| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
|
© 2004 European Association of Cardio-Thoracic Surgery
Improved total quality by monitoring of a cardiothoracic unit. Medical, administrative and economic data followed for 9 yearsDepartment of Cardio-Thoracic Surgery, Heart Center, The European Association for Cardio Thoracic Surgery, University Hospital of Northern Sweden, S-90185 Umeå, Sweden
* Corresponding author. Tel.: +46-90-785-3600; fax: +46-90-785-3601 Received May 21, 2003; received in revised form September 12, 2003; accepted September 15, 2003
To describe monitoring of a cardio-thoracic department from a total quality aspect point of view and to follow the development over 9 years. During the time period 1994–2002 a total of 10,828 cardio-thoracic operations were performed. Capacity, demographic, risk, quality, outcome and economic data were prospectively collected in various registries and analysed. Mean (and median) age increased from 64.2 to 65.3 (66–67). Patients above 70 years increased from 33.6 to 38.7% and above 80 from 2.9 to 5.5%. Operative mortality was unchanged over the time periods at slightly over 2%, with 1-year mortality 6–7%. Mortality for primary, elective coronary artery bypass grafting was 0.26% during the last 3 years. The rate of postoperative complications remained unchanged or decreased with few exceptions: Patients with postoperative confusion increased from 5.0 to 8.1% and patients with a need for face mask ventilation increased from 2.4 to 4.0%. Mean postoperative ventilation time was unchanged at around 22 h, whereas the median decreased from 9.5 to 5.3 h. The workload created by elderly patients was especially noticeable in the intensive care unit (ICU) as both number of postoperative deviations and ICU hours increased as a function of age. Cost per operation decreased by 11%. Total medical rationalisation was higher as salaries increased over time. Mean length of stay decreased by 3 days. Hospital staff hours per operation decreased whereas hospital staff hours per patient hour increased. Physician cost per operation was unchanged. Patient, staff and referring physician satisfaction was high. Several areas for improvement have been found. Monitoring and general feedback of total quality factors has shown itself a powerful tool to detect and follow large and subtle changes in the practice of cardio-thoracic surgery. Most followed factors show improvement in spite of an increase in mean and median age. Several areas may be defined where further development might decrease the trauma to the patient. Aiming at a total quality and patient safety system, monitoring is an essential prerequisite.
Key Words: Monitoring; Cardiac surgery; Time trends; Mortality; Morbidity; Costs
In modern delivery of medical services, emphasis is on quality, capacity and costs. These factors are well known within a cardio-thoracic department on a macro level (i.e. crude mortality, waiting lists and budget). However, in order to obtain a deeper insight into the detailed components of these factors and their development, continuous monitoring of various parameters has to be performed and analysed.
The Cardio-Thoracic Surgical department of University Hospital, Ume 1.1. General organisation The Department serves the northern half of Sweden's area (Population around 895,000). Eighty percent live along the Baltic coast. The distance from the southernmost to the northernmost point of the capture area is 980 km. There are 13 hospitals, with one University Hospital. Children with congenital heart disease and transplantation candidates are sent to other centres. The Department is a stand alone unit with operating facilities, intensive care unit and surgical ward with dedicated staff, anaesthetists and surgeons. The accounting system follows the organisation.1.2. Activities The Speciality in Sweden is Cardio-thoracic. Around 1200 operations are performed annually. The operations may be divided into pure coronary artery bypass grafting (CABG), other cardiac and thoracic (63, 28 and 9%, respectively and unchanged over the years). The normal operative schedule is six operations per working day.1.3. Staff 1.3.1. SurgeonsThe number of staff surgeons (i.e. specialists) has varied between six and eight with a performance of around 120 operations per surgeon and year. An additional variable number [4–6] of junior doctors in training has also been present. The minimal number of doctors that have to be present in order to operate the daily schedule is nine. The operations follow generally accepted principles. The use of off-pump surgery does not exceed 5%. 1.4. Anaesthetists The anaesthetists work only within the Cardio-thoracic Department. Apart from anaesthesia in the operating room (OR), they are responsible for long term patients in the intensive care unit (ICU) and take a large part in the on-call schedule. The number of anaesthetists has been around ten, (120 operations per anaesthetist and year).
