JAC Advance Access originally published online on January 31, 2006
Journal of Antimicrobial Chemotherapy 2006 57(3):546-550; doi:10.1093/jac/dki483
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Factors associated with adherence to infectious diseases advice in two intensive care units
1 Service d'Infectiologie, Hôpital l'Archet 1, Centre Hospitalier Universitaire de Nice, Nice, France; 2 Département de Santé Publique, Hôpital l'Archet 1, Centre Hospitalier Universitaire de Nice, Nice, France; 3 Réanimation polyvalente, Hôpital St Roch, Centre Hospitalier Universitaire de Nice, Nice, France; 4 Réanimation médicale, Hôpital l'Archet 1, Centre Hospitalier Universitaire de Nice, Nice, France
* Corresponding author. Tel: +33-4-92-03-55-15; Fax: +33-4-93-96-54-54; E-mail: pulcini.c{at}chu-nice.fr
Received 3 August 2005; returned 31 August 2005; revised 5 December 2005; accepted 12 December 2005
 |
Abstract |
|---|
 Top Abstract IntroductionPatients and methodsResultsDiscussionTransparency declarationsReferences |
|---|
Objectives: Several studies have reported that infectious diseases (ID) consultations improve the quality of antibiotic prescription, providing adherence is adequate. The aim of this study is to determine the factors associated with adherence to such therapeutic advice.
Patients and methods: We conducted a prospective study in two intensive care units (ICUs) over a 6 month period. Systematic bedside diagnostic and therapeutic ID advice was delivered for all patients receiving antibiotic therapy.
Results: A total of 381 consultations for 195 patients were recorded, 244 (64%) in ward A and 137 (36%) in ward B. The median SAPS score was 45 and the mortality rate was 23%. Infections accounted for 220 (58%) admissions. A diagnostic discrepancy between ID and intensive care specialists was noted in 125 (33%) cases. The ID specialist advised continuation of the same antibiotic therapy in 138 (36%) cases, a change in 154 (41%) and withdrawal in 89 (23%). Adherence to ID therapeutic advice was recorded for 326 (86%) cases. Multivariate analysis identified two factors independently associated with adherence: ward B [odds ratio (OR), 4.9; 95% confidence interval (CI), 2.0–12.1] and proposition to pursue the same therapy (OR, 4.8; 95% CI, 1.6–14.5).
Conclusions: Patients' characteristics and antibiotic therapy modalities do not influence adherence to ID consultation. In contrast, the ward and its characteristics play a major role in adherence to ID advice.
Keywords: antibiotic prescriptions , ID physicians , ICUs
|
Introduction |
|---|
|
TopAbstractIntroductionPatients and methodsResultsDiscussionTransparency declarationsReferences |
|---|
Antibiotic therapy is widely used in our modern hospitals. Various studies have indicated that control policies could significantly reduce the volume of prescriptions without altering the prognosis.1–5 These institutional policies led to an almost 50% reduction in antibiotic prescriptions but did not specifically deal with the quality of the prescription. The extent of this antibiotic reduction suggested that some of these prescriptions were not justified, probably because infections were excessively diagnosed.6,7 Accordingly, studies conducted by infectious diseases (ID) specialists reported improvement in the quality of antibiotic prescriptions through more appropriate diagnosis and, consequently, antibiotic therapy.8–12 As an example, in a study evaluating treatment for Staphylococcus aureus bacteraemia, the prognosis was improved when the physician in charge followed the ID specialist's advice (79% versus 64% of cure).12 This study highlighted the importance of adherence to ID advice, which varies from 46 to 94% in the literature.11–14
For 6 years we have been giving systematic ID consultations to physicians in several departments within our institution, i.e. providing regular advice without any specific request from the physician in charge. We give such advice in particular in intensive care units (ICUs), considering that more than half of the patients receive at least one antibiotic prescription during their hospitalization in ICUs.15 The systematic ID consultation allows a larger number of patients to benefit from specialized advice, and at an earlier stage of their management. We demonstrated that such ID consultations decreased antibiotic prescriptions from 30 to 50%, essentially because of diagnosis improvement.13,16
Taken all together, these studies showed that ID advice is beneficial for nearly all aspects of care of infected patients. However, these benefits depend on adherence to the recommendations. To the best of our knowledge, only one study dealt with the factors that affect adherence to ID advice, showing that therapeutic recommendations led to higher adherence than diagnostic ones, as did advice delivered in medical wards in private hospitals compared with surgery departments in public hospitals.14 This study was conducted in the context of formal ID consultations, which is different from our practice, i.e. systematic bedside consultations. We therefore conducted a prospective cohort study of consecutive ID systematic bedside consultations in two ICUs to identify factors associated with adherence to therapeutic recommendations.
|
Patients and methods |
|---|
|
TopAbstractIntroductionPatients and methodsResultsDiscussionTransparency declarationsReferences |
|---|
This prospective study was conducted in two ICUs A and B (a total of 26 beds) in which both surgical and medical adult patients were admitted. Two different medical teams worked separately in these departments, with a total of 10 full-time senior physicians and six residents. These two wards are not located in the same hospital, but both belong to the Nice University Hospital, a 1778-bed teaching medical centre. Systematic bedside ID advice involving the same ID specialist was delivered orally during the morning round for all patients receiving antibiotic therapy. At least one consultation per week was completed during 6 months, from 1 November 2003 to 30 April 2004.
