Multicentre insights into OPAT: a comprehensive analysis of Türkiye’s unstructured system

Abstract

Background/Objective

Outpatient parenteral antimicrobial therapy (OPAT) is a well-established, cost-effective treatment modality delivered through structured, multidisciplinary programs worldwide. In Türkiye, no standardized OPAT framework exists; patients continue therapy through emergency departments, primary care facilities, or privately funded nursing services. This is the first multicentre study to evaluate clinical outcomes, treatment adherence, and patient satisfaction with OPAT in Türkiye.

Methods

This prospective multicentre study enrolled adult patients receiving OPAT for chronic osteomyelitis, spondylodiscitis, prosthetic joint infection, or urinary tract infection across 20 hospitals in Türkiye between March and August 2024.

Results

A total of 265 patients were included (53.2% male; mean age 61.4 ± 13.5 years). Gram-positive pathogens predominated (60.2%); the most frequently used antibiotics were teicoplanin (70.5%), ceftriaxone (16.6%), and aminoglycosides (6.8%). The clinical success rate was 94.3%. ADEs led to treatment discontinuation in only 3 patients (1.1%). OPAT initiated at university hospitals was significantly more likely to be continued via emergency departments than at state hospitals (p = 0.010). Treatment-related adverse events occurred significantly more frequently outside dedicated OPAT units (18% vs. 3.1%; p = 0.030). Patient satisfaction was high overall; however, urinary tract infection patients reported significantly greater satisfaction than those with bone and joint infections across multiple domains. Additionally, patients treated in dedicated OPAT units reported significantly higher satisfaction than those treated in emergency departments across all domains assessed.

Conclusion

The absence of a structured OPAT system in Türkiye increases emergency department burden and elevates complication risk. Despite this, high clinical success rates and patient satisfaction support OPAT’s potential as a viable alternative to inpatient care. A standardized, multidisciplinary OPAT framework is needed to improve patient safety and optimize healthcare resource utilization.

Supplementary Information

The online version contains supplementary material available at https://doi.org/10.1186/s12879-026-14269-4.

Background

Outpatient parenteral antibiotic therapy (OPAT) is a treatment in which antimicrobials are administered intravenously (IV) on an outpatient basis as an alternative to hospitalization for specific indications and patient groups. OPAT is a treatment option that is widely accepted and considered to be safe. It is selected for patients with infections requiring long-term intravenous antibiotic therapy. It is evident that there is a substantial body of evidence supporting the clinical efficacy and cost-effectiveness of OPAT, given its utilisation in developed countries for a period exceeding four decades [1–4]. The implementation of OPAT within a structured programme has been demonstrated to engender a number of potential benefits, including a reduction in hospitalisation times, a decrease in the number of occupied beds, a decline in out-of-pocket expenses for patients, an enhancement in patient satisfaction, and a decrease in hospital-acquired infections [5–7]. In a significant number of cases of infections requiring parenteral antibiotic therapy, even if clinical stability is achieved, patients often remain in hospital solely for the continuation of intravenous or intramuscular antibiotic therapy. This is due to the lack of an appropriate sequential oral treatment option or insufficient oral bioavailability. The aforementioned issue has been shown to exert a significant strain on hospital bed capacity, resulting in an increase in workload and a disruption to patients’ ability to continue their daily lives. This is particularly evident in cases requiring long-term treatment, such as chronic osteomyelitis, spondylodiscitis, and prosthetic joint infections.

In Türkiye, no national standardised OPAT framework exists. Patients continue treatment through emergency departments, family medicine clinics, or privately funded nursing services, resulting in fragmented care and limited outcome data. Public hospitals in Türkiye are organised into secondary-care (district and state) hospitals, tertiary-care (training-and-research) hospitals, and university hospitals, the latter two managing more complex referrals with greater subspecialty and diagnostic capacity. Antimicrobial availability in the outpatient setting is governed by the national Health Services Regulation, which specifies the agents permissible for administration and thereby shapes the antibiotic options available for OPAT.

This multicentre study evaluates clinical characteristics, implementation practices, and patient-reported outcomes of OPAT across centres from different regions of Türkiye, aiming to provide evidence to support the development of a standardised national OPAT framework.