1.4.1. On call schedule 1.5. Facilities The operating suite consists of four fully equipped operating theatres, a small laboratory, a corridor and staff rooms for a total area of 300 m2. Mean time from arrival of the patient to beginning of surgery is 69 min. Change-over time (surgery to surgery) between cases is around 2 h. The staff consists predominantly of nurses/equivalent, who are dedicated for cardio-thoracic surgery only.The ICU has ten beds in two rooms for four patients and two isolation rooms. There is a small laboratory and storage and staff facilities for a total of around 600 m2. The staff consists of nurses in a primary nursing system. The cardio-thoracic ward has gradually reduced the number of beds, to 32 at present. There are six rooms for four patients, one room for three and five rooms for one patient. There is also a dining room, a gymnasium for pre- and postoperative physiotherapy, treatment rooms, staff facilities and nurses’ headquarters for a total of around 1550 m2. The administration consists of one nurse developer, one data manager, one staff manager, one clerk and one chief secretary. 1.6. Registries Data capture is performed by all staff in their respective area of competence. Several registries report formally at weekly, monthly, bi-annual or annual intervals.
1.6.1. Material
1.6.2. Methods In the deviation registry [1], everything that is not perceived as normal is registered. The registry contains three levels, the first being mainly organ systems, the second giving detailed information (see Table 3). The term ‘number of deviations’ used in Table 5 refers only to level one deviations, i.e. mainly organ systems. Financial data from the hospital's accounting system are converted to money value of 2002. Costs do not include X-ray, laboratory, physiotherapy and administration, nor the service given to the department from the hospital. To arrive at a ‘true’ costs level, the figure given should be increased by around 21%.
Through the initiative from the hospital, surveys of staff patient and referring physician satisfaction have been conducted.
1.6.3. Statistics
Demography, risk scores and mortality, Table 1. Mean age has increased by one and a half year, median age by 1 year. However, the number of patients aged 80 and above has doubled. The rate of second or more operations has decreased as a result of an aggressive percutaneous coronary intervention (PCI) attitude in these situations. Hospital mortality has been slightly over 2%, 30 day mortality is around 2.5%. One year mortality is 6–7% with no tendency for change. The large group with isolated CABG deserves special attention. A total of 6859 operations were performed with a hospital mortality of 1.27%. Subdividing the large CABG group into first elective, first emergent, repeat elective and repeat emergent, 30-day mortality has decreased.
In the Perfusion Incident Report registry all that is not normal before, during and after a perfusion is entered (Table 2) [2]. Incidents decreased during the last 6 years. Of the subsets, oxygenator incidents still occur at a rate of 1%. Oxygenator has been exchanged in around two patients in thousand, nowadays no longer during actual perfusion. Mechanical incidents have increased. (The heater-coolers used approached the end of their life span). Surgical incidents occur in less than 1% with no change over time. In six patients, a severe perfusion incident coincided with adverse patient outcome (death or stroke), giving an incidence of half a promille.
