Journal of Oncology Pharmacy Practice

Journal of Oncology Pharmacy Practice http://opp.sagepub.com/

Economic assessment of aseptic compounding rooms in hospital pharmacies in five European countries Bérengère Dekyndt, Bertrand Décaudin, Damien Lannoy and Pascal Odou J Oncol Pharm Pract published online 10 February 2014 DOI: 10.1177/1078155214520820 The online version of this article can be found at: http://opp.sagepub.com/content/early/2014/02/10/1078155214520820

Published by: http://www.sagepublications.com

Additional services and information for Journal of Oncology Pharmacy Practice can be found at: Email Alerts: http://opp.sagepub.com/cgi/alerts Subscriptions: http://opp.sagepub.com/subscriptions Reprints: http://www.sagepub.com/journalsReprints.nav Permissions: http://www.sagepub.com/journalsPermissions.nav

>> OnlineFirst Version of Record - Feb 10, 2014 What is This?

Downloaded from opp.sagepub.com at SCD Lille II - Secteur on May 21, 2014

XML Template (2014) [7.2.2014–12:18pm] //blrnas3/cenpro/ApplicationFiles/Journals/SAGE/3B2/OPPJ/Vol00000/140001/APPFile/SG-OPPJ140001.3d

(OPP)

[1–9] [INVALID Stage]

Journal of

Oncology Pharmacy Practice

Original Article

Economic assessment of aseptic compounding rooms in hospital pharmacies in five European countries

J Oncol Pharm Practice 0(0) 1–9 ! The Author(s) 2014 Reprints and permissions: sagepub.co.uk/journalsPermissions.nav DOI: 10.1177/1078155214520820 opp.sagepub.com

Be´renge`re Dekyndt1, Bertrand Dećaudin1,2, Damien Lannoy1,2 and Pascal Odou1,2

Abstract Purpose: The aims of the study are to make an inventory of fixtures of aseptic compounding structures, to compare, using real examples, the design and operating costs of controlled atmosphere area (CAA) with isolators and CAA with laminar flow biological safety cabinets (BSCs) in order to determine the most economical scheme in hospitals and to give a final facilities cost calculated for one workstation. Methods: Forty-three hospitals were interviewed (21 French and 22 from four European countries) over seven months. Hospital pharmacists completed a form with 390 items. Hospitals are compared according to their workstation type: BSCII or BSCIII (group B) and isolator (group I), using Mann and Whitney’s statistical test and Monte-Carlo modeling. Results: Twenty-one hospitals responded (11 French and 10 from other European countries). All European compounding unit organizations are not significantly different. The study compared items such as infrastructure cost, equipment cost, staff cost, consumable cost, cleaning cost and control cost. A synthesis of all costs has been drafted to calculate an estimated preparation cost which seemed to be higher for group B than for group I when staff costs were included ($46 and $31, respectively, in study conditions). Conclusions: The different costs studied have revealed little significant difference between group B and I. The preparation cost in group B appears higher than in group I. This pilot study has resulted in the calculation of an estimated manufactured preparation cost but this work should be completed to help optimize resources and save money.

Keywords Cost analysis, drug compounding, pharmacy service, hospital/standards, equipment and supplies, hospital/standards

Introduction A controlled atmosphere area (CAA) is an environment where a low level of environmental pollutants such as dust, airborne microbes, aerosol particles and chemical vapors is maintained. Sterile products that are routinely compounded in hospital pharmacy CAA include cytotoxic drugs, total parenteral nutrition (TPN), eye drops and central intravenous additive service (CIVAS). The level of pollutants must be kept low in order to contribute to production of a sterile drug. When the compounded product is toxic, the design of the CAA unit must protect the technicians. The objective of a good CAA design is to provide control of some parameters with the highest quality and conformity to

design requirements while maintaining reasonable construction and production costs. Several recommendations or mandatory regulations to hospitals and pharmacies exist. In recent years, the drug manufacturing conditions in hospitals have been increasingly demanding.

