Childhood flu vaccination: experiences of a new programme in England Richard Pebody PHE Respiratory Diseases Department, London 28 èmes Rencontres sur la grippe et sa prévention, Lyons, November 2015
UK has long-standing annual seasonal trivalent inactivated vaccine (TIV) programme All high risk groups under 65 years, including pregnant women All 65+ year olds Aim to provide direct protection to target groups Problems: effectiveness of TIV in elderly and very young is moderate at best most vulnerable groups are elderly and very young
Influenza vaccine uptake (GP Patient survey) by year for England 100% 90% 80% 65 and over Under 65 at risk WHO 2010 target Vaccine uptake (%) 70% 60% 50% 40% 30% 65,4% 67,4% 68,6% 71,0% 71,5% 39,9% 75,3% 48,0% 73,9% 42,1% 73,5% 45,3% 74,1% 47,1% 72,4% 51,6% 72,8% 50,4% 74,0% 51,6% 73,4% 51,3% 73,2% 52,3% 72,7% 50,3% 20% 10% 0% 2000/01 2001/02 2002/03 2003/04 2004/05 2005/06 2006/07 2007/08 2008/09 2009/10 2010/11 2011/12 2012/13 2013/14 2014/15 Survey Year Influenza and vaccination programme, JCVI, June 2015
Influenza vaccine uptake (%) in the under 16 years at-risk by clinical risk group and age (end January 2015) 100 90 80 6 months to under 2 years 2 years to under 5 years 5 years to under 16 years 70 Vaccine Uptake (%) 60 50 40 30 20 10 21,4 47,5 31,6 51,6 50,1 24,4 56,7 44,6 49,8 49,2 34,9 46,9 36,3 55,9 55,6 24,5 45,4 40,5 43,9 43,9 24,7 63,5 60,5 68,2 68,1 29,6 55,4 47,6 55,7 55,4 17,9 47,3 33,1 51,5 50,4 32,8 47,2 31,3 37,3 36,8 0 Chronic Heart Disease Chronic Respiratory Disease Chronic Kidney Disease Chronic Liver Disease Diabetes Immunosuppression CND incl.stroke/tia, Cerebral Asplenia Palsy or dysfunction or MS of the sple Based on 97.6% (7,625/7,809) of GP practices providing data across the optional risk group categories for the final January 2015 survey Clinical Risk Group Influenza and vaccination programme, JCVI, June 2015
Case fatality ratio (deaths / 1000 influenza admissions) by age and risk group
Incidence of influenza admission by age and risk group/100,000 (2000/01 to 2007/08)
Flu vaccination of children a new paradigm? Evidence to suggest children are main drivers of influenza transmission Newly licensed live attenuated influenza vaccine with higher effectiveness in children in Europe Vaccination of healthy children has potential to provide: - Direct protection to children themselves - Indirect protection to other vulnerable persons in population (by reduction in transmission) Surveillance of LAIV programme, May 2015
Effectiveness of flu vaccine: meta-analysis Evidence of flu VE (TIV and LAIV) TIV in adults LAIV in children Lancet ID Osterholm 2012
Loeb et al JAMA 2010
Modelling and economic evaluation of current and extended programme Estimate current burden of influenza Built transmission model to estimate impact of the selective programme direct and indirect effects of various programme extensions costs of different programme extensions savings in health care costs and QALYs
Extensions to current programme Extend to low-risk: 0.5-4 years 50-64 years 5-16 years 0.5-4 & 50-64 years 0.5-4 & 5-16 years 0.5-4 & 5-16 & 50-64 years 0.5-64 years Increasing cost Net additional cost 14.2m 92.7m Coverage assumed to be 30% in low-risk groups
Annual cost and impact of current selective programme Mean estimate of annual costs & QALYs gained of programme Some years little benefit Estimated to be very few cases even in absence of vaccination Current programme highly cost-effective
Cost-effectiveness of various additions to current flu programme All additions involving vaccination of children were cost effective even with low uptake (30%) Little additional benefit by adding older age groups but big increase 70 000 60 000 50 000 40 000 30 000 20 000 10 000 in programme costs 0 6m 4 yrs 5 16yrs 6m 16 yrs 6m 16 and 50 64 yrs 50 64 yrs 6m 64 yrs Cost in per QALY gained
JCVI statement, November 2012 Study found current influenza vaccination programme is highly likely to be cost effective; Study also suggested extending programme to low risk children is likely to be cost effective as provides: - direct protection lowering impact of flu on children; - indirect protection lowering flu transmission from children to other children, adults and those in the clinical risk groups of any age Recommended introduction of flu vaccination of healthy children aged 2 to 16 years of age with newly licensed live attenuated influenza vaccine