2.1. Morbidity (Table 3) Many deviations have decreased. Reoperations for bleeding decreased from 4.4 to 2.6% and for other causes, from 3.0 to 1.9%. Total rate of infections decreased from 8.7 to 5.8%. Other deviations have been reduced from 15.8 to 10.4%. The rate of a few deviations has increased; confusion (from 5.0 to 8.1%), arhythmia (from 28.3 to 29.9%) and the need for face mask ventilation (from 2.4 to 4.0%). Neurology (apart from confusion) has shown a gratifying decrease of both permanent and temporary deviations. The rate of deviations from the heart and circulation has remained unchanged. Within the group, arhythmias have increased, whereas inotrope usage decreased. Myocardial infarction is included in the registry according to a pre-set definition. However, the criteria for diagnosis of a myocardial infarction have changed during the time period, precluding comparable data. Deviations from the operative field have decreased from 18.6 to 15.1%. Postoperative bleeding with coagulation disturbances has increased during the last 3 years, possibly as a result of the use of new antiplatelet drugs in unstable angina. The use of tracheostomy has increased as well as use of face mask ventilation. On the other hand, total deviations from the respiratory organs have decreased as well as the rate of patients requiring respirator treatment over 16 h. The infection rate has decreased from 8.7 to 5.8%. Sepsis rate remains unchanged at one percent. Infections of the chest wound also seem unchanged. The main change has occurred in infections of the legs, lungs, urinary tract and other locations. Abdominal deviations decreased. However, the main part of the decrease is within the category of malaise, which of course is not necessarily due to abdominal organs. The rate of more severe abdominal organ complications seems to be unchanged below 1%. The rate of patients with unusual pain has decreased, whereas deviations from the peripheral circulation remain at slightly over one percent. Other deviations have decreased. Patients without any deviations, i.e. ‘normal’ patients, have increased from 37.2 to 42.5%. Some administrative and economic data (Table 4). Total length of stay in surgical department decreased by 2.9 days. Length of postoperative stay on the ward decreased by 2.5 days. Department staff hours (Physicians excluded) decreased by 12 h per operation.
Skin-to-skin time decreased by 0, 68 h. Hours per operation from operation suite staff (physicians excluded) have diminished by 1, 1 h. Hours worked per anaesthesia hour have increased from 8.5 to 9.2 and hours worked per hour of surgery have increased from 13.4 to 14.4 h (including on call schedule, administration, cleaning, etc.). ICU time is unchanged at a mean of 40 h with a median of around 21 h. Time on ventilator is unchanged at a mean of around 22 h, whereas the median time has markedly diminished by 4.2 h. ICU staff working hours per operation decreased to 58.3 whereas the ICU working hours per patient stay hour decreased from 1.59 to 1.39. The rate of patients leaving the ICU within 24 h increased from 77 to 79.5%. Patients with prolonged ICU stay have increased. Ward staff hours per operation decreased by 6.9 h whereas ward staff hours per patient day on ward increased by around 2 h. Cost per operation corrected for inflation (money value of 2002) has decreased by 11.9 tkr or 11%. Mean cost per operation for department staff has increased slightly whereas mean cost per work hour for department staff has increased by around 10% Table 5 shows the dependence of mean number of deviations and ICU hours on age and also that the deviations have decreased over time in each age category.
3.1. Limitations of the study The data are derived from routine registries. In spite of the known shortcomings of routine registries, these are the best data available. A distinction may be made between soft and hard data. In our experience, the harder the data, the more reliable they are. For instance rate of re-operations for bleeding are hard data with little cause for ambiguity. They may furthermore be checked against other registries. More soft data (such as unusual pain, or malaise) are open to personal interpretation in spite of the fact that definitions are available at the computer when entering data. This may possibly explain also the increase in some deviations through the mechanism of heightened awareness (in for instance confusion, where we have learned that some confused patients are difficult to diagnose. From experience, hard data are almost exact, soft data have to be interpreted cautiously. The task of a cardiothoracic unit is to produce cardiac and thoracic operations. It thus seems intuitively correct to relate the use of resources to this factor and to disregard all the minor, necessary and/or ancillary operations. In many of the above calculations, the denominator has been ‘operations’. In this entity we include all major operations, most often encompassing a sternotomy or thoracotomy. However, an operation is not a given entity. Thus, in order to justify the use of ‘operation’ as denominator in lieu of a much more thorough analysis based on individual data, the issue of case mix has to be addressed. By increasing the sample size, chance variations are minimised. This in turn increases the time span and makes the calculations more susceptible to case mix changes over time. In the case of an adult cardio-thoracic unit, proportions between the major groups operated (CABG, other cardiac, thoracic) should be unchanged over time. This has been the case in our unit. Another