1 Otitis Media Diagnosis and Management Guidelines: Implications for Research Peter S Morris (with thanks to Amanda Leach, Katherine Jarosz, Sam Brophy-Williams, Jess Sommer, Paul Torzillo and the OM TAG) for the Child Health Division of the Menzies School of Health Research, Darwin, Australia.

2 Acknowledgment of CountryI would like to acknowledge the Kombumerri people, the traditional custodians of this land where we are meeting today. I pay my respects to their elders past and present and extend that respect to all Aboriginal and Torres Strait Islander people meeting with us today.

3 Disclosure Slide This work has been supported by project grants awarded by the Australian National Health and Medical Research Council. The update of the OATSIH OM Guidelines is being funded by the Centre of Research Excellence in Ear and Hearing Health in Aboriginal and Torres Strait Islander Children (CIA Prof Amanda Leach). The speaker has received funding from Pfizer/Wyeth Vaccines and Glaxo SmithKline to conduct investigator led studies and participate in Vaccine Advisory Group meetings.

4 Outline of Talk “No meaning without context”Recent evidence-based guideline history OM Guidelines- subgroups and outcome measures New developments- making the GRADE Practical issues for researchers targetting a global audience

5 Life at the “Top”…… NT 240,000 Total <1.5m ATSI 72,000QLD 1,000,000 ATSI 65,000 WA 110,000 ATSI 30,000 NT Yang Yin The other 22m

6 1967 Referendum

7 Paediatrics in the NT Health Gains Planning Information Sheet, Dec 2010

9 The 1995 RAOM meeting…… Sylvan Stool (Co-Chair) et al. Judith Boswell

10 The 1st EBM OM GuidelinesClinical Practice Guideline. Managing Otitis Media with Effusion in Young Children. AHCPR Publication “…The following recommendations for managing otitis media with effusion are tempered by the failure to find rigorous, methodologically sound research to support the theory that untreated otitis media with effusion results in speech/language delays or deficits.”

11 OM Guidelines

12 Important EBM influences

13 The EBM / GRADE approach……Population- what are the features of the specific condition where you might apply the intervention? Intervention- how do you apply the intervention in practice? Is it a simple or complex intervention? Comparison- what is the standard treatment in this population? Does it vary? Outcome- what are the important outcomes for this condition? Are any critical?

14 From Understanding to ApplicationBackground Biology Foreground Policy and practice

15 The importance of RCTs……

16 Subgroups- the challenge“I will you take you to the best restaurant in the UK if you can provide 5 examples of qualitative differences in the relative effect of an intervention in different populations!” Dave Sackett ~1994

17 The truth about EBM….

18 What happens if…… A 12 month old infant develops a typical episode of one of the types of otitis media (AOM with irritability/pain, OME with hearing loss, or CSOM with ear discharge) and….. They live in a developed country They live in a developing country They are at high risk of worse outcomes than usual (subgroup alert)

19 A friend from work calls……Her 12 month old daughter who has had runny nose and fever for 2 days. She is irritable, not sleeping well, and slightly reduced activity level and appetite (AOM- SOS 8/14). However, when she is not febrile she is alert, interactive, has normal appearance and no breathing problems. She asks you if she should go to the doctor now: A) Yes B) No- she can wait and go if her daughter gets any worse, or if no better in 2 days, or if still febrile in 3 days.

20 She goes to the doctor……On examination, her axillary temperature is 38.0o (100.4o). Her other observations are normal. She has a runny nose, a red throat without any exudate, and bilateral bulging tympanic membranes. The rest of her examination is unremarkable. The doctor gives advice about pain relief. Should the doctor give antibiotics now: A) Yes B) No- she can recommend waiting and starting antibiotics if she gets any worse or if she is no better in 2 days.

