WHAT IS THE FORMULA FOR SUCCESSFUL ANTIBIOTIC GUIDELINES. Dilip Nathwani

Transcription

1 WHAT IS THE FORMULA FOR SUCCESSFUL ANTIBIOTIC GUIDELINES Dilip Nathwani

2 GUIDELINES OBJECTIVES Generic overview of need for evidence based guideline Importance of international guideline development methodology and linking evidence to recommendation Review of dissemination/implementation strategies Core components of national antibiotic policy for hospitals- Scottish Overview

3 What are guidelines? Clinical guidelines are systematically developed statements which assist in decision making about appropriate health care for specific clinical conditions Clinical Resource and Audit Group (1993)

4 Purpose of Guidelines To promote effective health care by reinforcing good clinical practice To promote change in professional practice where this does not comply with current best practice

5 Why are Guidelines needed? Clinical care increasingly complex Continual investment in research but lag between evidence and practice Increasing pressure to provide cost effective health care Limited choices - hard choices Increasing shared care Need for effective dissemination of effective practice methods Clinical Resource and Audit Group (1993)

6

7 Reasons for Variations in CAP Management Reasons Many microbial causes Place of management eg home, general ward, ICU Many specialities involved, eg general practice, general physicians, respiratory physicians, infectious disease specialists, intensivists Huge choice of available antibiotics Different health-care systems eg effect of availability of antibiotics, costs

8 Method of Guideline Country Year Process Development France 1991 Consensus conference Italy 1995 Two authors, pneumology society approval Spain 1992 Expert committee UK 1993 Expert committee, British Thoracic Society approval USA 1998 Expert committee, IDSA approval First one with evidence grading

9 Developing Guideline Recommendations GOBSAT method Formal consensus techniques (consensus conference, Delphi technique, etc) Explicit linkage of recommendations and evidence

10 GOBSAT method Developing Recommendations Group of experts meet (usually without systematic review) to develop guidelines based on their own clinical judgement and experience Lack of explicit decision making process Small group processes likely to have major influence on decisions Usually difficult for guideline user to assess how recommendations were derived Source: Department of Public Health, University of Aberdeen

11 Practice guidelines developed by speciality societies ( ) 1998) inclusion of professional with different speciality: 28 % (mostly epidemiologist: 26 %) information on systematic literature search: 13 % criteria to grade the strength of evidence: 18 % overall 54 % of guidelines did not meet any criterion Lancet 2000; 355:

12 Method of Guideline Country Year Process Development France 1991 Consensus conference Italy 1995 Two authors, pneumology society approval Spain 1992 Expert committee UK 1993 Expert committee, British Thoracic Society approval USA 1998 Expert committee, IDSA approval First one with evidence grading

13 Practice Guideline Development: Consensus Conferences a broad-based panel listens to scientific data presented by experts, weighs the information and then composes a consensus statement that addresses a set of questions previously posed to the panel sources of bias: type of questions, composition of the panel, selection of experts, literature, etc small group processes likely to have major influence on decision as much a social as a scientific process

14 Clinical practice guidelines: candidemia, unstable patient, no prior fluco consensus conference fluco ampho B ampho B + 5 FC ampho B lipid ampho B + fluco fluco + 5 FC neutropenia 2/20 1/20 10/20 2/20 4/20 1/20 solid organ TX 7/18 4/18 4/18 2/18 1/18 0/18 non-neutropenic 5/20 4/20 4/20 2/20 5/20 0/20 Clin Infect Dis 1997; 25:

15 Clinical practice guideline: consensus conference candidemia dose of fluco (mg) < >800 stable 1/20 18/20 1/20 0/20 deteriorating 0/20 1/20 18/20 1/20 dose of ampho B (mg/kg) < >1.0 stable 1/20 14/20 5/20 0/20 deteriorating 0/20 2/20 14/20 4/20 Clin Infect Dis 1997; 29:

16 GUIDELINE DEVELOPMENT METHODOLOGY

17 1. Selection of topics Criteria for selection of topics: Existence of variation in practice Evidence of effective practice Burden of disease Priority area for national health service - cardiovascular disease, cancer, mental health, HAI/ANTIBIOTIC PRESCRIBING

18 Development of E-BE guidelines

19 IDSA clinical practice guidelines Quality of evidence I evidence from at least one RCT II III evidence from at least one well-designed CT without randomisation; from cohort or case-controlled analytic studies; from multiple time-series studies; from dramatic results in uncontrolled experiments evidence from opinions of respected authorities, based on clinical experience, descriptive studies or reports of expert committees Clin Infect Dis 1994; 18: 421.

