Cost analysis of disease including treatment with dalbavancin in a Spanish hospital: ECODAL ANALYSIS

Abstract

Introduction and aim

Dalbavancin is an antibiotic with activity against gram-positive bacteria that allows early discharge of patients requiring intravenous therapy. Outpatient treatment helps offset hospitalisation costs associated with standard intravenous treatment. Our objective was to assess the cost of disease management, including treatment with dalbavancin, in a Spanish hospital for 1 year, and the hypothetical costs associated with treatment with other therapeutic alternatives to dalbavancin.

Methods

A single-centre, observational, retrospective post-hoc analysis was conducted based on electronic medical records analysing all patients who received dalbavancin treatment throughout 1 year; cost analysis was performed for the whole process. In addition, three scenarios designed on the basis of real clinical practice by clinical experts were hypothesised: (i) individual therapeutic alternative to dalbavancin, (ii) all patients treated with daptomycin, and (iii) all days of dalbavancin as outpatient treatment transformed into hospital stay. Costs were obtained from the hospital.

Results

Thirty-four patients were treated with dalbavancin; their mean age was 57.9 years, and 70.6% were men. The main reasons for dalbavancin use were outpatient management (61.7%, n = 21) and ensuring treatment adherence (26.5%, n = 9). The main indications were: osteoarticular infection (32.4%) and infective endocarditis (29.4%). One-half (50%) of the infections were due to Staphylococcus aureus (23.5% were methicillin resistant). All patients achieved clinical resolution, and no costs associated with dalbavancin-associated adverse events or re-admissions were reported. The mean total cost of treatment was 22,738€ per patient, with the greatest expenditures in interventions (8,413€) and hospital stay (6,885€). The mean cost of dalbavancin treatment was 3,936€; without dalbavancin, this cost could have been increased to 3,324–11,038€ depending on the scenario, mainly due to hospital stays.

Main limitation

Limited sample size obtained from a single centre.

Conclusion

The economic impact of the management of these infections is high. The cost of dalbavancin is offset by the decreased length of stay.

Keywords:

• Dalbavancin
• health care costs
• gram-positive infections
• cost analysis

JEL CLASSIFICATION CODES:

• I10
• I1
• I
• I19

Introduction

Infections caused by methicillin-resistant Staphylococcus aureus (MRSA) are associated with increased risk of morbidity, mortality, and healthcare costs for hospitalised patients¹. Epidemiological analysis in Europe revealed that 28–43% of acute bacterial skin and skin-structure infections (ABSSSIs) in hospitalised patients are caused by MRSA^2,3. In Spain, MRSA prevalence in invasive isolates has been reported at 15–19%⁴.

Dalbavancin is a lipoglycopeptide with activity against gram-positive bacteria, including MRSA. It is approved in Europe for ABSSSIs in adults, with a recommended dose of 1,500 mg administered as either a single infusion of 1,500 mg or as a 1,000-mg loading dose followed 1 week later by 500 mg.⁵ However, real-life experience reveals that dalbavancin is being used in serious infections requiring long-term treatment^6–9. Dalbavancin represents an alternative to antibiotics such as glycopeptides, daptomycin, tigecycline, linezolid, or ceftaroline for the treatment of serious infections^10,11 and its use may reduce the length of hospital stay (LOS) and outpatient parenteral antibiotic treatment (OPAT) in real clinical practice^7,9.

The aim of this study was to calculate disease-associated costs, including treatment with dalbavancin, in patients treated in a Spanish tertiary-care hospital, including all required resources in the management of these patients, by contextualising the treatment cost related to the general healthcare management (direct medical costs). The impact of the use of dalbavancin compared with other therapeutic alternatives was estimated using hypothetical scenarios based on clinical practice.

Methods

Design

A single-centre, observational, retrospective, post-hoc study with cost grouping analysis was conducted using an electronic institutional healthcare database. All patients treated with at least one dose of dalbavancin in 2018 (1st January–31st December) were included. Inclusion criteria were age ≥18 years and dalbavancin treatment administered in-hospital or as OPAT. Patients with incomplete medical records were excluded.

Data collection

Data were collected in an electronic case record form following a pre-established protocol. The included variables were demographic characteristics, underlying diseases, diagnostic tests and interventions performed, hospital stays, number of outpatient medical consultations, and outcomes. Data regarding the infectious episode included acquisition place (nosocomial, healthcare-related, or community-acquired), source of infection, microorganism, phenotypic resistance, and antimicrobial treatments. Regarding dalbavancin use, the reason for prescription, dosages, administration regimen and duration of therapy were recorded. Antimicrobials administered previously, during, or after dalbavancin treatment were also recorded. Treatment-related adverse events (AEs) for all antibiotics were recorded in addition to their associated tests and additional LOS or treatment required.

All costs were provided by the hospital pharmacy and analytical accounting department (Table 1). Treatment costs were calculated based on the real ex-factory price plus VAT and Royal Decree 08/2010. Dalbavancin price was provided by the local representative of the Marketing Authorisation Holder.

Table 1. Diagnostic tests and intervention costs.

