تجزیه و تحلیل هزینه مستقیم واحد مراقبت های ویژه اقامت در چهار کشور اروپایی: استفاده از متدولوژی هزینه یابی استاندارد

Objectives The objective of the present study was to measure and compare the direct costs of intensive care unit (ICU) days at seven ICU departments in Germany, Italy, the Netherlands, and the United Kingdom by means of a standardized costing methodology. Methods A retrospective cost analysis of ICU patients was performed from the hospital's perspective. The standardized costing methodology was developed on the basis of the availability of data at the seven ICU departments. It entailed the application of the bottom–up approach for “hotel and nutrition” and the top–down approach for “diagnostics,” “consumables,” and “labor.” Results Direct costs per ICU day ranged from €1168 to €2025. Even though the distribution of costs varied by cost component, labor was the most important cost driver at all departments. The costs for “labor” amounted to €1629 at department G but were fairly similar at the other departments (€711 ± 115). Conclusions Direct costs of ICU days vary widely between the seven departments. Our standardized costing methodology could serve as a valuable instrument to compare actual cost differences, such as those resulting from differences in patient case-mix.

Although intensive care unit (ICU) beds comprise less than 10% of hospital beds, ICU departments consume 22% of total hospital costs in the United States [1]. Also, the costs of ICU departments in the Netherlands have been estimated to represent approximately 20% of the total hospital budget, with the costs per day between three- and fivefold greater in ICU departments than in general wards [2] and [3]. Therefore, several studies have assessed the costs of ICU services. Cost estimations of ICU stay vary extensively. From a multicenter German study, Moerer et al. [4] reported the total costs per ICU day to be €855 (inflated to 2008). At the other extreme, the total costs per day at ICU departments in the United States were found to be €3221 (inflated to 2008) [5]. A number of studies have tried to explain actual cost differences between ICU departments [2,6,7]. The patient case-mix is considered to have an important effect on the actual costs of ICU days. Other potential factors influencing actual cost differences include variations in study setting (e.g., bed occupancy rate, density of acute care beds, and staff composition), variations in medical practice (emergency retrievals, referral pattern, and use of mechanical ventilation), the availability of health-care resources (e.g., the presence of a High Dependency Unit), the hospital payment system (e.g., public/private-mix and insurance payment), and relative and absolute prices between countries [2] and [6]. It has been argued, however, that some of the observed cost differences are as a result of the methodologies used to estimate the costs rather than being as a result of actual differences [7] and [8]. The application of a standardized costing methodology enables a meaningful comparison of actual cost differences between health-care services [9] and [10]. This way cost differences can be attributed to the health-care services under consideration, rather than to differences in the costing methodology [11] and [12]. Yet, standardized costing methodologies are often restricted by the availability and quality of data. Resource quantities for individual patients are generally not available with the same level of precision, even within a single health-care provider's clinical costing system and systems vary markedly between health-care providers [13]. Several studies have made recommendations on the application of standardized costing methodologies and potential bias for the comparability between health-care services at different health-care providers [13], [14] and [15]. For example, Ritzwoller et al. [15] attempted to identify comparable measures from the hospital information systems of seven health-care providers in the United States to compare the health-care utilization of smokers, former smokers, and never smokers. They found a substantial variation in both the content and the capture of data across all health-care providers and across all cost components. Recommendations on the application of standardized costing methodologies have also extensively been made in the field of ICU stay [2], [6] and [16]. A systematic literature review by Elliot [7] demonstrated that the costing methodologies employed to calculate costs of ICU stay are diverse and make comparative analyses between studies difficult. In their narrative review, Pines et al. [8] have argued that despite considerable progress in costing methodologies, critical care studies have not adequately implemented these techniques. The objective of the present study was to measure and compare the direct costs of ICU days at seven ICU departments in Germany, Italy, the Netherlands, and the United Kingdom by means of a standardized costing methodology. Only one study in the field of ICU stay has earlier applied a standardized costing methodology across jurisdictions. Negrini et al. [17] developed a standardized costing methodology and tested its feasibility at ICU departments in France, Germany, Hungary, and the United Kingdom. A recognized limitation of their study was that estimates of costs were permitted for some cost components, because cost data were not always available at the ICU departments under consideration. To overcome this limitation, the present study used a standardized costing methodology that was based on the availability of data at the ICU departments under consideration.

