QJM Advance Access originally published online on August 26, 2005
QJM 2005 98(10):745-752; doi:10.1093/qjmed/hci114
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Pharmacy-implemented guidelines on switching from intravenous to oral antibiotics: an intervention study
From the Departments of 1Pharmacy and 4Medicine, Stobhill Hospital, Glasgow, 2Infectious Diseases Unit, Brownlee Centre, Gartnavel General Hospital, Glasgow, and 3Department of Pharmaceutical Sciences, University of Strathclyde, Glasgow, UK
Address correspondence to Dr R.A. Seaton, Consultant Physician, Brownlee Centre, Gartnavel General Hospital, 1053 Great Western Road, Glasgow, G12 0YN. email: andrew.seaton{at}northglasgow.scot.nhs.uk
Received 4 March 2005 and in revised form 12 July 2005
 |
Summary |
|---|
 Top Summary IntroductionMethodsResultsDiscussionReferences |
|---|
Background: A high proportion of medical in-patients in the UK receive intravenous (IV) antibiotic therapy. This may be inappropriate in non-severe infections, or unnecessarily prolonged.
Aim: To assess the impact of guideline implementation on IV antibiotic prescribing in medical admissions to a general hospital.
Design: Observational intervention study.
Methods: Data relating to infection and antibiotic therapy were collected for 4 weeks pre-intervention (group 1) and 4 weeks post intervention (group 2). Six months later, data were collected for a further 4 weeks following a second intervention (group 3). Interventions consisted of pharmacy-led implementation of guidelines incorporating criteria for IV therapy and switching to the oral route. The second intervention also included pharmacy-initiated feedback on prescribing. The main outcome measures were IV antibiotic duration, and appropriateness of the IV route and switching.
Results: Of 2365 admissions, 757 (32%) had 806 treated episodes. IV therapy was used in 40%, 46% and 36% (groups 1, 2 and 3, respectively) and was appropriate in 92% vs. 100% (group 1 vs. 2). In groups 2 and 3, oral switch timing was appropriate in 90% and 88%, vs. 17% in group 1 (p<0.001). Between groups 1 and 2, median duration of IV therapy was reduced from 3 to 2 days (p = 0.01). More patients in group 2 received appropriate exclusively IV therapy (65% vs. 96%, p<0.01). Duration of stay in IV-treated patients reduced from 13 to 10 days in groups 2 and 3 (p = 0.047). IV antibiotic expenditure reduced by 13% per patient admitted between groups 1 and 2.
Discussion: Pharmacy-led introduction of antibiotic guidelines appears to result in clinically appropriate reductions in IV therapy.
|
Introduction |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionReferences |
|---|
Antibiotics are the most commonly used and costly drug group in hospital. On average one third of patients receive antibiotic therapy,1–3 40% of whom receive an intravenous (IV) agent.2 Up to 40% of prescriptions are incorrectly prescribed or inappropriate,4–6 often reflecting over-use of expensive broad-spectrum IV agents. Escalation in empirical antibiotic usage is associated with microbial resistance,7 increased duration of hospital stay, morbidity, mortality and increased pressure on infection control resources. Indirect costs of IV therapy include time-consuming preparation, administration and monitoring of injections or infusions, which impact on nursing, medical or pharmacy time.
In June 2000, a multidisciplinary Antibiotic Management Team in North Glasgow developed antibiotic guidelines for junior doctors in general medicine and surgery. Modelled on previously developed guidelines,8 they reflect principles outlined by the Scottish Infection Standards and Strategies (SISS) group9 and incorporated British Thoracic Society (BTS) guidelines for the management of community-acquired pneumonia.10 This pilot study was undertaken opportunistically to coincide with the planned introduction of the guidelines. The study set out primarily to assess the short-term effectiveness of the pharmacy-led introduction of these guidelines on the quality of IV antibiotic use within medical units.
