Le Infezioni in Medicina, n. 3, 284-293, 2025

doi: 10.53854/liim-3303-5

ORIGINAL ARTICLES

Attitudes towards antibiotic use in end-of-life care: a nationwide Italian physician survey

Matteo Moroni1, Raffaella Antonione2, Eleonora Taddei3, Luigi Patil4, Massimo Fantoni3,5

1Struttura Semplice Dipartimentale Cure Palliative Ravenna, 48022 Lugo di Romagna, Italy;

2Rete Cure Palliative e Hospice Area Giuliana, Azienda Sanitaria Universitaria Giuliano-Isontina, 34129 Trieste, Italy;

3UOC Malattie Infettive, Fondazione Policlinico Universitario Agostino Gemelli IRCCS, 00168 Roma, Italy;

4Dipartimento Tecnologie - Servizio Soluzioni Sanità Digitale, Azienda Provinciale per i Servizi Sanitari Provincia Autonoma di Trento, 38123 Trento, Italy;

5Dipartimento di Sicurezza e Bioetica, Università Cattolica del Sacro Cuore, 00168 Roma, Italy.

Article received 2 June 2025 and accepted 1 August 2025

Corresponding author

Eleonora Taddei

E-mail: eleonora.taddei@policlinicogemelli.it

SummaRY

Background: Many patients receive antibiotics at the end of their lives: prudent use in this setting is essential to limit side effects and selective pressure. Evaluating benefits and harms of antibiotics in this context is complex, with many factors influencing final decisions, including transition to end-of-life care and diagnosis of impending death. We aimed to investigate attitudes towards antibiotic prescription among key specialists involved in end-of-life care, also as part of a consensus on antibiotic prescription at the end of life.

Materials and Methods: An anonymous ten-question survey was conducted in January 2025, over a one-month period, on the websites of the Italian Society of Infectious and Tropical Diseases (SIMIT), the Italian Society of Palliative Care (SICP), the Italian Society of General Practitioners (SIMG), and the Italian Federation of Associations of Internal Medicine Physicians (FADOI), with a denominator of 9224 potential respondents. Specialists were invited to respond questions about usual practice with both end-stage oncological and non-oncological patients. A descriptive analysis of the aggregated data was performed.

Results: 880 physicians (9.5% of the denominator) participated (59.2% female), most working in hospitals (56.7%) and internal medicine departments (39.9%). A plurality (40.8%) had less than 10 years of work experience. Attitudes towards infection management and antibiotic prescription varied widely. More than half of the respondents reported treating infections differently in oncological versus non-oncological patients. Only a minority used antibiotics to treat respiratory secretions. Most could not rely on institutional guidelines for antibiotic prescription in terminally ill patients.

Conclusion: The attitude towards antibiotic prescription at the end of life is influenced by many factors, including the physician’s background and setting. Case-by-case decision-making and advanced care planning could be significantly supported by guidelines based on studies that provide real-life data and effectively stratify patients and scenarios.

Keywords: End-of-life care, antibiotic therapy, medical futility, palliative care, survey study.

INTRODUCTION

With the advent of antibiotics, treatment of infections shifted from supportive measures to curative possibilities, making bacterial infections the prototype of curable diseases. However, bacterial infections are common in the final stages of life and may even be considered part of the normal dying process [1, 2].

Evidence on the benefits of antibiotic therapy in reducing discomfort for suspected or confirmed infections at the end of life is conflicting and guidelines are scarce.

Patients often experience high levels of antibiotic exposure at the end of life [6, 7], even in palliative care or hospice settings, suggesting that antibiotic prescription in this context could benefit from stewardship interventions [6-9].

This study aimed to investigate the attitudes of physicians towards managing infections and prescribing antibiotics for end-of-life care, identifying key topics for guidelines and targets for antibiotic stewardship. This work is preparatory to a consensus on antibiotic prescription at the end of life by the Società Italiana di Malattie Infettive e Tropicali (SIMIT), Società Italiana di Cure Palliative (SICP), Società Italiana di Medicina Generale (SIMG), and Federazione delle Associazioni dei Dirigenti Ospedalieri Internisti (FADOI).

