Abstract
Background
Little is known about antimicrobial prescribing when patient care is transitioned to comfort measures only (CMO). We used a multidisciplinary survey and retrospective cohort study to gain insight into antimicrobial prescribing in this population at an academic medical center to inform future antimicrobial stewardship interventions.
Methods
A survey focusing on antimicrobial prescribing during the transition to CMO was electronically distributed to providers in medical subspecialities and responses were compared across specialties. A retrospective chart review was performed of patients admitted to an academic medical center in 2020 who were on antimicrobials in the 48 hours prior to CMO. We investigated the percentage of patients who remained on antimicrobials after the transition to CMO and rationale for continuing antimicrobials.
Results
We received 113 survey responses (35% response rate). Forty-one percent of respondents indicated that they “sometimes” or “often” continued antimicrobials during the transition to CMO. Patient/family preference and symptom palliation were the most common factors cited by respondents when deciding whether to continue antimicrobials in this population. Of the 546 patient charts reviewed, 140 (26%) patients were alive 48 hours after CMO order, and 19 (14%) of those patients remained on antimicrobials. Five of 19 (26%) patients had documentation that antimicrobials were continued due to patient/family preference and 5 of 19 (26%) patients had documentation that antimicrobials were continued for palliation of symptoms.
Conclusions
Patient/family preference and symptom palliation are important factors in prescribing antimicrobials when patient care is transitioned to CMO. More evidence is needed regarding palliative benefits of antibiotics to inform provider discussions of benefits and harms of antimicrobial use in this setting.
Keywords: antimicrobial stewardship, comfort measures only, end of life, palliative care
In a multidisciplinary survey and retrospective chart review at an academic medical center, patient/family preferences and symptom palliation were important factors when providers continued or recommended continuing antimicrobials during transition to comfort measures only.
Antibiotics are often a part of end-of-life (EOL) care [1], including patients who are transitioned to comfort measures only (CMO) [2–4]. Prior survey studies have also shown that some physicians favor prescribing antibiotics to patients receiving comfort care when they have an infection [5–7], and no difference was found between medicine subspecialties when asked about withholding antimicrobials when patient care is transitioned to CMO [7].
Though antimicrobials are sometimes utilized in patients receiving comfort-focused care, the evidence for their effectiveness when used in a palliative capacity is mixed. A systematic review on the effectiveness of antimicrobials for symptom management in patients receiving palliative or hospice care was unable to draw conclusions due to limitations in the included observational data [8]. A more recent retrospective study of hospitalized patients in the last 14 days of life found that antibiotics were prescribed for symptom relief in 44.1% of patients, but only helped achieve symptom relief in 22.8% of cases [9]. In contrast, another observational study of patients receiving home or skilled nursing facility hospice care found that 60% of antimicrobial prescriptions that met a minimum use criteria led to symptom resolution [10]. Older, observational studies have suggested that antibiotics may be more helpful at alleviating symptoms caused by urinary tract infections (UTIs) compared with other infections at EOL [11–13].
Antimicrobial use at EOL, particularly for patients on CMO, also exposes patients to unwelcome medication side effects such as nausea and diarrhea (including Clostridioides difficile infection) [2, 14–16], as well as more indirect harms such as discomfort associated with peripheral line insertion [17]. Moreover, there are additional potential harms of antimicrobials to consider; for example, a recent study showed that antimicrobial use was associated with increased length of stay in patients with advanced cancer who were transitioned to CMO [18]. Antibiotic use at EOL also has been associated with the development of antimicrobial resistance [19], and their use contributes to the significant costs of healthcare at EOL [16].
In the setting of the uncertain benefits and potential harms associated with antimicrobial use at EOL, relatively little is known about the rationale for prescribing antimicrobials in this population. A retrospective study of antimicrobials prescribed for patients discharged to hospice found that only a small minority were prescribed for palliative reasons, and antimicrobials were more commonly prescribed to cure, suppress, or prevent an infection [20]. However, in prior survey studies, the desire to palliate symptoms and patient and family preference have emerged as important factors in continuing antimicrobials at EOL [6, 7].
