Although the distribution of Candida species varies between institutions and geographic areas, Candida albicans remains the most frequently isolated species in invasive candidiasis in non-neutropenic critically-ill patients.21,38 However, an increase in invasive candidiasis due to non-C. albicans Candida species, such as Candida parapsilosis and Candida glabrata, has been observed,40,43 which is due, although partially, to the widespread use of fluconazole prophylaxis in critically ill patients.46

The majority (89.3%) of the experts consulted agreed that the incidence of invasive infection due to non-C. albicans Candida species has increased in Critical Care Units in our country. In particular, and on a scale of 0 to 10 points in which 10 represents the maximum degree of agreement, 26 of the 31 specialists granted 7 or more points to this statement. A high level of consensus was achieved (Top 4>70%).

The recognition of co-morbidities and risk factors in population at risk for invasive candidiasis due to non-C. albicans Candida species is vital to define the more effective antifungal drug patients should receive with the less toxicity.9,10,45,49

The vast majority of the panel members (93.5%) considered that patients’ risk factors or co-morbidities should be taken into account since they may be related to invasive fungal infections due to non-C. albicans Candida species. Specifically, and on a scale of 0 to 10 points in which 10 represents the highest degree of agreement, 29 specialists granted 7 or more points to this statement, for which a high level of consensus was reached (Top 4>70%).

Considering the relative frequency of fluconazole resistance to C. albicans and the increasing frequency of invasive candidiasis due to C. parapsilosis and C. glabrata,43,49 knowledge of the local etiology of invasive candidiasis is necessary to optimize the antifungal treatment44 that can improve patient survival when administered early and adequately.17

Due to its benefits in optimizing initial empirical treatment, the vast majority of the experts (93.5%) granted great importance to the knowledge of the local epidemiology of Candida infections. Specifically, and on a scale of 0 to 10 points in which 10 represents the maximum level of significance, 29 of the 31 experts gave 7 or more points to the statement, establishing a consensus agreement (Top 4>70%).

The isolation of Candida from clinical specimens other than blood is an important challenge for clinicians, as it can represent anywhere from a single colonization, multiple sites colonization or local infection to severe invasive candidiasis. Early detection of invasive candidiasis through active research is important since it enables the determination of local epidemiology, the implementation of antifungal treatment and the improvement of patients’ prognosis.21,27,38,54

Experts’ level of consensus was high (Top 4>70%) when recommending active research of Candida colonization to determine local epidemiology. Specifically, and on a scale of 0 to 10 points in which 10 represents the highest level of agreement, 83.9% of the experts granted 7 or more points to this statement.

In the last few years, in vitro susceptibility of Candida to fluconazole has decreased.19 This decrease could be partly due to the rise of infections due to C. glabrata.43 Furthermore, there has been a significant increase of Candida tropicalis and other rare species resistant to fluconazole.2 In order to optimize the empirical antifungal treatment, this situation requires identifying the local profile of the etiology of invasive candidiasis and the susceptibility patterns to fluconazole and other current antifungal drugs of the predominant species.55

The great majority of the specialists (90.3%) confirmed they find it necessary to know the local antifungal susceptibility patterns of the different Candida species causing invasive candidiasis. In particular, and on a scale of 0 to 10 in which 10 represents the highest level of agreement, 28 of the panel members granted 7 or more points to this statement, establishing a high level of consensus (Top 4>70%).

There is a considerable amount of literature that evidence the need for implementing stewardship programs for antimicrobial use in hospitals.3,25,48,51,53,56,57 In 2007, the Infectious Diseases Society of America (IDSA) and the Society for Healthcare Epidemiology of America (SHEA) published guidelines underlining that the creation of a responsible multidisciplinary team to carry out the tasks of an institutional program to enhance the Program for Optimizing the Use of Antimicrobial Agents (PROA, Acronym in Spanish) on antifungal treatment in invasive candidiasis, is a key strategy for the program's success.12 The 2012 consensus document drafted by the Study Group of Nosocomial Infections (GEIH) of the Spanish Society of Infectious Diseases and Clinical Microbiology (SEIMC), the Spanish Society of Hospital Pharmacy (SEFH) and the Spanish Society of Preventive Medicine, Healthcare and Hygiene (SEMPSPH) establishes that for the development of an institutional program to enhance antimicrobial stewardship programs, the core of the multidisciplinary team should at least consist of a specialist in infectious diseases or a clinician expert in infectious diseases, a clinical pharmacist specialist in antimicrobial agents and a microbiologist specialist in antimicrobial drug resistance.50 To date, it has been established that the minimum team for a PROA on antifungal treatment in invasive candidiasis should be made up by a specialist in infectious diseases, a clinical pharmacist specialist in infectious diseases, a microbiologist, a specialist in intensive care medicine and a member of the reanimation unit.50

