bs_bs_banner
Journal of Evaluation in Clinical Practice ISSN 1365-2753
Problems related to antifungal prescription: a qualitative study of the views of intensivists Tânia Pereira Salci MSc,1 Adelia P. Batilana,2 Ricardo Pietrobon MD PhD3 and Silvana Martins Caparroz-Assef PhD4 1
PhD Student, Post Graduate Program in Biosciences Applied to Pharmacy, Universidade Estadual de Maringá, Maringá, Paraná, Brazil Master Student, Research on Research Group, Duke University Health System, Durham, North Carolina, USA 3 Associate Professor, Department of Surgery, Duke University Health System, Durham, North Carolina, USA 4 Associate Professor, Pharmacology and Therapeutic Department, Universidade Estadual de Maringá, Maringá, Paraná, Brazil 2
Keywords antifungal, barriers to prescription, intensive care unit, intensivist opinion, qualitative research, quality of health care Correspondence Mrs Tânia Pereira Salci Post Graduate Program in Biosciences Applied to Pharmacy Universidade Estadual de Maringá Av. Colombo, n. 5790, bl. K68, sl. 102b Maringá, Paraná 87020-900 Brazil E-mail: [email protected] Accepted for publication: 17 April 2014 doi:10.1111/jep.12181
Abstract Rationale and objective The choice of the appropriate antifungal medication is essential for therapeutic success. Although guidelines are available in the literature that regulate the consistent use of antifungal, no previous qualitative studies have addressed the difficulties related to the use of antifungal medication, especially in the intensive care unit (ICU). Our objective was to qualitatively investigate how intensivists consider antifungal prescriptions in an adult ICU. Methods The Grounded Theory analytical method was used for the data analysis. Physicians who worked in the adult ICU and prescribed antifungal medications were individually interviewed. A semi-structured interview was used to ask core questions, followed by follow-up questions at the discretion of the interviewer. Results Our analysis generated eight main emerging themes that were classified into three related groups. The main insights were that various interconnected reasons were given for the lack of conformity with regard to prescription patterns for antifungals. A negative cycle was perceived based on issues related to prescriptions and the search for knowledge. If problems related to individual actions and multidisciplinary team integration are resolved and local protocols are implemented based on local epidemiology, then barriers to proper prescriptions can be overcome when intensivists are faced with the unusual practice of prescribing antifungals. Conclusions Our investigation indicates the need for prescription assistance with support from a well-trained multidisciplinary team and consensus among its members and the importance of well-designed protocols.
Introduction Fungi are among the main causal agents of infection in the intensive care unit (ICU). Candida spp., in particular, are among the leading causes of bloodstream infections and sepsis [1]. Fungal infections are associated with mortality rates of up to 70% [2] and high morbidity rates because of the difficulty establishing an early diagnosis and effective treatment [3]. The choice of the appropriate antifungal drug, early treatment, and correct dosage and duration of treatment are essential aspects for therapeutic success [4]. However, approximately 80% of hospitalized patients receive inadequate antifungal treatment, leading to increases in hospital stays and health care costs of up to USD$17 000 per patient [5]. Considering the information available in the literature to guide antifungal therapy, some problems have been identified, such as 460
limited diagnostic options [6], difficulty establishing early treatment [7], difficulty determining the correct dosage and duration of treatment [8], microorganism resistance [9], high cost of treatment [5], and high mortality rate [10]. Thus, rigorous treatment protocols are needed that seek to minimize such problems [11]. The Infectious Diseases Society of America (IDSA) has created evidence-based guidelines for the management of invasive candidiasis to aid therapeutic choices [4]. Furthermore, the authors stress the importance of implementing local protocols for the treatment and control of fungal infections, especially invasive candidiasis, with the aims of diminishing the indiscriminate and irrational use of this class of drugs, reducing costs, and providing greater safety and efficacy [4,11]. The inadequate use of drugs has been discussed in qualitative studies that investigate the factors that influence the prescription of
Journal of Evaluation in Clinical Practice 20 (2014) 460–466 © 2014 John Wiley & Sons, Ltd.
T.P. Salci et al.
Problems related to antifungal
different classes of drugs [12,13] and barriers that limit compliance with standardized protocols [14]. Antifungal drugs are not beyond the scope of this context because problems related to the use of these medications have been demonstrated [15]. However, no previous qualitative studies have addressed the difficulties related to the use of antifungal medication, especially in the ICU. Considering this gap in the literature, the aim of the present study was to qualitatively investigate how intensivists assess fungal infection and antifungal prescriptions, and perceive barriers to the adequate prescription of antifungal drugs in an adult ICU.
Methods Ethics statement This study received approval from the Institutional Review Board of Universidade Estadual de Maringá, Brazil. All of the subjects were informed that their statements would be used for research purposes only, and written informed consent was obtained before the interview.
Study design A qualitative research strategy was used to gain insights into organizational and cultural issues within the ICU. Qualitative research methods have been developed in social sciences to enable researchers to better understand people and the social and cultural contexts within which they live. Studying educational settings and processes is particularly useful [16]. The Consolidated Criteria for Reporting Qualitative Research was used to provide guidelines in the design of the study [17]. The Grounded Theory analytical method [16] was used as the theoretical framework, which intended to explore the phenomena of meaning and action in the context in which they occur. The data were collected from July to November 2012.
Research team and reflexivity The researchers involved in the present study were a female pharmacist MSc researcher (TS), a female PhD pharmacist (SA), a female researcher (AB) and a male MD/PhD (RP). All of the authors are active clinical researchers who are involved in studies to improve protocols for drug prescriptions. Researchers TS and SA have been involved in antifungal research since 2009. RP and AB are research methodologists with no previous work or preconceptions on the use and misuse of antifungal medications. All of the respondents were interviewed by the same interviewer (TS). The respondents and interviewer (TS) had no prior contact
Table 1 Topics covered by the interview guide
© 2014 John Wiley & Sons, Ltd.
• • • • • •
and no common interests or working relationship. No conflict of interest was declared by any of the researchers with regard to this area of research.
Recruitment All of the physicians who worked in the adult ICU at the university hospital were asked to participate in the study. Eight physicians were individually interviewed face-to-face, and three physicians refused to participate in the study without providing a reason for non-participation. No exclusion criteria were applied. For the selection of respondents, contact was made with the director of the ICU, who provided a list of physicians. The first contact with the physicians was made during working hours. The respondents were informed of the identity of the interviewer, the importance and objectives of the study, and how the interview would be conducted. The participant then decided either to be interviewed at that time or to schedule a more convenient time.
Interview procedure The interviews occurred at the workplace in a reserved room. Only the interviewer and respondent were present during the interview. Following the provision of informed consent, the interviews lasted approximately 40 minutes and were digitally recorded. Field notes were made after each interview to avoid interruptions during the interview itself.
Data collection The interview began with questions about age, gender, years of experience, number of hours worked per week and number of institutions where the respondent worked. A semi-structured interview was used (Table 1) to ask core questions verbatim, followed by additional follow-up questions at the discretion of the interviewer. To enhance consistency, all of the transcriptions were performed by the interviewer (TS) immediately after data collection [17]. Transcriptions were recorded and available to the other researchers so they all could review and confirm the information. The identification of the respondents was removed from the transcriptions to ensure confidentiality and anonymity. A field log was kept to assist the coding of the data. The research team discussed the data, and inferences were made after each interview to address the research topic.
Data analysis The research team (SA and TS) analysed all of the transcribed interviews individually according to the steps of Grounded Theory: (i) initial coding; (ii) focused coding; and (iii) axial coding [18].
Reasons for initiating antifungal medication; Resources that support the prescription; Frequently required information; Assistance from colleagues regarding prescription; Confidence in prescribing antifungal medication; and Suggestions for contributing to and/or facilitating the prescription of antifungal medication.