The aim of this study was to characterize antimicrobial use and rationale for antimicrobial prescribing when patient care was transitioned to CMO at an academic medical center and to use these data to inform future antimicrobial stewardship approaches and interventions in this population.
METHODS
We first conducted a cross-sectional survey at a single tertiary-care academic medical center. Respondents included attending physicians, advanced practice providers, and fellows in the internal medicine subspecialties that we believed would be most involved in antimicrobial decision making when patient care was transitioned to CMO. These specialties included infectious disease, hematology/oncology, palliative care, pulmonary/critical care, and hospital medicine.
A “faculty champion” from each specialty was identified who provided email listservs of providers from their respective specialty. The survey was administered using Research Electronic Data Capture (REDCap). Participation was voluntary and anonymous, and no financial incentive was provided.
The survey was created by the authors and included questions about the practices and attitudes of prescribing antimicrobials during the transition to CMO. Responses were graded using 4- to 5-point Likert scales and there was also an open-ended question for respondents to share other thoughts regarding antimicrobial prescribing in this population. Respondents were also asked for their training level and specialty. The survey was pilot tested on individuals outside of the authors' institutions in the represented medicine subspecialties to obtain feedback, and the final 42-item survey was distributed on 31 January 2022 via email. Differences among specialties were compared using Kruskal-Wallis tests in Stata/BE 17.0, and responses to the open-ended question were reviewed for representative comments.
A retrospective chart review was then performed for patients admitted to Beth Israel Deaconess Medical Center (BIDMC) from 1 January 2020 to 31 December 2020 who had an order for CMO placed during a hospitalization and who had at least 1 active antimicrobial order in the 48 hours prior to CMO order. Patients were subsequently excluded if their initial CMO order was reversed. Information extracted from the electronic medical record (EMR) included demographic information, comorbidities, antimicrobial orders, and antimicrobial indication. The authors (J. L. and L. B.) who performed manual extraction from clinician documentation in the EMR used a shared extraction tool. The extraction tool was piloted on 10 sample charts. Overall, 118 of the 130 items (91%) were rated concordantly by the reviewers. Cohen kappa statistic was 0.73–1 for individual variables. The authors involved in chart review used this pilot to address remaining areas of disagreement to further improve concordance.
We investigated patient and antimicrobial status 48 hours after CMO order placement. For patients who were still alive and remained on antimicrobials 48 hours after CMO order, we noted if antimicrobials were prescribed for symptom relief or patient/family preference. Data and descriptive statistics were tabulated and analyzed using Microsoft Excel. The BIDMC institutional review board deemed this study exempt from review.
RESULTS
Behaviors Around Antimicrobials and CMO Transition
One hundred thirteen of the invited 326 providers in infectious disease, palliative care, hematology/oncology, pulmonary/critical care, and hospital medicine completed the survey for a response rate of 35%. Two of the 113 did not indicate a specialty and thus were excluded from further analysis.
Hospital medicine had the most respondents (32/111 [29%]), followed by hematology/oncology (30/111 [27%]), infectious disease (20/111 [18%]), pulmonary/critical care (18/111 [16%]), and palliative care (11/111 [10%]). The majority of respondents were attending physicians (83/111 [75%]); a smaller proportion were fellows (21/111 [19%]), and advance practice providers (5/111 [5%]). Two respondents did not specify a practice level.
Overall, 96 of 111 (86%) respondents reported being at least sometimes involved in the care of patients when care was transitioned to CMO. A similar number of respondents (91/111 [82%]) specified that they at least sometimes discussed antimicrobial decision making with patients, family members, or primary teams when care was transitioned to CMO. There was no statistical difference between specialties (P = .1564). Forty-six of 111 (41%) respondents reported they at least sometimes continued or recommended continuing antimicrobials when patient care was transitioned to CMO. There was a statistically significant difference between specialties (P = .0135), with the median response of hematology/oncology providers indicating they more often continued antimicrobials in this population.