The level of consensus of the panel members was high (Top 4>70%) when considering that the statement regarding the minimum team required for the development of an efficient program on the use of antifungal drugs for non-neutropenic critically-ill patients requires the inclusion of a pharmacist, a specialist in anesthesiology and reanimation, a specialist in intensive care medicine, a microbiologist, and a specialist in infectious diseases. Specifically, and on a scale of 0 to 10 in which 10 represents the greatest level of agreement, 87.1% of the experts gave 7 or more points to this statement.

The consensus document drafted in 2012 by the Study Group of Nosocomial Infections (GEIH) of the Spanish Society of Infectious Diseases and Clinical Microbiology (SEIMC), the Spanish Society of Hospital Pharmacy (SEFH) and the Spanish Society of Preventive Medicine, Healthcare and Hygiene (SEMPSPH) establishes that for the development of an institutional program to enhance antimicrobial stewardship programs, the core of the multidisciplinary team should at least consist of a specialist in infectious diseases or a clinician specialist in infectious diseases, a clinical pharmacist specialist in antimicrobials and a microbiologist specialist in antimicrobial drug resistance.50 No matter how ambitious it may be, no PROA on antifungal treatment in invasive candidiasis can achieve the objectives unless the program is shared by a member of the management team of the hospital.26

None of the coordinators’ proposed response options achieved 50% required to establish a consensus agreement – preventivist (35.5%), hematologist (32.2%), administrator/medical management (48.4%). This question was selected for the second phase of the study, in which the majority of the panelists (92.1%) considered the inclusion of an Administrator/Medical Management to the minimum team in order to elaborate an efficient program on microbial use in non-neutropenic critically-ill patients. In contrast, no consensus was achieved regarding the preventivist (28.6%) and the hematologist (21.4%).

Different types of minimum teams required for an efficient antifungal stewardship program are published in the literature.3,25,48,51,57

Again, the answers showed the lack of expert consensus with each and every one of the response options – daily (19.4%), weekly (25.8%), biweekly (9.7%), monthly (29.0%), and quarterly basis (16.1%). The question was reformulated – How often do you consider the core team should meet to take decisions or make recommendations? – and was selected for the second phase of the study, in which the specialists agreed with the appropriateness of establishing a monthly meeting of the core team, therefore reaching a consensus agreement (≥50%). In contrast, no agreement was reached with the other response options: daily (7.1%), weekly (39.3%), biweekly (0%), and quarterly basis (3.6%).

The literature contains many studies in which the roles and responsibilities of the multidisciplinary team for antimicrobial stewardship programs in hospitals are defined.3,25,48,53,56,57 Thus, and as a guide for action, some papers describe stewardship programs for antibiotic treatment similar to PROA on antifungal treatment in invasive candidiasis.51 Multidisciplinary teams for monitoring antifungal treatment can reduce pharmacological costs without reducing quality of care.29

Experts reached a consensus agreement (Top 4>70%) for each and every one of the response options proposed by the Coordinators. Specifically, the order of the options rated with 7 or more points was the following: raising awareness of the problem of inadequate use of antimicrobials, 100%; improve suitability of empirical antifungal treatment, 100%; optimize dosing regimens based on PK-PD indices of antifungal drugs, 96.8%; discontinue antifungal treatment in the absence of proven fungal infection, 96.8%; and convey the need to form teams to lead the PROAs to improve the quality of care, 87.1%.