Materials and methods

The design, setting and participants of the study

This prospective multicentre study enrolled adult patients at 20 voluntarily participating hospitals across six regions of Türkiye (nine university, seven secondary-care, four tertiary-care hospitals) between 1 March and 1 August 2024, with follow-up completed by 1 October 2024. Patients with chronic osteomyelitis, spondylodiscitis, prosthetic joint infection, or urinary tract infection who initiated OPAT were monitored throughout treatment and for 30 days thereafter; primary outcomes included adverse effects, rehospitalisation, surgical debridement, and treatment non-response. Eligible patients were aged ≥ 18 years, clinically stable (afebrile ≥ 48 h, normal vital signs and blood glucose, no need for close monitoring or concurrent surgical intervention), with confirmed susceptibility to the prescribed antibiotic, and receiving agents permissible under the national Health Services Regulation (e.g. teicoplanin, ceftriaxone, aminoglycosides, piperacillin-tazobactam). Patients under 18 years, those receiving a single parenteral dose, or those treated outside guideline-specified antibiotics were excluded. All participants provided written or verbal informed consent.

Data collection and analysis

Clinical data (physical examination, medical history, laboratory tests, culture results, imaging) were extracted from hospital automation systems by research assistants.

In addition to data obtained from clinical records, patients’ treatment adherence during the OPAT process, the manner in which OPAT was administered, and conditions of access to treatment were assessed using a structured 10-questions patient questionnaire developed for this study (an English-language version is provided as Supplementary File 1). The questionnaire consisted of closed-format items; although an ‘other’ option was available for a few questions, these entries were not analysed as free-text data and no narrative patient commentary was collected. No remuneration or incentive was provided to patients, and participation was entirely voluntary.The questionnaire comprised items relating to the occurrence of any issues during treatment and the nature of these issues, the reasons for opting for outpatient treatment, adherence to dosage and timing, reasons for non-adherence, the status of payment for treatment access and the method of payment, as well as dressing requirements and the individual responsible for providing the dressings. The response formats were recorded as binary (yes/no), ordinal categorical, or multiple-choice options, depending on the question type. In questions allowing multiple responses, each option was coded as a separate variable. The assessment of treatment adherence was not confined exclusively to patient self-reports; it was interpreted in conjunction with inferential indicators derived from treatment flow and administration records entered into the database by physicians/research assistants at the centres (e.g. planned treatment duration, administration setting, records of interruptions or delays in dose administration). In instances where substantial discrepancies were identified between patient-reported data and data derived from medical records, data verification was conducted by contacting the relevant centre.

Patient satisfaction was measured using a 10-item, 5-point Likert-scale survey (1 = strongly disagree; 5 = strongly agree), adapted from previously published OPAT patient-satisfaction instruments [8, 9] and the Patient Satisfaction Questionnaire-3 (PSQ-3) [10], covering follow-up, examination, laboratory monitoring, daily activity continuity, and overall satisfaction, completed in person at treatment conclusion (an English-language version is provided as Supplementary File 2).

The data collection form, prepared by the principal investigator, the treatment adherence and implementation conditions questionnaire, the patient satisfaction questionnaire and the database file (Microsoft Office Excel^®) were shared via email with the research assistants at the centres wishing to participate in the study. The principal investigator was responsible for consolidating all data from the relevant centres and ensuring its consistency prior to analysis. Discrepancies or conflicts in the data were addressed through consultation with the respective centre’s researcher, with corrections being applied wherever feasible. The patient data were processed in an anonymized manner.

All clinical outcomes were evaluated during OPAT and within 30 days following its completion. Clinical response was defined as marked improvement or resolution of symptoms with no need for additional systemic antibacterial therapy; sustained response required the absence of clinical worsening, antibiotic escalation, or new antibacterial therapy during 30-day follow-up. Adverse drug events (ADEs) were defined per ICH E2A criteria [11]. Rehospitalisation was defined as unplanned emergency admission during or within 30 days after OPAT; elective admissions were excluded.