1

Biopharmacy, Galenic and Hospital Pharmacy Department, UFR Pharmacie, Universite´ Lille Nord de France, Lille, France 2 Pharmacy, Lille University Hospital, Lille, France Corresponding author: Bertrand Dećaudin, Faculte´ de Pharmacie, Laboratoire de Biopharmacie, Pharmacie Galeńique, Hospitalie`re (EA4481), 3, rue du Professeur Laguesse, BP 83, Lille 59006, France. Email: [email protected]



(OPP)


2

Journal of Oncology Pharmacy Practice 0(0)

Indeed, the issue of recommendations or mandatory regulations to hospitals and pharmacies has allowed pharmacists to rely on several texts to ensure the quality of their pharmaceutical preparations. These documents include: ISO standard: particularly with ISO 146441; Current Good Manufacturing Practices2 (cGMP); United State Pharmacopeia Chapter 7973 (USP 797); European Pharmacopeia4–6: Specific monographs of parenteral and sterile preparations; . Pharmaceutical Inspection Convention and Pharmaceutical Inspection Co-operation Scheme7 (PIC/S); and . CM/resAP 20118 resolution on quality and safety assurance requirements for medicinal products prepared in pharmacies for the special needs of patients.

. . . .

Moreover, French pharmacists refer to chapters 6 and 7 of the Bonnes pratiques de pre´paration9 (BPP; French Good Compounding Practices for Hospitals and Community Pharmacies). None of these texts describe a specific method for room design in order to maintain sterility and sterile manufacturing process. To prevent and reduce contamination risks, protection facilities available to us are: 1. Isolators (volume enclosed by walls materializing a physical separation between sterile area and nonsterile area and provided with a biodecontamination system), 2. Class II BSC (BioSafety Cabinet II: partially open area on the front with a laminar air flow materializing a barrier between the operator and preparation) and Class III BSC (physical enclosure barrier between the operator and preparation, with negative pressure and without biodecontamination system). Parallel to the emergence of these areas, centralized manufacturing has increased in hospital pharmacies. This centralization means that preparations are safer10–12 but also more economical.13–15 For a long time, cost accounting was reserved for industry alone. Various methods are described to correct this image and enhance the pharmaceutical activity in hospital pharmacies, in particular the compounding activity. The questions and problems are the same for each CAA design: what will be the most economical scheme for the number and type of products made? Many studies compare rooms equipped with BSCII and those equipped with an isolator but no study has ever compared these two systems of protected drug compounding from an economical point of view.

The aims of this study are: . To make an inventory of fixtures of aseptic compounding structures. . To compare the design and operating costs of a manufacturing room equipped with an isolator versus a room equipped with laminar flow biological safety cabinet (BSC), using real examples in order to determine the most economical scheme for a number and a type of given preparations. . To give a final facilities cost calculated for one workstation.

Methods The study was conducted in Europe over eight months from January 2011 to August 2011. Forty-three hospitals were interviewed (21 from France and 22 from other European hospitals in 9 different countries).

Survey form Data were collected through a form programmed in Visual Basic for Applications (VBA) completed by hospital pharmacists. The form contained 390 items covering: . General information about the hospital and its pharmaceutical activity. . Infrastructure cost. The infrastructure cost of the room was divided into the cost of building walls, cost of initial and operational qualifications of the room and the cost of the central air handling. . Equipment cost including cost of protective equipment (isolator or BSC), and other items: automated production tools (filling or sealing), control instruments (scales) and others (welding machines, refrigerators and freezers). The cost of these facilities included the costs of their installation and operational qualifications. . Staff cost. The staff was detailed in number of individuals in each occupational category (senior pharmacists, junior pharmacists, technicians, maintenance technicians, students) and average salaries in each category were provided by the hospitals. The staff training cost was the cost necessary to train all workers of the team for one year (hardware, cost of hours not worked in the unit and cost of external training). . Consumables cost including cost of dressing and compounding materials or devices. The dressing cost included the cost of masks, hair covers, gowns, sterile clothes (pyjamas), overshoes, subgloves, over-gloves, soap, shoes, hydro-alcoholic



(OPP)


Dekyndt et al.