The literature contains many studies in which the roles and responsibilities of the multidisciplinary team for antimicrobial stewardship programs in hospitals are defined.3,25,48,53,56,57 Thus, and as a guide for action, some reports describe stewardship programs for antibiotic treatment similar to the PROA on antifungal treatment in invasive candidiasis.51 Multidisciplinary teams for follow-up of antifungal treatment can reduce pharmacological costs without impairing quality of care.29

Again, the answers of the panel members showed a consensus agreement (Top 4>70%) for each and every one of the response options. In this case, the order of responses that rated 7 or more points (Top 4) was as follows: avoid drug combinations that may represent unnecessary overlap in antimicrobial spectra, 100%; avoid inadequate empirical antifungal treatments, 100%; avoid prescribing inadequate antifungal therapy in the treatment of confirmed fungal infections, 100%; development of institutional guidelines for the management of invasive candidiasis and candidemia, 96.8%; set the duration of antifungal treatment based on available scientific evidence, 93.5%; establish a treatment with the most cost-effective antifungal molecule when the fungal etiology and outcome of microbiological study are known, 90.3%; and avoid antifungal treatment in febrile processes in which no fungus has been isolated or in the event the microorganisms have been isolated from non-sterile specimens and were considered as contamination or colonization, 83.9%.

The EPICO 1.061 and EPICO 2.0 projects62 established a set of consensus recommendations for the diagnosis and treatment of invasive candidiasis. The ‘MAGICS Algorithm’ was produced from quantifiable measures contained in these recommendations (Fig. 1).

Fig. 1.

MAGIC algorithm for the diagnosis and treatment of invasive candidiasis.

(0.53MB).

The vast majority of the specialists (80.6%) considered performing the ‘MAGICS Algorithm’ (Fig. 1) necessary in the initial approach of non-neutropenic critically-ill patients with suspected invasive candidiasis. Specifically, on a scale of 0 to 10 in which 10 represents the maximum level of significance, 25 of the 31 experts granted 7 or more points to this consideration, reaching a consensus.

The EPICO 1.061 and EPICO 2.0 projects62 established a set of consensus recommendations for the diagnosis and treatment of invasive candidiasis. The ‘MAGICS Algorithm’ was produced from quantifiable measures contained in these recommendations (Fig. 1).

Panel members reached a consensus agreement when considering the inclusion of four of the response options proposed by the coordinators in the ‘MAGIS Algorithm’. Specifically, the following items reached consensus (≥50%): suspension of treatments due to evidence of no fungal infection, 93.5%; therapeutic de-escalation, 87.1%; dose and drug adjustment according to organic failures, 67.7%; and sequential therapy, 58.1%. In contrast, no consensus was reached in the case of monitoring plasma levels, whereas inclusion in the algorithm was only considered by 35.5% of the specialists.

The quality indicator is the quantitative measure used as a guide to monitor and assess the quality of important aspects of clinical practice. Its design should include a description of different sections to ensure their validity and reliability, process indicators (carried out with the available resources, protocols and scientific evidence) being those evaluating how clinical practice develops.52 The indicators proposed in the literature for monitoring the use of antibiotics,58 including those specific for critically ill patients,59 can be adapted to the use of antifungal drugs.3

Experts reached a consensus agreement for 13 of the 15 response options proposed by the coordinators. Specifically, the order of the items that reached consensus based on the percentage of responses rated with 7 or more points was: monitoring and control of the implementation of clinical protocols, 90.3%; patients on antifungal treatment by time frame, 87.1%; inadequate empirical antifungal treatments, 87.1%; patients on empirical therapy who actually suffered from invasive candidiasis, 87.1%; time delay of antifungal treatment from positive blood culture, 87.1%; use of antifungal drugs as directed therapy, 87.1%; modification of inadequate empirical antifungal treatments, 83.9%; modification of antifungal treatments due to ‘therapeutic escalation’, 83.9%; changes of antifungal drugs used as treatment, 83.9%; suspension of treatment due to evidence of no fungal infection, 80.6%; days without antifungal prophylaxis therapy, 74.2%; and patients with sepsis/septic shock treated with antifungal drugs during the first three hours, 74.2%.

In contrast, no consensus was reached in the cases of days without antifungal treatment in patients who have received such treatment and, the days without antifungal treatment in ICU. The percentages of responses in the Top 4 were established at 45.2% and 38.7%, respectively.