way of keeping track of the case mix is to monitor risk scores. With current risk scores, this is only applicable for cardiac cases where the mean Higgins score was unchanged. Thus we have some indirect indications that the entity ‘operation’ as the denominator in these broad calculations may be used -with caution. Some formal projects have been conducted to improve medical and total quality, re-operation rate, infection rate [3], thrombophlebitis rate and decreased postoperative ventilation. An active research programme about post-operative confusions has been instigated [4]. Routines are regularly discussed and changes are implemented after discussion between surgeons, anaesthetists and nurses. Some pharmacological changes have been made. Apart from these formal projects, data feed-back to all staff has been a constructive mechanism by which everybody is made aware of total quality features. The rate of re-operations for bleeding and other causes may be taken as an example. In the early period, the total re-operative rate was 7.4%. We then instigated a project with the explicit aim of reducing this figure. We went through the technical details of both hemostasis and of the closure of the sternum and made a feed-back mechanism to each individual surgeon. The results have been gratifying, but we still have a total re-operative rate of 4.5%, 2.6% re-operation for bleeding, 1.9% for unstable sternum or sternal infection. Computerised data collection and resources utilisation studies have become more common in the last decade, most often taking coronary surgery as a model operation. Ratcliffe et al. [5] presented a Java-based intranet program, used by nurses who collected key clinical pathway data. A short feedback time could detect deviations from the clinical pathway norm already after ten patients. London et al. [6] described in 1997 two cohorts of adult cardiac operation, before and after the instigation of a fast track system. They found a decrease in most factors studied: Median time to extubation (19.2–10.2 h); median ICU stay (96–49 h; total postop length of stay (222–167 h). Mortality did not change, morbidity decreased. They also give a wealth of information about many medical and administrative data. Pinna Pintor et al. have written a valuable review on the evolution of case-mix in heart surgery [7]. The main finding of this analysis is that many total quality indicators have improved over the 9 years studied. The rates of mortality and many instances of morbidity have now reached a level where investigations into cause (such as randomised studies) would require huge materials. A few indicators have deteriorated and may be traced to a shift in case mix. The ‘ordinary’ patient during period III requires less resources, whereas the patient with deviations requires much more. This may largely be explained by the entry into the practice of octogenarians [8]. Thus we have a growing group of patients that have no or few deviations and a much smaller but growing group of patients with multiple deviations which take longer time to resolve. These facts have had no consequencies for the mean Higgins score, which remained unchanged during the three periods. The known change in case mix that has occurred is the increased rate of elderly people. A calculation into how much this change amounts to reveals that the mean Higgins score would have increased with as little as 0.15 score points. Thus the Higgins score, calculated as the mean for the department is not very sensitive to changes in age. One finding is the decrease in operating time. This is probably due to the fact that in period I, we had quite a high load of education. In period III all responsible surgeons are well into their speciality, there is a lower load of education and the individual surgeon has developed his personal way of operating efficiently. Further development in this area is unlikely unless the time frame of the operations themselves change. The finding of higher cost per nurses working hour in Table 4 is due to the fact that salaries for nurses have increased above the inflation rate. The cost for physicians is also subject to extra-medical factors. In 1996 there were more doctors in junior positions. In 2002, our physicians have gained the necessary time and experience to obtain senior, more well paid positions. Thus the rationalisation and total quality improvement is much greater than what is expressed in the cost reduction of 11%. It is not easy to capture the ‘true’ rationalisation but it may be as large as 20%. 3.2. Length of stay has decreased There are several severe flaws with this factor. We believe that we have now reached a low level of postoperative time within the department and that it will not be possible to lower it further with totally reduced costs for society, especially from the background of long travelling distances within our area. The department may lower its cost by discharging patients earlier but the costs of healing will then have to be picked up by somebody else.The operating room is where the ground is laid for many complications. This is where the necessary surgical trauma is created. However, it is in the ICU that the consequencies of most of the major complications are taken care of. The ICU uses about 23% of the resources (physician cost excluded). Thus, avoidance of all complications and deviations from the normal postoperative course will result in a lighter burden on the ICU (and ward) with improved total quality. This work has to start pre-operatively and in the operating room. There are still many general avenues of