 255 (4.4%) complications and 82 (1.4%) deaths. The distribution of procedures among the surgical specialties is shown in Table 1. Those findings reported and discussed at the surgical M & M conference were compared to those identified and evaluated by other elements of our QA program. Practioner quality of care was evaluated by designating one
Surgeons have long recognized the need for peer review of quality of care [l-3]. Mistakes and misadventures are inevitable in medical practice but can promote learning if physicians accept appropriate responsibility and discuss such events with their peers [4]. Morbidity 97
0022-4804/92 $1.50 Copyright 0 1992 by Academic Press, Inc. All rights of reproduction in any form reserved.
98
JOURNAL
TABLE Invasive
Procedures
Performed
Specialty
OF SURGICAL
RESEARCH:
VOL.
52, NO. 2, FEBRUARY
1
1992
TABLE by Surgical Procedures
Identification
Specialty performed
General surgery Urology Orthopedic surgery Otolaryngology Ophthalmology Oral and maxi110 facial surgery Neurosurgery Thoracic surgery Plastic surgery Other
2526 1167 523 497 427 187 152 143 109 24
Total
5755
of three levels of care for that episode [6]. Level I implies that most experienced, competent practitioners would have handled the case similarly in all respects. Level II implies that most experienced, competent practitioners might have handled the case differently in some respect. Level III implies that most experienced, competent practitioners would have handled the case differently in some respect. Evaluation of care includes diagnostic workup, timeliness and appropriateness of diagnosis and treatment, and recognition and management of clinical deterioration. Our current QA program includes occurrence screening, wound infection surveillance, surgical case review, and continuous evaluation of instances of mortality and morbidity. Occurrence screening is conducted by a nurse assigned to the QA coordinator. The following events are identified: return to OR same admission; admission ~3 days after outpatient procedure, death t24 hr after admission and ~48 hr after operation, transfer to ICU t72 hr after operation, readmission 414 days after discharge. Planned reoperation, e,g., multiple staged procedures and scheduled admissions, are not considered further. Each case is evaluated by peer review by a physician from that specialty involved. The medical record is reviewed and a level of care is assigned. Wound infection surveillance is carried out prospectively by an Infection Control Nurse. A wound infection is defined as the presence of purulent drainage from the surgical wound with or without positive culture. Surgical case review includes all invasive procedures performed on the service using designated screening criteria. The weekly M & M conference is a discussion of complications and deaths occurring in all surgical specialties submitted voluntarily by the Chief Resident of each service. Level of care is assigned at this conference by consensus. Statistical comparison between groups were made with x2 and linear correlation analysis with P < 0.05 for significance.
2
of Adverse
Method of detection
Events identified
Occurrence screening Wound infection surveillance Surgical case review
Events Events reported at M&M conference
95
70 (74%)
69 14
32 (46%) 5 (35%)
RESULTS
Six hundred and two adverse events were initially detected by occurrence screening but in only 95 (16%) instances were problems in care identified by peer review. Of these only 70 (74%) had been submitted to M & M conference (Table 2). The most frequent finding was unplanned return to the operating room during the same admission (Table 3). Significantly fewer of these events were reported at M & M conference than other identified events (51% vs 18%, P < 0.05). All deaths identified by occurrence screening had been reported at M & M conference. Sixty-nine wound infections were identified during this period but 37 (54%) were identified via wound surveillance by the nurse and not initially the physician at M & M conference. Ninety-nine procedures failed initial screening during surgical case review. Forty-nine (49%) procedures failed criteria for justification or appropriateness and the remainder were discrepancies in either the preoperative and postoperative diagnosis or the postoperative and pathologic diagnosis. Fourteen adverse events (primarily error in diagnosis) were identified by subsequent peer review but only 5
TABLE Adverse
Events
Identified
3 by Occurrence M&M Reported
Screening
conference Not reported
Total
Return to OR same admission Admission