The quality indicator is the quantitative measure used as a guide to monitor and assess the quality of important aspects of clinical practice. Its design should include a description of different sections to ensure its validity and reliability, outcome indicators being those that quantify the consequences of the clinical practice, in terms of complications, mortality, missed opportunities, circuit failures, quality of life, etc.52 The indicators proposed in the literature for monitoring the use of antibiotics,58 including those specific for critically ill patients,59 can be adapted to the use of antifungal drugs.3

The responses showed a consensus agreement (Top 4>70%) reached by the panelists in 7 of the 8 response options proposed by the coordinators. In this case, the order of the items that reached consensus based on the percentage of responses rated with 7 or more points was as follows: antifungal resistance, 90.3%; consumption of antifungal drugs by Defined Daily Dose per 100 days-stay, 83.9%; mortality, 80.6%; average stay related to invasive candidiasis, 80.6%; cost per process, 77.4%; consumption of antifungal drugs by Days of Therapy per 100 days-stay, 77.4%; and cost per patient, 71.0%. Finally, in the case of budget compliance, the percentage of responses was 38.7%, so no consensus was established.

Education is considered a ‘persuasive intervention’.6 Education in safe and judicious antimicrobial prescribing is an important element of any antimicrobial stewardship program.12,13,26,29,39,51

All (100%) of the panel members agreed with both of the above statements, establishing in both cases a consensus agreement (Top 4>70%).

Active learning techniques have proven more effective in achieving a change in behavior of prescribers than mere passive diffusion of information.4,11,26,47

Experts achieved a consensus agreement (Top 4>70%) for each and every one of the response options proposed by the coordinators. Specifically, the order of the items that reached consensus based on the percentage of responses rated with 7 or more points was: external knowledge management strategies, 83.9%; individual learning strategy, 83.9%; internal communication strategies, 80.6%; improvement and innovation strategy, 80.6%; retention of expert's knowledge strategies, 77.4%; document management strategies, 77.4%; and organizational learning strategies, 74.2%.

Clinical guidelines based on scientific evidence are considered a key element of any stewardship program on the management of the antifungal treatment.14,18 A key objective of any stewardship program on the management of antifungal treatment is to increase the adherence of prescribers to the use of clinical guidelines.34,35 Adequate undergraduate training on the knowledge and treatment of infectious diseases is the basis for safe and appropriate prescribing of antimicrobial drugs.35 The establishment of an updated intranet web of the different institutions is as an excellent way to provide easy and convenient access to information of the stewardship programs on antifungal treatment.26

The answers show a consensus agreement among the specialists consulted (Top 4>70%) of 3 of the 4 statements proposed by the coordinators. Specifically the percentage of responses rated with 7 or more points were the following: clinical guidelines based on scientific evidence are considered a key element of any EXCELLENCE IN MANAGEMENT program, 96.8%; establish a strategy to manage learning and collaboration networks in order to identify opportunities of improvement, creativity and innovation, 90.3%; and a web or updated presence in social networks on the hospitals intranet is an essential tool to disseminate the professionals’ EXCELLENCE IN MANAGEMENT programs on antifungal treatment, 83.9%.

Meanwhile, as the percentage of panel members that agreed with the statement ‘Undergraduate training should be a primary goal of the EXCELLENCE IN MANAGEMENT programs on antifungal treatment’ was established at 67.7%, the statement was selected for inclusion in the second phase of the Delphi study, in which 74.1% of the specialists awarded 7 or more points to this consideration, establishing an agreement (Top 4>70%).

The positive experience with multidisciplinary teams for monitoring invasive candidiasis created in our country25,56 and other non-Spanish hospitals29,32 support the feasibility of creating these units.

Most panel members (74.2%) agreed in stating that the creation of multidisciplinary teams for monitoring invasive candidiasis in Spanish hospitals is completely feasible. Specifically, on a scale from 0 to 10 points in which 10 represents the highest level of importance level, 23 of the 31 specialists granted 7 or more points to this consideration, establishing a consensus agreement.

The creation of multidisciplinary teams for monitoring invasive candidiasis leads to a marked decrease in healthcare expenditure by facilitating the de-escalation and suspension of a considerable number of antifungal treatments.3,29,31,33,51,57

Most of the experts (80.6%) agreed that the investment of resources is always justified for the implementation of a program of EXCELLENCE IN MANAGEMENT to monitor invasive candidiasis. Specifically, and on a scale of 0 to 10 points in which 10 represents the highest degree of agreement, 25 specialists awarded 7 or more points to this consideration, establishing a consensus agreement.