Childhood Live Attenuated Influenza Vaccination (LAIV) Programme Annual i-move meeting, July 2015
UK surveillance plans for childhood flu programme 1. To measure the uptake of the new LAIV vaccine programme; 2. To measure the effectiveness of LAIV in children 3. To obtain a preliminary estimate of the overall impact of vaccinating school age children in terms of reduced: - Primary care consultations and positivity - Lab confirmed hospitalisations - ICU admissions - Excess mortality By comparing cumulative incidence in LAIV targeted and nontargeted age-groups in pilot with non-pilot areas. Annual i-move meeting, July 2015
Flu vaccine uptake Influenza and vaccination programme, JCVI, June 2015
Uptake for childhood flu programme Influenza and vaccination programme, JCVI, June 2015
School-age LAIV pilot areas in 2013/14 and 2014/15 Influenza and vaccination programme, JCVI, June 2015
Vaccine uptake in pilot areas, 2014-15 models of delivery 20 Influenza and vaccination programme, JCVI, June 2015
Flu vaccine uptake in 4-11 year olds by region deprivation, ethnicity and religion, 2013-14 70 60 50 % uptake 40 30 20 10 0 SE Essex Bury Gateshead Havering Newham Leicester <10.4 10.4 <17.3 17.3 <26.1 26.1 <39.9 39.9+ <5% 5 <12% 12 <34% 34% 0% >0% 0% 1 4% 5% Urban Rural Pilot site of school IMD 2010 score of school School lsoa % black or minority ethnicity Variable Influenza and vaccination programme, JCVI, June 2015 School lsoa % Jewish School lsoa % Muslim Rural/urban School status
Flu vaccine effectiveness, 2014-15 Influenza and vaccination programme, JCVI, June 2015
Weekly GP influenza-like illness rates for 2014/15 and past seasons, by age, England and Wales The RCGP weekly ILI consultation rate breached the pre-epidemic threshold in week 50 and remained at low levels of intensity until week 14 (dipping below threshold twice)
Weekly number of outbreaks by institution (A) and virological test results where available (B) by week of onset, 2014/15 UK Influenza and vaccination programme, JCVI, June 2015
Adjusted VE estimates for flu by age, type of vaccine in <18 year olds, UK, October 2014 April 2015 using TNCC (n= 701) Factor Level Adjusted VE a % (95% CI) by type Age 2,3,4 2,3,4 intranasal <18 intranasal <18 injection A A(H3N2) B 58.5 (-31.4, 86.9) 69.2 (-30.9, 92.7) 100 (-82.5, 100) b 52.5 (-54.3, 85.4) 65.7 (-50.1, 92.1) 100 (-112.8, 100) b 31.2 (-29.5, 63.4) 35.0 (-29.9, 67.5) 100 (17, 100) b -69.4 (-409.3, 43.7) -73.2 (-456.9,46.2) -123.7 (-1343, 65.3) Influenza and vaccination programme, JCVI, June 2015
Impact in 2013-14 Influenza and vaccination programme, JCVI, June 2015
Study designs for the evaluation of vaccine impact Adapted from Halloran et al
Influenza disease pyramid: sources of data for flu surveillance Deaths Hospitalised cases Community cases seen by a general practitioner Community cases not seen by a general practitioner Excess mortality monitoring Severe Influenza Surveillanc system (USISS) Emergency Departments (ED GP consultation rates Virological surveillance NHS 111 Telephone survey School surveillance
Methods To measure programme impact: compared disease incidence in targeted and non-targeted age-groups in pilot and non-pilot areas Range of clinical and laboratory-confirmed community, secondary care and mortality end-points; Additional swabbing GP practices, emergency departments and hospitals recruited in seven pilot areas; Cumulative incidence and positivity rates calculated in pilot and non-pilot areas for 2013-14 season based upon place of residence/catchment population; Influenza and vaccination programme, JCVI, June 2015
Influenza season 2013/14 Weekly proportion of sentinel GP samples positive for influenza Weekly influenza-like illness GP consultation rate 2014 Annual Influenza European Meeting, Vienna
Ratios* and 95% CI for cumulative influenza indicator activity in LAIV pilot vs non-pilot areas 2013/14 Increasing disease severity ILI swab positivity (RCGP) ILI GP consultations (RCGP) ILI GP out of hours consultations (ReSST) Respiratory ED admissions (EDSSS) Influenza confirmed hospitalisations (USISS sentinel) Influenza swab positivity (RDMS) 0 0.5 1 1.5 2 2.5 3 Ratio *Risk ratios calculated for rates with negative binomial regression. Odds ratios calculated for proportions with logistic regression, correcting for overdispersion.
