Eur J Plast Surg (1996) 19:14-17

European ] r ~ l

_Q

x oa'o'l f l l l S t l C

bur-gery

© Springer-Verlag 1996 q

Quality assurance applied to air-fluidized beds Our experience - part I P. Gilardino, M. Signorini, G. Pajardi, L. Donati Institute of Plastic Surgery,Milan UniversityMedical School, Milano, Italy

Abstract. The use of air-fluidized beds (AFBs) was introduced into our Burn Center in 1988. These devices have a significant cost, which is acceptable only if they permit objective clinical advantages. Five years later it was decided to carry out a critical retrospective analysis of the cases treated during the last decade, comparing patients nursed on air-fluidized beds with those nursed on conventional beds, to determine the real utility and effectiveness of such devices. Some parameters of each patient have been analyzed such as mortality, duration of hospitalization, number of surgical operations, and an index of risk (Roi index). The data shows there has been a significant benefit in using AFBs in treating severely burned patients, represented by a significant reduction of the percentage of mortality (39% vs. 27%), whereas no correlation was found in the length of hospitalization and number of operations. Even though the AFBs result in an increased cost for the Burn Center, data shows that this device remains cost-effective. The authors provide some guidance for their correct utilization. Key words: Air-fluidized beds - Quality assurance Cost-effectiveness - Burns

The history of the evaluation of medical care (quality assurance) began during the second half of the last century. Over the years many scientific societies have been founded at both international (ISQA: International Society for Quality Assurance in Health Care), and national level (AECCA in Spain, VRQ in Italy, etc.), which have the specific goal of exploring knowledge of quality evaluation, sharing experiences and, especially, promoting the more widespread use of these methodologies. Quality assurance (QA) [8, 15, 26, 27, 41] is not simply based on the limited results which can be obtained by controlling procedures, but on a dynamic proCorrespondence to: E Gilardino, Via Pianella, 4, 1-20148 Milano,

Italy

cess of performance control aimed at modifying the behaviour of health operators. The goal is to optimize resuits and the cost-benefit ratio, without reducing the effectiveness [18] and acceptability of the delivered medical services. One of the aspects that is commonly evaluated in QA is the cost-benefit ratio, and therefore the relationship between costs and quality. QA itself can be divided into two major components: the quality assessment which has been discussed so far and technology assessment (TA). TA can be defined as the process of planning and conducting studies aimed at evaluating and judging medical technology [1, 7, 10, 11, 14, 16, 22, 30]. The goal is to establish a scale of values for each technology in terms of its benefits and contraindications for different patient categories, which would define the limits and appropriateness of its use. At the Niguarda Burns Unit, a TA study of air-fluidized beds (AFBs) has been carried out [2, 13, 17, 19, 21, 24, 32, 34, 38]. They were first introduced in the practice in 1988. These devices were developed expressed for burn treatment by Hargest and Artz in 1968, and are now very well known throughout the world. From 1988 to the end of 1992, AFBs were used in 83 patients for a total of 2524 days. Initially, only very severe patients were treated [12, 29] but after the initial encouraging experience the indications were extended to intermediate patients (BSA>30% or mainly posterior lesion localization; in any case, the mean risk index of AFB patients has always been higher than the mean risk of patients treated in the previous five years: 47% vs. 29%). Because the cost of hiring these beds weighs significantly on the unit's budget [3], a specific cost-effectiveness analysis [4, 6, 9, 37] was undertaken in order to: 1. verify the true usefulness of AFBs in the treatment of severely burned patients, 2. establish precise indications for rationalizing their use as much as possible.