|
Methods |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionReferences |
|---|
The study was performed in 12 medical wards in Stobhill hospital, Glasgow. Patients were admitted via the medical receiving unit, and then were transferred to one of 11 general medical units. Empirical antibiotic prescribing was at the discretion of the on-call junior medical staff, with or without discussion with senior colleagues. A hospital formulary was in use, and the locally produced ‘junior doctors’ handbook’ provided a list of appropriate antimicrobials for selected conditions. Prior to the study, there were no guidelines on the use of antimicrobials, and in particular, there was no guidance on the route of administration or duration of IV therapy. Prior to and throughout the study period, there was no strategy for restricting selected antimicrobials. There was no infectious diseases physician on the study site, and therefore specialist clinical review of patients did not occur throughout the study. Advice on antimicrobials was generally provided ad hoc by telephone, if requested, by the on-call microbiologist, usually following a positive blood culture isolate. This practice continued throughout the study period. The Drugs and Therapeutics Committee, senior nursing and medical staff approved the study. Ethical approval was not required, as the intervention used standard tools used in everyday practice, and did not include identifiable patient data. It was a prospective, observational intervention study. Based on previous surveys,2 we estimated that 3–5 new patients per day would start IV antibiotic therapy (80–140 patients/month).
The study took place in three patient populations over 9 months and involved two interventions. Data were collected by the study pharmacist over three 4-week periods. Pre-intervention (group 1) and post-intervention (group 2) were separated by a 2-week intervention phase. A further intervention after 6 months was followed by data collection from group 3. During the pre-intervention phase (group 1), prescribing was monitored by the study pharmacist, but there was no intervention to alter prescribing. Interaction between prescribers and the study pharmacist was limited to answering prescribers’ questions regarding use of the guidelines. The study pharmacist did not pro-actively intervene in prescribing. Medical staff did not change during the study period.
The second intervention involved presentations to re-introduce the sepsis and switch guidelines, followed by direct feedback on the IV antibiotics prescribed, and promotion of timely IV to oral antibiotic switch.
In each phase of the project, the project pharmacist (CM) screened medication charts daily. All patients who commenced antibiotics during the study period were included, except those prescribed for prophylaxis, H. pylori eradication or within haemato-oncology, where specialist guidelines were already in place.
Sepsis and IV to oral antibiotic switch guidelines were launched (by RAS and CM) at three separate staff meetings following completion of the pre-intervention phase. The majority of junior doctors responsible for prescribing attended, as did all pharmacists and ward nursing managers. Background, rationale, aims and objectives of the study were discussed, including the importance of: recognition of sepsis parameters, indications for IV therapy, blood cultures in patients with sepsis, and recording information about infection in the case notes. The sepsis guidelines highlighted the importance of identifying the anatomical site and severity of infection, as well as the likely source (hospital- or community-acquired) and indications for IV therapy. Sepsis was defined as any infective episode with an associated systemic inflammatory response11 (two or more of: temperature >38°C or <36°C, Heart rate >90 bpm, respiratory rate >20/min and white cell count >12/mm3 or <3/mm3). Severe sepsis was sepsis plus organ hypoperfusion (hypotension, renal failure, acidosis, confusion or adult respiratory distress syndrome).11 Hospital-acquired infection was defined as any antibiotic episode where an antibiotic was commenced at least 72 h after admission,12 and assumes the infection was neither present nor incubating at the time of admission.
Copies of guidelines were distributed to all members of medical staff and pharmacists, and were prominently displayed in the ward doctors’ room and at the nursing station.