Impact in 2014-15 Influenza and vaccination programme, JCVI, June 2015
Cumulative primary care indicators in primary school pilot, secondary school pilot and non-pilot areas, week 40 2014 to week 14 2015 600 RCGP ILI 80 Sentinel positivity Consultation rate per 100,000 500 400 300 200 100 0 Positivity (%) 70 60 50 40 30 20 10 0 Primary school (5 10yrs) Secondary school (11 16yrs) <5yrs 17+yrs Primary school (5 10yrs) Secondary school (11 16yrs) <5yrs 17+yrs Age group Age group Influenza and vaccination programme, JCVI, June 2015
Impact of vaccinating primary and/or secondary school age children on range of indicators, 2014-15 Influenza and vaccination programme, JCVI, June 2015
Cumulative primary care consultations, in primary pilot and non-pilot areas and risk ratios, week 40 2014 until week 14 2015 Disease indicator Agegroup Non-pilot area Primary pilot area OR (95% CI) ILI GP consultations <5 253.1 84/33,192 26.1/100 000 (1/3826) 0.08 (0.01-1.02) 5-10 266.9 (104/38,969) 19.7/100 000 (1/5086) 0.06 (0.01-0.62) 11-16 371.2 (133/35,830) 112.6/100 000 (6/5330) 0.31 (0.10 0.95) 17+ 2,299 (452,461) 219.1/100 000 (143/65,260) 0.41 (0.19 0.86) -Also impact of primary care pilots for swab positivity, EDDS hospitalisations, ICU admissions, less for severe end-points; -Limited/no impact of secondary care pilots
Key findings Key role of routine surveillance in planning and evaluating roll-out of new childhood flu vaccine programme Uptake of LAIV programme in roll-out in school age-pilots achieved ~50% and ~40% in 2-4 year olds with lessons learnt regarding delivery and risk factors for low uptake; Key surveillance systems established to measure impact of pilot programme on influenza in targeted and non-targeted groups; Evidence of effectiveness of LAIV in 2014-15, despite circulation of drifted strain Data demonstrating: - Over two seasons, consistent, decreases in disease incidence and influenza positivity across a range of surveillance schemes in primary age pilot vs nonpilot areas, not secondary school age; On-going surveillance underway as LAIV programme is rolled out to additional age-cohorts (5-6 year olds) and geographically discrete pilots in 2015/16 (primary school age children) Surveillance of LAIV programme, May 2015
Acknowledgements PHE Respiratory Diseases Surveillance Team PHE Immunisation Department PHE Syndromic Surveillance team PHE Virus Reference laboratory: PHE SW Region laboratory RCGP Research and Surveillance Unit Health Protection Scotland Public Health Agency Northern Ireland Public Health Wales ALL THE LAIV PILOT SITES responsible for delivery of programme All contributing USISS and EDSSS NHS acute trusts and RCGP practices Office for National Statistics for the mortality data Annual i-move meeting, July 2015