Item	Unit cost (€)
Hemogram	10
Clotting time	0.9
Biochemistry: lipid profile (TG, CT, HDL, LDL)	10
Biochemistry: liver function tests (GOT, GPT, BT, GGT, LDH, FA)	6
Basic Biochemistry: (glucose, urea, creatinine, K, Na)	6
Biochemistry: renal profile (urea, creatinine, MDRD, uric acid)	13
C-reactive protein	3.89
Proteinogram	0.87
Erythrocyte sedimentation rate	0.87
Creatine Kinase	0.87
Lymphocite populations or subpopulations	64.88
Immunological tests	21
Procalcitonin	6
Arterial blood gas test	13
Clinical urine test	6
Blood culture	21
Urine culture	16
Stool culture	63
Urine toxicology test	38.93
S. pneumoniae/Legionella Antigenuria	7
Quantiferon-TB	30
Troponin T	43.25
Quantitative RNA PCR for HIV (viral load)	60
Iron profile tests	14.44
Atypical bacteria serology (Coxiella, Chlamydia, Mycoplasma)	6
16s universal PCR for bacteria in biopsy	71.42
Bacterial and fungal culture (any sample)	21
Cardiac valve 16s universal PCR	71.42
Transthoracic echocardiogram	215
Transesophageal echocardiogram	35.69
Abdominal echography	39.88
Panendoscopy	38.41
Computed tomography (with contrast)	190
Computed tomography (without contrast)	86
Any type of X-ray	12.75
Other non-abdominal echography	38.41
Magnetic resonance imaging	130
Magnetic resonance imaging (with contrast)	180
Positron emission tomography	1,290
Heart valve surgery	35,702
Trauma surgery*	5,000
Pacemaker/implantable cardioverter-defibrillator implantation or replacement	10,250
Peripherally inserted central catheter	58.95
Hickman catheter	55.58
Hospital stay, infectious diseases ward (day)	296.85
Hospital stay, general ward (day)	250
Hospital stay, ICU (day)	1,100
Home hospitalization (day)	200
Re-admission (day)	250
Day-care hospital (day)	277
Referral (h)	70
Nurse	69
Ambulatory visit (h)	115

*Mean price: knee, hip, and tibial surgery.

Analysis

Statistical analysis was performed using R software (4.0.2). Continuous data are presented as medians with interquartile ranges (IQRs) (minimum 25%, maximum 75%) and means with standard deviations (SD). Categorical data are presented as frequencies with percentages (%). Normality of distribution was assessed with the Shapiro-Wilk test and homoscedasticity with the Bartlett test. The Mann–Whitney U test was used for non-normally distributed variables, and a two-tailed independent samples t-test was used for normally distributed variables. Differences between categorical data were analysed using the χ2 test or Fisher's exact test. For each analysis, a two-sided probability value of p < 0.05 was considered statistically significant.

Base case scenario (with dalbavancin)

Cost analysis was based on data obtained from the medical records regarding LOS in the different departments, day-care hospital, or home hospitalisation, consultations, pharmacological treatment and administration costs, and other costs (diagnostic tests, interventions, and AEs). LOS was calculated from the dates of hospital admission and discharge. If re-admission due to the same infectious episode was required, its costs were also considered. Budget impact (BI) was calculated as mean cost per patient and multiplied by the total number of patients/year at the centre.

To produce plausible comparisons for the results, three scenarios were designed based on the clinical experience.

Hypothetical therapeutic alternative scenario

A hypothetical alternative treatment for each patient was proposed by a team of infectious diseases experts (physicians and hospital pharmacist): alternative drugs, posology, length of treatment, day-care hospital, hospital LOS, home hospitalisation, and requirement for central venous catheter or peripheral catheter for drug administration depending on the selected drug for treatment. As this was a theoretical exercise, costs for therapeutic failures or AEs were assumed to be equal to treatment with dalbavancin.

Hypothetical treatment with daptomycin scenario

All patients were hypothetically treated with daptomycin. Daptomycin dosage was adjusted according to the main infection, the treatment guidelines used in the centre (ABSSSI, 4 mg/kg; spondylodiscitis and osteomyelitis, 6 mg/kg; osteoarticular infection, 8 mg/kg; bacteriemia, thrombophlebitis or endocarditis, 10 mg/kg) and renal function of the patient (if CrCl <30 ml/min, administration was every 48 hours). Daptomycin was assumed to be administered through a peripheral venous catheter at the day-care hospital. A patient with necrotising pneumonia and empyema was assumed to be treated with tedizolid as daptomycin is not indicated for this syndrome.

Hypothetical scenario without treatment, assuming hospital stay

The length of dalbavancin OPAT was assumed to represent the days of hospital stay it prevented, and savings were estimated solely based on the reduction in LOS, not considering other treatments.

Probabilistic sensitivity analysis

An additional multivariable sensitivity analysis was performed for total costs, avoided costs, and for the three alternative scenarios by using a 10,000 simulations bootstrap.¹² Specifically, we simulated 10,000 patients for each approach by sampling from the original data with replacement. For each bootstrap subsample, we calculated the mean cost.