461
Problems related to antifungal
T.P. Salci et al.
remaining results describe these emerging themes and corresponding groups in detail (Table 3).
Coding One researcher (TS) coded each transcription. Discrepancies were discussed and settled among the team. The quotations used in the present report were translated from Brazilian Portuguese to English by the author (RP) and independently back-translated by the other bilingual researchers (SA, TS). The data were saturated with regard to new emerging themes, beginning with the seventh interview.
Results Demographics and participant characteristics The majority of the subjects were male and younger than 50 years of age. All of them worked at more than one institution and in other ICUs. The mean weekly working hours was 72.3 ± 23.3 hours. Their clinical experience in the ICU was a minimum of 13 years and a maximum of 23 years (mean, 15.6 ± 4.0 years; Table 2).
Emerging themes and related quotations Our analysis generated eight main emerging themes: the initiation of antifungal use, uncommon prescription practices, individual actions, team integration, computerized systems, local protocols, importance attached to obtaining knowledge and sources of knowledge. These themes were classified into three related groups: prescription, prescription support and search for knowledge. The
Table 2 Characteristics of the respondents Characteristics
Average ± standard deviation
Age (years) Gender Male* Female* n of institutions working n of ICU working Average week work (hours) Average years in practice Average experience in ICU
42.9 ± 5.4 6 (75.0) 2 (25.0) 3.4 ± 1.5 2.7 ± 1.2 72.3 ± 23.3 18.3 ± 5.3 15.6 ± 4.0
*Frequency (%).
Issues related to prescription Time to initiate the antifungal use The use of antifungal drugs is justified when the fungus is identified as an etiological agent of infection based on positive laboratory exams. First, we have a positive culture to treat. If there is a culture, we begin antifungal treatment. (Respondent 7) However, this conduct is not always possible. The problem is response time. Reliable (laboratorial) feedback takes a long time. How long? What fungus is it? What’s the antibiogram? This always takes a long time, which, in intensive care, can sometimes be crucial. (Respondent 7) Thus, the clinical evaluation of the patient and recognition of risk factors are as essential as comorbidities and invasive procedures. Most patients in the ICU have at least one risk factor for the development of candidiasis. The microbiological identification of fungi can come too late or may even be hindered by the prior use of an antimicrobial agent. Therefore, the empirical initiation of treatment is necessary. [. . .] patients who have risk factors, for which we request a culture, but the results haven’t come back yet and there are risk factors. So, we take the liberty to start them on an antifungal. (Respondent 7) Frequently, the first choice of treatment is a broad-spectrum antibiotic. Generally, the main clinical suspicion is a bacterial infection. Therefore, the physicians cite that the prescription of antifungal drugs is their last thought when there is prolonged and ineffective antibacterial treatment with no resolution of the signs and symptoms of the infectious condition. And in the case of fungi in general, there is also something that is very common in the ICU: you treat the patient for a bacterial infection and he doesn’t get better. You treat him for every type of bacteria and the patient doesn’t get better. So, you treat him for a fungus because the patient may have an undiagnosed fungal infection. So, antifungal treatment may also be the last alterative you have when the patient is really in a serious situation. (Respondent 5)
Table 3 Barriers to adequate antifungal prescription and facilitators of guideline implementation Barriers to adequate antifungal prescription Issues related to prescription Initiation of antifungal use Uncommon prescription practice Issues related to prescription support Individual actions Integration of the team for prescriptions Computer systems Development and use of protocols Issues related to search for knowledge Importance attached to obtaining knowledge Sources of knowledge
Facilitators of guideline implementation
Establishment of local epidemiology Local antifungal guidelines: to whom, when, which and how much Individual knowledge Pharmacist to enhance patient safety; laboratory and intensivist integration Therapeutic decision-making process in computer programs Protocols based on reputable guidelines (e.g. IDSA guidelines) Permanent education Reliable source of literature
IDSA, Infectious Diseases Society of America.
462
© 2014 John Wiley & Sons, Ltd.
T.P. Salci et al.
Problems with diagnosing infection may delay the initiation of antifungal treatment and hamper timely treatment. Moreover, the physicians admitted that clinical reasoning delays the insertion of antifungal use. We use it later . . . a septic patient . . . prolonged stay in the ICU, he already made use of the entire arsenal of antibiotics and has septic decline (we call it septic decline), what do you do? You broaden the coverage for fungi. (Respondent 1)
Uncommon prescription practice Some of the physicians reported not feeling secure with regard to prescribing antifungal medication. Because it is not a common practice, health professionals often have little knowledge and become insecure about indicating this type of drug. [. . .] we have very limited knowledge on fungi and antifungal medications. So, it’s hard to prescribe. (Respondent 4)
Issues related to prescription support Individual actions In individual actions, commitment is a determining factor to improve any group or system. Engaged professionals benefit the service. We work in a multidisciplinary team in the ICU and, for many people, this job ‘is THE job’ and they make it their lives and dedicate themselves more. For others, such as temporary duty at the ICU, it as JUST ANOTHER job and, as they are only there temporarily, are not as committed to improvements. (Respondent 7) The health professionals cannot always count on the presence of colleagues to discuss the case and will often have to make a decision that will only later be discussed with the team. The autonomy to make decisions in such cases could be crucial for patient survival. But the intensivist on duty is somewhat solitary. He’s the one who has to decide on the conduct and sometimes doesn’t even have a second to spare to get advice. [. . .] You can’t always wait for the HICC (Hospital Infection Control Commission); normally you have to make the decision by yourself. (Respondent 1)
Team integration Teamwork is fundamental to create the parameters of care for each patient. Without teamwork, we would probably become lost with regard to treatment because of the serious condition of patients in the ICU. Such patients are a puzzle. You have to put all the pieces together, which takes time. If you have help, less time is needed to understand the case and, in intensive therapy, time translates to life. The faster you can put the pieces of the puzzle together, the greater the chance the patient has to leave the ICU alive and leave the hospital alive, with the fewest possible sequelae. (Respondent 3) Another respondent reported the conflicts experienced with regard to orientation from other health professionals. Because
© 2014 John Wiley & Sons, Ltd.