Data collection and study definitions
All consecutive consultations were included, and a standardized form was recorded in real time by the ID physician for each advice delivered. The same patient could then be included several times. The form was not available for the ICU specialist, the ID advice being delivered orally during the discussion with the medical staff. All data were checked by an independent reviewer. We collected the following data:
- Patients' demographic characteristics, ICU ward, comorbidities, motives for hospitalization and for antibiotic therapy and simplified acute physiology score II (SAPS II). Underlying diseases notified in a patient's chart were considered as comorbidities. Nosocomial infections were defined as those acquired at 48 h or after 48 h after the admission. Septic shock was defined as the association of a recognized infection and of haemodynamic failure despite fluid resuscitation.17 Acute respiratory distress syndrome (ARDS) was defined mainly on both acute onset of bilateral pulmonary infiltrates and concomitant hypoxaemia without cardiac involvement.18 Lastly, trauma implied two organ injuries or more.19
- Information collected regarding antimicrobial treatment included the kind of drug, dose, route and data regarding the infecting organism. Types of antibiotic therapy were as follows: empirical antibiotic therapy was defined as based on clinical findings only, documented antibiotic therapy was based on microbiological identification and adapted antibiotic therapy was based on antibiotic susceptibility; multi-drug therapy was defined as antibiotic therapy including two or more drugs.
- Each ID advice could include other diagnostic and/or therapeutic propositions, which were also recorded. Continuation of the antibiotic therapy referred to pursuing the same treatment, in terms of the kind of drug, route and dose. A change of any of these items was considered as treatment modification. Discontinuation meant stopping antibiotic therapy.
Adherence and outcome measurement
Adherence to therapeutic propositions was evaluated during the subsequent ID consultation. Adherence implied that the physician's prescription was completely in agreement with the ID therapeutic advice, in terms of choice of the kind of drug, route and dose. Any deviation was defined as non-adherence.
Outcome was assessed by mortality rate on the day of release from the ICU. Only the first ID advice was taken into account for this calculation, thus limiting the analysis to the patients.
Statistical analysis
Data analysis was performed using the SPSS software, version 11 (SPSS). 
2 and Student's t-test were used for statistical analysis. Odds ratio (OR) and 95% confidence intervals (CI) were used to estimate the association between adherence and its possible determinants. Multiple logistic regression analysis was used to adjust OR on possible confounding variables. A survival estimate was obtained using Kaplan–Meier curves with a log-rank test. A two-sided P < 0.05 was considered statistically significant.
|
Results |
|---|
|
TopAbstractIntroductionPatients and methodsResultsDiscussionTransparency declarationsReferences |
|---|
A total of 381 ID consultations for 195 patients took place over 6 months, 244 (64%) in ward A and 137 (36%) in ward B. Of the patients 52% (102/195) received a single ID advice. For the others, 4.7 ± 2.6 consultations per patient were delivered on average.
Demographic and clinical characteristics of the patients hospitalized in the two wards
Data are presented in Table 1.
|
Antibiotic therapy prescribed by the intensive care physician
Among 381 antibiotic treatments, 198 (52%) were empirical, 66 (17%) documented, i.e. based on microbiological identification, and 117 (31%) adapted, i.e. based on antibiotic susceptibility. A single drug was prescribed in 156 (41%) cases, and more than one in the other 225 (59%).
Diagnostic and therapeutic ID specialist propositions
ID consultant propositions are summarized in Figure 1. The ID specialist agreed with the diagnosis in 256 (67%) of the cases, and suggested an alternative diagnosis in 125 (33%). Proposed alternative diagnoses were the absence of ongoing infection in 67 cases, infection other than that initially diagnosed in 23 cases, bacterial colonization in 21 cases and a non-infectious cause of fever in 14 cases. The main motives for stopping antibiotic therapy were a non-infectious disease diagnosis in 46% (41/89) and completion of treatment duration in 27% (24/89).
|
Overall, ID advice was followed for 326 (86%) consultations.
