ABSTRACT

Therapeutic drug monitoring (TDM) has as its main objective to ensure that the plasma drug concentration remains within the appropriate range. Regarding the economic dimension of TDM, it is known that there are gains in health outcomes; however, there is still little evidence for the benefit of this procedure performed by pharmacists within the hospital context. With this project, we aimed to create a matrix of cost avoidance associated with TDM performed by pharmacists and to quantify the total avoided costs in 1 year, by implementing a TDM process in a tertiary hospital. For the studied period, we collected 362 pharmaceutical interventions related to TDM of antibiotics performed in adults. As a result, considering these data, the total cost avoidance in 1 year was 371,018 ($416,584.58) at one medical center. We conclude that TDM is highly cost-avoidant and that the implementation costs by pharmaceutical services is clearly lower than the benefit achieved.

INTRODUCTION

In spite of the standardization efforts in product development and a number of drugs having been made available to prescription and to be readily given to patients, there are still cases in which further personalization of therapy is needed to maximize health outcomes and reduce the risk of adverse effects.[1]

Therapeutic drug monitoring (TDM) has as its main objective to ensure that drug concentration in the plasma remains within the appropriate range, maximizing the clinical benefit for the patient.[2] TDM is of the utmost importance in drugs with narrow therapeutic margins, where there is a thin margin between the desired benefit and the adverse events. This process is done using pharmacokinetic principles and software applications that use varied models, but in this case, mostly Bayesian calculation methods are used. This is currently a routine procedure in many hospitals and is usually performed by pharmaceutical services following a serum drug measurement (Figure 1).

Figure 1

Drug prescription process.

Figure 1

Drug prescription process.

Regarding the economic dimension of TDM, there are gains in health outcomes[3]; however, there is little evidence on the financial/economic benefit of this procedure that is performed by pharmacists[4] within the hospital context. With the progressive extension of therapeutic monitoring to innovative therapies (monoclonal, cytotoxic antibodies),[5,6] it becomes even more important to develop a reproducible matrix to assess the economic benefit of TDM and further develop and improve these processes within institutions.

Objective

There are two main direct objectives in this research. First, to create a matrix of costs associated with the main adverse events of three antibiotics: vancomycin, gentamicin, and amikacin in the Portuguese setting. And second, to quantify the total avoided costs in 1 year, by implementing a TDM process, performed by pharmacists in a tertiary care hospital and to assess its direct economic impact.

We depart from the role of TDM and its main objective to ensure that the plasma drug concentration remains within the appropriate range. We will assess if there is an economic dimension of TDM. And if there are gains in health outcomes, we will provide evidence for the benefit of this procedure performed by pharmacists within the hospital context. We aim to create a matrix of cost avoidance associated with TDM performed by pharmacists and to quantify the total avoided costs in 1 year, by implementing a TDM process in a tertiary hospital.

METHODS

To create the cost-avoidance matrix, we used the cost-benefit model Nesbit et al.[7] This model considers that the (pharmacist) intervention can eliminate the occurrence of an harmful event associated with a drug. Accordingly, the avoided cost with an intervention is the product of the probabilities of occurrence of the avoided events and the cost associated with treating those events. For each drug, we defined the two most common adverse events of severe impact, according to their summary of product characteristics (see Table 1) and their costs. We also considered the situation of subtherapeutic levels of drug. The probability of events was defined based on a literature review of cohort and case-control studies. The cost of the events was considered based on the Portuguese coding matrix of costs in the national health care system, version (DRG) 2018.[8]

Table 1

Adverse events defined for the study and their associated cost

Adverse events defined for the study and their associated cost
Adverse events defined for the study and their associated cost

Pharmaceutical interventions performed at a Portuguese hospital refer to the period between January and November 2019, from an internal pharmaceutical intervention database. Statistical analysis was done with R Studio, version 1.2.

RESULTS

For this study, we collected 362 pharmaceutical interventions related to TDM of antibiotics performed in adults. The median number of analytical measurements performed per patient per hospitalization is four, ranging between one and 22. Most of the antibiotic monitoring was done on vancomycin (62%), with the remaining done on gentamicin (30%) and on amikacin (8%).

Following the analysis of the most frequent adverse events, a cost matrix was constructed, by choosing the appropriate code for each event at the DRG framework in Portugal. The probability matrix was created using previous studies and meta-analyses[613] (Table 1).

After applying the cost-benefit model, the cost avoidance for TDM in each drug, per patient, was the following: vancomycin 900.92 ($990.97, USD), amikacin 1114.80 ($1226.82), and gentamicin 1257 ($1382.64). The cost of performing TDM in one patient (pharmacist time + laboratory test) is 35 ($38.50).

As a result, considering these data, the total cost avoidance in 1 year was 371,018 ($416,584.58) at one medical center. Because we are using data from January to November 2019, this is probably an underestimation of the true annual savings.