Decisions about continuing antimicrobials in patients transitioned to CMO varied based on the type of infection being treated. While 24%–30% of respondents indicated they would be likely or very likely to continue antimicrobials for skin/soft tissue infection, UTI, pneumonia, gastrointestinal infections, and meningitis/central nervous system infections, only 19% and 16% of respondents would be likely or very likely to continue antimicrobials for musculoskeletal infections and bacteremia, respectively. There was a statistically significant difference between specialties for all infections, with pulmonary/critical care respondents notably having a median response of “very unlikely” to continue antimicrobials for all infections.
Perception of Palliative and Adverse Effects of Antimicrobials
For all specified symptoms, respondents thought antimicrobials were less likely to be helpful if patients were already being treated with symptom-directed medications (ie, acetaminophen for fevers, narcotics for pain or dyspnea, guaifenesin for cough). There was a significant minority (29%–47% depending on the symptom) of the total respondents who felt antimicrobials would be at least somewhat helpful for treating the queried symptoms (Table 1). There was a statistically significant difference between specialties only for fever (P = .0089), with infectious disease and hematology/oncology having a median response that antimicrobials would be “somewhat helpful” compared with other specialties having a median response of “not helpful.”
Table 1.
Number and Percentage of Responses Indicating Antimicrobials Were at Least Somewhat Helpful for Specified Symptoms in Patients Already on Symptom-Directed Medications
| Specialty | Fever | Pain | Cough | Dyspnea | Dysuria |
|---|---|---|---|---|---|
| Total (N = 111) | 52 (47) | 38 (34) | 32 (29) | 35 (32) | 48 (43) |
| Hematology/oncology (n = 30) | 19 (63) | 13 (43) | 13 (43) | 15 (50) | 16 (53) |
| Infectious disease (n = 20) | 11 (55) | 5 (25) | 6 (30) | 5 (25) | 10 (50) |
| Palliative care (n = 11) | 1 (9) | 5 (45) | 4 (36) | >4 (36) | 6 (55) |
| Hospital medicine (n = 32) | 14 (44) | 13 (41) | 6 (19) | 7 (22) | 12 (38) |
| Pulmonary/critical care (n = 18) | 7 (39) | 2 (11) | 3 (17) | 4 (22) | 4 (22) |
| P value | .0089 | .1447 | .1454 | .0570 | .1039 |
Data are presented as No. (%) unless otherwise indicated. Specialties compared using Kruskal-Wallis test; P < .05 indicates significant difference between specialties.
Respondents also expressed concern about side effects or toxicities of antimicrobials, though the level of concern varied somewhat by side effect and provider group. Diarrhea and nausea were of greatest concern overall, though there were significant differences in the level of concern between specialties. Most respondents were also at least somewhat concerned about antimicrobial resistance, length of stay, and cost of antimicrobials, though there were no statistically significant differences between specialties (Table 2).
Table 2.
Number and Percentage of Responses Indicating Respondents Who Were at Least Somewhat Concerned About Antimicrobials Causing Specified Symptoms
| Specialty | Altered Mental Status | Diarrhea | Nausea | Rash | Antibiotic Resistance | Length of Stay | Cost |
|---|---|---|---|---|---|---|---|
| Total (N = 111) | 57 (51) | 100 (90) | 84 (76) | 60 (54) | 68 (61) | 55 (50) | 70 (63) |
| Hematology/oncology (n = 30) | 15 (50) | 27 (90) | 23 (77) | 15 (50) | 15 (50) | 19 (63) | 20 (67) |
| Infectious disease (n = 20) |
13 (65) | 20 (100) | 18 (90) | 14 (70) | 13 (65) | 11 (55) | 16 (80) |
| Palliative care (n = 11) |
7 (64) | 11 (100) | 11 (100) | 9 (82) | 10 (91) | 5 (45) | 8 (73) |
| Hospital medicine (n = 32) |
14 (44) | 29 (91) | 25 (78) | 16 (50) | 21 (66) | 13 (41) | 16 (50) |
| Pulmonary/critical care (n = 18) |
8 (44) | 13 (72) | 7 (39) | 6 (33) | 9 (50) | 7 (39) | 10 (56) |
| P value | .3909 | .0448 | .0055 | .1155 | .1368 | .3241 | .2024 |
Data are presented as No. (%) unless otherwise indicated. Specialties compared using Kruskal-Wallis Test; P < .05 indicates significant difference between specialties.