3

solutions and nailbrushes. The dressing cost only includes single-use devices. The compounding materials cost included the cost of needles, syringes, air venting filters, spikes, bags, tubing, filters, sterile drapes, needle collectors, trays of instruments, containers for incineration, crimp pliers, eye drops bottles, dropper nozzles, self-adhesive opaque bags, radio-sterilized bags, neoprene gloves (for Isolators), paper, labels and garbage. . Cleaning cost included the costs of household products and decontamination. . Controls cost. Controls cost covered consumables costs related to environment (air control, microbiological controls of surface and air and chemical contamination controls) or to preparations (sterility checks or analytical testing).

Costs analysis The cost of pharmaceutical active substances was not included in the calculation of costs. Costs were expressed as total cost and according to activity or infrastructure data (cost/m2, cost/workstation, cost/ hour, cost/preparation) and they were expressed in US dollars. A workstation was defined as the area necessary to perform the work for one operator (one protective equipment can consist of several workstations). We shared CAA into two groups: those equipped with isolators (Isolators group) and those fitted with other equipment for aseptic production included BSCII or BSCIII (BSC group). Synthesis calculation of cost per workstation was performed with straight-line depreciation of the room at 10 years, an equipment straight-line depreciation at 7 years and the annual number of various preparations made per workstation. Calculations corresponded to cost/workstation, cost/m2, or cost/preparation related to a standard room measuring 23.5 m2 and 23.18 m2, a number of staff per workstation of 3.1 and 2.4 agents/workstation and a number of preparations compounded per workstation of 4731 and 4725 preparations/workstation for BSC and isolators groups, respectively. For the calculation of the staff cost, the

average French payment cost was used. The staff levels may differ according to the country, so we calculated the staff cost as the product of the median staff cost/ year/agent (all levels combined) by the median number of agents necessary on one workstation. The Mann–Whitney U test was used to compare cost values with the null hypothesis that there is no difference in cost between the BSC and isolators groups. Results are expressed as median values (min–max) of cost values. Monte-Carlo16 modeling or permutations were used to compare proportions. The level of significance was established at 0.05.

Results We obtained results from 21 CAA (10 with BSC and 11 with isolators), namely a response rate of 48.8%. The hospitals included in the study were in France (11), Italy (1), Switzerland (5), Denmark (3) and Sweden (1). Characteristics of hospitals and workstations are summarized in Tables 1 and 2. Statistical differences were notified between the two groups about the number of agents per workstation and the number of hours worked annually per workstation. The median number of agents/workstation and the median number of hours worked annually/workstation were higher for group B than group I (3.1 (1.53–5) vs. 2.1 (0.50–5); p ¼ 0.04 and 6001 (2600–9100) vs. 3275 (777– 10,400); p ¼ 0.04). The construction costs are described in Table 3. The median cost/m2 of CAA was over twice the value for group B compared to group I ($10,375 (10,375–10,542) vs. $5035 (908–19,208); p ¼ 0.52). The median numbers of workstation per protective equipment were 1 (1–2) and 2 (2–4) (p < 0.01) for the BSC and isolators groups, respectively. Costs of equipment are summarized in Table 4. The median equipment cost per workstation was less than half for group B compared to group I ($36,420 (8513–88,646) vs. $74,414 (30,175–253,484); p ¼ 0.03). Dressing costs are described in Table 5. The median dressing cost per hour worked in the room seems different between the two groups ($1.16 (0.18–2.17) vs. $0.45 (0.17–1.38); p ¼ 0.19). The median cost of compounding materials/ preparation was similar for the two groups ($3.70 (0.34–5.09) vs. $2.88 (0.14–5.74); p ¼ 0.96). Out of the

Table 1. Hospitals characteristics (median values (min–max)). Characteristics

Global

BSC

Isolators

Teaching hospitals/hospitals Mean number of beds/hospital Time since commissioning of the room (years) Mean number of preparations/year

86% 1050 (450–3000) 6 (0.5–24) 15,500 (2597–60,700)

73% 960 (700–3000) 7 (0.5–24) 12,944 (2597–35,200)

100% 1450 (450–3000) 5 (2.5–11) 15,500 (2600–60,700)