Clinical response and adverse drug events were assessed by the treating infectious diseases physician at each participating centre, who was also the site investigator, according to these predefined criteria; a formal independent dual adjudication was not performed, but all records were reviewed and verified centrally by the principal investigator, with discrepancies resolved through direct consultation with the responsible site investigator. Laboratory monitoring was performed at scheduled control visits rather than on a fixed weekly schedule, reflecting the absence of a standardised national monitoring protocol; a complete blood count, biochemistry, and C-reactive protein were the tests most frequently obtained.

Statistical analysis

Continuous variables were presented as mean ± standard deviation or median (range), based on their distribution characteristics; categorical variables were presented as counts and percentages. Responses to the treatment adherence and implementation conditions questionnaire were analysed as categorical or ordinal categorical variables; for items allowing multiple responses, each option was coded as a separate binary variable, and counts and percentages were calculated for each option. Satisfaction survey items were treated as 5-point Likert-type (ordinal) scales, and response distributions for each item were reported as counts and percentages; in analyses deemed necessary, “positive responses” were presented by combining the “agree” and “strongly agree” options. In comparisons based on center types and/or the settings where OPAT was implemented; for the analysis of continuous variables, the Student’s t-test was used for independent groups when parametric conditions were met, and the Mann–Whitney U test when they were not; for categorical variables, the chi-square test or Fisher’s exact test was used; and logistic regression analysis was employed to identify independent parameters. Associations derived from logistic regression were expressed as odds ratios (ORs) with 95% confidence intervals. In all statistical tests, a p-value < 0.05 was considered statistically significant.

Results

Demographic and clinical characteristics

A total of 265 patients were included (53.2% male; mean age 61.4 ± 13.5 years; median 63, range 22–94). Comorbidities were present in 76.2%; the most common were diabetes (27.1%), hypertension (24.1%), and coronary artery disease (12.0%) (Table 1).

Table 1.

Comprehensive profile of OPAT patients: demographics and treatment characteristics

	n	(%)*
Age
Mean ± SD	61.4 ± 13.5
Median (min-max)	63 (22–94)
Gender
Male	141	53.2
Female	124	46.8
Comorbidity
Yes	202	76.2
No	63	23.8
Reason for continuing treatment via OPAT*
To reduce hospitalization costs	119	44.9
Insufficient hospital bed capacity	112	42.2
Patient preference	51	19.2
Other	5	1.8
OPAT indications*
Chronic osteomyelitis	105	39.6
Spondylodiscitis	24	9.1
Prosthetic joint infection	68	25.7
Urinary tract infection	68	25.7
Treatment facility
University hospital	86	32.5
Secondary care state hospital	75	28.3
Tertiary care state hospital	104	39.2
OPAT administration location*
OPAT treatment unit	66	24.9
Emergency department/state hospital	124	46.7
Emergency department/university hospital	11	4.1
Private healthcare facility	24	9
Home	15	5.6
Other	18	6.8
OPAT duration (days)
Mean ± SD	21.8 ± 11.9
Median (min-max)	14 (1-120)
OPAT application method
Intravenously	153	57.8
Intramuscularly	112	42.2
Reasons for OPAT completion
Upon achievement of culture negativity	21	7.9
Upon achievement of clinical response	156	58.8
Due to completion of treatment duration	109	41.1
Patient discontinued treatment voluntarily	15	5.6
Due to serious adverse effect	3	1.1
Other	19	7.1
Requirement for patient rehospitalization
Yes, rehospitalization required	37	13.9
No, rehospitalization not required	228	86.1

*Percentage of columns

**Note: Due to overlap among variables, total percentages may exceed 100%

Presenting infections

OPAT indications were chronic osteomyelitis (39.6%), prosthetic joint infection (25.7%), urinary tract infection (25.7%), and spondylodiscitis (9.1%) (Table 1).