Ethics

The analysis was approved by the Ethics Committee of Hospital General Universitario Gregorio Marañón, Madrid, Spain. No informed consent was needed; however, all patients were informed about the collection of their data.

Results

Overall, data from 34 patients treated with dalbavancin in 2018 were included.

Patient characteristics, clinical outcomes, and costs related to base case

Demographics

Twenty-four (70.6%) patients were male, and the mean age was 57.9 years (range 27–96 years). The most prevalent underlying diseases were cardiac pathology (32.3%), diabetes mellitus (26.5%) and liver disease (20.6%) (Table 2).

Table 2. Patient's baseline characteristics and main infections.

Baseline characteristics
Sex	29.4% Women	70.6% Men
Age	57.9 mean (±16.3)	55 median (IQR, 48.2–65.7)
Underlying diseases	N (%)
Cardiovascular disease	11 (32.3)
Diabetes mellitus	9 (26.5)
Liver disease	7 (20.6)
Child A	3 (42.9)
Child B	4 (57.1)
Meld score	Mean 12.4 (±8.8)
Child-Pugh (5–7)	4 (57.1)
Child-Pugh (7–9)	3 (42.9)
Respiratory disease	6 (17.6)
Neurologic disease	5 (14.7)
Immunosuppressive treatment/corticoids (on the previous month)	5 (14.7)
Active solid neoplasia	4 (11.8)
Chronic renal disease	3 (8.8)
Gastrointestinal tract disease	3 (8.8)
HIV infection	2 (5.9)
Haemodialysis	1 (2.9)
Solid organ transplant	1 (2.9)
Peritoneal dialysis	–
Hematopoietic stem cell transplantation	–
McCabe scale	Ultimately fatal, 5 (14.7)
Charlson comorbidity index	Mean 1.97 (±1.8)
No comorbidity (0–1)	15 (±44.1)
Low comorbidity (2)	8 (±23.5)
High comorbidity (+3)	11 (±32.3)
Main infection and aetiology
Osteoarticular infection (n = 11)*	CoNS, n = 3 MSSA, n = 2 CoNS + Enterococcus faecalis, n = 1 CoNs + Corynebacterium striatum, n = 1 CoNS + Cutibacterium acnes, n = 1 Enterococcus faecium, n = 1 Enterococcus faecalis + Corynebacterium striatum, n = 1 Enterococcus faecalis, n = 1
Endocarditis (n = 10)	MSSA, n = 4 Streptococcus spp., n = 2 CoNS, n = 2 MRSA, n = 1 Unknown, n = 1
Catheter related bacteraemia (n = 3)	MSSA, n = 1 MRSA, n = 1 CoNS + Enterococcus faecalis, n = 1
Thrombophlebitis (n = 3)	MSSA, n = 2 MRSA, n = 1
ABSSSI (n = 2)	CoNS, n = 1 MSSA, n = 1
Pacemaker/Intracardiac device infection (n = 2)	CoNS, n = 1 MSSA, n = 1
Bacteremia overall (n = 2)	MSSA, n = 2
Necrotising pneumonia and empyema (n = 1)	MRSA, n = 1

Abbreviations. ABSSSI: acute bacterial skin and skin structure infection; CoNS: coagulase-negative Staphylococci; HIV: human immunodeficiency virus; MRSA: methicillin resistant Staphylococcus aureus; MSSA: methicillin susceptible Staphylococcus aureus.

*Osteoarticular infection: comprising prosthetic joint infection (n = 5, 14.7%), osteomyelitis (n = 3, 8.8%), septic arthritis (n=2, 5.9%), and spondylodiscitis (n = 1, 2.9%)

Characteristics of the infection

Overall, 50% (n = 17) of the infections were nosocomial infections, 41.2% (n = 14) were community-acquired infections, and 8.8% (n = 3) were healthcare-related infections. The types of infection and aetiologies are described in Table 2. Some patients had more than one infection. Overall, 13 patients had bacteraemia (38.2%). The mean LOS was 31.4 days (SD 20.6 days), with a median of 27 days (IQR, 16–39).

As depicted in Table 2, the aetiological agent was S. aureus in one-half (n = 17, 50%) of the cases, with methicillin-susceptible S. aureus (MSSA) in 38.2% (n = 13) and MRSA in 11.8% (n = 4). Coagulase-negative staphylococci (CoNS) were involved in 32.3% (n = 11) of the infections. In four of the 11 CoNS cases, other microorganisms were isolated: Corynebacterium striatum (n = 1), Enterococcus faecalis (n = 2), and Cutibacterium acnes (n = 1). In one patient cultures were sterile.

All patients required multiple diagnostic tests, such as hemogram (mean 9.6, SD 7.7), blood gas test (mean 7.4, SD 10.9), blood cultures (mean 4.3, SD 5.7), and other tests. The mean cost per patient of these tests was 1,200.8€ (SD 712.3€), and the BI was 40,834€ (Table 3). Most of this cost was attributed to transthoracic echocardiogram (12.6%), with a mean cost of 151.8€/patient and a BI of 5,160€, in addition to biopsy polymerase chain reaction (8.4%), with a mean cost of 100.8€/patient and BI of 3,428€.