For patients commencing IV antibiotics, IV to oral switch therapy (IVOST) guidelines and a ‘REFER TO IVOST PROTOCOL’ sticker were inserted into medicine charts within 24 h. Nurses responsible for administering IV antibiotics were informed of the criteria and rational for switching from IV to oral antibiotics (Figure 1). The pharmacist discussed specific prescriptions with medical and nursing staff only if requested by prescribers in order to clarify the use of the guidelines. Guidelines were left in place in all medical units following this phase of the study.
|
Six months after the introduction of the guidelines, similar presentations were made to pharmacists and a new cohort of prescribers, reinforcing the importance of the guidelines. In addition each patient receiving IV therapy had a ‘pharmacy antibiotic feedback’ form inserted in their case notes. These were developed by the pharmacist and ID physician. In brief, these outlined proposed antibiotic therapy for the stated indication and criteria for IV to oral switch. Prescribers were encouraged by the project pharmacist to discuss differences in antibiotic management.
Groups 1, 2 and 3 were compared. Primary outcome measures were appropriateness and duration of IV therapy, and appropriateness of timing of IV to oral switch. Secondary outcome measures were total duration of antibiotic therapy, and length of hospital stay. Antimicrobial costs were calculated for groups 1 and 2.
The pharmacist collected and recorded data from case records, observation and medicine charts. Data included: demographics; co-morbidity; symptoms and signs of infection (including sepsis criteria); the site of the infection; whether the oral route of administration was compromised; antibiotic prescribing information and details of investigations performed. The study pharmacist assessed the appropriateness of the route of therapy and timing of IVOST daily, based on the presence or absence of the guideline clinical criteria for IV therapy, as recorded by nursing or medical staff (Figure 1). If a patient was administered an IV agent in the absence of sepsis or other IV indicators, the route of administration was recorded as inappropriate by the study pharmacist. The appropriateness of each IV prescription was then verified by a second observer (NB), blinded to the phase of the study, who checked appropriateness against the recorded clinical data. In circumstances where there was uncertainty, a third blinded observer (RAS), assessed the data. Duration of therapy and stay was rounded upwards to the nearest day. Antibiotic costs for groups 1 and 2 were crudely calculated by the study pharmacist by examining ward pharmacy issues. Total antibiotic expenditure, costs per patient treated for infection and cost per IV-treated episode were calculated. Hospital acquisition costs did not change during the study. Personnel costs were for whole-time D Grade Pharmacist (£27 754 per annum), assuming 50% of time on direct pharmaceutical care.
Data were entered and analysed on Microsoft Excel. Length and appropriateness of therapy were calculated for each infection episode. Duration of stay was calculated for the first episode of infection in each patient, in case patients were treated for multiple infections during the same admission. Non-normally distributed continuous data were compared using non-parametric tests (Mann-Whitney U test). Categorical data were compared using the 
2 test with Yates' correction for continuity. p values <0.05 were regarded as significant.
|
Results |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionReferences |
|---|
The study was performed from March to May 2001, and during November 2001, during which time 2365 patient admissions were screened for inclusion in the study: 757 (32%) patients treated with 806 antibiotic courses were included. Groups 1, 2 and 3 were comparable in age, sex, proportion with hospital-acquired infection, and proportion with chronic medical conditions (Table 1). Documented sites of infection were similar, the most common being respiratory tract (68%, 59% and 58%). A similar proportion of IV-treated patients had signs of the sepsis syndrome (74%, 82% and 81%) and severe pneumonia (43%, 48% and 56%), although a greater proportion of patients receiving IV therapy in groups 2 and 3 had parameters of severe sepsis (40% vs. 59% and 64.5%, p<0.01) (Table 1). In group 1, no infection source was documented in 43 (15%) episodes, compared to 8 (3%) in group 2 and 7 (2.5%) in Group 3 (p<0.001) (Table 1).
|
The proportion of IV-treated episodes was similar between groups (40%, 46% and 36%), and IV therapy without switch was used in 34%, 25% and 24%, respectively (p = 0.15 for groups 1 and 2 compared) (Table 2). The guidelines included guidance for most scenarios where IV therapy was required (92% in group 1, 89% in group 2 and 94% in group 3). Initial IV route of administration was appropriate in 92% in group 1 and all of patients in groups 2 and 3 (Table 2).