Factors Influencing Antimicrobial Prescribing at CMO
As seen in Figure 1, when asked about factors that contributed to continuing antimicrobials when patient care is transitioned to CMO, patient/family preference was specified by the most respondents (86/111 [77%]), followed by providing palliation of a symptom (69/111 [62%]). Ninety-five of 111 (86%) respondents indicated that antimicrobial decision making during the transition to CMO should be shared between patient and provider, while 10 of 111 (9%) indicated that decision making should be patient-directed and 6 of 111 (5%) indicated that decision making should be provider-directed. There was not a statistically significant difference between specialties (P = .3680). Participant responses to survey open-ended questions also suggested patient/family preference and symptom palliation were important drivers behind decisions to continue antimicrobials in this setting (Table 3).
Figure 1.
Percentage of respondents who indicated that the below items factored into decision making when deciding whether to continue antimicrobials during the transition to comfort measures only (CMO).
Table 3.
Factors in Antimicrobial Prescribing During the Transition to Comfort Measures Only and Representative Comments Identified From Survey Open Response Question
| Factors in Antimicrobial Prescribing at CMO | Representative Comments |
|---|---|
| Patients and family often influence antimicrobial decision making | “The decision to continue abx is usually driven by families who are unwilling to fully withdraw all measures … the attendings are aware of antimicrobial resistance and other associated issues, but feel that their hands are tied so long as there is shared decision making with patient's family.” “It feels like it's basically 95% family preference … I generally recommend against but have a low threshold to just go with family preference.” “The emotional burden of removing interventions is already very heavy, and we are trying to spare the family the burden of yet another decision.” “If a day of antibiotics allows us to otherwise move toward CMO, that is a reasonable tradeoff for the patient.” “I advise drawbacks, but don’t push back too hard … They [families] will sometimes let the antibiotics go in a day or two.” |
| Antimicrobial prescribing with goal of symptom palliation | “Is there data to indicate that administering antibiotics for XYZ infection helps palliate symptoms in general and/or at the end of life; alternatively, does cessation of antibiotics lead to an increase in symptoms?” “It would be helpful from an ID perspective to know if treatment of infection would improve QoL or symptoms and side effect risks.” “For an infection that will require a prolonged course of IV antibiotics I am more likely to recommend stopping as the antibiotics would likely cause more harm than benefit. However, if a patient had dysuria from a UTI and could complete a 5-day course of PO antibiotics with improved symptoms, then I would typically offer this.” |
Abbreviations: abx, antibiotics; CMO, comfort measures only; ID, infectious disease; IV, intravenous; PO, oral; QoL, quality of life; UTI, urinary tract infection.
Retrospective Chart Review
A total of 558 patient charts were initially identified; 12 were subsequently excluded as their initial CMO order was reversed. Demographic information and baseline characteristics of the cohort can be seen in Table 4. The use of broad-spectrum antimicrobials providing methicillin-resistant Staphylococcus aureus and Pseudomonas coverage were common in the cohort prior to CMO order placement. Intravenous vancomycin was the most prescribed antibiotic in the cohort (324/546 patients [59%]), followed by intravenous cefepime (238/546 patients [44%]), intravenous/oral metronidazole (140/546 patients [26%]), intravenous ceftriaxone (98/546 patients [18%]), and intravenous piperacillin-tazobactam (93/546 patients [17%]). Seventy-two patients (13%) had an active intravenous carbapenem order in the 48 hours prior to CMO. Micafungin was the most common antifungal prescribed, with 42 patients (8%) having an active order prior to CMO order placement.
Table 4.