Microbiological findings

Therapy was empiric in 162 patients (61.1%) and culture-directed in 103 (38.9%). Culture specimens were obtained from 211 patients; a causative organism was identified in 128, while the remaining 137 comprised 54 with no culture and 83 with negative cultures. Gram-positive organisms predominated. Staphylococcus aureus was isolated in 35 patients (27 MRSA, 8 MSSA), and methicillin-resistant coagulase-negative staphylococci in a further 16; overall, methicillin-resistant staphylococci accounted for 43 of the 128 culture-positive patients (33.6%). ESBL-producing Escherichia coli (n = 18) and Klebsiella spp. (n = 9) were the leading Gram-negative pathogens. Taken together, 70 of the 128 culture-positive isolates (54.7%) were ESBL-producing or methicillin-resistant. The distribution of isolated microorganisms is shown in Table 2.

Table 2.

Distribution of microorganisms isolated from culture-positive patients (n = 128)

Microorganism	n (%)
MRSA	27 (21.1)
MSSA	8 (6.2)
Methicillin-resistant CoNS	16 (12.5)
Other coagulase-negative staphylococci	8 (6.2)
ESBL-producing E. coli	18 (14.1)
E. coli (non-ESBL)	10 (7.8)
ESBL-producing Klebsiella spp.	9 (7.0)
Klebsiella spp. (non-ESBL)	4 (3.1)
Enterococcus spp.	13 (10.2)
Corynebacterium spp.	5 (3.9)
Other	10 (7.8)
Total	128 (100.0)

Antibiotics, route of administration, and administration setting

The most frequently used antibiotics were teicoplanin (70.5%), ceftriaxone (16.6%), and aminoglycosides (6.8%). Dosing was once-daily in 238 patients (89.8%) and less frequent (thrice-weekly or every 48 h) in 7, while only 20 (7.5%) required two or more daily doses. Administration was intravenous in 57.8% and intramuscular in 42.2% of patients. Mean OPAT duration was 21.8 ± 11.9 days (median 14, range 1–120); mean prior hospitalisation was 11.67 ± 9.59 days. In 78% of patients, the OPAT antibiotic matched the in-hospital regimen. The main reasons for outpatient continuation were cost reduction (44.9%) and insufficient bed capacity (42.2%) (Table 1).

The most common administration settings were the emergency department (46.7%) and dedicated OPAT units (24.9%). OPAT was nurse-administered in 96.6% of cases; however, 3 patients (1.1%) received administration at home by non-healthcare professionals (Table 1). Of the 153 patients (57.7%) who received intravenous therapy, access was peripheral in 150 and via a central catheter or implanted port in only 3; the remaining 112 patients (42.3%) received intramuscular therapy. Vascular access–related events were infrequent, comprising 9 access-site problems and 3 episodes of thrombophlebitis.

Clinical course, readmission, and safety

Clinical success was achieved in 94.3% of patients. Treatment completion was most commonly due to clinical response (58.8%) or planned duration (41.1%); ADEs led to discontinuation in only 3 patients (1.1%). Unplanned rehospitalisation occurred in 37 patients (14.0%), most commonly for surgical debridement (n = 19) or infection relapse (n = 7) (Table 1).

Antimicrobial resistance was not associated with worse outcomes: among the 128 culture-positive patients, clinical failure occurred in 0/70 (0.0%) with a resistant organism versus 3/58 (5.2%) with a susceptible organism (Fisher’s exact p = 0.090), and unplanned readmission in 9/70 (12.9%) versus 12/58 (20.7%) (p = 0.247). Neither MRSA (failure p = 1.00; readmission p = 0.390) nor ESBL-producing organisms were associated with worse outcomes; the lower readmission observed among ESBL isolates (0% vs. 20.8%, p = 0.007) reflected their almost exclusive derivation from urinary tract infections rather than a protective effect of resistance.

Access to treatment, challenges encountered during the process and treatment adherence

The outcomes of the process, as assessed based on patient reports, are presented in Table 3. While 205 patients (77.3%) reported experiencing no issues during the treatment process; among those reporting issues, procedure-related problems (17, 6.4%) and antibiotic-related adverse effects (10, 3.8%) were the most frequently reported issues; the most common drug-related adverse effect was the development of diarrhea (7 patients), though this diarrhea was not reported as being associated with Clostridioides difficile (Table 3). 236 patients (90.1%) reported not making any additional payments for OPAT; details regarding payment situations and who covered the cost of dressing changes are provided in Table 3 (among those requiring dressing changes, 44.5% were performed at home by family members, and 36.3% were performed at a healthcare facility for a fee) (Table 3).