|
Improvements were observed in the appropriateness of IVOST timing; 90% switched to oral therapy appropriately in group 2, compared to 17% in group 1 (p<0.001), and fewer switched late (6% in group 2 vs. 78% in group 1, p<0.001) (Table 2). Appropriateness of IV to oral switch was sustained at 6 months in group 3 (Table 2). Appropriateness of IV to oral switching was examined separately for each week prior to and following the intervention (Figure 2). This also showed marked differences pre- and post-intervention. However, there were non-significant improvements in the appropriateness of switch (p = 0.8) and in the reduction of late switching (p = 0.39) between weeks 3 and 4, prior to the implementation. The differences between week 4 (pre-intervention) and week 7 (post-intervention) remained significant for both switching on time (p<0.01) and for late switching (p<0.01). Appropriate switching was maintained following the second intervention (Figure 2).
|
Appropriate exclusive IV therapy (e.g. patients in whom oral route was compromised) was observed significantly more frequently in groups 2 and 3 (96% in group 2 and 100% in group 3 vs. 65% in group 1, p<0.01 for comparison between groups 1 and 2) (Table 2). Median duration of IV therapy reduced from 3 days (group 1) to 2 days in group 2 (p = 0.01), although overall length of antibiotic therapy (including oral treatment) was unchanged (Table 3). In group 3, median duration of therapy was 3 days. Median duration of stay for IV-treated patients reduced from 13 to 10 days (p = 0.047), and this was sustained at 6 months in group 3 (median 10 days, IQR 6–25 days) (Table 3). No differences were observed in the length of therapy or hospital stay in patients treated exclusively with oral antibiotics.
|
Expenditure on IV antibiotics in group 1 was £10 919 (£14.25 per patient admitted) vs. £8796 (£12.37 per patient admitted) in group 2, i.e. 13% reduction in IV costs per patient admitted. IV antibiotic costs per episode treated were reduced by 17% (£86 vs. £71 in group 2). Cost attributed to additional pharmaceutical care in group 2 period was £1156. When included with group 2 IV antibiotic costs, 9% reduction in total antibiotic expenditure was observed.
|
Discussion |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionReferences |
|---|
This pilot study evaluated the short-term impact of pharmacy-implemented guidelines on IV antibiotic use in medical units. Two separate strategies were used: firstly, introduction of sepsis and IVOST guidelines with minimal intervention by a clinical pharmacist; and secondly, reintroduction of guidelines followed by feedback on prescribing by a clinical pharmacist. Due to the opportunistic nature of the study, which was around the time of guideline introduction, and because of resource limitations, a simple before and after study was performed. Prior to the intervention, empirical IV therapy was shown to be appropriate. However, using defined criteria for IV to oral switch revealed that a high proportion of patients prior to the intervention were not switched to oral therapy, despite improvements in clinical signs of infection. Following implementation of the guidelines, a significant reduction in duration of IV therapy was observed. Importantly, this was associated with improvements in the appropriateness of IV to oral switch by the defined criteria, suggesting an improvement in the awareness of guidelines and their utility in the process of streamlining antibiotic prescribing. The follow-up study showed that the appropriateness of empirical IV therapy and IV to oral antibiotic switch had been maintained at 6 months. This however had followed guideline reintroduction, as well as the introduction of pharmacist-initiated antibiotic feedback. Others have demonstrated treatment guidelines promoting oral therapy reduce IV duration: from 5.7 days to 2.1 days in pneumonia;13 from 4.4 days to 3.7 days in patients with community-acquired infection treated on medical units;14 and from 6 to 4 days in a heterogeneous patient population.15 In our study, median length of IV therapy was 3 days prior to intervention (comparing favourably with the intervention arms of the above studies), but still significantly reduced to 2 days. Although median duration of IV therapy increased following the second intervention, IV to oral switch in these patients was timed appropriately.