MATERIALS AND METHODS

An anonymous survey was conducted from January 1 to January 31, 2025, over a one-month period, promoted via the institutional websites of the involved scientific societies and via email. At the time of survey dissemination, 9224 (cfr abstract) potential respondents were reached, as SIMIT had 1638 members, SICP 1126 members, SIMG 3450 members and FADOI 3010 members, based on the mailing lists of each society we had access to. The survey collected demographic data, working settings, and specialty areas of participants. Ten questions addressed the management of terminally ill patients, both with and without cancer (Table 1).

Table 1 - Results of the survey.

For the purpose of this survey, “end-stage” was defined, for cancer patients, as metastatic or locally advanced cancer with a performance status of ECOG ≥3 and, for non-cancer patients, based on the criteria of the Italian position paper on end-stage organ failures by SIAARTI [10].

The questionnaire was developed by a panel of experts from Infectious Diseases, Palliative Care, Internal Medicine, and General Practice specialties, based on the most relevant issues in the field of antibiotic use at the end of life (i.e. management of fever, withdrawal and withholding of antibiotic therapy, timespan of therapies, antibiotic stewardship, communication with patients, families and proxies, advanced care planning), as emerging from recent literature and guidelines (Figure 1).

Figure 1 - Graphs of responses from Question 1, Question 2, Question 3, Question 4 and Question 9 of the survey.

Descriptive statistics were applied to the aggregated results.

Consent was waived due to non-interventional nature of the study and guaranteed anonymity of physicians participating in the survey. Ethical approval (ID 7270) was obtained by the territorial Ethics Committee of the Center promoting the study (Comitato Etico Territoriale Lazio Area 3).

RESULTS

A total of 880 physicians completed the survey (59.2% females). Most of them were working in Internal Medicine (39.9%) and Palliative Care (24.8%) settings, with hospitals being the most common workplace (56.7%).

The majority agreed that antibiotics should be prescribed for suspected or confirmed bacterial infections in terminally ill patients. In the presence of fever, most participants considered underlying pathologies and prognosis for further management. More than half respondents would treat infections differently whether a terminally ill patient has end-stage neoplasm as compared to a non-oncologic disease; however, this was less frequent among physicians working in a palliative care setting.

The primary reason for not starting or withdrawing antibiotics were signs of imminent death, followed by severe side effects. Internal Medicine physicians tended to rely on family members’ consent more frequently for starting antibiotics, while General Practitioners and Palliative Care physicians often followed the patients’ choices.

Coherent with reasons for withholding/withdrawing antibiotics, the most frequent reason for prescribing antibiotics for prolonged fever (>72 hours) in a non-septic patient was a life expectancy of at least 48 hours.

Regarding therapy timespan, most prescribers were uncertain or disagreed with stopping pneumonia treatment after five days despite favorable response, except for Infectious Diseases physicians. Only a minority of physicians considered using antibiotics exclusively to reduce dyspnea and respiratory secretions; in particular, about half of physicians working in Palliative Care (PC) agreed with this use.

The preferred antibiotics duration for upper urinary tract infection treatment was seven days.

Finally, most respondents agreed with sharing care planning with a legal representative for non-competent patients and only a minority of physicians reported that their institutions had guidelines for antibiotic use in terminally ill patients.

Complete results of the survey are reported in Table 2.

Table 2 - Complete content of the survey

DISCUSSION

This survey aimed to investigate the attitude of physicians from different specialties towards the management of suspected or confirmed infections and antibiotics prescription at the end of life.

The results highlight the complexity of antibiotic usage as perceived by respondents, who appear to consider several factors in their decision-making.

Imminent death was a common reason to withdraw antibiotics in this survey, which is very reasonable but differs from what is reported in some real-life experiences: nearly half of terminally ill patients receive antibiotics up to the day of death, and deterioration discourages deprescribing, as reported in a recent antibiotic stewardship intervention study in Singapore [11].

Additionally, a previous survey conducted by Crispim et al., based on clinical vignettes, showed a definite tendency towards antibiotic prescription and escalation in end-of-life patients, even among palliative care physicians [12]. While stopping antibiotics due to severe side effects, a prevalent attitude in our survey, is common sense, imminent death seems to pose a greater dilemma for prescribers.