Baseline Characteristics of All Patients in the Cohort and Patients Who Remained in Hospital or Discharged on Antimicrobials 48 Hours After Comfort Measures Only Order Placement
| Characteristic | Total Cohort (N = 546) |
Patients in Hospital/Discharged With Active Antimicrobial 48 h After CMO Order (n = 19) |
|---|---|---|
| Average age, y | 72 | 75 |
| Sex | ||
| Male | 305 (56) | 12 (63) |
| Female | 241 (44) | 7 (37) |
| Race | ||
| White | 358 (66) | 15 (79) |
| Non-White | >188 (34) | 4 (21) |
| Ethnicity | ||
| Hispanic | 19 (3) | 1 (5) |
| Non-Hispanic | 527 (97) | 18 (95) |
| Primary service | ||
| ICU | 406 (74) | 4 (21) |
| Medicine | 98 (18) | 8 (42) |
| Solid oncology | 23 (4) | 6 (32) |
| Liquid oncology | 12 (2) | 0 (0) |
| Neurology | 4 (1) | 0 (0) |
| Surgery | 1 (0) | 1 (5) |
| Unknown/unclear | 2 (0) | 0 (0) |
| Comorbidities | ||
| Heart failure | 173 (32) | 3 (16) |
| Solid malignancy | 168 (31) | 8 (42) |
| Cirrhosis | 94 (17) | 2 (11) |
| COPD | 89 (16) | 0 (0) |
| ESRD on dialysis | 86 (16) | 0 (0) |
| Dementia | 64 (12) | 3 (16) |
| Liquid malignancy | 53 (10) | 1 (5) |
| Bone marrow/solid organ transplant | 10 (2) | 1 (5) |
| COVID-19 infection during hospitalization | 91 (17) | 0 (0) |
| Active palliative care consult | 129 (24) | 14 (74) |
| Active ID consult | 95 (17) | 5 (26) |
Data are presented as No. (%).
Abbreviations: CMO, comfort measures only; COPD, chronic obstructive pulmonary disease; COVID-19, coronavirus disease 2019; ESRD, end-stage renal disease; ICU, intensive care unit; ID, infectious disease.
Forty-eight hours after CMO order placement, 140 of the 546 patients in the cohort remained alive (26%). Of those 140, 19 (14%) remained on antimicrobials. Among CMO patients alive on antimicrobials at 48 hours, 7 of 19 (37%) patients were discharged from the hospital on at least 1 antimicrobial in the 48 hours after CMO order, and 12 of 19 (63%) patients remained in the hospital on at least 1 antimicrobial. Baseline characteristics of these 19 patients are shown in Table 4 and characterization of their infections and antimicrobials are shown in Table 5. Of note, 5 of 19 (26%) had documentation that antibiotics were prescribed to palliate symptoms and 5 of 19 (26%) also had documentation that antimicrobials were prescribed at least in part because of patient/family preference.
Table 5.
Characterization of Infections and Antimicrobials for Patients Who Remained in Hospital or Discharged on Antimicrobials 48 Hours After Comfort Measures Only Order Placement
| Characteristic | Patients in Hospital or Discharged With Active Antimicrobial 48 h After CMO Order |
|---|---|
| Antimicrobial indication (n = 19) | |
| Active infection | 18 (95) |
| Empiric for fever/sepsis | 1 (5) |
| Other | 0 (0) |
| Type of active infection (n = 18) | |
| Pneumonia | 4 (22) |
| Multiple | 4 (22) |
| Urinary tract | 4 (22) |
| Bacteremia | 3 (17) |
| Gastrointestinal | 2 (11) |
| Soft/soft tissue | 1 (6) |
| Antimicrobial used (n = 19) | |
| Amoxicillin-clavulanate (PO) | 4 (21) |
| Cefepime (IV) Metronidazole (PO/IV) Vancomycin (PO) |
3 (16) |
| Azithromycin (PO/IV) Ceftriaxone (IV) Daptomycin (IV) Vancomycin (IV) |
2 (11) |
| Ceftaroline (IV) Ciprofloxacin (PO) Doxycycline (PO) Levofloxacin (PO) Nitrofurantoin (PO) Piperacillin-tazobactam (IV) Trimethoprim-sulfamethoxazole (PO) |
1 (5) |
| Patient/family preference at least part of prescriber rationale | 5 (26) |
| Palliation of symptoms at least part of prescriber rationale | 5 (26) |
Data are presented as No. (%).
Abbreviations: CMO, comfort measures only; IV, intravenous; PO, oral.