Univariate associations with adherence
Results are presented in Table 2.
|
Multivariate model of adherence
In a multivariate analysis using a logistic regression model, only two factors remained independently associated with adherence to ID therapeutic propositions: ward B (OR, 4.9; 95% CI, 2.0–12.1) and continuation of the same antibiotic therapy (OR, 4.8; 95% CI, 1.6–14.5). Details of the analysis are presented in Table 2.
Patients' characteristics (age, severity of illness, presence of septic shock or ARDS, nosocomial infection, type of infection) and type of antibiotic therapy prescribed by the intensive care physician were not associated with adherence.
Outcome
The overall mortality rate was 23% (44/195). Adherence was not statistically linked with survival in the Kaplan–Meier analysis (P = 0.21).
|
Discussion |
|---|
|
TopAbstractIntroductionPatients and methodsResultsDiscussionTransparency declarationsReferences |
|---|
Adherence to ID therapeutic recommendations appeared to be linked to the ICU department and the continuation of the same antibiotic treatment, but not to patients' characteristics.
This prospective study of 381 ID consultations delivered in two ICUs included all the patients receiving antibiotics during 6 months. These two ICU wards had accepted systematic ID advice for 6 years, delivered by the same physician. It was then possible to precisely determine factors related to adherence to ID advice. One limitation of our study was its restriction to ICUs, because the level of adherence could differ in other specialities, reflecting different physicians' feelings. Also, diagnostic propositions were not specifically studied, and different ones may provoke non-adherence. Further studies are needed to confirm these hypotheses. Finally, advice was delivered by the same ID specialist, so we could not evaluate the relation between ID specialist and adherence. However, adherence was found to be independent of the ID physician in the study conducted by Lo et al.14 In the same way, we did not study the adherence rate for each ICU specialist, the decisions of treatment being taken jointly by the medical team during the visit.
Therapeutic recommendations provided by the ID consultant in the two wards were followed 86% of the time, which is concordant with the data published in the literature.11–14 Adherence was not related to patients' characteristics, particularly severity of the disease, as appreciated through the SAPS II scoring system, the presence of septic shock or ARDS, comorbidities, the community-acquired nature of the infection or the type of infection. Lo et al.14 reported similar results, but in the context of a specific request for formal ID consultation. Likewise, adherence differed between wards A and B. A difference in adherence had already been recorded between surgical and medical wards and between private and public institutions.14 But our study dealt with two departments of the same speciality in the same public institution.
The difference in adherence between the two ICU wards could have many explanations. Patient inclusions came much more from department A in which infections represent 67% of the recruitment, compared with 41% in department B, in which trauma is the main cause of hospitalization. The two wards (A and B) also present significant differences in terms of recruitment of patients, with older and more severe patients in ward A. But these different characteristics are not significantly related to adherence in multivariate analysis, so they cannot explain the difference in adherence between the two wards. This difference in adherence is thus probably the fact of a difference in behaviour of the two medical teams. This possibly includes prescription habits, difficulties with multidisciplinary team work or personal interest in ID advice. It is also possible that the physicians of ward A are less prone to change because they have to manage more severe patients, and because, given the burden of infected patients in their own recruitment, they feel more experienced in ID.
In fact, we are now experienced in managing this attitude, and we believe that non-adherence should be anticipated during confrontation between the ID specialist and the physician in charge. One way to avoid oppositions to ID advice is to propose short-term goals for the current antibiotic therapy (such as resolution of fever, decrease in C-reactive protein and bacteriological eradication). In case of non-achievement of these objectives, the physician would be more receptive to a new discussion.
In conclusion, the ICU medical team is the determinant factor of adherence to ID advice. This could be due in part to a difference in their approach to infection, thus requiring at least some degree of common training for ID specialists and physicians frequently in charge of infected patients in their daily practice. In order to improve adherence, ID consultants have to anticipate disagreement using up-to-date communication training.
|
Transparency declarations |
|---|
|
TopAbstractIntroductionPatients and methodsResultsDiscussionTransparency declarationsReferences |
|---|
None to declare.