An alternative explanation for the earlier switch in our study is differing case mix. However, in the IV-treated patients, proportions with significant co-morbidity, sepsis syndrome and severe pneumonia were the same. Interestingly, a greater proportion of IV-treated patients with severe sepsis were seen post-intervention. Since sepsis parameters and severity markers were not recorded in patients receiving oral therapy, this difference cannot easily be explained. It may be that a greater proportion of patients with severe sepsis were treated with oral therapy before the intervention. Another possible explanation for apparent earlier IV to oral switch is the fabrication of criteria to justify IV therapy. This is unlikely, as there were no prescribing restrictions and no other apparent motive or benefit to prescribers.
Since the same medical staff were present in the first phases of the study and it was not a randomized controlled study, it is not possible to exclude the possibility that prescribing was improving irrespective of the intervention. The final phase of the study was with a new cohort of prescribers and the improvements seen initially seemed to have been maintained. It is not possible to judge the impact of the second intervention, which was implemented on the background of an established guideline.
In the final week of the first observation period (group 1), prior to implementation of the guidelines, there was a slight but insignificant improvement in the appropriateness of IV to oral switch (Figure 2). This may have been accounted for by some overlap of patients on IV therapy pre-intervention switching during the intervention phase of the study when awareness of guidelines was potentially greatest, or may have represented simple variation in prescribing prior to the intervention. It is also possible that the daily presence of the study pharmacist on the medical wards reviewing case notes and prescriptions may have unintentionally influenced prescribers, who were aware that an antibiotic study was underway. Following the first intervention, there was sustained improvement in the appropriateness of the IV to oral switch over the 4 weeks of observation. Six months after completion of the study, a further pharmacy-led pro-active intervention was associated with maintenance of IV to oral switch standards. As data were not gathered prior to this second intervention, it is not possible to determine its true impact. Irrespective of this it, does demonstrate that standards can be maintained with repeated interventions.
A reduction in the duration of hospital admission for IV-treated patients from 13 to 10 days was observed between group 1 and groups 2 and 3 (Table 3), suggesting that the duration of IV therapy may influence decisions to discharge from hospital. Others have shown that in patients hospitalized with non-life-threatening lower respiratory tract infections, those randomized to receive oral rather than IV therapy were discharged from hospital more quickly.16 Our study population was situated in one hospital, and was well defined and closely scrutinized during the project, and it was clear that there were no other interventions or policy changes to encourage earlier discharge or earlier switch. It is noteworthy that patients treated with exclusive oral therapy were not discharged earlier post-intervention.
Recording of site of infection improved significantly after the first intervention. This may be explained by a better appreciation of infection syndromes, or alternatively, proportionally less patients without infection may have been given antibiotics in the post-intervention phase.
Financial savings were suggested between groups 1 and 2 overall and per infection treated, although spending may have not been representative of the individual patients studied, as pharmacy ward returns were used to estimate costs. Savings were due to reductions in IVs through reduced duration and/or through the use of cheaper but equivalent agents. Savings may have been under-estimated, as consumables (syringes, needles, IV giving sets, saline, etc.), fixed costs (professionals’ time preparing, administrating and monitoring antibiotics) and impact of reduced length of stay were not calculated. The attributable salary cost for pharmacy had a marginal effect on savings overall. Potentially, a dedicated full-time pharmacist could have a wider impact by proactively influencing prescribing in both medical and other patient groups, thereby contributing to greater cost savings in other areas. A full pharmaco-economic analysis was not undertaken, but would be beneficial.
The strengths of the study were that a large number of patients were studied, and continuous detailed clinical data were collected over a 2-month period. This was enhanced by the repeat survey after 6 months following a further intervention. Most previously published studies have focussed on the impact of intervention on specific agents or indications, whereas we assessed prescribing for medical patients with any infection. This was also a practical assessment of multi-faceted interventions to improve antibiotic prescribing, which are directly applicable to hospital practice.