Problems related to antifungal
these professionals work in different places, they perceive that unjustified conduct and the lack of local protocols make the pharmacology approach unreliable. Although important, the confidence placed on a colleague may lead to difficulties voicing different opinions. You first think that it’s safe to have someone who’s going to study the case and give you advice or someone already knows about the subject and you end up taking it for granted. Another aspect is that sometimes the people in whom you are placing this responsibility are not always looking for knowledge [. . .] you fall into a conflict when you think to use an antifungal that you saw used in the routine in other hospitals and the person in charge tells you to use another one. (Respondent 6) Sometimes, physicians in the HICC supervise and discuss cases with intensivists and offer orientation with regard to prescriptions. Only one of the respondents reported that he/she also relied on the pharmacist, in addition to the members of the HICC, for support in different aspects of the prescription. [. . .] we have recently counted more and more on pharmacologists. Pharmacologists have been crucial to remind us about the many drug interactions, since we often use 20, 25, or 30 items in a prescription, for which there is certainly going to be an interaction, bioavailability of the drug in different parts of the organism, and dependence on the serum protein concentration. (Respondent 2) Another participant stressed the importance of interactions among the team to minimize errors. There must be interaction of the team, the interaction of all health professionals. Sometimes you have to impede a prescription, for instance. ‘This prescription won’t do. It’s not right.’ But a lot of things are not impeded. It seems like sometimes there’s a fear of the physician and the wrong prescription is carried forward. This creates a culture of errors. (Respondent 1) However, according to some of the respondents, discussions only occur within the same professional category (i.e. between physicians) because they do not see other professionals as partners or capable of contributing to the team. The discussion on the type of antibiotic or antifungal drug that is going to be used [. . .] since the specialists in infectology are responsible for the HICC, it is basically these professionals, but we have others as well: physicians from nephrology, pneumologists, cardiologists, neurologists, neurosurgeons, general surgery, and the intensivists themselves. The infectologists are basically the most important ones. (Respondent 5)
Computer systems Some respondents reported feeling the need for improvements to facilitate the prescription process, such as computerized systems, because this kind of support can aid the therapeutic decisionmaking process, something that allows prescriptions based on protocols and the enforcement of such local protocols. I would very much like to have a medical information system on the computer that could help in the decision-making process that allowed prescribing based on a specific protocol and even forced the use of this protocol. (Respondent 3) 463
Problems related to antifungal
Local protocols With regard to hospital conduct, the respondents mentioned that the lack of feedback from the support team is a problem. The determination of local epidemiology with further divulgence of this standard would facilitate the decision-making process and support the health professionals. [. . .] sometimes we don’t have any feedback. For instance, I have difficulty knowing what the epidemiological standard of the university hospital is, because I don’t get any feedback. (Respondent 6) Guidelines on the choice of antifungal medication, when to initiate treatment, and the duration of treatment can effectively assist prescribing physicians. The lack of a local protocol that is in agreement with standard guidelines can have a negative effect on treatment because of the lack of consistent conduct and the time that elapses during the decision-making process until initiation of the indicated treatment. I think the first thing is systematization – a protocol, a routine . . . For patient x, y, and z, you use this medication. If the patient is stable, you use this medication. If the patient is unstable, you use so-and-so . . . when do you stop? If there are 14 days after the hemoculture, 15 days . . . make something formalized. (Respondent 6)
Issues related to search for knowledge Importance attached to obtaining knowledge The need for further knowledge was highlighted by one of the respondents. Support from the scientific literature was described, in which the participant understood the need for constant updating on measures for the diagnosis and treatment of fungal infections. Something that we must constantly seek out is whether there has been a change in the evidence supporting our view of the risk of fungal infection. (Respondent 2) Others do not prioritize the act of seeking knowledge and overlook the importance of this. Since it is not part of the daily routine for you to seek knowledge, unlike in the case of bacteria and sepsis, for instance, for which we have all the most recent knowledge on viruses, bacteria . . . Fungi, on the other hand, even though such cases are more and more common, it’s not . . . it’s not the current fashion. (Respondent 3) Some of the respondents admitted not having interest in seeking knowledge on antifungal treatment because of a feeling that the issue is not a priority for these individuals. [. . .] my current objective is not related to fungi [. . .] But, despite working in the ICU, I don’t have any . . . , let’s say, I don’t seek specific knowledge on fungi, but it’s because of my other current priorities, because I’m studying other things. (Respondent 6) Sources of knowledge The readily available knowledge from the pharmaceutical industry often makes this the first source of data on antifungal drugs. What comes to us first: The sales reps bring papers; they make a visit. You know, the ICU is visited by laboratory 464
T.P. Salci et al.
representatives every day. They bring specific literature, up-to-date literature, I mean, on what is happening. (Respondent 5) Moreover, the majority of the respondents stated that the first option in seeking knowledge is through technological resources. [. . .] The Internet. I have a cell phone with a program. If I have any questions, I search it, which is the ‘ABX guide’. (Respondent 4) Ah, today it’s the Internet. On the Internet, you get every type of study. (Respondent 6) When asked to be more specific about the Internet, the search tool option was considered a source of knowledge. [. . .] for example, Doctor Google. (Respondent 1) The most frequently searched questions are about dose, adjusting the dose, the choice of drug and toxicity. [. . .] dose, the ideal antifungal drug, when the fungus is identified, the ideal dose, whether adjustments need to be made due to kidney or liver failure. (Respondent 4)
Interconnection of themes In the system represented by our model (Fig. 1), the main insights come from various interconnected reasons for the lack of conformity with regard to prescription patterns for this class of drugs, issues that may be related to the three main themes discussed herein. A negative cycle was perceived based on issues related to prescriptions and searching for knowledge (i.e. the intensivists overlooked the importance of fungal infection, did not seek knowledge and consequently did not improve the prescription). If the problems related to individual actions and multidisciplinary team integration are resolved, with implementation of local protocols based on local epidemiology, then the barriers to proper prescriptions could be overcome. To establish antifungal prescriptions, such questions as ‘for whom’, ‘when’, ‘which (antifungal)’ and ‘how much’ should be answered with the purpose of reducing the prescription barriers that face the unusual practice of prescribing antifungals by intensivists.
Discussion To our knowledge, this is the first qualitative investigation of how intensivists regard fungal infections and antifungal prescriptions and their perceived barriers to the adequate prescription of antifungal drugs in an adult ICU. Despite the wide range of research on the epidemiology of invasive candidiasis and growing importance of ICU infections related to these microorganisms, some problems related to antifungal therapy are common. The first professional who is faced with situations of infection risk or the infection itself is often the intensivist. Because of the severity of these patients, action must be taken immediately to contain the spread of the infection. Qualitative methodology was chosen in the present study to understand this complex phenomenon. Therefore, various reasons were described for the lack of conformity with regard to prescription patterns for this class of drugs. The article ‘How to select an antifungal agent in critically ill patients’ was recently published. This paper reported the points to be followed by different ICUs to guide the selection of antifungal agents and proposed an algorithm to aid this choice in critically ill
© 2014 John Wiley & Sons, Ltd.
T.P. Salci et al.
Problems related to antifungal
Figure 1 The negative cycle that hampers the correct prescription of antifungal medication and points that should be considered to overcome these barriers. A central negative cycle is represented by a gear. Intensivists overlook the importance of invasive candidiasis, do not seek further knowledge and do not improve the prescription. The left arrow represents the actions that are necessary to stop this gear and overcome the barriers to proper antifungal prescription, including individual actions, multidisciplinary team integration and local protocol implementation (computer system utilization can be auxiliary). Finally, the right arrow represents the main questions that should be answered to establish and improve antifungal prescription.
patients [19]. Other authors have highlighted the need to know local epidemiology and establish local antifungal utilization guides [19,20]. Reinforcing this theory, a recent study by our group reported the presence of problems related to antifungal utilization [8], indicating the need for protocols that guide ‘to whom’, ‘when’, ‘which’ and ‘how much’. Identifying the need for antifungal therapy is not always easy. A fungal infection is proven in only 21.1% of cases that involve the use of antifungal drugs, and this may be attributable to the difficulty establishing an early diagnosis [21]. The respondents admitted that antifungal therapy is most often performed empirically and as an antimicrobial alternative when other antibacterial options have all been discarded. The respondents exhibited insecurity about the prescription of antifungal therapy because it is not a routine practice. Therefore, permanent education, feedback about local epidemiology, and statistics on the success or failure of therapeutics should be reported among ICU professionals [22]. Delays in the initiation of adequate antifungal therapy in critical patients have a significant impact on illness and mortality rates [15]. Indeed, the initiation of antifungal treatment appears to be a determinant for patient survival, in which each hour of delay in adequate treatment is associated with a mean reduction of 7.6% in patient survival [23,24]. These findings underscore the importance of early diagnosis, and early and effective treatment. Institutions should determine the distribution of fungal species and evaluate
© 2014 John Wiley & Sons, Ltd.