quality improvement that remain incompletely explored. Should we introduce a ‘black box’, i.e. an on line registry of events also for the surgical procedure as is done for anesthesia? Should we learn more from specialities outside our own, such as geriatrics, psychiatry and neurology? What general conclusions are we to draw from the research of Marc De Leval [9] who found that also minor incidents in the operating room exerted an influence on final outcome? How far can cardio-thoracic surgery reach in the quest for the distant vision of no complication surgery? This remains a great challenge for every surgeon within cardio-thoracic surgery. Monitoring of an activity gives the unit several advantages. Monitoring may serve as baseline for improvements. Furthermore, it is essential that the unit and its leader is the one most knowledgeable about the activities. Forewarned is forearmed. Changes may be easily followed. Projects for change are easy to justify and follow. Informed feedback may be given to staff, patients, hospital and society. However, it is also a potent tool for scientific development. It is easy in a scientific project to temporarily use the monitoring infrastructure in a true prospective way. Is the monitoring used cost effective? The cost has deliberately been kept low by making monitoring a semi-automatised feature of everyday work. Costs for monitoring involve investments in hardware, programming and maintenance of the registries. There is a need for somebody within the department to take full responsibility for them. Furthermore, data capture may take some more time than if that function was not demanded from the staff. We believe that the positive development of quality indices presented is largely the result of monitoring. Such factors as improved general education level by all staff, improved routines taken from the scientific development and specific projects have of course also played a role. However, access to data facilitates the constant quality work necessary in a developing organisation. We believe that this is the main value of monitoring. As such it is invaluable both for the individual patient, for the department and for the economic development.
ICVTS on-line discussion Author: Dr. Sven Beholz, Charité, Department for Cardiovascular Surgery, Schumannstr. 20/21, Berlin 10117, Germany Date: 30-Sep-2003 Message: The authors describe a well documented and very useful tool to gain information on the key data of an established cardiothoracic surgery unit with partly excellent results and remarkable improvements over time. Nevertheless the term "total quality" used several times is misleading. Total quality management as for example the excellence approach of the European Foundation for Quality Management (EFQM) includes lots of aspects of quality management as customer (e.g. patients, employees, cooperating physicians) satisfaction, responsibility of the management, coordination of resources, risk management and last but not least planning and documentation of all necessary activities for improvement. The database as a tool for monitoring data regarding the key processes is extremely helpful but must be accompanied by a lot of well documented activities to evaluate the real effect of a quality management system. Response Author: Prof. Torkel Aberg, Uema University Hospital, Secretary General, The European Association for Cardio-thoracic Surgery, Uema, S-90185, Sweden Date: 07-Oct-2003 Message: Thank you for your valuable comment. It is quite true that "Total Quality" encompasses all the mentioned features. Most if not all are implemented and used in our department. However, a presentation of all the results of various measurements such as patient satisfaction, staff satisfaction and referring physician satisfaction turned out to be cumbersome and was judged to be outside the scope of the present presentation. However, your point is very well taken. Monitoring is not enough but it is a valuable and essential tool to be able to act according to total quality principle. Author: Dr. Thomas Ferguson, Professor of Surgery, Emeritus, Washington University School of Medicine, Department of Surgery, Suite 3108 Queent Tower, Barnes-Jewish Hospital, St Louis, MO 63110, USA Date: 21-Oct-2003 Message: This study is a most valuable one, in that it represents an attempt to dissect the reasons for changes in patient outcomes beyond the traditional ones. Historically, surgeons have relied on mortality, then later morbidity (as reflected by length of stay) as outcome markers. In this paper, attention is given to the global environment of the surgical episode, certainly the next step in outcomes analysis. It would be most beneficial if large numbers of surgical units collected data of this type, and the findings pooled. We then could talk about outcomes in surgery in a meaningful way.
The authors would like to thank all staff of the Department for giving patients such an interested and excellent care but also for being such faithful and dependable collaborators in the necessary data collection. You are true professionals! doi:10.1016/S1569-9293(03)00218-4
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |
Top
Abstract
1. Introduction
, Sweden, started in 1988. The emphasis on quality, capacity and costs was consciously built into the organisation, which allowed the creation of various data capture systems 
berg T, Svenmarker S, Hohner P, Hentschel J. Routine registration of deviations from the norm in cardiac surgery: a potent clinical research tool and quality assurance measure. Eur J Cardiothorac Surg. 1997;11:10–16