20 Clinical practice guideline: candidemia (IDSA) C. albicans, C. tropicalis, C. parapsilosis: ampho B fluconazole C. glabrata: ampho B fluconazole C. krusei: ampho B 0.6 mg/kg/d (AI) 6 mg/kg/d (AI) 0.7 mg/kg/d (B III) 12 mg/kg/d in less-critically ill patients (BIII) 1.0 mg/kg/d (BIII) for 2 weeks after the last positive blood culture and resolution of signs and symptoms (A III) Clin Infect Dis 2000; 30:

21 Objectives of SIGN? SIGN was established in 1993 to sponsor and support the development of evidence based national clinical guidelines and to facilitate their implementation into local practice for the benefit of patients.

22 SIGN Guideline Development Methodology SIGN guidelines are developed using standardised methodology to maximise validity, involving - multidisciplinary development groups - systematic reviews - explicit linkage between recommendations and evidence

23 Evidence Good evidence from study designs which reduce bias and are well conducted and reported Bad evidence from studies which are badly designed or which are poorly conducted or reported

24 Excluded Studies 1. Descriptions of interventions to change antibiotic prescribing without measurement of the effect of these interventions on prescribing or other outcome measures. 2. Surveys of hospitals to establish the range of measures used to control or optimise antibiotic prescribing.

25 The grading of level of evidence Level Ia Ib IIa IIb III IV WP Type of evidence from meta analysis of randomised controlled trials from at least one randomised controlled trial from at least one well-designed controlled trial without randomisation from at least one other type of well-designed quasi-experimental study from well-designed non-experimental descriptive studies, correlation studies and case control studies from expert committee reports or opinions and/or clinical experience of respected authorities Opinion of this working party

26 Developing Recommendations Explicit linkage or recommendations and evidence Involves explicit statement about quality of supporting evidence alongside key recommendations As a result the decision making process is more explicit Source: Department of Public Health, University of Aberdeen

27 The grading of recommendations Grade A (evidence levels Ia and Ib) B (evidence levels IIa, IIb and III) C (evidence level IV) WP Recommendations Requires at least one randomised controlled trial as part of the body of literature of overall good quality and consistency addressing the specific recommendation Requires availability of well-conducted clinical studies but no randomised trials on the topic of the recommendation Requires evidence from expert committee reports or opinions and/or clinical experience of respected authorities (including the current working party). Includes absence of directly applicable studies of good quality Opinion of this working party

28 Recommended Indications for Surgical Antibiotic Prophylaxis Operation Recommendation Odds Ratio NNT Outcome Cardiac pacemaker insertion A AP is recommended Any infection Tonsillectomy C AP is NOT recommended There is no evidence of effectiveness of prophylaxis from RCTs Colorectal surgery A AP is highly recommended Infection Breast surgery C AP is recommended but local policy makers may identify exceptions One RCT showed a non significant treatment effect. Subsequent inclusion of patients not randomised to the original study enhanced the treatment effect Craniotomy A AP is recommended Wound infection Hysterectomy - abdo A AP is recommended but local policy makers may identify exceptions Wound infection Cataract surgery C AP is recommended but local policy makers may identify exceptions Effectiveness is inferred from 42evidence about other procedures involving insertion of prosthetic devices Total hip replacement A AP is highly recommended 0.27 Hip infection

29 SIGN guideline development Selection of Topic Composition of the Development Group 2 months Group meetings 12 months Open Review Meeting/ National Open Review Meeting/Peer review 15 months 21 months Final Document Schedule Review 2 years post

30 Are guidelines following guidelines? adherence to methodological standards (n=279 guidelines; ) development and format 51 % identification and summary of evidence 34 % formulation of recommendations 46 % overall improvement from 36.9 % (1985) to 50.4 % (1997), but identification and summary of evidence remained weak (36 %) JAMA 1999; 281:

31 Italian Physicians Perceptions of Practice Guidelines Persistence of Traditional View Personal experiences / > evidence based opinions Not transferable to patient/clinical situation Cost containment - highlighted and externally imposed Administrative > educational or informative No enthusiasm for multi-disciplinary involvement Threatened clinical freedom Primary care least favourable towards guidelines and non-physician group involvement Arch Int Med 2001, 161:

32 Judgement and Interpretation of Evidence: An experiemental study of determinants of Group Judgements in Clinical Guideline Development Raine R et al, Lancet 2004; 364:

33 Qualitative study: CBP, IBS, CFS 51% OF 192 SCENARIOS RESEARCHED AGREED WITH RESEARCH EVIDENCE MORE LIKELY TO AGREE IF GPS ONLY AS OPPOSED TO MIXED GROUPS, IF LITERATURE REVIEW PROVIDED, OR IF EVIDENCE WAS IN ACCORDANCE WITH CLINICIANS BELIEF CLINICAL EXPERIENCE HAD A MODIFYING EFFECT ON BELIEFS ABOUT RESEARCH EVIDENCE CLINICIANS JUDGEMENT AND PATIENTS PERCEPTIONS ARE AS IMPORTANT AS EVIDENCE IN THE CLINICAL GUIDELINE DEVELOPMENT PROCESS.

34 The AGREE (appraisal of guidelines, research, and evaluation collaboration ) guidelines appraisal instrument provides a validated, internationally agreed framework for assessing the quality of clinical practice guidelines. 1998: 14 EU, US, CAN; NZ;AUS

35 AGREE Instrument Examines 6 domains/aspects and guideline development Scope and purpose Stakeholder involvement Rigour of development Clarity and presentation Applicability Editorial independence

36 International Appraisal Tool for Assessing the Quality of Clinical Practice Guidelines Qual Saf Health Care 2003; 12: GUIDELINES, 11 COUNTRIES 95% of appraisers found intrument useful for assessing guidelines Reliability was acceptable for most domains Established programme guidelines had significantly higher scores editorial independence and rigour of development

37 2002 Guideline International Network (G-I-N) Global

38 GUIDELINE INTERNATIONAL NETWORK > 2000 INTERNATIONAL GUIDELINES GIN RELATED ORGANISATIONS CRITICAL APPRAISAL TOOLS FOR CLINICAL GUIDELINES IMPELEMENTAITON SUPPORT FOR EVERYDAY CLINICAL PRACTICE GUIDELINE DEVELOPMENT TOOLS AND RESOURCES NETWORKING CAPACITY NEWS & ACTIVITIES

39 Dissemination News Media / Public Education Managers Undergraduates Guidelines Hospital Senior Staff meetings Postgraduate trainees GPs Pharmacy Group

40 Dissemination strategies Guideline booklets Quick reference guides Peer reviewed journals Targeted distribution via clinical teams Electronic publishing