Ideally, a controlled trial would be undertaken, to minimize bias. Such a study design would require the participation of a number of different but equivalent hospitals. This was not feasible with the resources available for the current study. Others have demonstrated how interrupted time series can evaluate prescribing. This technique is useful in assessing longer-term trends in prescribing, and does not require a specific control group.17 However, this project was designed to judge the immediate impact of the introduction of guidelines, with detailed, continuous data collected prospectively. Such detailed continuous data collection over many months, particularly with regards to the appropriateness of IV to oral antibiotic switch, would require much greater resources than were available for the present study.
Day-to-day contact with prescribers and nursing staff by the project pharmacist reinforced adherence to the guidelines. The enthusiastic cooperation of ward staff was also critical. Nursing staff responsible for preparing and administrating IV therapy were particularly supportive of the intervention, as these affected their workload. Proactive guidance on prescribing by way of insertions into the antibiotic prescription chart has been effective at reducing IV clindamycin use in a multi-centre study.18 Contact between the pharmacist and prescriber was not systematic in the first part of the study, although presence of the pharmacist in the wards probably served as a reminder, and may have affected prescribing. There is potentially much greater scope for the pharmacist in promoting good prescribing in hospitals through immediate feedback19 or through multidisciplinary antimicrobial management teams with infection specialists.4,20 Such strategies may incorporate a more robust approach to prescribing, including prior authorization for some agents.21 Other agents may be amenable to automatic or pharmacy-initiated IVOST. It is likely that without the continual implementation of guidelines, clinical standards would deteriorate and financial savings would not be realized, particularly as turnover of junior medical staff is frequent. In the present study, the repeat survey after a lapse of 6 months and after further intervention suggested that standards may be maintained with repeated interventions.
The study has suggested that simple, locally developed guidelines may facilitate more appropriate use of IV antibiotics with earlier switching to oral therapy. This may reduce hospital stay and antibiotic expenditure. Every prescriber has a responsibility to prescribe antibiotics appropriately, although others in the clinical team play an important part in the administration of antibiotics and the review of therapeutic decisions. As such, the clinical pharmacist potentially has an important role in promoting and maintaining appropriate prescribing of IV antimicrobials in hospitals.
|
Acknowledgments |
|---|
We wish to thank Paul Burton for assistance with statistical analysis and Paul Cleary for critical review of the manuscript. Members of the North Glasgow Antimicrobial Management Group (Steve Alcock, Heather Black, Joyce Brown, Gail Caldwell, Giles Edwards, Brian Jones, Karen Ross) supported the study concept and reviewed the results. Funding was received from the Greater Glasgow Health Board. This study was presented in part at the 40th annual meeting of the Infectious Diseases Society of America, Chicago, October 2002, and the 18th International Conference on Pharmaco-epidemiology, Edinburgh, August 2002. The second pharmacy intervention was performed as part of CM's MSc thesis, and was presented at the 9th Federation of Infections Societies meeting in Manchester, 2002.
|
References |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionReferences |
|---|
1. Fraser GL, Stogsdil P, Dickens JD, et al. Antibiotic optimisation. An evaluation of patient safety and economic outcomes. Arch Int Med 1997; 157:1689–94.
2. Bodasing N, Cahill S, Seaton RA. Survey of antimicrobial prescribing in West Glasgow hospitals. J Infect 2002; 44:113.
3. Seaton RA, Nathwani D, Phillips G, et al. Clinical record keeping in patients receiving antibiotics in general medical wards. Health Bull 1999; 57:28–33.
4. Gross R, Morgan AS, Kinky DE, et al. Impact of a hospital-based antimicrobial management program on clinical and economic outcomes. Clin Infect Dis 2001; 33:289–95.[CrossRef][Web of Science][Medline]
5. Gomez J, Conde Cavero SJ, Hernanadez Cardona JL, et al. The influence of the opinion of an infectious disease consultant on the appropriateness of antibiotic treatment in a general hospital. J Antimicrob Chemother 1996; 38:309–14.