Table 3. Mean costs per patient and total budget impact (base case).

		Mean cost (€)	Budget impact* (€)
Diagnostic tests		1,201	40,834
Interventions		8,413	286,042
Stay	Hospital stays	6,885	234,090
	ICU stay	809	27,506
	Day care hospital stay	195	6,630
	Home hospitalization	94	3,196
	Re-admission	79	2,686
	Total	8,062	274,108
Pharmacological costs	Previous antibiotic treatment	880	29,939
	Dalbavancin	3,936	133,841
	Concomitant treatment	90	3,063
	Total	4,907	166,843
Consultations	Referred patients	13	442
	Nursing	25	850
	Outpatient visit	32	1,088
	Total	70	2,380
	Previous antibiotic administration	45	1,530
	Dalbavancin administration	41	1,394
	Adverse events	0	0
	Total	86	2,924
	Direct Costs	22,738	773,087

Abbreviation. ICU: intensive care unit.

*n: total cohort

A total of 15 patients (44.1%) required surgical interventions related to the infection: six required peripheral central catheter insertion, five required cardiac surgery, four required pacemaker/implantable cardioverter-defibrillator implantation or replacement, three required orthopaedic surgery, and one required Hickman catheter insertion. The mean cost of interventions was 8,413€ (SD 16,979€) and the BI was 286,042€ (Table 3).

Mean and median costs per patient’s baseline characteristics, type of infection and microorganism are detailed in Table 4.

Table 4. Costs per microorganism causing infection, type of infection, and patient's baseline characteristics.

Baseline characteristics		n	Mean	Lower CI95%**	Upper CI95%	SD (±)	Median	Min	Max	p-Value
Sex	Female	10	21,291	9,272	33,309	16,801	21,730	3,390	45,462	.75
	Male	24	23,341	14,281	32,401	21,457	14,260	2,859	95,142
Age	18–45	7	17,498	13	34,983	18,906	8,436	3,390	57,627	.23
	46–55	11	19,071	9,484	28,659	14,271	15,465	3,402	45,216
	56–65	7	27,170	0	57,496	32,790	10,604	2,859	95,142
	66–75	3	38,794	17,155	60,433	8,711	41,982	28,938	45,462
	76–85	2	37,473	0	163,896	14,071	37,473	27,523	47,423
	>86	4	14,825	562	29,088	8,963	12,089	7,369	27,754
Cardiovascular disease		11	29,662	10,270	49,053	28,864	15,143	4,192	95,142	.59
Chronic kidney disease		3	21,038	0	66,571	18,329	13,376	7,783	41,955	.95
Haemodyalisis		1	41,955	–	–	NA	41,955	41,955	41,955	NE
Respiratory disease		6	8,164	5,039	11,289	2,978	7,903	4,192	12,337	.01*
Diabetes Mellitus		9	22,525	9,860	35,191	16,478	23,318	2,859	47,423	.91
Neurologic disease		5	9,561	6,445	12,677	2,510	8,436	7,369	13,376	.10
Liver disease		7	18,624	2,531	34,718	17,401	12,337	7,783	57,627	.80
Gastrointestinal disease		3	26,978	458	53,499	10,676	27,754	15,936	37,246	.32
Active solid neoplasia		4	28,762	0	57,727	18,203	33,092	3,402	45,462	.42
HIV infection		2	13,740	0	31,569	1,984	13,740	12,337	15,143	.86
Type of infection		n	Mean	Lower CI95%**	Upper CI95%	SD (±)	Median	Min	Max	p-Value
Osteoarticular infection		11	24,242	14,825	33,660	14,018	23,318	7,783	45,216	.31
Endocarditis		10	33,433	13,384	53,481	28,026	21,610	7,369	95,142	.08
Catheter-related bacteraemia		3	7,844	0	25,281	7,019	4,192	3,402	15,936	.12
Infectious thrombophlebitis		3	21,210	0	65,863	17,975	11,411	10,263	41,955	1.00
ABSSSI		2	5,605	0	33,761	3,134	5,605	3,390	7,821	.07
CIED infection		2	17,492	0	162,930	16,187	17,492	6,046	28,938	.75
Bacteraemia		2	15,191	0	171,888	17,441	15,191	2,859	27,523	.47
Microorganism causing infection		n	Mean	Lower CI95%**	Upper CI95%	SD (±)	Median	Min	Max	p-Value
MSSA		13	26,706	10,675	42,737	26,528	15,465	2,859	95,142	.86
CoNS		7	29,481	11,098	47,865	19,877	41,982	3,390	47,423	.34
MRSA		4	10,768	4,641	16,896	3,851	9,884	7,369	15,936	.26
CoNS + Enterococcus faecalis		2	13,755	0	135,260	13,524	13,755	4,192	23,318	.80
Streptococcus spp.		2	20,045	0	117,992	10,902	20,045	12,337	27,754	.80
CoNS + Corynebacterium striatum		1	13,376	–	–	NA	13,376	13,376	13,376	NE
CoNS + Cutibacterium acnes		1	24,518	–	–	NA	24,518	24,518	24,518	NE
Enterococcus faecalis		1	10,801	–	–	NA	10,801	10,801	10,801	NE
Enterococcus faecium		1	37,246	–	–	NA	37,246	37,246	37,246	NE
Enterococcus faecalis + Corynebacterium striatum		1	7,783	–	–	NA	7,783	7,783	7,783	NE
Empirical treatment		1	15,143	–	–	NA	15,143	15,143	15,143	NE