41 SEPSIS MANAGEMENT PROTOCOL: EMPIRIC ANTIBIOTIC THERAPY INDICATIONS FOR IV USE 1.Serious or severe sepsis + 2.Febrile with neutropenia or immunosupression 3.Specific infections - endocarditis, abscess, meningitis septic arthritis or osteomyelitis 4.Oral route is compromised - nil by mouth or <50mls fluids orally - reduced absorption - mechanical swallowing disorder - unconsciousno oral formulation available Definition of Sepsis: Clinical Symptoms of Infection (Fever, sweats, chills or rigors, malaise, etc) Temperature >38 C <36 C Tachycardia >90 bpm Tachypnoea RR > 20/min WCC <4 or >12 +Serious/Severe: Sepsis associated with organ dysfunction, hypoperfusion (oliguria, acute alteration of mental state) or hypotension Community Acquired (Infection present or suspected on admission) Culture appropriate area: eg. Blood Cultures (8-10ml of blood into each of the culture bottles) urine, sputum, CSF, wound or venous access site Antibiotic (Route) IV vs Oral*? Antibiotic (Choice) Antibiotic dose Hospital Acquired (Infection 48 hours after admission) Dosing Antibiotic Oral IV Amoxycillin 500mg t.d.s 500 mg t.d.s Co-amoxiclav 375mg or 625mg t.d.s 1.2g t.d.s Erythromycin 500mg q.d.s 1g b.d. Clarithromycin 500mg b.d. 1g b.d. Benzylpenicillin Pen V 1g b.d. 1-2g q.d.s. Flucloxacillin 1g q.d.s. 1-2g q.d.s. Clindamycin 300mg t.d.s. 600mg q.d.s. Ciprofloxacin mg b.d. See advice (or see allergy section) Metronidazole 400mg t.d.s. 500mg t.d.s. Cefuroxime g t.d.s. Ceftriaxone - 1-2g o.d. Gentamicin - 7mg/kg o.d.** **except endocarditis, ascites/pregnancy (see aminog-lycoside protocol) COPD exacerbation Pneumonia UTI Skin & soft Intraabdominal Hepatobiliary CNS PUO PUO CNS Intraabdominal tissue/bone /joint Skin & soft tissue/bone/ joint UTI Pneumonia COPD exacerbation ORAL ORAL ORAL ORAL ORAL ORAL ORAL ORAL ORAL ORAL ORAL ORAL ORAL Amox. Amoxycillin Co-amoxiclav FlucloxacillinCo-amoxiclavCo-amoxiclav Not + + appropriate Co-amoxiclavCo-amoxiclav Not appropriate erythromycin** Pen V* Co-amoxiclav FlucloxacillinCo-amoxiclavCo-amoxiclav + Ciprofloxacin IV IV IV IV IV IV IV IV IV IV IV IV IV IV IV Amox. Amoxycillin Co-amoxiclav Fluclox. Cefuroxine Erythromycin**** Ben. Pen.* metronidazole Coamoxiclav +/- gentamicin *** Ceftriaxone Co-amoxiclav Co-amoxiclav +/- +/- Gentamicin Gentamicin *** Ceftriaxone (seek advice) Cefuroxime + metrondiazole +/- gentamicin Seek advice Co-amoxiclav +/- Gentamicin Co-amoxiclav +/- Gentamicin Ciproflox. Not appropriate * If Penicillin allergy use clindamycin ** If gastrointestinal disturbance with erythromycin consider clarithromycin *** Aciclovir IV may be indicated empirically **** If severe - IV cefuroxime + erythromycin IV See Pneumonia Protocol for further guidance Penicillin Allergy If patient gives a history of rash or anaphylaxis or there is reasonable doubt: for gram +ve cocci cover: Vancomycin IV for gram -ve bacilli cover: Ciprofloxacin Have you completed the antibiotic stamp/ sticker? (Green adhesives) If not available please write indication for antibiotic in case records If renal failure or dysfunction seek advice See Gentamicin Protocol for dosing + monitoring Complicated or unusual pathogen infection For further antibiotic sepsis advice contact On Call ID Physician - NWH Switchboard Duty Microbiologist - NWH Switchboard Duty Clinical Pharmacist Modify treatment according to laboratory sensitivity If patient is on intravenous antibiotics can you change them to oral therapy? April 1999 Review April 2003

42

43

44 IMPLEMENTATION STRATEGIES CONSIDER AND IDENTIFY AS PART OF GUDELINE DEVELOPMENT PROCESS

45 Implementation of SIGN Guidelines in Scotland Topic Full % Not % Cancer Stroke/CHD Dentistry Diabetes 35 4 Mental Health All others 23 24

46 Best Evidence to Best Practice IMPLEMENTATION OF CHANGE IN PATIENT CARE 54 PAPERS - ORIGINAL STUDIES AND SYSTEMATIC REVIEWS OF CHANGE INTERVENTIONS Grol R & Grimshaw J Lancet 2003; 262:

47 Best Evidence to Best Practice Mixed intervention best effect Most well designed interventions had a mean of ~ 10% impact Some interventions better for certain areas Education and information ~8% median improvement Audit and feedback Reminders and computer decision support Multi-professional collaboration Patient mediated Combined others

48 Implementation Strategies Facilitative strategies rely on regular feedback of information to prescribers about their own practice in comparison with peers and usually involves personnel communication as well as mailed information but compliance with the policy remains voluntary

49 Barriers to Guidelines 1 Professional Knowledge Unaware Unfamiliar Attitudes Do not agree with recommendations Unable to meet recommendations Inertia Cabana et al, JAMA Oct , 282:

50 Barriers to Guidelines 2 External Patient related Guideline related Inconvenient or confusing Contradictory Environmental Time and other resources Reimbursement Liability Cabana et al, JAMA Oct , 282:

51 Sequence of Behaviour Change Knowledge Attitudes Behaviour Cabana et al, JAMA Oct , 282:

52 Reasons for physician s nonadherence to the recommendations of practice guidelines, as identified in a systematic review of the literature Cabana MD, Rand CS, Powe NR et al JAMA 1999; 282: Known reasons for nonadherence to guidelines Knowledge Lack of awareness Lack of familiarity Attitudes Nonagreement with interpretation of evidence Applicability to own patients Cost-effectiveness Lack of outcome expectation (compliance with guideline will not lead to the expected outcome) Lack of self-efficacy (cannot perform recommendations) Inertia or lack of motivation Behaviour Patient (preferences opposed to guideline recommendations) Guidelines Not readily available Difficult format Contradictory Environment Insufficient time, insufficient resources, organizational barriers, and reimbursement problems Malpractice liability

53 CLINICIANS PERCEPTIONS OF THE PROBLEM OF ANTIMICROBIAL RESISTANCE IN HEALTH CARE FACILTIES Arch Intern Med 2004; 164:

54 Is Antibiotic Resistance a local (in organisations or practice) or national problem? 117 QUESTIONNAIRES 95% V 77% (p < 0.001) perceived a national problem as opposed to local problem 28 focus group - 93% v 46% (p < 0.001) perceived a national problem as opposed to local problem

55 Which 3 of the 12 CDC Campaign Steps were the most barriers to implementation? Treat infection, not colonisation (35%% Practice Antimicrobial Control (33%) Stop treatment, when infection is cured or not likely (35%)

56 Use of Local Data as a valuable prescribing tool 31.6% thought this was one of three least important strategies Only 5.1% thought it was one of three most important strategies

57 Adherence Barriers to Antimicrobial Treatment Guidelines in Dutch Teaching Hospital Emerging Infect Dis 2004; 10(3):

58 CAP GUIDELINE BARRIERS Guideline Dissemination Credibility of content Physician Readiness to change or use guideline( routine prescribing, no perceived resistance problem, autonomy, insufficient knowledge of cultures, low efficacy of streamlining Social and institutional context Resident not independent decision makers or supervisors Id consult secondary to supervisor Different guidelines between different department

59 Comparison of process and outcome measures for management of community-acquired pneumonia (CAP) in patients with severe or nonsevere CAP Relation to antibiotic protocol Patients with Patients with Outside Within severe CAP nonsevere CAP protocol protocol RR Characteristic (n = 48) (n = 148) (n = 58) (n = 146) (95%CI) Process or outcome of CAP management Died 20 (42) 10 (7) ( ) Received antibiotics within 29 (61) 73 (49) h of hospital adm. ( ) Received iv antibiotics 34 (71) 58 (39) within 24h of hospital adm. ( ) Blood gas levels determined 44 (92) 126 (85) within 24h of hospital adm. (o ) Blood gases or O 2 saturation 29/44 (66) 80/126 (63) repeated within 24h of ( ) hospital adm. n/n (%) Antibiotic therapy in relation to protocol Patient died or was (29) 25 (17) 0.58 readmitted to hospital ( )