Blood culture is the method of choice in diagnosing candidemia, although its usefulness is restricted by up to 50% of the cases.1 Thus, the availability of other non-culture-based microbiological diagnostic techniques may be useful for both diagnosis and de-escalation, even though currently none of them are fully standardized.5,30,31,51

The answers show the level of agreement of the panel members when assessing the usefulness of the availability of the technique for detecting beta-glucan as part of an integrated clinical process of management and resources for monitoring of invasive candidiasis. Specifically, 77.4% of the panelists granted 7 or more points to this assessment, so an agreement was achieved. By contrast, the non-culture-based microbiological techniques was valued by only 21 of the 31 experts consulted (67.7%), so no consensus was reached (Top 4<70%).

This question was selected for inclusion in the second phase of the Delphi study, in which the specialists were requested to assess the Usefulness of non-culture-based microbiological techniques, reaching a high level of consensus – 92.6% of the responses were rated with 7 or more points.

The creation of multidisciplinary teams for monitoring invasive candidiasis has greater utility in hospitals with a greater population at risk of acquiring this fungal infection, which is significantly more common in larger hospitals.25,51,56

The vast majority of the panel members (96.8%) stated that the implementation of EXCELLENCE IN MANAGEMENT programs for monitoring invasive candidiasis are cost-effective in hospitals with more than 500 beds. No consensus was however reached regarding cases of hospitals with less than 200 beds – response rate in the Top 4 of only 16.1%. No agreement was reached either in the case of hospitals with 200 – 500 beds – response rate in the Top 4: 67.7%, so this question was selected for inclusion in the second phase of the Delphi study in which a consensus agreement was reached – 77.8% of the responses were rated with 7 or more points.

Standardization of clinical practice decreases the disparity in the treatment of patients using a similar approach. The implementation of the protocols also improves the distribution of responsibilities among the physicians involved in clinical procedures. Procedures subject to protocols are mainly those involving a larger number of patients, high cost and affecting medical areas where risk patients are located.36

The experts reached a consensus agreement for each and every one of the statements proposed by the coordinators. Specifically, based on the responses rated with 7 points or more, the following order of the items was established: improvement and innovation, 93.9%; measurement, 83.9%; and learning and creativity, 87.1%.

Quality as such is meaningless, therefore different authors have proposed that improving quality of care requires a process involving the definition of this concept.20 However, at least sometimes, it is difficult to differentiate between processes that provide high levels of quality and those considered low quality.7 Thus, the quality of healthcare should be defined within a framework that provides ongoing improvement in this area.42

The panel members reached a consensus agreement (Top 4>70%) for each and every one of the response options provided by the coordinators. Specifically, the order of importance awarded to each of the responses based on the answers rated with 7 or more points were: patient safety, 100%; analysis of the results achieved and learning activities, 96.8%; assess, revise and refine the deployment process, 96.8%; quality assurance, 93.5%; epidemiological surveillance, 90.3%; applied research, 90.3%; comparison of indicators strategy (benchmarking) among different hospitals, 87.1%; and basic research, 77.4%.

Quality as such is meaningless, therefore different authors have proposed that improving quality of care requires a process involving the definition of this concept.20 However, at least sometimes, it is difficult to differentiate between processes that provide high levels of quality and those considered low quality.7 Thus, the quality of healthcare should be defined within a framework that provides ongoing improvement in this area.42

Experts reached a consensus agreement in 7 of the 8 measures proposed by the coordinators. In this case, the order based on the percentage of respondents that rated the measures with 7 or more points was: population health outcomes, 93.5%; process performance, 87.1%; process costs, 83.9%; technology, information and knowledge, 83.9%; perception of stakeholders (affected groups of interest), 80.6%; activity, 80.6%; and economic and financial management, 77.4%.

Meanwhile, the percentage of respondents that granted 7 or more points to ‘Performance of partners and suppliers’ was 61.3%, thus it was selected for inclusion in the second phase of the Delphi study where a consensus agreement was reached – 21 responses of the 28 (77.8%).