DISCUSSION

These results point to an important level of cost avoidance of the TDM performed by pharmacists on the selected antibiotics. It is also important to point out that the percentage of TDM performed for gentamicin and amikacin was significantly lower than vancomycin. One of the major limitations of this study is the usage of other cohort and case studies to define the probabilities of occurrence of the adverse events. In the future, it would be best to use real-world data from our center, as it could provide more accurate estimates in our setting.

CONCLUSIONS

Measuring the economic benefit of interventions by health professionals remains a challenge, largely because of the availability of data, selecting the appropriate variables and models, or the difficulty of translating avoided costs into direct savings. In this study, it is possible to frame the benefit of therapeutic monitoring in a cost matrix adjusted to the setting of a Portuguese hospital. We also construct a probability matrix based on the currently available studies. Future research should try to quantify and assess these phenomena and values, through robust studies, within the Portuguese reality.

We conclude that TDM is highly cost-avoidant. Furthermore, the implementation costs of TDM by the pharmaceutical services are clearly lower than the benefit achieved. In future studies, it will be important to measure the translation of this cost avoidance into direct savings, use this methodology in other settings, and populate the probability matrix with real-world evidence from local centers.

References

References
1.
Kang
J,
Lee
M.
Overview of therapeutic drug monitoring
.
Korean J Intern Med
.
2009
;
24
:
1
.
2.
Gross
A.
Best practice in therapeutic drug monitoring
.
Br J Clin Pharmacol
.
1998
;
46
:
95
99
.
3.
Monteiro
J,
Hahn
S,
Gonçalves
J,
Fresco
P.
Vancomycin therapeutic drug monitoring and population pharmacokinetic models in special patient subpopulations
.
Pharmacol Res Perspect
.
2018
;
6
:
e00420
.
4.
Ibrahim
M,
Abdelrahim
H,
Ab Rahman A. Cost evaluation of therapeutic drug monitoring of gentamicin at a teaching hospital in Malaysia
.
Pharm Pract (Granada)
.
2014
;
12
:
372
.
5.
Imamura
C.
Therapeutic drug monitoring of monoclonal antibodies: applicability based on their pharmacokinetic properties
.
Drug Metab Pharmacokinet
.
2019
;
34
:
14
18
.
6.
Morawska
K,
Goirand
F,
Marceau
L,
et al.
5-FU therapeutic drug monitoring as a valuable option to reduce toxicity in patients with gastrointestinal cancer
.
Oncotarget
.
2018
;
9
:
11559
11571
.
7.
Nesbit
T,
Shermock
K,
Bobek
M,
et al.
Implementation and pharmacoeconomic analysis of a clinical staff pharmacist practice model
.
Am J Health-Syst Pharm
.
2001
;
58
:
784
790
.
8.
Regulamentos e as Tabelas de Preços das Instituições e Serviços Integrados no Serviço Nacional de Saúde (SNS)
.
2018; Portaria n.° 254/2018.
9.
Bamgbola
O.
Review of vancomycin-induced renal toxicity: an update
.
Ther Adv Endocrinol Metab
.
2016
;
7
:
136
147
.
10.
Forouzesh
A,
Moise
P,
Sakoulas
G.
Vancomycin ototoxicity: a reevaluation in an era of increasing doses
.
Antimicrob Agents Chemother
.
2008
;
53
:
483
486
.
11.
Bakke
V,
Sporsem
H,
Von der Lippe
E,
et al.
Vancomycin levels are frequently subtherapeutic in critically ill patients: a prospective observational study
.
Acta Anaesthesiol Scand
.
2017
;
61
:
627
635
.
12.
Selby
N,
Shaw
S,
Woodier
N,
Fluck
R,
Kolhe
N.
Gentamicin-associated acute kidney injury
.
QJM
.
2009
;
102
:
873
880
.
13.
Xie
J,
Talaska
A,
Schacht
J.
New developments in aminoglycoside therapy and ototoxicity
.
Hear Res
.
2011
;
281
:
28
37
.
14.
Saunders
N,
Adams
D,
Lynn
W.
A prospective laboratory-based audit of gentamicin use and therapeutic monitoring
.
J Antimicrob Chemother
.
1995
;
36
:
729
736
.
15.
Jenkins
A,
Thomson
A,
Brown
N,
et al.
Amikacin use and therapeutic drug monitoring in adults: do dose regimens and drug exposures affect either outcome or adverse events? A systematic review
.
J Antimicrob Chemother
.
2016
;
71
:
2754
2759
.
16.
Duszynska
W,
Taccone
F,
Hurkacz
M,
Kowalska-Krochmal
B,
Wiela-Hojeńska
A,
Kübler
A.
Therapeutic drug monitoring of amikacin in septic patients
.
Crit Care
.
2013
;
17
:
R165
.

Competing Interests

Source of Support: None. Conflict of Interest: None.