Table 3.

Conditions affecting patient adherence to OPAT

Conditions	n	(%)*
Continuation of daily activities during OPAT
Yes, continued daily activities	233	87.9
No, did not continue daily activities	32	12.1
Issues experienced by the patient during OPAT
No issues experienced	205	78.2
Antibiotic-related adverse effect	10	3.8
Issues related to drug administration method	17	6.4
Inappropriate treatment course	1	0.4
High treatment cost	10	3.8
Other	16	6
Adherence to medication schedule
Yes, I adhered to all administration times	180	68.7
Except for 1–2 doses, I adhered to others	65	24.8
I improperly followed my 4–6 dose schedule	11	4.2
I mostly failed to adhere to administration times	6	2.3
Patient adherence to treatment schedule
Completed all doses	231	88.2
Missed 1–2 doses	26	9.9
Missed 4–6 doses	3	1.1
Missed less than half of doses	1	0.4
Missed more than half of doses	1	0.4
OPAT treatment expenses covered by patient
No, extra payment required from patient	236	90.1
Yes, payment made to private healthcare facility	16	6.1
Yes, payment provided to home care healthcare personnel	6	2.3
Other payment	4	1.5
Wound dressing application status in patients needing dressing
At home, administered by a healthcare professional for a fee	2	1.9
At home, administered by my relatives	49	44.5
At the healthcare facility, provided for a fee	40	36.3
At the healthcare facility, provided for free	18	16.3
My wound dressings were not done regularly	1	1

Percentages for questionnaire items are calculated over 262 respondents (3 patients did not complete the questionnaire)

Overall, 88.2% of patients completed all doses and 68.7% adhered fully to the dosing schedule; 87.9% continued daily activities during OPAT (Table 3).

Among non-adherent patients, the main barriers were distance to the treatment facility (34.3%), personal workload (24.3%), and difficulty finding an administration site (14.3%). Patients who had to pay out-of-pocket for OPAT administration were significantly more likely to be non-adherent than those who did not (8/26, 30.8% vs. 23/236, 9.7%; Fisher’s exact p = 0.005).

Patient satisfaction

Positive responses predominated across all satisfaction domains (Fig. 1). Patients with urinary tract infections reported significantly higher satisfaction than those with bone and joint infections across multiple domains, including follow-up regularity, physical examination, laboratory monitoring, physician access, perceived improvement, overall satisfaction, and willingness to recommend OPAT (all p ≤ 0.025; Table 4).

Fig. 1.

Patient satisfaction survey results graphic

Table 4.

Comparison of patient-reported outcomes between bone and joint infections (prosthetic joint infection, osteomyelitis, spondylodiscitis) and urinary tract infections (Independent Samples t-test Analysis)

Parameter	Mean ± SD		p value
	Bone and joint infections (n = 194)	Urinary tract infections (n = 68)
Regular follow-up	4.25 ± 0.87	4.68 ± 0.50	< 0.001
Thorough physical examination	4.25 ± 0.83	4.59 ± 0.58	0.002
Necessary laboratory investigations	4.31 ± 0.80	4.69 ± 0.50	< 0.001
Uninterrupted daily activities	3.99 ± 1.10	4.28 ± 0.90	0.056
Access to physician	4.09 ± 0.98	4.41 ± 0.79	0.017
Need for wound dressing	3.91 ± 1.22	4.00 ± 1.21	0.680
My condition improved	4.07 ± 1.10	4.51 ± 0.70	< 0.001
Treatment satisfaction	4.11 ± 1.06	4.43 ± 0.78	0.025
I would prefer OPAT	3.95 ± 1.14	4.31 ± 0.86	0.008
I would recommend OPAT to my relatives	4.02 ± 1.10	4.40 ± 0.85	0.010

Of greater relevance to service organisation, satisfaction differed markedly by administration setting: patients treated in a dedicated OPAT unit reported significantly higher satisfaction than those treated in emergency departments across every domain assessed (all p ≤ 0.041, Mann–Whitney U test), including overall satisfaction (mean 4.38 vs. 4.01, p = 0.002) and willingness to choose OPAT again (4.31 vs. 3.83, p < 0.001). This setting-based comparison bears more directly on our central conclusion than the comparison by infection type (Table 5).