To date, a number of terms that identify the quality of process and outcome indicators have been accepted. This is the case of integrity, which establishes whether the indicators are timely, reliable and accurate,37 and performance that is established by the ability of the indicators to achieve sustained trends, define objectives, external comparisons available and confidence in the understanding of the causes and effects of the results based on the deployed actions.22 Additionally, segmentation of results implies the need for these to not only be globally analysed, but also from the perspective that the overall outcomes come from the aggregation of the individual outcomes of each of the indicators that have been analysed separately.28

The responses showed a high level of agreement (Top 4>70%) among the specialists regarding the relevance of all of the response options proposed by the coordinators.

The order of the characteristics granted 7 or more points is: integrity, 100%; performance, 100%; scope and relevance; 93.5%; and segmentation, 90.3%.

The dissemination to the different services and to the hospital of the results obtained following the implementation of optimization programs in the management of antimicrobial therapy has proven to be a useful strategy in monitoring trends in the use of antimicrobial drugs and to identify the critical areas that require special attention to succeed in achieving proper prescription.24 In fact, doctors and staff related to antibiotic therapy are considered key elements that should be informed about the objectives and results of the antibiotic stewardship programs.8 Moreover, the dissemination of information on bacterial and fungal resistance is an essential element of training.8 However, disseminating the results of monitoring programs of certain infections and antibiotic use has not always translated into improved outcomes.16 In this context, in our country we have a study called ‘Study on the Prevalence of Nosocomial Infections in Spain’ (EPINE, Acronym in Spanish), whose results are disseminated to the participating hospitals,15 and the ‘Surveillance Study of Nosocomial Infections in Catalonia (VINCat, Acronym in Spanish), which is a standardized surveillance program that provides individualized information to each participating center and conducts benchmarking of specific results.60 Meanwhile, expert clinical decision support systems based primarily on software tools, have been proposed as elements to be considered to enhance the dissemination of the results of antibiotic stewardship programs.41

Panel members reached a consensus agreement (≥50%) for 6 of the 9 response options proposed by the coordinators. Specifically, the percentage of positively valued responses in items that reached consensus were: inform to the infection committee, 96.8%; to their team, 93.5%; to the critical care units, 87.1%; to the medical management, 83.9%; to all of the services implicated, 83.9%; and to the pharmacy committee, 71%.

On the other hand, the percentage of responses with positive assessment for the options ‘to the attendees of general hospital information sessions’, ‘to all who have access to the hospital web’ and ‘to the hospital board’ were respectively 48.4%, 35.5% and 32.3%, so no consensus was established in any of these cases.

As part of the ‘National Healthcare Safety Network’ (NHSN), the Centers for Disease Control and Prevention of the United States (CDC) have developed an application to introduce information regarding the days of therapy with antimicrobial drugs (DOT) for subsequent monthly exploitation. This information primarily has software maintenance. CDC will add a ‘risk adjustment’ function to the application to enable benchmarking.8

The response option, ‘every three months’ for the dissemination of results in the hospital, was considered suitable by 51.6% of the specialists, whereas a consensus agreement was established (≥50%). This was not the case for the other response options, since they were not considered adequate by at least half of the panel members – every 6 months, 25.8%; once a month, 22.6%; and annually, 0.0%.

The dissemination of information on bacterial and fungal resistance is an essential element of training, as well as refresher courses for improving the prescription of antimicrobial drugs.8 In this context, we have in our country a study called EPINE, which results are disseminated to the participating hospitals,15 and the VINCat study, which is a standardized surveillance program that provides individualized information to each participating center and conducts benchmarking of specific results.60

The vast majority of the experts (96.8%) agreed on disseminating the results outside the hospital, reaching a consensus agreement. Specifically, 71.3% agreed on disseminating the results to scientific journals, and 67.7% to Autonomous Community Administrations, 6.3% to scientific societies, and 51.6% to other hospitals to compare performance, so a consensus agreement was established for the dissemination to scientific journals.

In contrast, no consensus was reached in the case of communicating the results to the National Administration and to specialized press, where a positive response was only obtained by 25.8% and 12.9% of the experts, respectively. Moreover, in the case of non-specialized general press, this response option was completely ruled out by all panel members.

Although Law 14/1986 of April 25, includes several sections on actions of health services, no information is available about the need to report on issues related to this document.23

The vast majority of experts (96.7%) considered that the dissemination of results outside the hospital should be conducted annually, establishing a consensus agreement. Other response options proposed by the coordinators were not considered adequate for at least half of the members of the panel – every 6 months, 3.3%; every 3 months, 0%; and once a month 0%.