DISCUSSION
We found that decision making about antimicrobials is an important consideration for clinicians caring for hospitalized patients at the end of life, and almost half of survey respondents indicated that they at least sometimes continued antimicrobials when patient care is transitioned to CMO. However, we found that only 14% of patients who were alive 48 hours after CMO order placement remained on antimicrobials. This discrepancy may be in part due to survey participants recalling prescribing antimicrobials at EOL, but not necessarily those who have formal CMO orders. Our findings were similar to a retrospective cohort study performed at 2 academic centers in the United States where 16.5% of patients who were alive 48 hours after CMO initiation remained on antimicrobials [3]. There was a statistically significant difference between medicine subspecialties when asked how often antimicrobials are continued when patient care is transitioned to CMO with hematology/oncology having the highest median response, indicating more frequent antimicrobial continuation. Solid malignancy was also the most common comorbidity among patients who were alive with active antimicrobial orders 48 hours after CMO order placement.
Patient and family preference emerged as an important factor in antimicrobial decision making during the transition to CMO in our survey, which was also seen in previous survey studies about antimicrobial prescribing at EOL [6, 7]. The overwhelming majority of survey respondents also thought that antimicrobial decision making during the transition to CMO should be shared between patient and provider. The survey responses also reflected practices at our institution; 26% of the patients who were alive and on antimicrobial therapy after CMO order had documentation that antimicrobials were prescribed at least in part because of patient/family preference. Prior literature had found that a similar percentage of patients received antimicrobials due to patient/family preference on discharge to hospice [20]. Free text survey responses also suggested that providers were often reluctant to withdraw antibiotics if patients and/or families favor continuing them, even if they were not convinced of their benefit.
After patient/family preference, the desire to palliate symptoms was the next most common factor seen in survey responses for continuing antimicrobials when patient care is transitioned to CMO. The importance of the perception of symptom relief in antimicrobial prescribing at EOL was also seen in a prior survey study [6]. Again, our survey data reflected actual practices at our institution; 26% of the patients who were alive and on antimicrobial therapy after CMO order had documentation that antimicrobials were prescribed at least in part for symptom palliation. Our survey also found that, when taking into account symptom-directed therapy, providers thought antibiotics would be most helpful with fever and dysuria and less helpful with cough and dyspnea. Interestingly, however, about the same percentage of respondents indicated they would be likely or very likely to continue antibiotics for either UTI or pneumonia when patients are transitioned to CMO. This finding may suggest that providers are more likely to continue antibiotics if a short or defined course of antibiotics can be established.
Despite evidence that the desire to palliate symptoms is an important driver of antimicrobial prescribing at EOL, there is a paucity of data to support this approach in the EOL population [8]. The variability in symptom palliation with antimicrobials is likely driven both by the difficulty in diagnosing an acute bacterial infection at EOL as well as the heterogeneity of the EOL population. For example, for a patient with stage IV pancreatic cancer who is receiving home hospice and has an estimated prognosis of weeks to months, antimicrobials could feasibly help to relieve dysuria from a UTI. However, if that same patient developed hypoxic respiratory failure secondary to progression of metastatic pulmonary disease and has a prognosis of hours to days, antimicrobials are likely to be less effective at palliating symptoms from a postobstructive pneumonia. In general, we suspect that antimicrobials are less likely to be helpful when patients are closer to the active dying process, which was also seen in a prior observational study [21]. This is especially important to consider in the CMO population at our institution, as most patients died within 48 hours of CMO initiation.
Our findings have important implications for antimicrobial stewardship in the CMO population. In clinical scenarios in which providers do not think antimicrobials will realistically improve symptoms at EOL, especially in patients in the active dying process (ie, expected prognosis is hours to days), care should be taken to carefully frame the anticipated lack of benefits alongside the potential harms of antimicrobial use to patients, families, and other providers. Providers should keep in mind that patients and families may overestimate the potential benefits of antimicrobials at EOL [22], though providers themselves may not always be aware of the risks associated with antimicrobial therapy [7]. If no clinical benefit of antimicrobial use is suspected, it may be reasonable to strongly recommend discontinuation of antibiotics while reassuring patients and families that antimicrobials would be very unlikely to offer benefit and may, in fact, prolong suffering from their underlying disease. Moreover, there may be nonantimicrobial medications that may offer a better chance at symptom improvement. Patients and families understandably may not as be motivated by length of stay, antimicrobial resistance, and healthcare costs, but providers should also take these factors into account.