|
References |
|---|
|
TopAbstractIntroductionPatients and methodsResultsDiscussionTransparency declarationsReferences |
|---|
1. Wolf S, Leitritz L, Rupp C et al. Cost reduction after introduction of a multidisciplinary infectious disease service at a German university hospital. Infection 2000; 28: 379–83.[CrossRef][Web of Science][Medline]
2. Lemmen SW, Häfner H, Kotterik S et al. Influence of an infectious disease service on antibiotic prescription behavior and selection of multiresistant pathogens. Infection 2000; 28: 384–7.[CrossRef][Web of Science][Medline]
3. John JF, Fishman NO. Programmatic role of the infectious diseases physician in controlling antimicrobial costs in the hospital. Clin Infect Dis 1997; 24: 471–85.[Web of Science][Medline]
4. Clinton White A, Atmar RL, Wilson J et al. Effects of requiring prior authorization for selected antimicrobials: expenditures, susceptibilities and clinical outcomes. Clin Infect Dis 1997; 25: 230–9.[Web of Science][Medline]
5. Gould IM. Antibiotic policies and control of resistance. Curr Opin Infect Dis 2002; 15: 395–400.[Web of Science][Medline]
6. Erbay A, Bodur H, Akinci E et al. Evaluation of antibiotic use in intensive care units of a tertiary care hospital in Turkey. J Hosp Infect 2005; 59: 53–61.[CrossRef][Web of Science][Medline]
7. Roger PM, Martin C, Taurel M et al. Motives for the prescription of antibiotics in the emergency department of the University Hospital in Nice: a prospective study. Presse Med 2002; 31: 58–63.[Web of Science][Medline]
8. Luk WK, Wong SSY, Yuen KL et al. Inpatient emergencies encountered by an infectious disease consultative service. Clin Infect Dis 1998; 26: 695–701.[Web of Science][Medline]
9. Byl B, Clevenbergh P, Jacobs F et al. Impact of infectious diseases specialists and microbiological data on the appropriateness of antimicrobial therapy for bacteremia. Clin Infect Dis 1999; 29: 60–6.[Web of Science][Medline]
10. Fluckiger U, Zimmerli W, Saw H et al. Clinical impact of an infectious disease service on the management of bloodstream infection. Eur J Clin Microbiol Infect Dis 2000; 19: 493–500.[CrossRef][Web of Science][Medline]
11.
Gomez J, Conde Cavero SJ, Hernandez Cardona JL et al. The influence of the opinion of an infectious disease consultant on the appropriateness of antibiotic therapy in a general hospital. J Antimicrob Chemother 1996; 38: 309–14.
12. Fowler VG, Sanders LL, Sexton DJ et al. Outcome of Staphylococcus aureus bacteremia according to compliance with recommendations of infectious diseases specialists: experience with 244 patients. Clin Infect Dis 1998; 27: 478–86.[Web of Science][Medline]
13. Roger PM, Farhad R, Pulcini C et al. Elderly patients presenting with fever and respiratory problems in an intensive care unit. Diagnostic, therapeutic and prognostic impact of a systematic infectious disease consultation. Presse Med 2003; 32: 1699–704.[Web of Science][Medline]
14. Lo E, Rezai K, Evans AT et al. Why don't they listen ? Adherence to recommendations of infectious disease consultations. Clin Infect Dis 2004; 38: 1212–8.[CrossRef][Web of Science][Medline]
15. Warren MM, Gibb AP, Walsh TS. Antibiotic prescription practice in an intensive care unit using twice-weekly collection of screening specimens: a prospective audit in a large UK teaching hospital. J Hosp Infect 2005; 59: 90–5.[CrossRef][Web of Science][Medline]
16. Roger PM, Hyvernat H, Verleine-Pugliese S et al. Short-term impact of systematic infectiology consultation on antibiotic use in an intensive care unit. Presse Med 2000; 29: 1640–4.[Web of Science][Medline]
17. Levy MM, Fink MP, Marshall JC et al. 2001 SCCM/ESICM/ACCP/ATS/SIS International Sepsis Definitions Conference. Crit Care Med 2003; 31: 1250–6.[CrossRef][Web of Science][Medline]
18. Bernard GR, Artigas A, Brigham KL et al. The American – European Consensus Conference on ARDS. Definitions, mechanisms, relevant outcomes, and clinical trial coordination. Am J Respir Crit Care Med 1994; 149: 818–24.[Abstract]
19. Muckart DJ, Bhagwanjee S. American College of Chest Physicians/Society of Critical Care Medicine Consensus Conference definitions of the systemic inflammatory response syndrome and allied disorders in relation to critically injured patients. Crit Care Med 1997; 25: 1779–85.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  E. Sellier, P. Pavese, S. Gennai, J.-P. Stahl, J. Labarere, and P. Francois Factors and outcomes associated with physicians' adherence to recommendations of infectious disease consultations for inpatients J. Antimicrob. Chemother., November 12, 2009; (2009) dkp406v1. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Pulcini Comment on: Non-adherence to infectious diseases consultations: are surgeons to blame? J. Antimicrob. Chemother., November 1, 2006; 58(5): 1095 - 1095. [Full Text] [PDF]
|
|
|
|
|
|
V. De Souza, A. MacFarlane, A. W. Murphy, B. Hanahoe, A. Barber, and M. Cormican A qualitative study of factors influencing antimicrobial prescribing by non-consultant hospital doctors J. Antimicrob. Chemother., October 1, 2006; 58(4): 840 - 843. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. G. Madariaga Non-adherence to infectious disease consultations: are surgeons to blame? J. Antimicrob. Chemother., May 1, 2006; 57(5): 1019 - 1020. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||