Accurate prediction of impending death (i.e., last few days of life) is essential in terms of supporting patients, families, caregivers, and healthcare professionals to clarify goals of care, promote shared decision-making process, avoid aggressive care, and achieve a good death and best possible quality of life.

In addition, a proper diagnosis of impending death improves a better communication of prognosis with patients and families and helps in End-of-Life care, also to provide better symptom management.

Although many international studies have been conducted to investigate signs and symptoms of impending death both in cancer patients (CPs) and in non-cancer patients, uncertainty in prognostication is still a major issue [13-15]. Further research is needed to allow a better identification of end-of-life patients and a better diagnosis of impending death. Dissemination of some skills from Palliative Care Physicians to other specialists is desirable.

When investigating reasons to antibiotic prescription for terminally ill patients, Durand et al. found that it was frequently reported as symptom driven [16]. We found that many respondents consider peak body temperature as a reason to treat fever with antibiotics, perhaps for the same palliative reason. It has been shown that fever response is often blunted in elderly people, but no correlation between higher body temperature and bacterial etiology or worse prognosis has been established [17]. Prescribing antibiotics to manage respiratory secretions was uncommon among all categories of respondents. In fact, this is debated in literature, with conflicting results about the benefits even in confirmed lower respiratory tract infections [3, 4].

The setting of care was the third most-commonly mentioned factor when deciding to start antibiotics for fever. Respondents were not asked to justify their responses further, but this is coherent with “No administration route” being the fourth most common reason to withdraw antibiotics. Additionally, physicians working in the hospitals (ID and IM) seem to value diagnostic testing before antibiotic prescription, possibly because they generally have more resources compared to those working outside of hospitals.

The setting of infection treatment is indeed very relevant, as it may limit feasible care, truncating the decision-making process from the beginning. For example, in a modern scenario of growing bacterial resistance, especially in our country, antibiotic therapies available outside hospitals are becoming predictably ineffective for many pathogens. In this context, starting an antibiotic without microbiological data may be questionable.

IM physicians are most represented among those who share decisions about antibiotic therapies with families, while GP and PC physicians more often share decisions with the patient. This is coherent with PC physicians being specifically trained to comprehensively assess and manage the physical, psychological and spiritual distress of patients and their social needs, determining goals of care, recognizing and respecting their cultural values; likewise, GPs often support patients through their whole lifespan and disease trajectory, becoming a crucial counterpart in decision making, including during advanced disease and terminal phases of life. On the other hand, IM physicians may often lack the opportunity to build a long-lasting relationship with their patients, due to in-hospital care dynamics (shorter inpatient stay, focus on acute care) as well as they may lack specific training in addressing end-of-life conversations.

Another survey by Larnard et al. showed that patient/family preference was a leading factor in prescribing antibiotics during the transition to comfort measures only [18]. A retrospective study by Servid S et al. found that almost one in four antibiotic prescriptions for terminally ill patients was attributed to patients/family requests [19]. The ability to share difficult decisions directly with patients is a skill worth cultivating, even beyond end-of-life care settings. At the same time, complying with patients and caregivers’ requests should be balanced with the physician’s view of what is the best decision. Physicians and Palliative Care teams should indeed be prepared to provide guidance to share responsibility and relieve all parties from uncertainty and guilt as much as possible [20].

As previously stated in literature, promoting Advanced Care Planning (regardless of the setting) can help both healthcare professionals and patients (together with caregivers, or the proxy in case of non-competent patients) to share a decision-making process to avoid medical futility and to respect patients’ wills and needs [21-24]. Conversely, timely discussion of goals of care appears infrequent, as reported in a previous survey conducted in USA nursing homes [25].

Another interesting finding of our survey is that more than half of the respondents in all categories - except for PC physicians - declared to consider oncologic and non-oncologic terminally ill patients differently when deciding the management of a suspected infection. This may be due to the hard skills and expertise in end-of-life care by palliative care specialists, which is worth disseminating as a soft skill among healthcare professionals in the modern context of an aging population.

Regarding the timespan of therapies, ID physicians seem to show a stronger expertise of shorter antibiotic courses as compared to other specialists.