60 SEPSIS MANAGEMENT PROTOCOL: EMPIRIC ANTIBIOTIC THERAPY INDICATIONS FOR IV USE 1.Serious or severe sepsis + 2.Febrile with neutropenia or immunosupression 3.Specific infections - endocarditis, abscess, meningitis septic arthritis or osteomyelitis 4.Oral route is compromised - nil by mouth or <50mls fluids orally - reduced absorption - mechanical swallowing disorder - unconsciousno oral formulation available Definition of Sepsis: Clinical Symptoms of Infection (Fever, sweats, chills or rigors, malaise, etc) Temperature >38 C <36 C Tachycardia >90 bpm Tachypnoea RR > 20/min WCC <4 or >12 +Serious/Severe: Sepsis associated with organ dysfunction, hypoperfusion (oliguria, acute alteration of mental state) or hypotension Community Acquired (Infection present or suspected on admission) Culture appropriate area: eg. Blood Cultures (8-10ml of blood into each of the culture bottles) urine, sputum, CSF, wound or venous access site Antibiotic (Route) IV vs Oral*? Antibiotic (Choice) Antibiotic dose Hospital Acquired (Infection 48 hours after admission) Dosing Antibiotic Oral IV Amoxycillin 500mg t.d.s 500 mg t.d.s Co-amoxiclav 375mg or 625mg t.d.s 1.2g t.d.s Erythromycin 500mg q.d.s 1g b.d. Clarithromycin 500mg b.d. 1g b.d. Benzylpenicillin Pen V 1g b.d. 1-2g q.d.s. Flucloxacillin 1g q.d.s. 1-2g q.d.s. Clindamycin 300mg t.d.s. 600mg q.d.s. Ciprofloxacin mg b.d. See advice (or see allergy section) Metronidazole 400mg t.d.s. 500mg t.d.s. Cefuroxime g t.d.s. Ceftriaxone - 1-2g o.d. Gentamicin - 7mg/kg o.d.** **except endocarditis, ascites/pregnancy (see aminog-lycoside protocol) COPD exacerbation Pneumonia UTI Skin & soft Intraabdominal Hepatobiliary CNS PUO PUO CNS Intraabdominal tissue/bone /joint Skin & soft tissue/bone/ joint UTI Pneumonia COPD exacerbation ORAL ORAL ORAL ORAL ORAL ORAL ORAL ORAL ORAL ORAL ORAL ORAL ORAL Amox. Amoxycillin Co-amoxiclav FlucloxacillinCo-amoxiclavCo-amoxiclav Not + + appropriate Co-amoxiclavCo-amoxiclav Not appropriate erythromycin** Pen V* Co-amoxiclav FlucloxacillinCo-amoxiclavCo-amoxiclav + Ciprofloxacin IV IV IV IV IV IV IV IV IV IV IV IV IV IV IV Amox. Amoxycillin Co-amoxiclav Fluclox. Cefuroxine Erythromycin**** Ben. Pen.* metronidazole Coamoxiclav +/- gentamicin *** Ceftriaxone Co-amoxiclav Co-amoxiclav +/- +/- Gentamicin Gentamicin *** Ceftriaxone (seek advice) Cefuroxime + metrondiazole +/- gentamicin Seek advice Co-amoxiclav +/- Gentamicin Co-amoxiclav +/- Gentamicin Ciproflox. Not appropriate * If Penicillin allergy use clindamycin ** If gastrointestinal disturbance with erythromycin consider clarithromycin *** Aciclovir IV may be indicated empirically **** If severe - IV cefuroxime + erythromycin IV See Pneumonia Protocol for further guidance Penicillin Allergy If patient gives a history of rash or anaphylaxis or there is reasonable doubt: for gram +ve cocci cover: Vancomycin IV for gram -ve bacilli cover: Ciprofloxacin Have you completed the antibiotic stamp/ sticker? (Green adhesives) If not available please write indication for antibiotic in case records If renal failure or dysfunction seek advice See Gentamicin Protocol for dosing + monitoring Complicated or unusual pathogen infection For further antibiotic sepsis advice contact On Call ID Physician - NWH Switchboard Duty Microbiologist - NWH Switchboard Duty Clinical Pharmacist Modify treatment according to laboratory sensitivity If patient is on intravenous antibiotics can you change them to oral therapy? April 1999 Review April 2003

61

62 Process of Care Nov 02 Dec 02 Jan 03 Feb 03 Mar 03 (to 13/03) Overall Mean time to 1 st dose of antibiotic (minutes ) Appropriate antibiotics Appropriate antibiotics within 4-hours % 84% 83% 91% 100% 83% 42% 52% 50% 62.5% 60% 52%

63 Quality Indicator for CAP Number of patients receiving appropriate antibiotic therapy within 4 hours of admission Time to Ist dose of antibiotic given in hospital- within 4 hours of admission

64 Improving the management of CAP: CBA evaluation with CE analysis Pre intervention v post-intervention (TAYCAPP pathway) Indicator: Number of patients receiving appropriate antibiotics within 4 hours of admission Pre Intervention Study 33% Control 32% Post intervention Study 56% Control 36% Adjusted absolute change 17% p =0.035 Barlow, Nathwani, Williams et al 2004

65 The Clinical Effectiveness Cycle Health intervention New Research Systematic Reviews Guideline development Compliance measurement Guideline implementation Standard setting Minimum data set Clinical audit criteria

66 AUDIT & FEEDBACK Quantity: Antibiotic consumption Quality Restricted agents where use, breeches of protocol Point prevalence survey Episodic audit of specific disease areas in rotation depending on need Feedback Regular, group, individual, retrospective, proactive, concurrent specific feedback

67 CONCLUSIONS Antibiotic guidelines methodology should be based on evidence and the recommendations should link to evidence Define dissemination & implementaiton strategies Adopt generic national template of key components of policy but locally adapted National framework for antimicrobial policy e.g SMC Development of national preformance indicators and intergration into the quality improvement & governance strategy