(OPP)


4


Table 2. Characteristics of workstation in European included hospitals (median values (min–max)). Characteristics

Global

BSC

Isolators

Type of preparation Cytotoxics Total parenteral nutrition CIVAS Others

67% 5% 5% 24%

50% 10% 10% 30%

82% 0% 0% 18%

Antibiotics, mouthwashes, eye drops, therapy based enzymes reconstitution, etc. 2

Work area/workstation (m ) Number of agents/workstation* Number of hours worked annually/workstation* Number of preparations annually manufactured/workstation

23.18 (8.75–90) 2.7 (0.50–5) 4914 (777–10,400)

23.50 (11.82–38.41) 3.1 (1.53–5) 6001 (2600–9100)

23.18 (8.75–90) 2.1 (0.50–5) 3275 (777–10,400)

4725 (1300–30,350)

4731 (1484–18,000)

4725 (1300–30,350)

*Significant statistical difference between BSC group and isolators group.

Table 3. Construction cost (in USD) (median values (min–max)). Characteristics

Global

BSC

Isolators

Construction cost Installation qualification (IQ) Operational qualification (OQ) Air handling unit Total construction cost Construction cost/m2

313,313 (4158–1,549,248) 2662 (387–54,093) 7513 (387–155,477) 189,127 (60,716–619,700) 735,670 (68,107–2,176,223) 10,375 (908–19,208)

335,849 (173,514–970,230) 3993 (1248–11,535) 11,778 (3588–33,159) 267,241 (139,034–579,022) 735,670 (373,844–1,593,948) 10,375 (10,375–10,542)

313,313 (4158–1,549,248) 2614 (387–54,093) 5396 (387–155,477) 116,474 (60,716–619,700) 634,658 (68,107–2,176,223) 5035 (908–19,208)

Table 4. Cost of protective equipment relative to one technician’s workstation (in USD) (median values (min–max)). Characteristics

Global

BSC

Isolators

Number of protective equipment Cost of protective equipment Cost of other equipment Total equipment cost Equipment cost/workstation*

1 (1–4) 145,717 (15,730–715,471) 16,132 (1032–662,785) 182,037 (17,028–826,161) 60,528 (8513–253,484)

1 (1–4) 51,556 (15,730–308,459)* 15,289 (1032–29,387) 105,210 (17,028–309,491)* 36,420 (8513–88,646)*

1 (1–2) 196,377 (84,024–715,471)* 17,744 (9283–662,785) 242,112 (120,699–826,161)* 74,414 (30,175–253,484)*


14 hospitals responding to the form, which prepare cytotoxic drugs; five of them used closed transfer systems. In these installations, these systems were not used for the entire production. Cleaning costs are described in Table 6. Costs of consumables for cleaning the interior of the protective device per workstation per preparation made were similar for the two groups ($0.59 (0.08–0.97) vs. $0.41 (0.01–4.58); p ¼ 0.95). The median annual costs of staff were $318,482 (96,120–834,097) and $358,284 (134,419–448,133) for

groups B and I, respectively (p ¼ 0.81), i.e. a cost of staff per preparation of $36 (4–80) and $19 (6–174) (p ¼ 0.65) for groups B and I, respectively. The median annual cost of staff training was higher for group B than for group I ($32,553 (4179–294,128) vs. $16,325 (68–26,183); p ¼ 0.04). Controls are described in Table 7. Analytical assays of the preparations were performed in 24% of the hospitals surveyed (each in an internal control department of the hospital pharmacy). Two of these assays were made on preparations containing cytotoxics.



(OPP)


Dekyndt et al.