Abbreviations. CIED: cardioelectronic implantable electric device (pacemaker/cardioverter-defibrillator); CoNS: coagulase-negative Staphylococci; CI: confidence interval; MRSA: methicillin resistant Staphylococcus aureus; MSSA: methicillin susceptible Staphylococcus aureus; NE: Not evaluable.

*p values <0.05; **Negative values were switch to zero.

Antibiotic therapy

Prior therapy

Thirty-one patients (91.2%) received antibiotics prior to dalbavancin: 10 (29.4%) received one antibiotic, 15 (44.1%) received two, and six (17.6%) received three or more. Altogether, patients received an average of 1.74 (SD 0.92) antibiotics prior to dalbavancin.

The most common first-line antibiotics were cloxacillin (n = 7, 22.6%), vancomycin (n = 6, 19.3%), and teicoplanin (n = 5, 16.1%). Regarding second-line treatment, daptomycin was the most used (n = 5, 21.7%). In the third-fourth treatment lines the treatments of choice were ampicillin, cefazolin, cloxacillin, or linezolid. Rifampicin was the antibiotic used as fifth line in one patient. The total costs of these five antibiotic lines were 292.4€, 284.9€, 252.3€, 49.7€, and 1.2€, respectively, with a total cost 880.5€ (SD 1,870€) and a BI of 29,939€ (Table 3). The mean total cost of administration of these previous therapies was 44.87€ (SD 26.71€) per patient.

Dalbavancin therapy: reasons for use, dosage and concomitant treatments

Overall, 73.5% patients received dalbavancin after one or two previous antibiotics. The main reasons were OPAT management (n = 21, 61.7%), to ensure treatment adherence (n = 9, 26.5%), previous therapeutic failure (n = 3, 8.8%), and refusal of daily intravenous treatment (n = 1, 2.9%).

Three patients (8.8%) received dalbavancin as a first line therapy to treat ABSSSI, bacteraemia, and prosthetic joint infection. In these cases, its use was motivated by outpatient management, microorganism resistance to other alternatives, the presence of a non-removable prosthetic material, and to ensure treatment adherence.

Eleven patients (22.3%) required a single dose, while 23 patients (67.6%) required more than one dose of dalbavancin. The mean number of administrations was 3.6 (SD 4.3, median 2 [IQR 1–3], minimum 1, and maximum 21). The mean duration of therapy was 51.7 days (SD 76.6 days), median 28 days (IQR 14–49.5); most patients (61.8%) required long-term treatment, (>2 weeks) (mean 10.7 weeks). Overall, the cohort required a mean dose of 4,400 mg (minimum 1,100 mg, maximum 18,000 mg). Details of dalbavancin posology are presented in Supplemental Table 1.

Nine patients (26.5%) required concomitant antibiotic treatment: ciprofloxacin (n = 3, 33.3%), rifampicin (n = 2, 22.2%), fosfomycin (n = 1, 11.1%), metronidazole (n = 1, 11.1%), tedizolid (n = 1, 11.1%), and trimethoprim/sulfamethoxazole (n = 1, 11.1%).

The mean time from diagnosis to dalbavancin prescription was 26.1 days (SD 24.7 days), median 23 days (IQR 9.7–30.7) and the mean time from prescription to hospital discharge was 2.5 days (SD 15.5 days), median 1 day (IQR 0.2–4). After discharge, the mean OPAT length was 47.4 days (SD 75.6 days), median 16.5 days (IQR 14–47.2).

The mean costs of dalbavancin were 3,936€ per patient (SD 4,211€) (Table 3), of which 782.8€ (SD 1,226.61€) corresponded to within-hospital administration, and the remaining 3,153.2€ corresponded to OPAT.

Clinical outcomes

Most patients (93.9%) achieved clinical success with dalbavancin treatment in combination with another antibiotic, and 6.1% achieved clinical success with dalbavancin monotherapy. In one case, the outcome was unknown due to loss of follow-up.

One patient required re-admission for the same cause after achieving clinical success with dalbavancin and another antibiotic. This re-admission lasted 9 days in the infectious diseases ward and had a cost of 2,672€. Intensive care unit (ICU) admission was not required.

Safety data

Only one patient had a mild AE (localised evanescent rash on the back of the neck) that was likely related to dalbavancin treatment 28 days after the first dose. No pharmacological treatment was required.