This study is the first to our knowledge to use a multidisciplinary survey to gain insight into antimicrobial prescribing when patient care is transitioned to CMO. We were also able to verify survey respondent perceptions of antimicrobial prescribing in patients on CMO by performing a comprehensive chart review of this patient group over 1 year. There are also several limitations to our study. Our findings are from a single academic tertiary care center in the United States, and may not be reflective of other medical institutions in the United States and elsewhere in the world. Our retrospective chart review was also limited by reliance on provider documentation, especially as documented preferences about life-sustaining treatments are not always documented in the medical record.
Overall, we found that a significant number of providers indicated that they at least sometimes continued or recommended continuing antimicrobials when patient care is transitioned to CMO, and 14% of patients in our cohort who were alive 48 hours after CMO order remained on antimicrobials. Patient and family preference and a desire to palliate symptoms emerged as important influences in antimicrobial decision making in this population. Given the limited evidence base supporting the efficacy of antimicrobials to palliate symptoms for many patients receiving comfort-focused care, providers should carefully frame the expected benefits alongside the potential risks of continued antimicrobial therapy with patients and families. Future studies are needed to establish whether antimicrobials provide symptom palliation for specific infections in patients at EOL who are already receiving comfort-focused care. This information could help guide future antimicrobial stewardship interventions and, more importantly, prescribers and patients/families as they make these difficult management decisions together.
Contributor Information
Jeffrey Larnard, Division of Infectious Disease, Beth Israel Deaconess Medical Center, Boston, Massachusetts, USA.
Leslie Blackshear, Department of Medicine, Beth Israel Deaconess Medical Center, Boston, Massachusetts, USA.
Matthew Shou Lun Lee, Division of Infectious Disease, Beth Israel Deaconess Medical Center, Boston, Massachusetts, USA.
Mary K Buss, Department of Palliative Care, Tufts Medical Center, Boston, Massachusetts, USA.
Wendy Stead, Division of Infectious Disease, Beth Israel Deaconess Medical Center, Boston, Massachusetts, USA.
Notes
Acknowledgments. This work was conducted with support from the Harvard Catalyst/Harvard Clinical and Translational Science Center (National Center for Advancing Translational Sciences, National Institutes of Health [NIH] award number UL1 TR002541) and by financial contributions from Harvard University and its affiliated academic healthcare centers.
Patient consent. The design of this work has been reviewed and determined exempt by the Beth Israel Deaconess Medical Center institutional review board.
Disclaimer. The content of this work is solely the responsibility of the authors and does not necessarily represent the official views of Harvard Catalyst, Harvard University and its affiliated academic healthcare centers, or the NIH.
All authors have submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Conflicts that the editors consider relevant to the content of the manuscript have been disclosed.