5

Table 5. Total cost of dressing per hour worked in the room (in USD) (median values (min–max)). Characteristics

Global

BSC

Isolators

Use of Mask* Hair covers* Under-gloves Over-gloves Total dressing cost Dressing cost/hour worked in the room

78% 82% 100% 54% 7733 (1367–51,948) 0.61 (0.17–2.17)

100% 100% 100% 44% 8094 (2118–51,948) 1.16 (0.18–2.17)

55% 64% 100% 64% 5681 (1367–20,450) 0.45 (0.17–1.38)


Table 6. Cost of consumables for cleaning the interior and exterior of the protective device (in USD) (median values (min–max)). Characteristics

Global

BSC

Isolators

Equipment cleaning agent

Peracetic acid; ethanol-70%

89% ethanol-70%

40% peracetic acid; 20% ethanol-70%

Hydrogen peroxide; peracetic acid; mixture of the products 38% 43.08 (108.6)

–

20% 41.70 (114.7)

27% hydrogen peroxide; 36% peracetic acid; 36% mixture of the products 55% 44.27 (108.3)

24,837 (2581–174,784) 0.59 (0.01–4.58)

24,837 (14,958–34,716) 0.59 (0.08–0.97)

69,524 (2581–174,784) 0.41 (0.01–4.58)

79.88 (7.12–909.82)

24.07 (7.12–197.03)

145,58 (7.58–909.82)

Biodecontamination Agent for equipment biodecontamination Room biocleaning Mean room biocleaning frequency (times per year) (standard deviation) Total cleaning cost Cost of consumables for cleaning the interior of the protective device/workstation/ preparation made Cost of consumables for cleaning the outside of protective device per m2

Fifty percent of the units handling cytotoxic drugs participated at least once in a research study on environment and chemical contamination by personnel. One of them seemed to do these checks regularly. The molecules targeted were 5 fluoro uracil (5FU), methotrexate (MTX) and cyclophosphamide. A synthesis calculation is displayed in Table 8 (with all costs). The infrastructure cost represents 9–13% of the preparation cost, the equipment cost 6–11%, staff cost represents 66–67%, the consumables cost 9–10%, the cleaning cost 1.5–3.6% and the control cost 0.5–1.4%.

Discussion To our knowledge, this is the first study to assess the design and operating costs of CAAs and to compare, with real costs, CAAs equipped with isolators with CAAs with BSCs in order to determine the most economical scheme in hospitals. We found that the total

cost per preparation is higher for the BSC group than for the isolators group. Distribution between hospitals equipped with isolators and those equipped with other safety devices was rather equal in this study. BSC group included only BSC II except one hospital with BSC III in a class ISO5 room. Hospital activity characteristics were similar in the two groups. Nevertheless, the number of hours worked annually per workstation was significantly higher for the BSC group than for the isolators group while the number of preparations manufactured annually was similar between these two groups. Further research studies should be carried out, breaking down the time allocated to each production stage in order to determine at what stage the agents save time by using isolators. The number of agents (manipulator agent, manipulator assistant, pharmacist, control staff) per workstation was higher for rooms provided with BSC than for isolators. In the recommendations texts, it is advisable to bring as few people as possible into



(OPP)


6


Table 7. Total cost of controls (in USD) (median values (min–max)). Characteristics Environment microbiological control Frequency per year Preparation microbiological control Frequency per year Analytical preparation controls Frequency per year Preparation controls Double visual inspection Gravimetric system Spectrometry Osmometry Total cost of microbiological controls Total cost of microbiological controls/preparation made

Global

BSC

Isolators

52 (2–365)

52 (12–365)

19 (1–52)

2 (1–260)

1,5 (1–260)

2 (1–52)

4315 (40–13,000)

131 (40–8500)

13,000 (n ¼ 1)

81% 33% 10% 10% 3.023 (58–19.202) 0.31 (0.04–0.97)

70% 25% 0% 10% 4.629 (58–19.202)* 0.31 (0.13–0.93)

91% 40% 18% 9% 1.704 (111–9855)* 0.16 (0.04–0.97)


Table 8. Simulation table of the study cost/workstation (expressed in US$)a. Cost/Workstation

BSC

Infrastructure cost Construction cost (straight-line depreciation 10 years) Room qualification (IQ + OQ) cost (straight-line depreciation 10 years) Room qualification (PQ + maintenance) cost Equipment cost Equipment cost (straight-line depreciation 7 years) Equipment qualification (IQ + OQ) cost (straight-line depreciation 7 years) Equipment qualification (PQ + maintenance) cost Staff cost Staff cost Staff training cost Consumables cost Compounding materials Dressing cost Cleaning costs Room cleaning cost Equipment cleaning cost Control costs Room microbiological control cost Preparation microbiological control cost Total annual cost/workstation Total annual cost/workstation (without staff cost) Total cost/preparation Total cost/preparation (without staff cost)