Hospital stays, home hospitalisation and day-care hospital

Most patients were admitted to general medical or surgical wards (n = 30, 88.2%), with a mean LOS of 21.9 days (SD 19.6 days) and a mean cost of 5,471€ (SD 4,898€). Five of these patients were admitted to the infectious diseases ward, with a mean LOS of 32.4 days (SD 12.5 days) and a mean cost of 9,618€ (SD 3,705€). The mean cost for the total population (n = 34) was 1,414€ (SD 3,690€). One patient was admitted to the ICU for 25 days, with a cost of 27,500€.

Nine patients required day-care hospital (26.5%), for a mean 2.7 days (SD 2.8 days), and the mean cost was 739€ (SD 771€). One patient required home hospitalisation for 16 days, with a total cost of 3,200€.

Overall, the 34 patients required a mean 28.6 days (SD 20.4 days) of hospitalisation, independently of home hospitalisation, day-care hospital or other department admission. The mean total cost for the general stay (including stays in general medical or surgical wards, infectious diseases ward, ICU, day-care hospital or home hospitalisation) of these patients was 8,062€ (SD 6,614€) (Table 3).

Total cost of the base case

The mean total cost per patient in the base case was 22,738€, with the greatest cost derived from interventions (8,413€) and stays (8,062€) (6,885€for general hospital stay). The mean total cost of dalbavancin was 3,936€ and 41€ for its administration (Table 3).

Costs analysis for the hypothetical scenarios

Hypothetical therapeutic alternative scenario

Daptomycin was chosen as an alternative in most cases (38%), followed by linezolid (24%) (Table 5). Days on treatment ranged from 14 to 270 days. On average, patients had 32.9 days of home hospitalisation, 16.8 days of hospital stay, and 2.85 days of day-care hospital (52.6 days in total) (Table 5).

Table 5. Characteristics of the hypothetical alternative treatment.

Pharmacological treatment
First alternative antibiotic	% Patients	Mean days on treatment	Standard deviation (±)	Mean daily dose (mg)	Standard deviation (±)
Daptomycin	38	30.38	33.9	765	153.3
Linezolid	24	54.13	87.6	1,200	–
Tedizolid	12	91.00	54.2	200	–
Vancomycin	5.9	21.00	9.9	714	404.1
Cefazolin	5.9	21.00	9.9	6,000	–
Ceftriaxone	5.9	14.50	0.7	3,000	1,414.2
Levofloxacin	2.9	168.00	–	500
Trimethoprim- sulfamethoxazole	2.9	14.00	–	160/800	–
Daptomycin + Levofloxacin	2.9	14.00	–	750	–
Second alternative antibiotic
Trimethoprim- sulfamethoxazole	40	15	–	160/800	–
Ampicillin	20	84	–	12000	–
Tedizolid	20	40	–	200	–
Levofloxacin	20	14	–	500	–
Resource use
Days	Patients	Mean	Standard deviation
Hospital stays	15	16.80	6.3
Home hospitalization	44	32.93	45.2
Day care hospital	38	2.85	2.3
Units	Patients	Mean	Standard deviation
CVC	38	24.85	27.2
PICC	18	9.67	6.7
Peripheral catheter	3	14.00	–

Abbreviations. CVC: central venous catheter; PICC: peripherally inserted central catheter.

The mean pharmacological costs were 4,832€ (SD 6,778€) vs 4,907€ (SD 5,490€) for patients who received dalbavancin (Table 6). The mean total cost was 26,062€ (SD 20,646€), which was mean 3,324€ (SD 6,129€) more expensive than treatment with dalbavancin mainly due to an increase in LOS, home hospitalisation, and day-care hospital.

Table 6. Mean cost per patient for base case scenario, hypothetical therapeutic alternative scenario, treatment with daptomycin scenario and scenario without treatment assuming hospital stay.

		Base case (€)	Hypothetical therapeutic alternative scenario (€)	Hypothetical treatment with daptomycin scenario (€)	Hypothetical scenario without treatment, assuming hospital stay (€)
Diagnostic tests		1,201	1,201	1,201	1,201
Interventions		8,413	8,413	8,413	8,413
Stay	Hospital stays	6,885	7,503	6,885	20,959
	ICU stays	809	809	809	809
	Day-care hospital stays	195	301	12,229	195
	Home hospitalization	94	2,906	–	94
	Re-admission^#	79	–	79	79
	Total	8,062	11,519	20,002	22,136
Pharmacological costs	Previous antibiotic treatment	880	880	880	880
	Antibiotic treatment: Dalbavancin	3,936	–	–	–
	Antibiotic treatment: others	–	3,772	2,995	–
	Concomitant treatment	90	180	90	90
	Total	4,907	4,832	3,965	970
Consultations	Referred patients	13	13	13	13
	Nursing	25	25	25	25
	Outpatient visit	32	32	32	32
	Total	70	70	70	70
Other costs	Previous antibiotic administration	45	45	45	45
	Drug administration	41	52	82	0
	Adverse events	0	0	0	0
	Total	86	97	127	45
	Direct costs	22,738	26,062	33,776	32,835
Mean difference between scenario and base case		–	3,324	11,038	10,097

Abbreviation. ICU: intensive care unit; ^#re-admission not associated with dalbavancin.