6. Byl B, Clevenbergh P, Jacobs F, et al. Impact of infectious diseases specialists and microbiological data on the appropriateness of antimicrobial therapy for bacteraemia. Clin Infect Dis 1999; 29:60–6.[Web of Science][Medline]
7. Dancer SJ. The Problem with Cephalosporins. J Antimicrob Chemother 2001; 48:463–78.
8. Seaton RA, Nathwani D. Rationale for sepsis management in immunocompetent adults. Proc R Coll Phys Edin 2000; 30:11–19.
9. Scottish Infections Standards and Strategies (SISS group) Good practice guidance for antibiotic prescribing in hospital. J R Coll Phys Edin 2003; 33:281–4.
10. British Thoracic Society. Guidelines for the management of community acquired pneumonia in adults. Thorax 2001; 56(suppl. IV):1–55.
11. Bone RC, Balk RA, Cerra FB, et al. Definitions for sepsis and organ failure and guidelines for the use of innovative therapies in sepsis. Chest 1992; 101:1644–55.
12. Emmerson AM, Enstone JE and Kelsey MC. The second national prevalence survey of infection in hospitals: methodology. J Hosp Infect 1995; 30:7–29.[CrossRef][Web of Science][Medline]
13. Al-Eidan FA, McElnay JC, Scott MG, et al. Use of a treatment protocol in the mangement of community acquired lower respiratory tract infection. J Antimicrob Chemother 2000; 45:387–94.
14. Laing RBS, MacKenzie AR, Shaw H, Gould IM, Douglas JG. The effect of intravenous to oral switch guidelines on the use of parenteral antimicrobials in medical wards. Journal of Antimicrobial Chemotherapy 1998; 42:107–11.
15. Sevinc F, Prins JM, Koopmans RP, et al. Early switch from intravenous to oral antibiotics: guidelines and implementation in a large teaching hospital. J Antimicrob Chemother 1999; 43:601–6.
16. Chan R, Hemeryck L, O'Regan M, Clancy L, Feely J. Oral versus intravenous antibiotics for community acquired lower respiratory tract infection in a general hospital: randomised controlled trial. Br Med J 1995; 310:1360–2.
17. Ansari F, Gray K, Nathwani D, et al. Outcomes of an intervention to improve hospital antibiotic prescribing: interrupted time series with segmented regression analysis. J Antimicrob Chemother 2003; 52:842–8.
18. Martinez MJ, Freire A, Castro I, et al. Clinical and economic impact of a pharmacist-intervention to promote sequential intravenous to oral clindamycin conversion. Pharm World Sci 2000; 22:53–8.[CrossRef][Web of Science][Medline]
19. Seto W-H, Ching T-Y, Kou M, et al. Hospital antibiotic prescribing successfully modified by immediate concurrent feedback. Br J Clin Pharm 1996; 41:229–34.[Web of Science][Medline]
20. Knox K, Lawson W, Dean B, Holmes A. Multidisciplinary antimicrobial management and the role of the infectious diseases pharmacist—a UK perspective. J Hosp Infect 2003; 53:85–90.[CrossRef][Web of Science][Medline]
21. White AC, Atmar RL, Wilson J, et al. Effects of requiring prior authorisation for selected antimicrobials: expenditures, susceptibilities and clinical outcomes. Clin Infect Dis 1997; 25:230–9.[Web of Science][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  D. Mertz, M. Koller, P. Haller, M. L. Lampert, H. Plagge, B. Hug, G. Koch, M. Battegay, U. Fluckiger, and S. Bassetti Outcomes of early switching from intravenous to oral antibiotics on medical wards J. Antimicrob. Chemother., July 1, 2009; 64(1): 188 - 199. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||