15 M a t e r i a l and m e t h o d s

Table 1. Percentage mortality by BSA category

This study retrospectively considered all patients over the age of ten treated at the Burns Unit between 1 January 1983 and 31 December 1992; younger patients were excluded because they are often admitted in spite of relatively small lesions. The enrolled patients were divided into two groups: the 1983-1987 patients treated on conventional beds, and the 1988-1992 patients treated in part on AFBs and in part on conventional beds. A retrospective analysis [5, 33] was preferred because it offered the immediate availability of a significant number of patients and, more importantly, because it was the only format possible for the study; a prospective study seemed to be unethical [23, 39] as the non-treated group would be denied what was considered to be the objective benefits of AFBs. As in all retrospective studies, the main defect of this study was the lack of comparable randomized groups: the patients were not selected for specific features and, even more importantly, the treatment did not differ only in terms of the tested technology. Nevertheless, the study was considered to be acceptable and significant because no other relevant therapeutic changes have been introduced over the past ten years. The use of AFBs is thus considered to be the main, if not the only, therapeutic variable distinguishing the two considered fiveyear intervals: 1983-1987 and 1988-1992. The comparability of the two groups was verified in terms of age, sex, BSA and cause of burn. The data selected as AFB efficacy indicators was collected: length of hospitalization, number of surgical procedures, and mortality. These parameters were selected as significantly correlated factors in terms of the rate of wound healing and the incidence of local or systemic infection [28, 35, 36, 40]. A second investigation was then undertaken on a systematic sample of two-thirds of the patients (i.e.: two out of every three patients included in this study), considering all of the patients in the first five-year group (1983-1987), and only those treated using a AFB in the second five year group (1988-1992). In addition to the previously examined parameters, the percentage of full-thickness burn, concomitant pathologies, and the number of repeated operations on the same anatomic site because of graft failure were considered. This enabled a mortality risk index to be calculated for each patient (Roi index) [20, 31], and to compare the results with those in the international literature. The Z2-test and Student's t-test were used to verify statistically significant differences.

BSA

1983/87

1988/92

Significance

0-20% 2040% 40-60% 60-80% 80-100% 0-40% 40-100%

12.5% 32% 53% 88% 100% 26% 69%

6.9% 17% 18% 62% 92% 14% 43%

n.s. P<0.053 P<0.0051 n.s. n.s. P<0.023 P<0.01

Results

c a l l y significant difference in m o r t a l i t y b e t w e e n the two groups w h e n c o m p a r i n g those patients with 0 - 4 0 % B S A and those with 4 0 - 1 0 0 % . However, if the studied p o p u l a t i o n is b r o k e n up into five s u b g r o u p s o f i n c r e a s i n g B S A , statistically significant differences are p r e s e n t o n l y in 2 0 - 4 0 % and 4 0 - 6 0 % groups. D u r i n g this p e r i o d o f observation, the L D 5 0 i n c r e a s e d f r o m 45 to 70% (group 1 vs. group 2). In the s e c o n d part o f the study, a total o f 196 patients were e x a m i n e d : 137 o f the first f i v e - y e a r group (all nursed on c o n v e n t i o n a l beds), and 59 o f the s e c o n d group (all o f w h o m h a d b e e n treated using an A F B ) . In this part o f the study, the patients in group 2 cannot be c o n s i d e r e d c o m p a r a b l e with those in group 1 in terms o f severity. In Table 2, the differences b e t w e e n group 1 and group 2 are reported. The d e c r e a s e in m o r t a l i t y was even greater than in the first part o f the study, and was p a r t i c u l a r l y significant for the 2 0 - 8 0 % Roi i n d e x risk group (Table 3). Similarly, the L D 5 0 c a l c u l a t e d on B S A - r e l a t e d m o r t a l i t y i n c r e a s e d f r o m 45% (group 1) to 75% (group 2).

D u r i n g the s t u d i e d t e n - y e a r period, a total o f 396 patients w e r e a d m i t t e d to the Burns Unit: 214 in the five y e a r s and 182 in the second. O f the latter, 83 were treated using A F B s . The two groups w e r e c o m p a r a b l e in terms o f average age (38 vs. 40 y e a r s ) and B S A (36.5% vs. 35.9%); the cause and related area distribution o f the burns were similar. T h e average l e n g t h o f h o s p i t a l stay i n c r e a s e d f r o m 40 d a y s in the first group to 52 d a y s in the second; the average n u m b e r o f surgical p r o c e d u r e s u n d e r g o n e b y each patient also r o s e f r o m 0.9 in the first to 1.3 in the s e c o n d group. M o r e i m p o r t a n t and significant w e r e the m o r t a l i t y data o f the two groups during the course o f c o m p a r a b l e B S A burns: there was a drop f r o m 39% to 23% b e t w e e n the first and s e c o n d study periods. In Table 1, the results o f the two groups have b e e n c o m p a r e d using two different criteria. T h e r e is a statisti-

Table 2. Statistical data concerning sampled population

No. of patients Average age Average B SA Average R.I Average R.I survivors Average FTB Associated pathologies Average hospital stay No. of interventions Mortality

1983/87

1988/92

137 39.4 36% 29% 12% 9.6% 25% 48 gg. 1.4 39%

59 47.3 41% 47% 34% 20% 45% 60 gg. 2 27%

Significance

P<0.0007 P<0.005 P<0.00004 P<0.002 P<0.03 P<0.04

Table 3. Correlation between mortality and the Roi index R.I

1983/87

1988/92

0-20% 20 40% 40-60% 60-80% 80-100% 0-20% 20-80% 80-100%

14% 55% 50% 80% 94% 14% 63% 94%

6% 8% 11% 60% 68% 6% 16% 68%

Significance

n.s.