Table 5.

Patient satisfaction by administration setting: dedicated OPAT unit versus emergency department (Mann–Whitney U test)

Satisfaction domain	OPAT unit (mean)	Emergency dept. (mean)	p
Regularity of follow-up	4.51	4.19	0.012
Thoroughness of examination	4.48	4.20	0.041
Adequacy of laboratory monitoring	4.62	4.26	0.001
Physician accessibility	4.43	4.04	0.002
Perceived clinical improvement	4.28	4.05	0.024
Overall satisfaction	4.38	4.01	0.002
Willingness to choose OPAT again	4.31	3.83	< 0.001
Willingness to recommend OPAT	4.31	3.93	0.008

Comparative analyses

It was found that OPAT, when initiated in university hospitals, was continued via the emergency department at a significantly higher rate compared to state hospitals (p = 0.010).

The incidence of adverse drug reactions was found to be higher among patients treated outside the OPAT unit compared to those treated within the OPAT unit (p = 0.030).

Discussion

This study provides the first multicentre real-world data on OPAT in Türkiye. The absence of a standardised national framework results in heterogeneous practices across settings. Critically, adverse drug reactions were significantly more common outside dedicated OPAT units, consistent with IDSA guidelines emphasising systematic monitoring as mandatory for safe implementation [12]. A meta-analysis of 13 studies confirmed that organised systems are prerequisite for equivalence with inpatient care [13], and inclusion of an infectious disease pharmacist was shown to reduce serious ADE risk by 74% (OR: 0.26; p = 0.01) [14]. Clinician-supervised OPAT achieves higher success rates with lower readmission and medication-change rates [15–17], underscoring that organised OPAT is a fundamental requirement for patient safety.

Emergency department overreliance is a major structural concern. University hospital-initiated OPAT was disproportionately continued via emergency departments, likely reflecting limited bed capacity. Turkish emergency departments managed over 129 million visits in 2021 [18], making them poorly suited to planned long-term OPAT follow-up. Organisational deficiencies in OPAT are associated with increased emergency visits and readmissions [19]; even emergency department-based models shown feasible in paediatric populations operated within structured frameworks not present in our setting [20].

Despite structural limitations, high clinical success (94.3%), low readmission (14.0%), and overall high patient satisfaction support OPAT’s viability as an inpatient alternative. Satisfaction rates of 83–97% are consistently reported internationally [8], and OPAT participants demonstrate improved quality of life within four weeks of discharge [21, 22].

The significantly higher satisfaction in UTI patients compared to those with bone and joint infections likely reflects the distinct nature of these conditions. UTIs are typically acute with short, predictable treatment courses, whereas chronic osteomyelitis, spondylodiscitis, and prosthetic joint infections involve prolonged therapy, high recurrence risk, and complex wound care that adversely affect quality of life and satisfaction [23]. To our knowledge, comparative satisfaction analyses by OPAT indication remain scarce, and our findings offer a novel contribution to the literature.

More relevant to service organisation than this infection-type difference, however, was the finding that satisfaction was significantly higher among patients treated in dedicated OPAT units than among those managed through emergency departments, across every domain assessed. This setting-based difference speaks directly to the central argument of this study: structured, dedicated delivery yields a superior patient experience relative to fragmented emergency-department care, reinforcing the case for a coordinated national OPAT framework. Financial burden also appeared to compromise adherence: patients who paid out-of-pocket for administration were significantly more likely to be non-adherent than those who did not (30.8% vs. 9.7%, p = 0.005), underscoring that a funded, structured OPAT pathway is important not only for safety but also for treatment completion.