Hypothetical treatment with daptomycin scenario

Assuming daptomycin as the alternative treatment, the total mean cost was 33,776€ (SD 25,602€). The main increase in costs was due to increased day-care hospital utilisation, as it is the administration standard for daptomycin. The mean total pharmacological cost (3,965€, SD 4,943€) was lower than in the base case (4,907€, SD 5,490€), with the mean total cost of the drug being 2,995€ (SD 3,483€) compared with 3,936€ (SD 4,211€) for the base case (Table 6).

Hypothetical scenario without treatment, assuming hospital stay

Assuming that each dalbavancin OPAT day prevented one hospital-stay day, the mean total cost of hospital stay would have been 20,959€ (SD 23,626€), and the mean total costs avoided due to decreased LOS would have been 13,710€/patient (Table 6). In this scenario, no alternative treatment was considered, and the final mean total cost per patient was 32,835€ (SD 29,362€). Mean LOS could have been reduced in 47.4 days (SD 75.6 days) using dalbavancin, representing a total reduction of 1,612 days of hospital-stay and a total BI of 343,298€.

Probabilistic sensitivity analysis for the economic burden of the management of patients treated with dalbavancin

The results of 10,000 simulations for all the scenarios and avoided costs were in line with previous results of the analysis.

The mean total cost for the base case was 22,717€ (SD 3,400€), and the mean cost of the hypothetical therapeutic was 26,062€ (SD 2,646€), compared with the base case, the hypothetical scenario was a mean 3,324€ (SD 6,129€) more expensive.

Simulations for the daptomycin scenario revealed a mean cost of 33,747€ (SD 4,350€), entailing a mean increase of 11,032€ (SD 2,639€) compared with the base case.

The mean total cost without dalbavancin, assuming hospital stay, was 32,794€ (SD 4,983€); this scenario was a mean 10,077€ (SD 3,248€) more expensive than the base case.

Discussion

In this retrospective analysis, dalbavancin was shown to be effective and safe for the treatment of severe gram-positive infections such as osteoarticular infections and endocarditis, even after previous treatment failure. Dalbavancin was proven to be safe, as many patients required long-term treatment and only one suffered a mild AE not requiring treatment discontinuation or treatment-associated costs. Other drugs for treating gram-positive infections could potentially have related AEs, such as nephrotoxicity caused by vancomycin,¹³ or myelosuppression caused by linezolid.¹⁴

In our analysis, all patients with a known outcome who were treated with dalbavancin as monotherapy or in combination achieved clinical success. These results are more favourable than those reported in other settings, such as in an analysis performed in French hospitals in which 20% of patients experienced treatment failure.¹⁵

The reduction in patient hospitalisation during treatment with dalbavancin has previously been reported, with a reduction of 3.4 days described for ABSSSI in Spain.¹⁶ Longer reductions were observed in our analysis due to the inclusion of more severe infections that required longer treatments.

Regarding resource consumption, a published meta-analysis and cost analysis in the United States concluded that there are additional benefits for patients treated with dalbavancin for ABSSSIs, such as fewer catheter-related complications, decreased in-hospital stays, and optimisation of treatment adherence.¹⁷ These results were further reinforced in a publication supporting the use of long-acting lipoglycopeptides in deep skin infections, infective endocarditis, and osteomyelitis; the benefits outweighed other traditional approaches, especially in vulnerable populations with adherence problems,¹⁸ increasing hospital admissions and additional treatments.¹⁹

A study in an Italian hospital reported results that were aligned with ours; the authors reported savings of around 8,000€/patient when using dalbavancin, with a BI of 998,368€ (1,358 days of hospitalisation), mainly due to reduced LOS.²⁰ Furthermore, a study conducted in Spanish hospitals reported savings of almost 7,000€/patient and a mean reduction in LOS of 14 days/patient, entailing a total of 1,193 days.²¹

Savings for the healthcare system have been confirmed in previous economic evaluations in Italy, Romania, Austria, and Spain for the use of dalbavancin in ABSSSIs. Data from Spain reported an increase in drug-acquisition costs of 42.3% that was balanced by a decrease in other costs (-41.4%). Additional general savings have been calculated to be between 233.991€ and 810.650€per year in Spain.^16,22 However, these studies were based on clinical trial evidence, and the two analyses by Marcellusi et al. focused exclusively on ABSSSIs.^16,22 The authors claimed that real-world evidence would be needed to evaluate the real clinical practice. In Spain, the DALBUSE study⁹ described the use of dalbavancin in 29 Spanish centres in 2016, and despite its inclusion of costs and savings estimates, no specific data were collected for this purpose.