Isolators

24,383 684 4032

11,669 164 1942

5203 1405 6906

10,631 3361 2462

135,133 11,622

94,202 4000

17,512 4597

12,367 1183

566 2791

3374 1937

1530 1464 217,828 71,073 46 15

80 666 148,038 49,836 31 11

a The surface of a room used for one workstation: 23.5 m2 and 23.18 m2, the number of staff per workstation: 3.1 and 2.4 agent/workstation and the number of preparations compounded per workstation: 4 731 and 4 725 preparations/workstation. A median number of 10 years is used for straightline depreciation.



(OPP)


Dekyndt et al.

7

the environment.2,7,9,17 Manipulation with an isolator may require less manipulator assistance than manipulation with a BSC. The cost of sterile compounding rooms is rarely published. One study estimated in 1995 at $31,400 the cost of construction for a room of 33.3 m2 (architectural and engineering fees, general contractor and construction fees, heating ventilation and air conditioning, electricity, plumbing, seamless flooring, ceiling tiles cleanroom, fixtures and epoxy wall paint).18 This cost would rise to $45,720 by adding inflation since 1995. Even discounting the construction cost with the inflation rate, it is well below the costs identified in our study ($342,396 for a room of 33.3 m2). There was a superiority (not significant) of the median construction costs for the BSC group, which could be explained by the level of control of the room. A CAA equipped with a BSC must, according to the recommendations, be classified ISO 5 while a CAA provided with an isolator may be classified ISO 7 (or ISO 8).9 Extreme values ($178,867 and $246,471) for the cost equipment/workstation were due to CAAs provided with an isolator composed of several parts and one transfer isolator. The transfer isolator does not generate a workstation, so the cost of the transfer isolator really increases the cost of protective equipment per workstation. Thus, all the high values of this cost category were probably due to the presence of the transfer isolator. In addition, a class III BSC presents an area closed by physical barriers unlike a Class II BSC, and has a ventilation system different from the BSC II. So, the costs of BSC III seem higher than those of BSC II. The same remarks can be made for the cost of protective equipment with costs of initial and operational qualifications. This may be due to the different characteristics of the equipment, which nowadays is customized. Equipment qualifications consist, among others, of checks of filter integrity or tightness of the aseptic compounding enclosure. So as equipment complexity increases, qualification costs increase. CAAs equipped with isolators used fewer masks and hair covers because these rooms are generally classified as ISO 7 so are less demanding than those of the BSC group. The use of isolators leads to a higher consumption of over-gloves. Indeed, over-gloves can protect not only the operator but also the neoprene sub-gloves. The difference in dressing cost could be explained by the room class, like the design costs. Dressing to meet the limits of particulate and microbiological contamination is more complete so more expensive in room class ISO 6 than in class ISO 7 or 8. In this study, only single use devices have been included in costs but it is certain that other very high costs such as clothing sterilization should be taken into account.

