Article Text

Download PDFPDF

2SPD-016 Data-driven selection of a medication management model in hospitalisation wards
  1. A Pérez1,
  2. V Correa2,
  3. MI Martínez3,
  4. R Borràs1,
  5. R López4,
  6. L Estrada1,
  7. E Terricabras1,
  8. S Aulet1,
  9. S Fernàndez5,
  10. C Miret3,
  11. C Quiñones1
  1. 1Hospital Universitari Germans Trias I Pujol, Pharmacy Department, Badalona, Spain
  2. 2Apex Consultoria, Apex Consultoria, Sant Quirze del Vallès, Spain
  3. 3Hospital Universitari Germans Trias I Pujol, Projects and Innovation Unit, Badalona, Spain
  4. 4Hospital Universitari Germans Trias I Pujol, Information Systems Unit, Badalona, Spain
  5. 5Hospital Universitari Germans Trias I Pujol, Nurse Direction, Badalona, Spain


Background and Importance Optimal dispensing and distribution management model of drugs reduces inefficiencies and increase drug safety.

Aim and Objectives To select best medication management model (centralised in pharmacy vs decentralised in hospitalisation wards (HW)) based on medication consumption pattern of different HW, in context of the redesign of medication management system in a high-complexity hospital.

Material and Methods Applying Pareto principles, an ABC-XYZ matrix was designed using medication consumption data from HW in January 2022. This data, obtained from the hospital’s management system, included medications not listed in a pharmacotherapeutic guide (PTG). Information analysed included medication, guide inclusion situation, dispensed quantities, and HW. Within each HW, medications were categorised according to quantity (ABC) and variability (XYZ), with ‘A’ denoting highest consumption and ‘Z’ signifying maximum variability in consumption.


  • A. x ≤ 80,0% (x medications ordered from maximum to lowest consumption)

  • B. 80,0% < x ≤ 95,0%

  • C. 95,0% < x ≤ 100,0%


  • X. CV < 0,3

  • Y. 0,3 ≤ CV ≤ 0,75

  • Z. CV > 0,75

Coefficient of variability (CV) was obtained by dividing standard deviation by the mean. Outliers were removed. ABC-XYZ combination defined consumption pattern of each medication for each HW, associated with a management system.

  • GROUP 1: AX, AY, BX, CX – High consumption, low variability. Decentralisation and replenishment based on standard minimums.

  • GROUP 2: BY, AZ – Moderate volume and variability. Decentralised with replenishment based on criticality or consumption peaks.

  • GROUP 3: BZ, CY, CZ – High variability, regardless of consumption. Centralised in pharmacy or decentralised with systematic monitoring of expiration dates.

  • GROUP 4: zero consumption.

Results 13 units and 826 references were analysed, 37 not included in PTG. Consumption pattern was similar across HW. In HW, ‘A’ account for 56–75 medications, ‘B’ for 63–99 and C for 105–151. A 39–96 [18%-32%] of the references belonged to Group 1, 54–62 [19%-24%] to Group 2, and 116–182 [48%-58%] to Group 3. Each HW only consumed 25%-36% of total references used in the hospital.

Conclusion and Relevance Optimal medication management model was determined by consumption pattern of each reference in each HW, rather than one-size-fits-all approach for entire hospital. However, data supports decentralising medications with monitoring of specific references.

Conflict of Interest No conflict of interest.

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.