# Antibiotic Consumption and Gram-Negative Resistance Dynamics in the ICU: A Five-Year Autoregressive Integrated Moving Average (ARIMA) Analysis

**Authors:** Lejla Rakovac Tupkovic, Nijaz Tihic, Jasmina Smajic, Hedim Osmanovic, Mirela Basic Denjagic, Predrag Jovanovic, Emir Becirovic, Minela Becirovic

PMC · DOI: 10.7759/cureus.102615 · Cureus · 2026-01-30

## TL;DR

This study shows that increased antibiotic use in hospital ICUs is linked to rising resistance in gram-negative bacteria over five years.

## Contribution

The study uses ARIMA modeling to demonstrate a temporal link between antibiotic consumption and resistance in ICU settings.

## Key findings

- Fluoroquinolone use correlates with increased resistance in Acinetobacter baumannii and Klebsiella pneumoniae.
- Carbapenem consumption is linked to higher Escherichia coli resistance rates.
- ICU antibiotic consumption is significantly higher than hospital-wide averages.

## Abstract

Aim: The aim of this study was to evaluate the impact of hospital antibiotic consumption on the rate of antimicrobial resistance (AMR) of gram-negative bacteria, specifically the Enterobacteriaceae family and the genus Acinetobacter, in the University Clinical Center (UCC) Tuzla, Bosnia and Herzegovina.

Methods: A five-year retrospective, observational, pharmacoepidemiological study was conducted (2014 to 2018). Antibiotic consumption was calculated using the WHO Anatomical Therapeutic Chemical/defined daily dose (ATC/DDD) methodology and expressed as DDD per 100 bed-days (BD). Microbiological data were obtained for Klebsiella pneumoniae, Escherichia coli, Proteus mirabilis, and Acinetobacter species. The temporal associations between consumption and resistance were analyzed using linear regression and autoregressive integrated moving average (ARIMA) models.

Results: The total antibiotic consumption at UCC Tuzla significantly increased from 61.35 to 73.51 DDD/100 BD (p=0.003). Consumption in intensive care units (ICUs) was significantly higher than the hospital-wide average (p<0.001), reaching up to 178.53 DDD/100 BD. ARIMA modeling confirmed significant positive correlations between the use of fluoroquinolones (J01MA) and resistance in Acinetobacter baumannii (beta = 7.678, p=0.006) and K. pneumoniae (beta = 18.368, p<0.001)9. A similar correlation was found for carbapenems (J01DD) and E. coli resistance (beta = 14.066, p=0.004).

Conclusion: The study demonstrates a significant temporal association between the volume of broad-spectrum antibiotic consumption and the escalation of AMR. The high selective pressure in ICUs identifies these units as primary reservoirs for multidrug-resistant pathogens. These findings highlight the importance of multidisciplinary antimicrobial stewardship programs and restricted use of reserve antibiotics to preserve therapeutic efficacy.

## Linked entities

- **Chemicals:** carbapenems (PubChem CID 134085)
- **Species:** Klebsiella pneumoniae (taxon 573), Escherichia coli (taxon 562), Proteus mirabilis (taxon 584), Acinetobacter (taxon 469)

## Full-text entities

- **Genes:** extended-spectrum beta-lactamase [NCBI Gene 13982007]
- **Diseases:** infectious disease (MESH:D003141), MRSA (MESH:D013203), UCC (MESH:C563594), bacterial infections (MESH:D001424), gram-negative bacteria (MESH:D016905), AMR (MESH:D060467), COVID-19 (MESH:D000086382), type IV hypersensitivity (MESH:D006968), Infections (MESH:D007239), DRESS syndrome (MESH:D063926), critically ill (MESH:D016638), CLSI (MESH:D007757), gram (MESH:D016908), bacterial co-infections (MESH:D060085), Clostridioides difficile infections (MESH:D003015), MDR (MESH:D018088)
- **Chemicals:** methicillin (MESH:D008712), vancomycin (MESH:D014640), glycopeptides (MESH:D006020), carbapenem (MESH:D015780), cephalosporins (MESH:D002511), Fluoroquinolones (MESH:D024841), UCC (-)
- **Species:** Enterobacteriaceae (enterobacteria, family) [taxon 543], Acinetobacter baumannii (species) [taxon 470], Pseudomonas aeruginosa (species) [taxon 287], Bacteria Latreille et al. 1825 (Bacteria stick insect, genus) [taxon 629395], Staphylococcus aureus (species) [taxon 1280], Proteus mirabilis (species) [taxon 584], Homo sapiens (human, species) [taxon 9606], Klebsiella pneumoniae (species) [taxon 573], Escherichia coli (E. coli, species) [taxon 562]

## Full text

_Full body text omitted from this summary view._ Fetch the complete paper as Markdown: https://tomesphere.com/paper/PMC12949948/full.md

## References

34 references — full list in the complete paper: https://tomesphere.com/paper/PMC12949948/full.md

---
Source: https://tomesphere.com/paper/PMC12949948