The use of closed transfer devices is not yet common according to the responses obtained in this study. The use of such devices increases the total compounding cost and gives a slight lengthening of the manufacturing time, but the benefit in terms of personal protection is not negligible. It would be interesting to create a score for staff and preparation protections (allocating a number to every device: needles, air venting filters and spikes, according to their ability to avoid contamination) in order to objectively compare security between all devices. And the high values of the cost of preparation materials are probably due to a regular and frequent change of equipment such as sterile drapes, needles, or syringes. Low values could be due to the production batches (the same needle or sterile drapes, for example, are used for several preparations). Cleaning products used in the CAA are very different. The cleaning products used for the BSC group were more economical (70% ethanol or DismozonÕ from Bode–Science–Competence company for the environment and alcohol at 70% for the inside of the BSC), while the products used for the isolators group were more expensive (SurfaniosÕ and Surfa’safeÕ from Anios company and IncidinÕ from Ecolab company for environment and hydrogen peroxide, peracetic acid and SoproperÕ from Schulke company for the inside). Cheap disinfectants such as 70% ethanol cannot be used in isolators because they are incompatible with the PVC barriers. There are no recommendations concerning the cleaning of these areas, so the clean-up costs are dispersed. Indeed, each hospital uses the product of its choice until the required limits of microbiological contamination of classified rooms are met. However, biocleaning seemed less frequent in the BSC group than in the isolators group. Contrary results might have been expected. The CAA of the BSC group have a higher level of requirements in terms of microbiological contamination than that of the isolators group. Biocleaning should be more common for hospitals in the BSC group. These results could be explained by a lack of understanding of these issues by the hospital pharmacists because this result is not logical. Neither the time required for the decontamination or sterilization of devices entering an ISO classified room, nor the time needed for staff dressing were included in this study. This time can be considerable for ISO 6 room and therefore may have an impact on costs. The difference in staff costs can be explained by the fact that wages were different between European countries. One study19 investigated the cost of TPN bags in Europe and reveals a difference in cost between European countries due to the working time. The BSC group was mainly composed of non-French European hospitals (70%) while the isolators group



(OPP)


8


was mainly composed of French hospitals (73%) with difference in annual salary per agent. Thus, this may explain the difference in cost. A calculation relating the staff cost to the standard of living in the country concerned should homogenize results. Staff training seemed shorter and less expensive for agents working with isolators. This is due to a preconceived idea that it is not necessary to be as careful in handling with an isolator as with a BSC, because the isolator has a physical barrier. However, chemical environment contamination around an isolator exists and can be due to improper handling inside it.20 So handling with any protective equipment should be the same for both groups and training as well. The controls carried out on the preparation may be multiple. Indeed, the double visual inspection was sometimes associated with a gravimetric system in some hospitals interviewed. Similarly, there were some controls that have not been mentioned because they were not proposed in the form. Thus, this preparation control item can be simply an overview of the use of certain checks such as the double visual control or the gravimetric system, but the reply list is not really exhaustive. The higher frequency of environment microbiological monitoring in the BSC group and the significant difference of costs could still be explained by the median of the ISO class room in the BSC group, which was lower than that of the isolators group. Thus, the limits of contamination are more stringent and controls are more frequent. Microbiological control of preparations was the same for both groups of protective device because the ability to maintain sterility is the same according to the different protective equipment. Internal staff usually carries out these controls. This can surely reduce the costs. They are still charged to the pharmacy and registered in the accounting process. Few units still practice assay on each preparation and staff and environment’s chemical contaminations controls (30% of hospitals interviewed have done these assays at least once, but only one does them regularly). These practices are still evolving with the development of new analytical techniques or kits for the determination of the contamination. Analytical control costs were not included in the synthesis calculation, unlike microbiological controls that do not depend on internal procedures, but on good practices. This study is a pilot study that should be pursued on a larger scale. French facilities are mainly represented. Other countries should be included to give more data on isolator technology. There are major differences between minimum and maximum costs, so it would be interesting to repeat a study targeting specific types of preparations (i.e. cytotoxic) performed in centers with a similar volume of activity. With a larger

sample of hospitals, another approach would be to compare the data according to the class of BSC and the design of isolator or according to the room classification, which depends on country regulations and preparations process. The number of hospitals was still too low to obtain statistically significant results. In addition, most of the results are from statements of pharmacists interviewed and a degree of subjectivity persists. Completing this study would allow the cost of a room to be planned according to the equipment placed inside thereby optimizing the profitability of this type of facilities.

Conclusion The organization of compounding unit seems to be homogeneous in European hospitals. The different costs studied have revealed little significant difference between groups B and I. The synthesis table of median costs of various European hospitals revealed higher preparation costs for group B than for group I. This pilot study should be completed to help to optimize resources and save money. Acknowledgment Authors would like to thank Mr Didier Meyer (Getinge Group) for his assistance.

Funding This study was funded by intramural funds from the department of biopharmacy, galenic and hospital pharmacy, EA 4481, IFR114, Universite´ Lille Nord de France, Lille.