susceptibility to clinical isolates. Geographical variations in the incidence of fungal species and the associated decrease in susceptibility raise concerns about treatment options and more aggressive dosing strategies for early empirical antifungal approaches [4]. The availability of such information can contribute to the appropriate use of antifungal drugs by prescribers, thus minimizing the problems identified in our studies. The present study observed certain difficulty recognizing the necessary assistance of other health professionals beyond the different specialties of the physicians. These results contrast with evidence that demonstrates that the integration of a multidisciplinary health care team reduces the mortality rate in the ICU, especially among patients in high-risk groups, such as those with sepsis [25]. Indeed, teamwork and the assistance of health professionals with different backgrounds are needed [26]. With regard to prescriptions and interdisciplinary support, numerous studies reported the importance of a pharmacist and integration with laboratory staff to enhance patient safety, ensure therapeutic efficacy, reduce costs and improve the quality of care offered [27]. Although studies and guidelines indicate the correct use of antifungal medications, conduct that diverges from that which has been established in the literature is often seen in clinical practice. Regardless of the quality of the health care offered, errors still occur in the use of medications. Insecurity about prescribing this class of drugs was described by some of the respondents. 465
Problems related to antifungal
Understanding the dimensions of the problems that arise from the inadequate use of antifungal drugs will permit the determination of measures needed to foster rational use, thus creating a setting for discussions of proper conduct and implementation of necessary changes [11]. These measures mainly involve the development of conduct to facilitate the appropriate use of antifungal drugs by prescribers, which must be conducted according to guidelines, such as those proposed by the IDSA, and suited to local epidemiological data. The present study has limitations that should be addressed, the most important of which was the limited sample size (n = 8). Notably, however, the eight intensivists who were recruited surpassed the minimum criterion for the methodology, and the data generated answered the research question, with a saturation of the data, with no further emerging topics. Moreover, the interviews occurred in a single hospital, and the central objective of the present study was related to the intensivists’ point of view. Therefore, future studies should consider physicians who work in multiple locations to provide a better overview of various ICUs. Notwithstanding these limitations, the present findings can assist in drafting plans for treatment optimization. This investigation also demonstrated the need for prescription assistance with support from a well-trained, multidisciplinary team with agreement among its members and importance of well-designed local protocols. In conclusion, numerous aspects influence the adequate prescription of antifungal medication. Knowledge of these aspects can assist in drafting utilization measures, providing educational interventions and ending the negative cycle. Such knowledge can also contribute to the implementation of fungal infection control strategies and development of local protocols to improve the quality of treatment in the hospital setting, especially in the ICU.
T.P. Salci et al.
9.
10. 11.
12.
13.
14.
15.
16.
17.
18. 19.
Acknowledgement We thank the Brazilian Agency CAPES for financial support.
References 1. Vincent, J. L., Rello, J., Marshall, J., et al. (2009) International study of the prevalence and outcomes of infection in intensive care units. JAMA, 302 (21), 2323–2329. 2. Yap, H. Y., Kwok, K. M., Gomersall, C. D., et al. (2009) Epidemiology and outcome of Candida bloodstream infection in an intensive care unit in Hong Kong. Hong Kong Medical Journal, 15 (4), 255–261. 3. Alangaden, G. J. (2011) Nosocomial fungal infections: epidemiology, infection control, and prevention. Infectious Disease Clinics of North America, 25 (1), 201–225. 4. Pappas, P. G., Kauffman, C. A., Andes, D., et al. (2009) Clinical practice guidelines for the management of candidiasis: 2009 update by the Infectious Diseases Society of America. Clinical Infectious Diseases, 48 (5), 503–535. 5. Arnold, H. M., Micek, S. T., Shorr, A. F., et al. (2010) Hospital resource utilization and costs of inappropriate treatment of candidemia. Pharmacotherapy, 30 (4), 361–368. 6. Patterson, T. F. (2005) Advances and challenges in management of invasive mycoses. Lancet, 366 (9490), 1013–1025. 7. De Pauw, B. E. (2006) Increasing fungal infections in the intensive care unit. Surgical Infections, 7 (Suppl. 2), S93–S96. 8. Salci, T. P., Gimenes, M., dos Santos, C. A., Svidzinski, T. I. E. & Caparroz-Assef, S. M. (2013) Utilization of fluconazole in an
466
20.
21.
22. 23.
24.
25.
26.
27.
intensive care unit at a university hospital in Brazil. International Journal of Clinical Pharmacy, 35 (2), 176–180. Lee, I., Fishman, N. O., Zaoutis, T. E., et al. (2009) Risk factors for fluconazole-resistant Candida glabrata bloodstream infections. Archives of Internal Medicine, 169 (4), 379–383. Shoham, S. & Marwaha, S. (2010) Invasive fungal infections in the ICU. Journal of Intensive Care Medicine, 25 (2), 78–92. Swoboda, S., Lichtenstern, C., Ober, M. C., et al. (2009) Implementation of practice guidelines for antifungal therapy in a surgical intensive care unit and its impact on use and costs. Chemotherapy, 55 (6), 418–424. Wood-Mitchell, A., James, I. A., Waterworth, A., Swann, A. & Ballard, C. (2008) Factors influencing the prescribing of medications by old age psychiatrists for behavioural and psychological symptoms of dementia: a qualitative study. Age and Ageing, 37 (5), 547–552. Ljungberg, C., Lindblad, A. K. & Tully, M. P. (2007) Hospital doctors’ views of factors influencing their prescribing. Journal of Evaluation in Clinical Practice, 13 (5), 765–771. Cavazos, J. M., Naik, A. D., Woofter, A. & Abraham, N. S. (2008) Barriers to physician adherence to nonsteroidal anti-inflammatory drug guidelines: a qualitative study. Alimentary Pharmacology and Therapeutics, 28 (6), 789–798. Garey, K. W., Rege, M., Pai, M. P., et al. (2006) Time to initiation of fluconazole therapy impacts mortality in patients with candidemia: a multi-institutional study. Clinical Infectious Diseases : an Official Publication of the Infectious Diseases Society of America, 43 (1), 25–31. Corbin, J. M. & Strauss, A. L. (2008) Basics of Qualitative Research: Techniques and Procedures for Developing Grounded Theory. Los Angeles: Sage Publications. McLellan, E., MacQueen, K. M. & Neidig, J. L. (2003) Beyond the qualitative interview: data preparation and transcription. Field Methods, 15 (1), 63–84. Charmaz, K. (2009) Constructing Grounded Theory: A Practical Guide through Qualitative Analysis. Los Angeles: Sage Publications. Dimopoulos, G., Antonopoulou, A., Armaganidis, A. & Vincent, J. L. (2013) How to select an antifungal agent in critically ill patients. Journal of Critical Care, 28 (5), 717–727. Lepak, A. & Andes, D. (2011) Fungal sepsis: optimizing antifungal therapy in the critical care setting. Critical Care Clinics, 27 (1), 123– 147. Alvarez-Lerma, F., Palomar, M., León, C., et al. (2004) [Indications for antifungal treatment in intensive care unit patients]. Enfermedades Infecciosas Y Microbiologia Clinica, 22 (5), 279–285. Lewis, R. E. (2011) Current concepts in antifungal pharmacology. Mayo Clinic Proceedings, 86 (8), 805–817. Kumar, A., Roberts, D., Wood, K. E., et al. (2006) Duration of hypotension before initiation of effective antimicrobial therapy is the critical determinant of survival in human septic shock. Critical Care Medicine, 34 (6), 1589–1596. Falagas, M. E., Apostolou, K. E. & Pappas, V. D. (2006) Attributable mortality of candidemia: a systematic review of matched cohort and case-control studies. European Journal of Clinical Microbiology and Infectious Diseases, 25 (7), 419–425. Kim, M. M., Barnato, A. E., Angus, D. C., Fleisher, L. A., Fleisher, L. F. & Kahn, J. M. (2010) The effect of multidisciplinary care teams on intensive care unit mortality. Archives of Internal Medicine, 170 (4), 369–376. Collins, S. A. & Currie, L. M. (2009) Interdisciplinary communication in the ICU. Studies in Health Technology and Informatics, 146, 362– 366. Erstad, B. L., Haas, C. E., O’Keeffe, T., Hokula, C. A., Parrinello, K. & Theodorou, A. A. (2011) Interdisciplinary patient care in the intensive care unit: focus on the pharmacist. Pharmacotherapy, 31 (2), 128–137.
© 2014 John Wiley & Sons, Ltd.