References
- 1. Marra AR, Puig-Asensio M, Balkenende E, Livorsi DJ, Goto M, Perencevich EN. Antibiotic use during end-of-life care: a systematic literature review and meta-analysis. Infect Control Hosp Epidemiol 2021; 42:523–9. [DOI] [PubMed] [Google Scholar]
- 2. Thompson AJ, Silveira MJ, Vitale CA, Malani PN. Antimicrobial use at the end of life among hospitalized patients with advanced cancer. Am J Hosp Palliat Care 2012; 29:599–603. [DOI] [PubMed] [Google Scholar]
- 3. Merel SE, Meier CA, McKinney CM, Pottinger PS. Antimicrobial use in patients on a comfort care protocol: a retrospective cohort study. J Palliat Med 2016; 19:1210–4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Oh DY, Kim JH, Kim DW, et al. Antibiotic use during the last days of life in cancer patients. Eur J Cancer Care 2006; 15:74–9. [DOI] [PubMed] [Google Scholar]
- 5. Zanartu C, Matti-Orozco B. Comfort measures only: agreeing on a common definition through a survey. Am J Hosp Palliat Care 2013; 30:35–9. [DOI] [PubMed] [Google Scholar]
- 6. Gaw CE, Hamilton KW, Gerber JS, Szymczak JE. Physician perceptions regarding antimicrobial use in end-of-life care. Infect Control Hosp Epidemiol 2018; 39:383–90. [DOI] [PubMed] [Google Scholar]
- 7. Datta R, Topal J, McManus D, et al. Education needed to improve antimicrobial use during end-of-life care of older adults with advanced cancer: a cross-sectional survey. Palliat Med 2021; 35:236–41. [DOI] [PubMed] [Google Scholar]
- 8. Rosenberg JH, Albrecht JS, Fromme EK, et al. Antimicrobial use for symptom management in patients receiving hospice and palliative care: a systematic review. J Palliat Med 2013; 16:1568–74. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Tagashira Y, Kawahara K, Takamatsu A, Honda H. Antimicrobial prescribing in patients with advanced-stage illness in the antimicrobial stewardship era. Infect Control Hosp Epidemiol 2018; 39:1023–9. [DOI] [PubMed] [Google Scholar]
- 10. Clark MD, Halford Z, Herndon C, Middendorf E. Evaluation of antibiotic initiation tools in end-of-life care. Am J Hosp Palliat Care 2022; 39:274–81. [DOI] [PubMed] [Google Scholar]
- 11. Reinbolt RE, Shenk AM, White PH, Navari RM. Symptomatic treatment of infections in patients with advanced cancer receiving hospice care. J Pain Symptom Manage 2005; 30:175–82. [DOI] [PubMed] [Google Scholar]
- 12. White PH, Kuhlenschmidt HL, Vancura BG, Navari RM. Antimicrobial use in patients with advanced cancer receiving hospice care. J Pain Symptom Manage 2003; 25:438–43. [DOI] [PubMed] [Google Scholar]
- 13. Clayton J, Fardell B, Hutton-Potts J, Webb D, Chye R. Parenteral antibiotics in a palliative care unit: prospective analysis of current practice. Palliat Med 2003; 17:44–8. [DOI] [PubMed] [Google Scholar]
- 14. Helde-Frankling M, Bergqvist J, Bergman P, Björkhem-Bergman L. Antibiotic treatment in end-of-life cancer patients—a retrospective observational study at a palliative care center in Sweden. Cancers (Basel) 2016; 8:E84. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Juthani-Mehta M, Malani PN, Mitchell SL. Antimicrobials at the end of life: an opportunity to improve palliative care and infection management. JAMA 2015; 314:2017–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Baghban A, Juthani-Mehta M. Antimicrobial use at the end of life. Infect Dis Clin North Am 2017; 31:639–47. [DOI] [PubMed] [Google Scholar]
- 17. Morrison RS, Ahronheim JC, Morrison GR, et al. Pain and discomfort associated with common hospital procedures and experiences. J Pain Symptom Manage 1998; 15:91–101. [PubMed] [Google Scholar]
- 18. Datta R, Zhu M, Han L, Allore H, Quagliarello V, Juthani-Mehta M. Increased length of stay associated with antibiotic use in older adults with advanced cancer transitioned to comfort measures. Am J Hosp Palliat Care 2020; 37:27–33. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Levin PD, Simor AE, Moses AE, Sprung CL. End-of-life treatment and bacterial antibiotic resistance: a potential association. Chest 2010; 138:588–94. [DOI] [PubMed] [Google Scholar]
- 20. Servid SA, Noble BN, Fromme EK, Furuno JP. Clinical intentions of antibiotics prescribed upon discharge to hospice care. J Am Geriatr Soc 2018; 66:565–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Nakagawa S, Toya Y, Okamoto Y, et al. Can anti-infective drugs improve the infection-related symptoms of patients with cancer during the terminal stages of their lives? J Palliat Med 2010; 13:535–40. [DOI] [PubMed] [Google Scholar]
- 22. 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; 32:537–43. [DOI] [PubMed] [Google Scholar]