Conflict of interest The authors declare no conflict of interest relating to this work.

References 1. Standard 14644-3. Cleanrooms and associated controlled environments – part 3: test methods, 1st edn. Geneva, Switzerland: International Organization for Standardization, 2005. 2. Sterile Drug Products Produced by Aseptic Processing. In: Guidances (drugs) – Current Good Manufacturing Practices (CGMPs)/Compliance. Center for Drug Evaluation and Research – FDA 2004. 3. Pharmaceutical compounding-sterile preparations (general information chapter 797). In: The United States Pharmacopeia, 28th rev., and the national formulary. 23rd edn. Rockville, MD: United States Pharmacopeial Convention, 2004, pp.2461–2477. 4. Preparations for irrigation (monography 01/2008:1116). In: The European Pharmacopeia, rev.7.5. European Pharmacopeial Convention, 2008, pp.731–732.



(OPP)


Dekyndt et al.

9

5. Eyes preparations (monography 01/2008:1163). In: The European Pharmacopeia, rev.7.5. European Pharmacopeial Convention, 2008, pp.710–712. 6. Method of preparation of sterile products (monography 5.1.1). In: The European Pharmacopeia, rev.7.5. European Pharmacopeial Convention, 2008, pp.503–504. 7. Guidelines on the Standards Required for the Sterile Preparation of Medicinal Products. In: PIC/S GMP GUIDE. The Pharmaceutical Inspection Cooperation Scheme – PIC/S – Publications, 2009. 8. Resolution CM/ResAP (2011)1 on quality and safety assurance requirements for medicinal products prepared in pharmacies for the special needs of patients. In: The 1103rd meeting of the Ministers. The Committee of Ministers. 19 January 2011, Deputies. 9. Bonnes Pratiques de Pre´paration. AFSSAPS: Agence française de sećurite´ sanitaire des produits de sante´, 2007. 10. Wernli M, Bo¨gli F, Mu¨hlebach S, et al. Centralized preparations of cytostatic agents: a method for quality control. Schweiz Med Wochenschr 1994; 124: 1962–1965. 11. Limat S, Drouhin JP, Demesmay K, et al. Incidence and risk factors of preparation errors in a centralized cytotoxic preparation unit. Pharm World Sci 2001; 23: 102–106. 12. Cazin JL and Gosselin P. Implementing a multiple-isolator unit for centralized preparation of cytotoxic drugs in a cancer center pharmacy. Pharm World Sci 1999; 21: 177–183.

13. Martı´ nez-Tutor MJ. Evaluacioń econo´mica de la centralizacioń de mezclas intravenosas. Farm Hosp 2006; 30: 351–358. 14. Legat C, Limat S, Coutet J, et al. Impact ećonomique de la pre´paration centraliseé des me´dicaments anticance´reux. J Pharm Clin 2003; 22: 181–185. 15. Rohrbach P, Collinot J-P, Arth A, et al. Reconstitution centraliseé des cytotoxiques en isolateur: incidence ećonomique. J Pharm Clin 1999; 18: 273–276. 16. Sham PC and Curtis D. Monte Carlo tests for associations between disease and alleles at highly polymorphic loci. Ann Hum Genet 1995; 59: 97–105. 17. Guidelines for using the test for sterility (monography 5.1.9.). In: The European Pharmacopeia, rev.7.5. European Pharmacopeial Convention, 2009, pp.519–520. 18. Schumock GT, Kafka PS and Tormo VJ. Design, construction, implementation, and cost of a hospital pharmacy cleanroom. Am J Health Syst Pharm 1998; 55: 458–463. 19. Walter E, Liu FX, Maton P, et al. Cost analysis of neonatal and pediatric parenteral nutrition in Europe: a multi-country study. Eur J Clin Nutr 2012; 66: 639–644. 20. Crauste-Manciet S, Sessink PJM, Ferrari S, et al. Environmental contamination with cytotoxic drugs in healthcare using positive air pressure isolators. Ann Occup Hyg 2005; 49: 619–628.