Journal of Evaluation in Clinical Practice ISSN 1365-2753
Problems related to antifungal prescription: a qualitative study of the views of intensivists Tânia Pereira Salci MSc,1 Adelia P. Batilana,2 Ricardo Pietrobon MD PhD3 and Silvana Martins Caparroz-Assef PhD4 1
PhD Student, Post Graduate Program in Biosciences Applied to Pharmacy, Universidade Estadual de Maringá, Maringá, Paraná, Brazil Master Student, Research on Research Group, Duke University Health System, Durham, North Carolina, USA 3 Associate Professor, Department of Surgery, Duke University Health System, Durham, North Carolina, USA 4 Associate Professor, Pharmacology and Therapeutic Department, Universidade Estadual de Maringá, Maringá, Paraná, Brazil 2
Keywords antifungal, barriers to prescription, intensive care unit, intensivist opinion, qualitative research, quality of health care Correspondence Mrs Tânia Pereira Salci Post Graduate Program in Biosciences Applied to Pharmacy Universidade Estadual de Maringá Av. Colombo, n. 5790, bl. K68, sl. 102b Maringá, Paraná 87020-900 Brazil E-mail: [email protected] Accepted for publication: 17 April 2014 doi:10.1111/jep.12181
Abstract Rationale and objective The choice of the appropriate antifungal medication is essential for therapeutic success. Although guidelines are available in the literature that regulate the consistent use of antifungal, no previous qualitative studies have addressed the difficulties related to the use of antifungal medication, especially in the intensive care unit (ICU). Our objective was to qualitatively investigate how intensivists consider antifungal prescriptions in an adult ICU. Methods The Grounded Theory analytical method was used for the data analysis. Physicians who worked in the adult ICU and prescribed antifungal medications were individually interviewed. A semi-structured interview was used to ask core questions, followed by follow-up questions at the discretion of the interviewer. Results Our analysis generated eight main emerging themes that were classified into three related groups. The main insights were that various interconnected reasons were given for the lack of conformity with regard to prescription patterns for antifungals. A negative cycle was perceived based on issues related to prescriptions and the search for knowledge. If problems related to individual actions and multidisciplinary team integration are resolved and local protocols are implemented based on local epidemiology, then barriers to proper prescriptions can be overcome when intensivists are faced with the unusual practice of prescribing antifungals. Conclusions Our investigation indicates the need for prescription assistance with support from a well-trained multidisciplinary team and consensus among its members and the importance of well-designed protocols.
Introduction Fungi are among the main causal agents of infection in the intensive care unit (ICU). Candida spp., in particular, are among the leading causes of bloodstream infections and sepsis [1]. Fungal infections are associated with mortality rates of up to 70% [2] and high morbidity rates because of the difficulty establishing an early diagnosis and effective treatment [3]. The choice of the appropriate antifungal drug, early treatment, and correct dosage and duration of treatment are essential aspects for therapeutic success [4]. However, approximately 80% of hospitalized patients receive inadequate antifungal treatment, leading to increases in hospital stays and health care costs of up to USD$17 000 per patient [5]. Considering the information available in the literature to guide antifungal therapy, some problems have been identified, such as 460
limited diagnostic options [6], difficulty establishing early treatment [7], difficulty determining the correct dosage and duration of treatment [8], microorganism resistance [9], high cost of treatment [5], and high mortality rate [10]. Thus, rigorous treatment protocols are needed that seek to minimize such problems [11]. The Infectious Diseases Society of America (IDSA) has created evidence-based guidelines for the management of invasive candidiasis to aid therapeutic choices [4]. Furthermore, the authors stress the importance of implementing local protocols for the treatment and control of fungal infections, especially invasive candidiasis, with the aims of diminishing the indiscriminate and irrational use of this class of drugs, reducing costs, and providing greater safety and efficacy [4,11]. The inadequate use of drugs has been discussed in qualitative studies that investigate the factors that influence the prescription of
Journal of Evaluation in Clinical Practice 20 (2014) 460–466 © 2014 John Wiley & Sons, Ltd.
T.P. Salci et al.
Problems related to antifungal
different classes of drugs [12,13] and barriers that limit compliance with standardized protocols [14]. Antifungal drugs are not beyond the scope of this context because problems related to the use of these medications have been demonstrated [15]. However, no previous qualitative studies have addressed the difficulties related to the use of antifungal medication, especially in the ICU. Considering this gap in the literature, the aim of the present study was to qualitatively investigate how intensivists assess fungal infection and antifungal prescriptions, and perceive barriers to the adequate prescription of antifungal drugs in an adult ICU.
Methods Ethics statement This study received approval from the Institutional Review Board of Universidade Estadual de Maringá, Brazil. All of the subjects were informed that their statements would be used for research purposes only, and written informed consent was obtained before the interview.
Study design A qualitative research strategy was used to gain insights into organizational and cultural issues within the ICU. Qualitative research methods have been developed in social sciences to enable researchers to better understand people and the social and cultural contexts within which they live. Studying educational settings and processes is particularly useful [16]. The Consolidated Criteria for Reporting Qualitative Research was used to provide guidelines in the design of the study [17]. The Grounded Theory analytical method [16] was used as the theoretical framework, which intended to explore the phenomena of meaning and action in the context in which they occur. The data were collected from July to November 2012.
Research team and reflexivity The researchers involved in the present study were a female pharmacist MSc researcher (TS), a female PhD pharmacist (SA), a female researcher (AB) and a male MD/PhD (RP). All of the authors are active clinical researchers who are involved in studies to improve protocols for drug prescriptions. Researchers TS and SA have been involved in antifungal research since 2009. RP and AB are research methodologists with no previous work or preconceptions on the use and misuse of antifungal medications. All of the respondents were interviewed by the same interviewer (TS). The respondents and interviewer (TS) had no prior contact
Table 1 Topics covered by the interview guide
© 2014 John Wiley & Sons, Ltd.
• • • • • •
and no common interests or working relationship. No conflict of interest was declared by any of the researchers with regard to this area of research.
Recruitment All of the physicians who worked in the adult ICU at the university hospital were asked to participate in the study. Eight physicians were individually interviewed face-to-face, and three physicians refused to participate in the study without providing a reason for non-participation. No exclusion criteria were applied. For the selection of respondents, contact was made with the director of the ICU, who provided a list of physicians. The first contact with the physicians was made during working hours. The respondents were informed of the identity of the interviewer, the importance and objectives of the study, and how the interview would be conducted. The participant then decided either to be interviewed at that time or to schedule a more convenient time.
Interview procedure The interviews occurred at the workplace in a reserved room. Only the interviewer and respondent were present during the interview. Following the provision of informed consent, the interviews lasted approximately 40 minutes and were digitally recorded. Field notes were made after each interview to avoid interruptions during the interview itself.
Data collection The interview began with questions about age, gender, years of experience, number of hours worked per week and number of institutions where the respondent worked. A semi-structured interview was used (Table 1) to ask core questions verbatim, followed by additional follow-up questions at the discretion of the interviewer. To enhance consistency, all of the transcriptions were performed by the interviewer (TS) immediately after data collection [17]. Transcriptions were recorded and available to the other researchers so they all could review and confirm the information. The identification of the respondents was removed from the transcriptions to ensure confidentiality and anonymity. A field log was kept to assist the coding of the data. The research team discussed the data, and inferences were made after each interview to address the research topic.
Data analysis The research team (SA and TS) analysed all of the transcribed interviews individually according to the steps of Grounded Theory: (i) initial coding; (ii) focused coding; and (iii) axial coding [18].
Reasons for initiating antifungal medication; Resources that support the prescription; Frequently required information; Assistance from colleagues regarding prescription; Confidence in prescribing antifungal medication; and Suggestions for contributing to and/or facilitating the prescription of antifungal medication.