P<0.0005 n.s.

16 Finally, the incidence of repeated surgical procedures due to previous graft failure decreased from 22% (group 1) to 16% in the group treated using the AFBs.

Discussion In the cost-effectiveness evaluation of a given technology, its costs must be compared with the benefits it offers in terms of patient care. The daily rental and maintenance costs of AFBs are Lit. 150,000. The beds were used for an average of 30 days per patient, over an average total hospitalization of 60 days. The increase in cost they generated was about 10%. There was no evidence of a reduction in the period of hospitalization of the patients in the second five-year group, especially those treated using AFBs, nor was there a reduction in the number of surgical procedures per patient to compensate for the increase in costs. In order to explain the length of stay (LOS) of group 2, a multivariate model was developed with LOS as a dependent variable. This showed that the number of surgical procedures was the most closely associated factor, while mortality was obviously associated with a shorter LOS. An unexpected result was that the use of an AFB seems to increase LOS, although, since the RI and the number of operations seem to be significantly higher in patients using AFBs, this probably means that their condition was more severe even if the RI does not fully explain the difference in severity between the two patients groups [25]. It could possibly be due to the fact that the burned patients nursed on AFBs were clinically more severe than the average control patients in the previous five-year group, in terms of the size and depth of the lesion, in addition to their age and general conditions (pre-existing or associated pathologies). Only a prospective study would clearly answer the question. Nevertheless, the clinical advantage of AFBs appears unquestionable to us. The increase in LD50 (the average burn size related to a 50% chance of survival) from 45% to 75% BSA, although partly due to other therapeutic and organizational innovations was mainly due to the use of AFBs. As can be seen in Table 3, the mortality levels of group 1 were comparable with the expected risk calculated for each patient (only in the 20-40% risk group was the average recorded mortality higher, 55%); but in group 2, the patients nursed on an AFB had a mortality rate which decreased to less than the minimum predicted by the Roi index. This reduction is statistically significant for the 20-80% risk group and is also present, but without statistical significance in the low (<20%) and very high (>80%) risk group. In conclusion, we believe that the use of AFBs has fulfilled the criteria of cost-effectiveness: a technology is considered superior to another in terms of cost-effectiveness if it is more effective and more expensive, providing the benefit is worth the extra cost. In our group, the daily cost of treatment per patient increased by about 10% with AFB technology; however, if the total cost of treatment during the second five-year period is consid-

ered, the beds account for less than one-twentieth of the economic burden and lead to a significantly higher chance of survival (39% vs. 27%). The study also establishes the limits beyond which no further benefit can be obtained, only an additional cost. Having seen the results, the indications for use have been objectively determined, restricting its use to patients with a Roi index between 20 and 80%, or with significant posterior burns. It is believed that this will allow a more rational use of AFBs, reducing costs as much as possible while still obtaining the greatest possible benefit.

References 1. Allan JD, Hall BA (1988) Challenging the focus on technology: a critique of the medical model in a changing health care system. Adv Nuts Sci 10:22-34 2. Antczak-Bouckoms A, Burdick E, Klawansky S, Mosteller F (1991) Introduction. Using medical registries and data sets for technology assessment. Int J Technol Assess Health Care 7: 123-128 3. Berenson RA (1984) Intensive care units: clinical outcomes, cost, and decisionmaking (Health Technology Case Study 28). Office of Technology Assessment, Washington DC 4. Brook RH, Lohr KN (1985) Efficacy, effectiveness, variations and quality: boundary-crossing research. Med Care 23: 710-722 5. DerSimonian R, Charette LJ, Mcpeek Bet al. (1882) Reporting on methods in clinical trials. N Engl J Med 306: 1332-1337 6. Detsky AS, Naglie G (1990) A clinician's guide to cost-effectiveness analysis. Ann Intern Med 113:147-154 7. Donabedian A (1988) The assessment of technology and quality: comparative study of certainties and ambiguities. Int J Technol Assess Health Care 4:487-496 8. Donabedian A (1990) La qualit5 dell'assistenza sanitaria. NIS Ed 9. Doubilet P, Weinstein MC, McNeil BJ (1986) Use and misuse of the term "cost effective" in medicine. N Engl J Med 314: 253-256 10. Eddy DM (1990) Should we change the rules for evaluating medical technologies? In: Gelijns AC (ed) Medical innovation at the crossroads: modern methods of clinical investigation. National Academy Press, Washington, DC, pp 117-134 11. Feeny D (1986) New health technologies: their effect on health and the cost of health care. In: Feeny D, Guyatt G, Tugwell P (eds) Health care technology: Effectiveness, efficiency, and public policy, Institute for research on public policy, Monreal Quebec, pp 5-24 12. Feller I, Claudella AJ (1987) The national burn information exchange. The use of a national burn registry to evaluate and address the burn problem. Surg Clin North Am 67:167-189 13. Freeman R, Ryan DW (1988) In: Ryan DW (ed) The medical fluidized bed. J Int Therapy Clin Monit 9:258-260 14. Fuchs VR, Garber AM (1990) The new technology assessment. N Engl J Med 323:673-677 15. Ghetti V (1989) La valutazione della qualitS: una nuova frontiera del sistema sanitario nazionale. Fondazione SmithKline Ed 16. Goodman C (1987) A practical guide to technology assessment. Business Health 4:14 19 17. Hargest TS, Artz CP (1969) A new concept in patient care: the airfluidized bed. AORN 10:50-53 18. Heithoff KA, Lohr KN (eds) (1990) Effectiveness and outcomes in health care. National Academy Press, Washington, DC