Economic evaluations can provide health-care decision makers with valuable information on the relative efficiency of alternative health-care services, health-care services at different health-care providers, and health-care services across jurisdictions. Because of the wide range of costing methodologies applied, however, cost assessments from different economic evaluations are often not readily comparable or cannot be adjusted to a different context [24] and [25]. Therefore, the present studies aimed to measure and compare the costs of ICU days by means of a standardized costing methodology. Direct costs per ICU day were €1383 ± 398 but varied between €1168 (department B) and €2025 (department G). Labor was the key cost driver and entirely explained the increased costs at department G (€1629 compared with an average €711 at the other departments). Labor costs have previously been demonstrated to be higher in the United Kingdom compared with other European countries [17] and [26]. Negrini et al. [17] found the proportion of labor to be 67% of the direct costs per day (compared with 61% in our study), with hospitals in the United Kingdom being far more costly than the hospitals in France, Germany, and Hungary. The standardized costing methodology was based on the availability of data at the seven ICU departments. We believe, however, that our sample of ICU departments is sufficiently representative to assume that the established methodology is consistent as well as generalizable to other settings. Although the bottom–up approach is generally believed to be the preferred methodology [20] and [21], it was not applicable to “diagnostics,” “consumables,” and “labor.” For “diagnostics,” patient-level data were not structurally available. For “consumables” and “labor,” patient-level data were subject to differences in coding systems. For example, drugs were registered by drug category at some departments and by drug name in others. Also, the same item, with the same name, may have represented an entirely different concept at different departments. With respect to “labor,” wide variability existed in terms of training of both ICU specialists and ICU nurses. For example, ICU specialists at departments B and C generally have more responsibilities compared with ICU specialists at the other departments because there was a relative shortage of ICU nurses. At departments D, E, and F, ICU specialists included ICU residents. Therefore, the standardized costing methodology entailed the application of the bottom–up approach for “hotel and nutrition” and the top–down approach for “diagnostics,” “consumables,” and “labor.” If researchers studying other settings have access to data at the patient level (bottom up), they must consider whether the benefit of a more accurate cost estimate justifies restricted comparability/generalizability incurred in using detailed information. Earlier studies have proven that the top–down approach results in fairly accurate cost estimates [20], [21] and [22]. To compare the cost estimates resulting from the bottom–up and top–down approaches, we additionally determined the costs for “medical imaging services” by using the bottom–up approach for patients admitted between January 2006 and April 2006 at departments B and C. This exercise did not result in significant cost differences for medical imaging services (P = 0.265). Likewise, the costs for the 25 most expensive drugs and disposables in terms of total expense were additionally determined by using the bottom–up approach at department A, but no significantly different results were observed (Pmedications = 0.373 and Pdisposables = 0.285). The costs for “labor” were additionally determined by using the bottom–up approach at department D (P = 0.462). Thus, there is reason to believe that the application of our standardized costing methodology resulted in robust cost estimates. The use of a standardized costing methodology is required to be able to explain actual cost differences between ICU departments in a straightforward way [7], [8] and [17]. The patient samples of the recruited ICU departments showed some actual differences at baseline that reflect the daily clinical practice. Some of these actual differences may explain the actual cost differences we observed, for example, the patient case-mix, variations in study setting (e.g., density of acute care beds), and variations in medical practice (e.g., use of mechanical ventilation [MV]). ICU departments in Germany are used for postsurgery care. This was reflected by our German department A, which included a higher proportion of patients with gastrointestinal diseases (31% compared with 12% on average in the other countries). In addition, the density of acute care beds per 1000 inhabitants is relatively high in Germany (6.4 compared with an average 2.9 in the other countries [23]). Similarly, it is likely that patients admitted to the ICU in the United Kingdom are more severely ill and in need of more intensive care (density of acute care beds: 2.3 [23]), which was reflected by the higher Simplified Acute Physiology Score II at our British department G (42 ± 20). As a result, actual differences might partly explain the lower costs at German department A and the higher costs at British department G. First ICU days have previously been demonstrated to be much more expensive than subsequent ICU days [27]. Our study, however, did not distinguish between first and subsequent ICU days, which may have falsely elevated cost variation between departments. The ICU length of stay varied widely between the seven departments (range: from 3.8 at department F to 7.8 at department A). Regression analyses performed at departments A, B, and D, however, suggest that the ICU length of stay was unable to predict daily ICU costs (R2 = 0.004). Furthermore, several recent studies suggest that MV is associated with increased costs in the ICU [16] and [28]. Approximately 60% of identified patients were mechanically ventilated at some point during their ICU stay, but this share varied between 33% at department C and 82% at department B. The actual difference in MV rates, however, was not virtually reflected by actual cost differences. In Germany, patients requiring MV were 20% more expensive than patients not requiring MV. These cost increases amounted to 44% in Italy, 34% in the Netherlands, and 39% in the United Kingdom. The country of treatment was clearly the most important factor to explain the actual cost differences observed in this study. To draw robust conclusions about actual cost differences between Germany, Italy, the Netherlands, and the United Kingdom, however, a larger sample of ICU departments per country is needed. Even though we attempted to select ICU departments that were representative of the overall setting and treatment patterns in the respective countries, recruitment was restricted by both hospitals' willingness to participate and time constraints. Random sampling could also enhance the reliability of our results. Still, cross-country cost comparisons in the field of ICU stay are scarce and thus we believe that our study provides valuable insight into the relative costs in different European countries. Our results may be viewed as preliminary data to support the funding of future studies to determine the generalizability of our results to other ICU departments within and beyond our sample of countries. Our standardized costing methodology could serve as a valuable instrument to determine the generalizability of our results.