461
Problems related to antifungal
T.P. Salci et al.
remaining results describe these emerging themes and corresponding groups in detail (Table 3).
Coding One researcher (TS) coded each transcription. Discrepancies were discussed and settled among the team. The quotations used in the present report were translated from Brazilian Portuguese to English by the author (RP) and independently back-translated by the other bilingual researchers (SA, TS). The data were saturated with regard to new emerging themes, beginning with the seventh interview.
Results Demographics and participant characteristics The majority of the subjects were male and younger than 50 years of age. All of them worked at more than one institution and in other ICUs. The mean weekly working hours was 72.3 ± 23.3 hours. Their clinical experience in the ICU was a minimum of 13 years and a maximum of 23 years (mean, 15.6 ± 4.0 years; Table 2).
Emerging themes and related quotations Our analysis generated eight main emerging themes: the initiation of antifungal use, uncommon prescription practices, individual actions, team integration, computerized systems, local protocols, importance attached to obtaining knowledge and sources of knowledge. These themes were classified into three related groups: prescription, prescription support and search for knowledge. The
Table 2 Characteristics of the respondents Characteristics
Average ± standard deviation
Age (years) Gender Male* Female* n of institutions working n of ICU working Average week work (hours) Average years in practice Average experience in ICU
42.9 ± 5.4 6 (75.0) 2 (25.0) 3.4 ± 1.5 2.7 ± 1.2 72.3 ± 23.3 18.3 ± 5.3 15.6 ± 4.0
*Frequency (%).
Issues related to prescription Time to initiate the antifungal use The use of antifungal drugs is justified when the fungus is identified as an etiological agent of infection based on positive laboratory exams. First, we have a positive culture to treat. If there is a culture, we begin antifungal treatment. (Respondent 7) However, this conduct is not always possible. The problem is response time. Reliable (laboratorial) feedback takes a long time. How long? What fungus is it? What’s the antibiogram? This always takes a long time, which, in intensive care, can sometimes be crucial. (Respondent 7) Thus, the clinical evaluation of the patient and recognition of risk factors are as essential as comorbidities and invasive procedures. Most patients in the ICU have at least one risk factor for the development of candidiasis. The microbiological identification of fungi can come too late or may even be hindered by the prior use of an antimicrobial agent. Therefore, the empirical initiation of treatment is necessary. [. . .] patients who have risk factors, for which we request a culture, but the results haven’t come back yet and there are risk factors. So, we take the liberty to start them on an antifungal. (Respondent 7) Frequently, the first choice of treatment is a broad-spectrum antibiotic. Generally, the main clinical suspicion is a bacterial infection. Therefore, the physicians cite that the prescription of antifungal drugs is their last thought when there is prolonged and ineffective antibacterial treatment with no resolution of the signs and symptoms of the infectious condition. And in the case of fungi in general, there is also something that is very common in the ICU: you treat the patient for a bacterial infection and he doesn’t get better. You treat him for every type of bacteria and the patient doesn’t get better. So, you treat him for a fungus because the patient may have an undiagnosed fungal infection. So, antifungal treatment may also be the last alterative you have when the patient is really in a serious situation. (Respondent 5)
Table 3 Barriers to adequate antifungal prescription and facilitators of guideline implementation Barriers to adequate antifungal prescription Issues related to prescription Initiation of antifungal use Uncommon prescription practice Issues related to prescription support Individual actions Integration of the team for prescriptions Computer systems Development and use of protocols Issues related to search for knowledge Importance attached to obtaining knowledge Sources of knowledge
Facilitators of guideline implementation
Establishment of local epidemiology Local antifungal guidelines: to whom, when, which and how much Individual knowledge Pharmacist to enhance patient safety; laboratory and intensivist integration Therapeutic decision-making process in computer programs Protocols based on reputable guidelines (e.g. IDSA guidelines) Permanent education Reliable source of literature
IDSA, Infectious Diseases Society of America.
462
© 2014 John Wiley & Sons, Ltd.
T.P. Salci et al.
Problems with diagnosing infection may delay the initiation of antifungal treatment and hamper timely treatment. Moreover, the physicians admitted that clinical reasoning delays the insertion of antifungal use. We use it later . . . a septic patient . . . prolonged stay in the ICU, he already made use of the entire arsenal of antibiotics and has septic decline (we call it septic decline), what do you do? You broaden the coverage for fungi. (Respondent 1)
Uncommon prescription practice Some of the physicians reported not feeling secure with regard to prescribing antifungal medication. Because it is not a common practice, health professionals often have little knowledge and become insecure about indicating this type of drug. [. . .] we have very limited knowledge on fungi and antifungal medications. So, it’s hard to prescribe. (Respondent 4)
Issues related to prescription support Individual actions In individual actions, commitment is a determining factor to improve any group or system. Engaged professionals benefit the service. We work in a multidisciplinary team in the ICU and, for many people, this job ‘is THE job’ and they make it their lives and dedicate themselves more. For others, such as temporary duty at the ICU, it as JUST ANOTHER job and, as they are only there temporarily, are not as committed to improvements. (Respondent 7) The health professionals cannot always count on the presence of colleagues to discuss the case and will often have to make a decision that will only later be discussed with the team. The autonomy to make decisions in such cases could be crucial for patient survival. But the intensivist on duty is somewhat solitary. He’s the one who has to decide on the conduct and sometimes doesn’t even have a second to spare to get advice. [. . .] You can’t always wait for the HICC (Hospital Infection Control Commission); normally you have to make the decision by yourself. (Respondent 1)
Team integration Teamwork is fundamental to create the parameters of care for each patient. Without teamwork, we would probably become lost with regard to treatment because of the serious condition of patients in the ICU. Such patients are a puzzle. You have to put all the pieces together, which takes time. If you have help, less time is needed to understand the case and, in intensive therapy, time translates to life. The faster you can put the pieces of the puzzle together, the greater the chance the patient has to leave the ICU alive and leave the hospital alive, with the fewest possible sequelae. (Respondent 3) Another respondent reported the conflicts experienced with regard to orientation from other health professionals. Because
© 2014 John Wiley & Sons, Ltd.