21 What happens if…… A 12 month old infant develops a typical episode of AOM and….. They live in a developed country They live in a developing country They are at high risk of worse outcomes than usual (subgroup alert)

22 Information can help!

23 Antibiotics for AOM Control rate=180/1134=16%Venekamp, Saunders, Del Mar, Glasziou, Rovers, 2015

24 Antibiotics for AOM Effect Size 0.70 (0.57,0.86)Venekamp, Saunders, Del Mar, Glasziou, Rovers, 2015

25 Antibiotics for AOM Fixed Effect I2=0%Venekamp, Saunders, Del Mar, Glasziou, Rovers, 2015

26 Fixed vs Random EffectsA fixed effect meta-analysis assumes all studies are estimating the same (fixed) treatment effect A random effects meta-analysis allows for differences in the treatment effect A prediction interval can provide a more complete summary of a random effects meta-analysis from study to study

27 Antibiotics for AOM Certainty=High Outcome=ImportantVenekamp, Saunders, Del Mar, Glasziou, Rovers, 2015

28 Evidence to RecommendationsCertainty of Evidence High     Moderate    Low   Very Low  Outcome Effect Size Standard RR 0.7 Large RR <0.5 Very Large RR <0.2

29 Certainty…. High     Further research is very unlikely to change our confidence in the estimate of effect Moderate    Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate Low   Further research is likely to have an importance impact on our confidence in the estimate of effect and is likely to change the estimate

30 What researchers can do….

31 1. Subgroups Make life more complicated but this is what everyone wants (ideally a subgroup of 1) Can be identified by clinician at time of diagnosis (specific condition or child features) Have a qualitatively different outcome to other children (in terms of the absolute difference) Need a specific treatment approach (linked to the relevant supporting data)

32 2. Outcome measures Death Disability Life threatening complicationSeverity of illness (pain / discomfort) Duration of illness (pain / discomfort) Clinical failure (needs more treatment)- patient Clinical failure (needs more treatment)- doctor

33 3. Mind the Gap…. Evidence – Practice Gap _______________ Evidence – Decision Making Gap Evidence – Feasibility Gap Evidence – Evidence Gap Evidence – Recommendation Gap Evidence – Population Subgroup Gap Evidence – Complex Intervention Gap Evidence – Clinical Outcome Gap

34 EBM vs Preferences!

35 Guideline Update ProcessTrying to make more “user-friendly” Adding estimates of effect Linking recommendations to GRADE Summary of Findings Tables Increasing transparency Cochrane Reviews provide the easiest access to data The perception is that they also provide the most reliable data

36 Important findings 1 High Certainty of EffectConjugate pneumococcal vaccine to prevent AOM (very small benefit) Antihistamines and decongestants for OME (no difference) Antibiotics for AOM (small benefit)

37 Important findings 2 Large EffectsTopical analgesics for AOM pain at 10 mins Antibiotics for AOM to prevent perforation Topical antibiotics for TT otorrhea Oral antibiotics for TT otorrhea Topical antibiotics (vs oral) for CSOM Topical antibiotics (vs cleaning) for CSOM

38 Implications for research 1High certainty of evidence and important benefit or harm- identify “evidence-practice gaps” Nothing in this category High certainty of evidence and no difference- identify “evidence-practice gaps” or new subgroups Antihistamines and decongestants for OME

39 Implications for research 2Moderate certainty of evidence and important benefit or harm- add to evidence or identify new subgroups Antibiotics for AOM to prevent perforation Topical antibiotics for TT otorrhea Topical antibiotics for CSOM Combined treatments for CSOM Low certainty of evidence and important benefit or harm- prioritise well-designed study Topical analgesics for AOM pain

40 Find the Gap…. Evidence – Practice Gap Pneumatic Otoscopy _______________ Evidence – Decision Making Gap Antibiotics for AOM Evidence – Feasibility Gap Surgery for OME Evidence – Evidence Gap Rx for high risk OME Evidence – Recommendation Gap Steroids for OME Evidence – Population Subgroup Gap Rx for AOMwiP Evidence – Complex Intervention Gap Rx for CSOM Evidence – Clinical Outcome Gap Rx for AOM

41 The future…. Even more collaboration at all levelsBetter access to important data Cochrane Reviews and GRADE Tables A global approach to healthcare Even more RCTs

42 Conclusions Otitis media guidelines have made an important contribution to health care Recommendations on diagnostic methods have had a limited impact on practice in Australia There are still few interventions supported by high quality evidence that improve outcomes Within the spectrum of OM, CSOM is the condition where effective interventions will be most beneficial Important subgroups are likely to appear over time and the use of combinations of therapies and more complex interventions will increase

43 Psalms, Chapter 23 Verse 4…

44 Thanks for helping….. To all the families who have participated in the Menzies studies over the past 30 years.

45 Thanks for visiting….