Finally, the majority of institutions of respondents did not have internal guidelines for antibiotic prescription in terminally ill patients; this constitutes a gap that deserves action.

The main limitation of this study lies in the survey methodology. The fact that physicians participated on a voluntary basis may have resulted in a respondent population unbalanced toward those with a particular sensitivity to appropriate antibiotic prescribing at the end of life, so generating a convenience sample bias. This nevertheless provides a picture of physicians engaged in this topic. Nonetheless, this approach could still be considered valid, as it provides a snapshot of current practices regarding antibiotic prescription at the end of life.

Multiple-response options were included for some items to reflect the diverse approaches reported in the literature.

It also must be mentioned that the study did not address at least two important categories of clinicians dealing with end-of-life patients, namely Oncologists and Intensive Care Specialists. Oncologists were not included in this study primarily because our focus was on EOL - regardless of the underlying disease- so we addressed physicians with a broader span of different patients as compared to oncologists. It must also be noted that, according to the Global Atlas of Palliative care, cancer accounts for 23% of deaths in the European region [26]. As to intensivists, we believe that their major focus on acute care and the fact that the great majority of terminally ill patients die outside of an ICU (at home, in hospice, long-term care facilities and medicine wards) justifies their exclusion. Further studies should address these peculiar categories of prescribers.

Finally, we acknowledge that doctors were asked to respond for themselves only, not considering a team-based approach to problems. Finally, the cultural setting is likely very relevant to Palliative Care, so results from Italy may be difficult to generalize to other contexts.

CONCLUSIONS

There is still much difficulty in correct identification of “End of life”, especially in non-oncological, co-morbid patients and in chronically progressive diseases. Greater involvement of Palliative Care specialists could promote appropriateness in care, increase in shared and advanced care planning with the patient and a reduction in disproportionate and futile treatments, including antibiotic treatment. Effective communication with patients and families is considered crucial, so that communication skills are key in this setting. Further research is needed to address knowledge gaps, such as better definition of “end-of-life” and of desirable outcomes for terminally ill patients.

Physicians involved in end-of-life care recognize the challenges of antibiotic prescription and consider multiple factors when deciding to start or stop therapy. Protocols for the treatment of infections at the end of life, drafted by Infectious Disease specialists and taking into account the specific microbial resistance pattern, could undoubtedly help in the management of terminally ill patients.

Conflict of interest

Nothing to declare.

Funding

No funding was received for this work.