Problems related to antifungal
these professionals work in different places, they perceive that unjustified conduct and the lack of local protocols make the pharmacology approach unreliable. Although important, the confidence placed on a colleague may lead to difficulties voicing different opinions. You first think that it’s safe to have someone who’s going to study the case and give you advice or someone already knows about the subject and you end up taking it for granted. Another aspect is that sometimes the people in whom you are placing this responsibility are not always looking for knowledge [. . .] you fall into a conflict when you think to use an antifungal that you saw used in the routine in other hospitals and the person in charge tells you to use another one. (Respondent 6) Sometimes, physicians in the HICC supervise and discuss cases with intensivists and offer orientation with regard to prescriptions. Only one of the respondents reported that he/she also relied on the pharmacist, in addition to the members of the HICC, for support in different aspects of the prescription. [. . .] we have recently counted more and more on pharmacologists. Pharmacologists have been crucial to remind us about the many drug interactions, since we often use 20, 25, or 30 items in a prescription, for which there is certainly going to be an interaction, bioavailability of the drug in different parts of the organism, and dependence on the serum protein concentration. (Respondent 2) Another participant stressed the importance of interactions among the team to minimize errors. There must be interaction of the team, the interaction of all health professionals. Sometimes you have to impede a prescription, for instance. ‘This prescription won’t do. It’s not right.’ But a lot of things are not impeded. It seems like sometimes there’s a fear of the physician and the wrong prescription is carried forward. This creates a culture of errors. (Respondent 1) However, according to some of the respondents, discussions only occur within the same professional category (i.e. between physicians) because they do not see other professionals as partners or capable of contributing to the team. The discussion on the type of antibiotic or antifungal drug that is going to be used [. . .] since the specialists in infectology are responsible for the HICC, it is basically these professionals, but we have others as well: physicians from nephrology, pneumologists, cardiologists, neurologists, neurosurgeons, general surgery, and the intensivists themselves. The infectologists are basically the most important ones. (Respondent 5)
Computer systems Some respondents reported feeling the need for improvements to facilitate the prescription process, such as computerized systems, because this kind of support can aid the therapeutic decisionmaking process, something that allows prescriptions based on protocols and the enforcement of such local protocols. I would very much like to have a medical information system on the computer that could help in the decision-making process that allowed prescribing based on a specific protocol and even forced the use of this protocol. (Respondent 3) 463
Problems related to antifungal
Local protocols With regard to hospital conduct, the respondents mentioned that the lack of feedback from the support team is a problem. The determination of local epidemiology with further divulgence of this standard would facilitate the decision-making process and support the health professionals. [. . .] sometimes we don’t have any feedback. For instance, I have difficulty knowing what the epidemiological standard of the university hospital is, because I don’t get any feedback. (Respondent 6) Guidelines on the choice of antifungal medication, when to initiate treatment, and the duration of treatment can effectively assist prescribing physicians. The lack of a local protocol that is in agreement with standard guidelines can have a negative effect on treatment because of the lack of consistent conduct and the time that elapses during the decision-making process until initiation of the indicated treatment. I think the first thing is systematization – a protocol, a routine . . . For patient x, y, and z, you use this medication. If the patient is stable, you use this medication. If the patient is unstable, you use so-and-so . . . when do you stop? If there are 14 days after the hemoculture, 15 days . . . make something formalized. (Respondent 6)
Issues related to search for knowledge Importance attached to obtaining knowledge The need for further knowledge was highlighted by one of the respondents. Support from the scientific literature was described, in which the participant understood the need for constant updating on measures for the diagnosis and treatment of fungal infections. Something that we must constantly seek out is whether there has been a change in the evidence supporting our view of the risk of fungal infection. (Respondent 2) Others do not prioritize the act of seeking knowledge and overlook the importance of this. Since it is not part of the daily routine for you to seek knowledge, unlike in the case of bacteria and sepsis, for instance, for which we have all the most recent knowledge on viruses, bacteria . . . Fungi, on the other hand, even though such cases are more and more common, it’s not . . . it’s not the current fashion. (Respondent 3) Some of the respondents admitted not having interest in seeking knowledge on antifungal treatment because of a feeling that the issue is not a priority for these individuals. [. . .] my current objective is not related to fungi [. . .] But, despite working in the ICU, I don’t have any . . . , let’s say, I don’t seek specific knowledge on fungi, but it’s because of my other current priorities, because I’m studying other things. (Respondent 6) Sources of knowledge The readily available knowledge from the pharmaceutical industry often makes this the first source of data on antifungal drugs. What comes to us first: The sales reps bring papers; they make a visit. You know, the ICU is visited by laboratory 464
T.P. Salci et al.
representatives every day. They bring specific literature, up-to-date literature, I mean, on what is happening. (Respondent 5) Moreover, the majority of the respondents stated that the first option in seeking knowledge is through technological resources. [. . .] The Internet. I have a cell phone with a program. If I have any questions, I search it, which is the ‘ABX guide’. (Respondent 4) Ah, today it’s the Internet. On the Internet, you get every type of study. (Respondent 6) When asked to be more specific about the Internet, the search tool option was considered a source of knowledge. [. . .] for example, Doctor Google. (Respondent 1) The most frequently searched questions are about dose, adjusting the dose, the choice of drug and toxicity. [. . .] dose, the ideal antifungal drug, when the fungus is identified, the ideal dose, whether adjustments need to be made due to kidney or liver failure. (Respondent 4)
Interconnection of themes In the system represented by our model (Fig. 1), the main insights come from various interconnected reasons for the lack of conformity with regard to prescription patterns for this class of drugs, issues that may be related to the three main themes discussed herein. A negative cycle was perceived based on issues related to prescriptions and searching for knowledge (i.e. the intensivists overlooked the importance of fungal infection, did not seek knowledge and consequently did not improve the prescription). If the problems related to individual actions and multidisciplinary team integration are resolved, with implementation of local protocols based on local epidemiology, then the barriers to proper prescriptions could be overcome. To establish antifungal prescriptions, such questions as ‘for whom’, ‘when’, ‘which (antifungal)’ and ‘how much’ should be answered with the purpose of reducing the prescription barriers that face the unusual practice of prescribing antifungals by intensivists.
Discussion To our knowledge, this is the first qualitative investigation of how intensivists regard fungal infections and antifungal prescriptions and their perceived barriers to the adequate prescription of antifungal drugs in an adult ICU. Despite the wide range of research on the epidemiology of invasive candidiasis and growing importance of ICU infections related to these microorganisms, some problems related to antifungal therapy are common. The first professional who is faced with situations of infection risk or the infection itself is often the intensivist. Because of the severity of these patients, action must be taken immediately to contain the spread of the infection. Qualitative methodology was chosen in the present study to understand this complex phenomenon. Therefore, various reasons were described for the lack of conformity with regard to prescription patterns for this class of drugs. The article ‘How to select an antifungal agent in critically ill patients’ was recently published. This paper reported the points to be followed by different ICUs to guide the selection of antifungal agents and proposed an algorithm to aid this choice in critically ill
© 2014 John Wiley & Sons, Ltd.
T.P. Salci et al.
Problems related to antifungal
Figure 1 The negative cycle that hampers the correct prescription of antifungal medication and points that should be considered to overcome these barriers. A central negative cycle is represented by a gear. Intensivists overlook the importance of invasive candidiasis, do not seek further knowledge and do not improve the prescription. The left arrow represents the actions that are necessary to stop this gear and overcome the barriers to proper antifungal prescription, including individual actions, multidisciplinary team integration and local protocol implementation (computer system utilization can be auxiliary). Finally, the right arrow represents the main questions that should be answered to establish and improve antifungal prescription.
patients [19]. Other authors have highlighted the need to know local epidemiology and establish local antifungal utilization guides [19,20]. Reinforcing this theory, a recent study by our group reported the presence of problems related to antifungal utilization [8], indicating the need for protocols that guide ‘to whom’, ‘when’, ‘which’ and ‘how much’. Identifying the need for antifungal therapy is not always easy. A fungal infection is proven in only 21.1% of cases that involve the use of antifungal drugs, and this may be attributable to the difficulty establishing an early diagnosis [21]. The respondents admitted that antifungal therapy is most often performed empirically and as an antimicrobial alternative when other antibacterial options have all been discarded. The respondents exhibited insecurity about the prescription of antifungal therapy because it is not a routine practice. Therefore, permanent education, feedback about local epidemiology, and statistics on the success or failure of therapeutics should be reported among ICU professionals [22]. Delays in the initiation of adequate antifungal therapy in critical patients have a significant impact on illness and mortality rates [15]. Indeed, the initiation of antifungal treatment appears to be a determinant for patient survival, in which each hour of delay in adequate treatment is associated with a mean reduction of 7.6% in patient survival [23,24]. These findings underscore the importance of early diagnosis, and early and effective treatment. Institutions should determine the distribution of fungal species and evaluate
© 2014 John Wiley & Sons, Ltd.