What can be added to the classical approach of cost savings driven by the reduced LOS is that, in the hypothetical therapeutic alternative exercise, dalbavancin was associated with slightly lower costs than the other choices, especially due to outpatient management. OPAT could entail further benefits, such as reducing the incidence of nosocomial infections, which have been reported as the fourth leading cause of death in terms of years of life lost, causing 13,000 deaths per year in Spain at a cost of more than 600 million euros.²³

One of the most important results of our analysis is that dalbavancin treatment in real clinical practice at a Spanish hospital allowed patients to be treated on an outpatient basis for a total of 1,612 days, thus reducing hospital stay and the risk of in-hospital medical complications. Despite the limited sample size from a single centre analysed in this work, which reduces the generalisability of the study, the data remain relevant for the Spanish healthcare system, including academic and non-academic medical centres.

Our scenario with daptomycin showed that both drugs have comparable costs; however, total costs would have been higher if patients were treated with daptomycin. This finding was previously reported by Turco et al. in real-world data from patients in the United States.²⁴ Interestingly, the hypothetical scenarios presented in the present analysis showed different therapeutic alternatives designed on the basis of real clinical practice in a renowned Spanish hospital; therefore, the treatment approaches corresponded to the available therapeutic arsenal and the management guidelines followed in the centre, which might differ from other hospitals.

Our analysis suggests that dalbavancin decreased hospital LOS, thereby reducing the total cost of management for severely ill patients. Furthermore, early hospital discharge may improve patient satisfaction.²⁵

There is a lack of economic information about the use of dalbavancin in different types of infections and in different therapeutic lines from the Spanish healthcare perspective. Herein, we evaluated the costs associated with treating patients with dalbavancin in real practice as a first stage that will facilitate future analytical modelling approaches. Of note, more than 90% of the patients included in our study had received at least one antibiotic treatment prior to dalbavancin. This sequential nature of the therapeutic strategies recommended for ABSSSI management makes a minimum cohort of each patient type mandatory to properly conduct subsequent cost studies. Therefore, the type of analysis described in this work may pave the way for subsequent studies on dalbavancin use that would help to improve perceptions among managers, funding entities, and decision makers in terms of the prescription of the most appropriate treatment.

Limitations

This cost analysis aimed to estimate the costs of disease management, including treatment with dalbavancin, in real clinical practice, analysing information from a single centre. However, there were several limitations regarding this analysis. Despite reporting a reduction in costs associated with dalbavancin use, future studies evaluating cost-effectiveness or cost-minimisation for this therapeutic strategy are required to properly assess the impact of dalbavancin in the Spanish healthcare system. This work provides the basis to further explore the added value of dalbavancin in the management of severe infections as an alternative to standard care.

In addition, our results might have been biased by the patient profile (severity, type of infection, frailty), which must be considered when interpreting the results. Propensity score matching was considered; however, it was limited by the sample size. The main limitation was that the data were obtained from a single hospital and cannot be extrapolated to hospitals with different characteristics. Furthermore, the days of dalbavancin treatment were assumed to be avoided days of hospital stay and thus avoided costs in the scenario without treatment (dalbavancin savings). To provide a more realistic comparison, the hypothetical alternative therapy approach was performed based on a previous study in our country and our results were in line with those reported.⁹ Furthermore, we did not address indirect costs, such as productivity loss, that could have increased the avoided cost.

Conclusions

Dalbavancin could represent a safe and effective treatment for severe gram-positive infections that would otherwise require prolonged hospital stays. In this single-centre analysis in a Spanish hospital, the total mean cost associated with managing patients treated with dalbavancin was 22,738€, with the cost of dalbavancin representing 17.3% of the total. However, in this setting, the cost of dalbavancin was compensated by the reduction in LOS. These outcomes were contrasted by a comprehensive exercise of scenarios with plausible therapeutic alternatives built from real clinical experience that suggested that, compared with dalbavancin, the cost associated with managing these patients with other antibiotics would have been higher. However, these results should be further confirmed with real-world data.

This analysis is the first step to better defining the potential of dalbavancin for the treatment of infections from an economic perspective, although subsequent studies are needed to confirm the cost consequences of employing this antibiotic in real practice.

Transparency

Declaration of funding

This analysis was sponsored by Angelini Pharma España S.L.U.

Declaration of financial/other interests

Axentiva Solutions SL is a consultancy firm working for Angelini Pharma España. Irantzu Rosselló is an employee of Angelini Pharma España.

Author contributions

Conceptualization: [Carlos Crespo, Patricia Muñoz, Irantzu Rosselló, Maricela Valerio], Methodology: [Carlos Crespo, Mónica Cerezales], Collection of data: [Maricela Valerio, Cristina Veintimilla, Carmen Rodríguez, Sofía de la Villa, Mar Sánchez Somolinos, Sara Rodríguez, Iván Adán, Ester Chamorro, Patricia Muñoz], Formal analysis and investigation: [Carlos Crespo, Mónica Cerezales], Writing – original draft preparation: [Mónica Cerezales, Carlos Crespo, Irantzu Rosselló]; Writing – review and editing: [all]

Previous presentations: This work has been previously presented at ISPOR 2021 as a poster.

Reviewer disclosures

Peer reviewers on this manuscript have no relevant financial or other relationships to disclose.

The Editor in Chief helped with adjudicating the final decision on this paper.

Acknowledgements

None stated