REFERENCES

  1. Herzig CTA, Dick AW, Sorbero M, et al. Infection Trends in US Nursing Homes, 2006-2013. J Am Med Dir Assoc. 2017; 18(7): 635.e9-635.e20.
  2. Abdel-Karim IA, Sammel RB, Prange MA. Causes of death at autopsy in an inpatient hospice program. J Palliat Med. 2007; 10(4): 894-898.
  3. Van Der Steen JT, Pasman HR, Ribbe MW, et al. Discomfort in dementia patients dying from pneumonia and its relief by antibiotics. Scand J Infect Dis. 2009; 41(2): 143-151.
  4. Givens JL, Jones RN, Shaffer ML, et al. Survival and comfort after treatment of pneumonia in advanced dementia. Arch Intern Med. 2010; 170(13): 1102-1107.
  5. White PH, Kuhlenschmidt HL, Vancura BG, et al. Antimicrobial use in patients with advanced cancer receiving hospice care. J Pain Symptom Manage. 2003; 25(5): 438-443.
  6. D’Agata E, Mitchell SL. Patterns of antimicrobial use among nursing home residents with advanced dementia. Arch Intern Med. 2008; 168(4): 357-362.
  7. Wi YM, Kwon KT, Hwang S, et al. Antimicrobial Stewardship Research Committee of Korean Society for Antimicrobial Therapy. Use of Antibiotics Within the Last 14 Days of Life in Korean Patients: A Nationwide Study. J Korean Med Sci. 2023; 38(9): 66-76.
  8. Marra AR, Puig-Asensio M, Balkenende E, et al. Antibiotic use during end-of-life care: A systematic literature review and meta-analysis. Infect Control Hosp Epidemiol. 2021; 42(5): 523-529.
  9. Barlam TF, Cosgrove SE, Abbo LM, et al. Implementing an Antibiotic Stewardship Program: Guidelines by the Infectious Diseases Society of America and the Society for Healthcare Epidemiology of America. Clin Infect Dis. 2016; 62(10): 51-77.
  10. SIAARTI, Grandi insufficienze d’organo “end stage”: cure intensive o cure palliative? “documento condiviso” per una pianificazione delle scelte di cura, available at: https://www.biodiritto.org/ocmultibinary/download/2554/24383/7/1eb56593f0926dc1537883d780290e66.pdf/file/INSUFFICIENZE_CRONICHE_END_STAGE.pdf [Last accessed on July 24, 2025].
  11. Hung KC, Lee LW, Liew YX, et al. Antibiotic stewardship program (ASP) in palliative care: antibiotics, to give or not to give. Eur J Clin Microbiol Infect Dis. 2022; 41(1): 29-36.
  12. Crispim DH, da Silva IO, de Carvalho RT, et al. End-of-life use of antibiotics: a survey on how doctors decide. Int J Infect Dis. 2022; 114: 219-225.
  13. Hosoi T, Ozone S, Hamano J, et al. Prediction Models for Impending Death Using Physical Signs and Vital Signs in Noncancer Patients: A Prospective Longitudinal Observational Study. Palliat Med Rep. 2021; 2(1): 287-295.
  14. Hui D, Hess K, dos Santos R, et al. A diagnostic model for impending death in cancer patients: Preliminary report. Cancer. 2015; 121(21): 3914-3921.
  15. Hosoi T, Ozone S, Hamano J. Variations in vital signs at the end of life in non-cancer patients: a retrospective study. Ann Palliat Med. 2020; 9(5): 2678-2683.
  16. Durand M, Forestier E, Gras Aygon C, et al. L’intergroupe GInGer (SPILF SFGG). Determinants of doctors’ antibiotic prescriptions for patients over 75 years old in the terminal stage of palliative care. Infect Dis Now. 2021; 51(4): 340-345.
  17. Norman DC. Fever in the elderly. Clin Infect Dis. 2000; 31(1): 148-151.
  18. Larnard J, Blackshear L, Lee MSL, et al. Perceptions and Reality of Antimicrobial Prescribing During the Transition to Comfort Measures Only at an Academic Medical Center. Open Forum Infect Dis. 2022; 10(1): ofac692-699.
  19. Servid SA, Noble BN, Fromme EK, et al. Clinical Intentions of Antibiotics Prescribed Upon Discharge to Hospice Care. J Am Geriatr Soc. 2018; 66(3): 565-569.
  20. Karlin D, Pham C, Furukawa D, et al. State-of-the-Art Review: Use of Antimicrobials at the End of Life. Clin Infect Dis. 2024; 78(3): 27-36.
  21. Kim H, Flieger SP. Barriers to Effective Communication about Advance Care Planning and Palliative Care: A Qualitative Study. J Hosp Palliat Care. 2023; 26(2): 42-50.
  22. Albrecht JS, McGregor JC, Fromme EK, et al. A nationwide analysis of antibiotic use in hospice care in the final week of life. J Pain Symptom Manage. 2013; 46(4): 483-490.
  23. Yao CA, Hsieh MY, Chiu TY, et al. Wishes of Patients With Terminal Cancer and Influencing Factors Toward the Use of Antibiotics in Taiwan. Am J Hosp Palliat Care. 2015; 32(5): 537-543.
  24. Bernard C, Tan A, Slaven M, et al. Exploring patient-reported barriers to advance care planning in family practice. BMC Fam Pract. 2020; 21(1): 94-102.
  25. Tark A, Estrada LV, Tresgallo ME, et al. Palliative care and infection management at end of life in nursing homes: A descriptive survey. Palliat Med. 2020; 34(5): 580-588.
  26. Connor SR. Global Atlas of Palliative Care (2nd ed.). Worldwide Hospice Palliative Care Alliance. 2020. Available at: http://www.thewhpca.org/resources/global-atlas-on-end-of-life-care. [accessed on February 04, 2025].