susceptibility to clinical isolates. Geographical variations in the incidence of fungal species and the associated decrease in susceptibility raise concerns about treatment options and more aggressive dosing strategies for early empirical antifungal approaches [4]. The availability of such information can contribute to the appropriate use of antifungal drugs by prescribers, thus minimizing the problems identified in our studies. The present study observed certain difficulty recognizing the necessary assistance of other health professionals beyond the different specialties of the physicians. These results contrast with evidence that demonstrates that the integration of a multidisciplinary health care team reduces the mortality rate in the ICU, especially among patients in high-risk groups, such as those with sepsis [25]. Indeed, teamwork and the assistance of health professionals with different backgrounds are needed [26]. With regard to prescriptions and interdisciplinary support, numerous studies reported the importance of a pharmacist and integration with laboratory staff to enhance patient safety, ensure therapeutic efficacy, reduce costs and improve the quality of care offered [27]. Although studies and guidelines indicate the correct use of antifungal medications, conduct that diverges from that which has been established in the literature is often seen in clinical practice. Regardless of the quality of the health care offered, errors still occur in the use of medications. Insecurity about prescribing this class of drugs was described by some of the respondents. 465
Problems related to antifungal
Understanding the dimensions of the problems that arise from the inadequate use of antifungal drugs will permit the determination of measures needed to foster rational use, thus creating a setting for discussions of proper conduct and implementation of necessary changes [11]. These measures mainly involve the development of conduct to facilitate the appropriate use of antifungal drugs by prescribers, which must be conducted according to guidelines, such as those proposed by the IDSA, and suited to local epidemiological data. The present study has limitations that should be addressed, the most important of which was the limited sample size (n = 8). Notably, however, the eight intensivists who were recruited surpassed the minimum criterion for the methodology, and the data generated answered the research question, with a saturation of the data, with no further emerging topics. Moreover, the interviews occurred in a single hospital, and the central objective of the present study was related to the intensivists’ point of view. Therefore, future studies should consider physicians who work in multiple locations to provide a better overview of various ICUs. Notwithstanding these limitations, the present findings can assist in drafting plans for treatment optimization. This investigation also demonstrated the need for prescription assistance with support from a well-trained, multidisciplinary team with agreement among its members and importance of well-designed local protocols. In conclusion, numerous aspects influence the adequate prescription of antifungal medication. Knowledge of these aspects can assist in drafting utilization measures, providing educational interventions and ending the negative cycle. Such knowledge can also contribute to the implementation of fungal infection control strategies and development of local protocols to improve the quality of treatment in the hospital setting, especially in the ICU.
T.P. Salci et al.
9.
10. 11.
12.
13.
14.
15.
16.
17.
18. 19.
Acknowledgement We thank the Brazilian Agency CAPES for financial support.
References 1. Vincent, J. L., Rello, J., Marshall, J., et al. (2009) International study of the prevalence and outcomes of infection in intensive care units. JAMA, 302 (21), 2323–2329. 2. Yap, H. Y., Kwok, K. M., Gomersall, C. D., et al. (2009) Epidemiology and outcome of Candida bloodstream infection in an intensive care unit in Hong Kong. Hong Kong Medical Journal, 15 (4), 255–261. 3. Alangaden, G. J. (2011) Nosocomial fungal infections: epidemiology, infection control, and prevention. Infectious Disease Clinics of North America, 25 (1), 201–225. 4. Pappas, P. G., Kauffman, C. A., Andes, D., et al. (2009) Clinical practice guidelines for the management of candidiasis: 2009 update by the Infectious Diseases Society of America. Clinical Infectious Diseases, 48 (5), 503–535. 5. Arnold, H. M., Micek, S. T., Shorr, A. F., et al. (2010) Hospital resource utilization and costs of inappropriate treatment of candidemia. Pharmacotherapy, 30 (4), 361–368. 6. Patterson, T. F. (2005) Advances and challenges in management of invasive mycoses. Lancet, 366 (9490), 1013–1025. 7. De Pauw, B. E. (2006) Increasing fungal infections in the intensive care unit. Surgical Infections, 7 (Suppl. 2), S93–S96. 8. Salci, T. P., Gimenes, M., dos Santos, C. A., Svidzinski, T. I. E. & Caparroz-Assef, S. M. (2013) Utilization of fluconazole in an
466
20.
21.
22. 23.
24.
25.
26.
27.
intensive care unit at a university hospital in Brazil. International Journal of Clinical Pharmacy, 35 (2), 176–180. Lee, I., Fishman, N. O., Zaoutis, T. E., et al. (2009) Risk factors for fluconazole-resistant Candida glabrata bloodstream infections. Archives of Internal Medicine, 169 (4), 379–383. Shoham, S. & Marwaha, S. (2010) Invasive fungal infections in the ICU. Journal of Intensive Care Medicine, 25 (2), 78–92. Swoboda, S., Lichtenstern, C., Ober, M. C., et al. (2009) Implementation of practice guidelines for antifungal therapy in a surgical intensive care unit and its impact on use and costs. Chemotherapy, 55 (6), 418–424. Wood-Mitchell, A., James, I. A., Waterworth, A., Swann, A. & Ballard, C. (2008) Factors influencing the prescribing of medications by old age psychiatrists for behavioural and psychological symptoms of dementia: a qualitative study. Age and Ageing, 37 (5), 547–552. Ljungberg, C., Lindblad, A. K. & Tully, M. P. (2007) Hospital doctors’ views of factors influencing their prescribing. Journal of Evaluation in Clinical Practice, 13 (5), 765–771. Cavazos, J. M., Naik, A. D., Woofter, A. & Abraham, N. S. (2008) Barriers to physician adherence to nonsteroidal anti-inflammatory drug guidelines: a qualitative study. Alimentary Pharmacology and Therapeutics, 28 (6), 789–798. Garey, K. W., Rege, M., Pai, M. P., et al. (2006) Time to initiation of fluconazole therapy impacts mortality in patients with candidemia: a multi-institutional study. Clinical Infectious Diseases : an Official Publication of the Infectious Diseases Society of America, 43 (1), 25–31. Corbin, J. M. & Strauss, A. L. (2008) Basics of Qualitative Research: Techniques and Procedures for Developing Grounded Theory. Los Angeles: Sage Publications. McLellan, E., MacQueen, K. M. & Neidig, J. L. (2003) Beyond the qualitative interview: data preparation and transcription. Field Methods, 15 (1), 63–84. Charmaz, K. (2009) Constructing Grounded Theory: A Practical Guide through Qualitative Analysis. Los Angeles: Sage Publications. Dimopoulos, G., Antonopoulou, A., Armaganidis, A. & Vincent, J. L. (2013) How to select an antifungal agent in critically ill patients. Journal of Critical Care, 28 (5), 717–727. Lepak, A. & Andes, D. (2011) Fungal sepsis: optimizing antifungal therapy in the critical care setting. Critical Care Clinics, 27 (1), 123– 147. Alvarez-Lerma, F., Palomar, M., León, C., et al. (2004) [Indications for antifungal treatment in intensive care unit patients]. Enfermedades Infecciosas Y Microbiologia Clinica, 22 (5), 279–285. Lewis, R. E. (2011) Current concepts in antifungal pharmacology. Mayo Clinic Proceedings, 86 (8), 805–817. Kumar, A., Roberts, D., Wood, K. E., et al. (2006) Duration of hypotension before initiation of effective antimicrobial therapy is the critical determinant of survival in human septic shock. Critical Care Medicine, 34 (6), 1589–1596. Falagas, M. E., Apostolou, K. E. & Pappas, V. D. (2006) Attributable mortality of candidemia: a systematic review of matched cohort and case-control studies. European Journal of Clinical Microbiology and Infectious Diseases, 25 (7), 419–425. Kim, M. M., Barnato, A. E., Angus, D. C., Fleisher, L. A., Fleisher, L. F. & Kahn, J. M. (2010) The effect of multidisciplinary care teams on intensive care unit mortality. Archives of Internal Medicine, 170 (4), 369–376. Collins, S. A. & Currie, L. M. (2009) Interdisciplinary communication in the ICU. Studies in Health Technology and Informatics, 146, 362– 366. Erstad, B. L., Haas, C. E., O’Keeffe, T., Hokula, C. A., Parrinello, K. & Theodorou, A. A. (2011) Interdisciplinary patient care in the intensive care unit: focus on the pharmacist. Pharmacotherapy, 31 (2), 128–137.
© 2014 John Wiley & Sons, Ltd.