17 19. Laurence JC, Lilly HA (1983) Are fluidized bead beds a hazard in the management of burns? In: Proceedings of the British Burn Association 16th Meeting. Newcastle upon Tyne 20. Lemeshow S, Teres D, Avrunin MS et al. (1987) A comparison of methods to predict mortality of intensive care unit patients. Crit Care Med 15:715-722 21. Lioret N, Lesage D, Grange B, Saisy R, Baux S (1982) Seven years experience in use of fluidized beds for treatment of burn patients. In: Proceedings 6th International Congress of Burns, San Francisco 22. Lohr KN, Rettig RA (eds) (1988) Quality of care and technology assessment. National Academy Press, Washington, DC 23. Marsden C (1992) An ethical assessment of intensive care. Int J Technol Assess Health Care 8:408-418 24. McNabb LJ, Hyatt J (1987) Effect of an air fluidized on insensible water loss. Crit Care Med 15:161-163 25. Munoz E, Josephson J, Tenenbaum N e t al. (1989) Diagnosis related groups, costs and outcomes for patients in the intensive care unit. Heart Lung 18:627-633 26. Perraro F (1985) Cure mediche e assistenza sanitaria: verifica e revisione della qualit5 (VRQ). Federazione Medica 38:749 27. Perraro F (1987) Qualit~ dell'assistenza. Promozione, valutazione, revisione. Federazione Medica 40:75 28. Philips LG, Heggers JR Robson MC et al. (1989) A comparative evaluation of infection rates of burn patients treated on an air-fluidized bed compared with standard therapy. Surg Forum 620-622 29. Rath T, Berger A (1982) Treatment of severe burn cases in the air-fluidized bed. Burns 9:115-117

30. Reerink E (1988) Qualit~ dell' assistenza sanitaria e valutazione della tecnologia: similitudini e differenze. Bollettino dell'ordine dei medici di Bologna 4:30 31. Roi LD, Flora JD, Davis TM, Wolfe RA (1983) Two new burn severity indices. J Trauma 23:1023-1029 32. Ryan DW (1983) The influence of environmental temperature (32 ° C) on catabolism using the clinitron fluidized bed. Intensive Care Med 9:279-281 33. Sacks H, Chalmers TC, Smith H (1982) Randomized versus historical controls for clinical trials. Am J Med 72:233-240 34. Scheulen JJ, Munster AM (1983) Clinitron air-fluidized support: an adjunct to burn care. J Burn Care Rehabil 4:271-275 35. Sharbaugh RJ, Hargest TS (1971) Bacteriocidal effects of the air-fluidized bed. Am Surg 37:583-586 36. Sharbaugh RJ, Hargest TS, Wright FA (1973) Further studies on the bacteriocidal effects of the air-fluidized bed. Am Surg 39:253-256 37. Siegel DM (1987) The high cost of medical technology; getting at the heart of the matter. Med Care 25:979-987 38. Smoot EC (1986) Clinitron bed therapy hazards. Plast Reconstr Surg 77:165 39. Sorenson JH (1990) Ethics and technology medicine: an introduction. Theoretical Med 11:81-85 40. Vesley D, Hankinson SE, Lauer JL (1986) Microbial survival and dissemination associated with an air-fluidized therapy unit. Am J Infect Control 14:35-40 41. Vuori H (1982) Quality assurance of health services: concepts and methodology. Public Health in Europe. WHO Regional Office for Europe, Copenhagen, No 16