The infection distribution and predominant causative organisms are consistent with the OPAT literature. The high prevalence of bone and joint infections reflects the necessity for prolonged parenteral therapy in these conditions [24]. Gram-positive predominance and teicoplanin as the most used antibiotic align with established glycopeptide preference in bone and joint infections, where once-daily dosing supports adherence [25].

Clinical success exceeding 90% and a readmission rate of 14% are consistent with large cohort studies reporting success rates up to 86% in bone and joint infections [26, 27].

This is among the first multicentre prospective studies to evaluate OPAT organisation and clinical outcomes nationwide in Türkiye, enabling reliable assessment of process indicators including ADEs, treatment adherence, and patient experience under real-world conditions.

This study has several limitations. The observational, single-arm design without a comparator group precludes conclusions regarding the efficacy of OPAT relative to inpatient care. Clinical outcomes and adverse events were determined by the treating site investigator without independent dual adjudication, which may introduce assessment bias, and questionnaire completion under physician supervision may have introduced response bias. Although laboratory monitoring was routinely performed, longitudinal laboratory trends were not analysed as a safety outcome. Operative source-control procedures were not systematically recorded unless they were documented as a reason for rehospitalisation. Therefore, the total number of patients with chronic osteomyelitis or prosthetic joint infection who underwent operative debridement before or during OPAT could not be determined.Subgroup sample sizes were modest and may limit statistical power, although findings were consistent across centres. A formal cost analysis could not be performed given the heterogeneous treatment settings. Finally, external validity is limited: our findings derive from a healthcare system without a structured OPAT framework and reliant on a specific national formulary (predominantly teicoplanin, ceftriaxone, and aminoglycosides), and may not be directly generalisable to systems with established OPAT programmes or different prescribing patterns.

In conclusion, this study demonstrates that OPAT treatments in Türkiye are not conducted within a standardized program, and that this is reflected in clinical outcomes. The fact that a significant portion of these treatments is administered through emergency departments increases the burden on emergency departments, makes effective follow-up difficult, and raises the risk of complications—which can be directly linked to the higher incidence of adverse drug reactions in settings outside of dedicated OPAT units. However, high clinical success rates, low readmission rates, and overall high patient satisfaction indicate that OPAT can serve as a genuine and effective alternative to hospitalization when appropriate conditions are met. The establishment of a standardized, multidisciplinary OPAT framework could improve the patient experience, increase treatment adherence, and contribute to the creation of a safer and more accessible antimicrobial treatment system across Türkiye by reducing healthcare costs.

Supplementary Information

Below is the link to the electronic supplementary material.

Abbreviations

OPAT

Outpatient parenteral antimicrobial therapy

IV

Intravenous

ADE

Adverse drug event

UTI

Urinary tract infection

ESBL

Extended-spectrum β-lactamase

IDSA

Infectious Diseases Society of America

PSQ-3

Patient Satisfaction Questionnaire-3

OR

Odds ratio

SD

Standard deviation

IQR

Interquartile range

Author contributions

E.A.B. contributed to all stages of the work, including the conception, design, data collection, analysis, interpretation, drafting, and critical revision of the manuscript. E.A. served as the supervisor, contributed to the study design, and substantively revised the manuscript text. B.C., H.A., Y.G., A.T., F.N.A.B., L.N.A., B.M., P.K., C.S., S.Ö.A., E.Y.İ., Ç.M., A.Ş.K., M.N.Ö., N.S., G.D., U.Ö., A.D., M.S.T., M.Ç., D.Ç.Ö., S.K., S.K., Y.Ç.K., Ö.E.K., H.D., and B.T. contributed to the acquisition of data for the work. All authors reviewed, edited, and approved the final version of the manuscript to be published.

Funding

This study received no external funding.

Data availability

The datasets generated and/or analysed during the current study are available from the corresponding author, Elif Agüloğlu-Bali (elifaguloglubali@gmail.com), on reasonable request.

Declarations

Ethics approval and consent to participate

This study was conducted in accordance with the Declaration of Helsinki. Ethics approval was obtained from the Etlik City Hospital Ethics Committee (Decision No.: AEŞH-EK1-2024-0011, Date: 14.02.2024). All participants provided written or verbal informed consent prior to participation.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.