HOUSE TO HOUSE MOBILIZATION FOR SUCCESSFUL MEASLES SIAs

HOUSE TO HOUSE MOBILIZATION FOR SUCCESSFUL MEASLES SIAs: SITREP AFTER FIVE YEARS IN AFRICA Bob Davis Measles/Health Delegate American Red Cross

WHOM DO WE HAVE TO REACH TO STOP MEASLES TRANSMISSION? § § § Ethno-linguistic minorities and slum dwellers Marginalized and ‘floating’ populations, both urban, peri-urban, and rural In a nutshell: Those who don’t watch CNN, don’t listen to the BBC, and don’t read the New York Times • Solution, for both polio and measles: next slide Photo, Prof. Stanley Foster 1

THE HOUSE TO HOUSE STRATEGY § § For GPEI, OPV SIA policy since 2001 PAHO policy in Latin America for measles SIAs: H 2 H mobilization in campaigns which vaccinated from fixed posts and fixed mobile posts H 2 H mobilization a best practice, UN supported measles SIA, Ethiopia, 2010 Used in Red Cross supported campaigns in 10 African countries: Benin, Burundi, C. A. R. , Kenya, Mali, Mozambique, Namibia, Senegal, Tanzania, Uganda 2

COMMUNITY MONITORING AS PART OF H 2 H MOBILIZATION Wherever possible, line list the 9 - to 59 -month-olds in the week before the campaign, using RC volunteers, then trace defaulters after Day 1 of the campaign to bring them in from home. Example from Tanzania: 3

HOUSE TO HOUSE MOBILIZATION (cont. ) § § § Does H 2 H mobilization produce better results in measles SIAs? Probably: Traditional mass media approaches may miss the least readily accessible populations, even in urban areas. Herd immunity is more easily achievable when we systematically reach populations who lack, e. g. , radio and TV. Data from 8 countries tend to support this hypothesis. 4

ADMINISTRATIVE COVERAGE ESTIMATES IN DISTRICTS WITH AND WITHOUT KENYA RED CROSS HOUSE TO HOUSE CANVASSING, 2009 MEASLES CAMPAIGN, NAIROBI DISTRICTS WITH CANVASSING DISTRICTS WITHOUT CANVASSING KIBERA, 57% DAGORETTI, 64% KASARANI, 83% WESTLANDS, 62% EMBAKASI, 80% CENTRAL, 68% PUMWANI, 58% MAKADARA, 53% UNWEIGHTED AVERAGE, 73 % UNWEIGHTED AVERAGE, 61%

COMPARATIVE CAMPAIGN COVERAGE, NAMPULA PROVINCE, MOZAMBIQUE, 2008 Red Cross Districts § § § Target 413, 005 Vaccinated 401, 604 Coverage 97. 2% Others § § § Target 214, 481 Vaccinated 188, 064 Coverage 87. 7% 6

COMPARATIVE CAMPAIGN COVERAGE, BAMAKO, MALI, 2011 Red Cross Zones § § Target 660, 000 Coverage 93. 6% Others § § Target 210, 317 Coverage 87. 8% Do these percentages make a difference? Yes, when herd immunity starts at > 90% 7

RESULTS OF SITE INTERVIEWS WITH CAREGIVERS, TWO RURAL PROVINCES COVERED BY BURUNDI RC, 2012 SIA HOUSE VISITS RADIO CHURCHES ALL RED CROSS POPULATION OTHER VOLUNTEERS Ruyigi 23 19 7 15 505710 562 Gitega 32 31 23 6 920136 1022 Total 55 50 30 21 1425846 1584

ADMIN COVERAGE ESTIMATES, BURUNDI’S 2012 SIA, NATIONWIDE AND IN THE FOUR REGIONS WITH H 2 H MOBILIZATION NATIONWIDE AVERAGE GITEGA MAKAMBA MUYINGA RUYIGI 103% 104% 116% 106% 115% AVE + 13 AVE + 12 AVE + 8 IN H 2 H REGIONS, BASED ON WEIGHTED AVERAGE

SOURCES OF INFORMATION CITED BY MOTHERS, ABOMEY, BENIN, SEPTEMBER 2011: 1/5 OF ALL VOLUNTEER MENTIONS FROM THE 4 PERCENT OF VOLUNTEERS WITH MEGAPHONES! SOURCE OF INFO Public Criers Red Cross Volunteers CUMULATIVE FIGURES Monday, 53 mentions by mothers and other caregivers 49 House to house Mentions volunteers, 28 House to house volunteers with megaphones, 15 Volunteers at fixed posts, 6 Radio 37 mentions Wednesday, 65 mentions 30 Friday, 63 mentions 104 20 10 37 5 87 110

H 2 H EVALUATED AS BEST PRACTICE, BENIN CAMPAIGN, 2011 CRITERIA Effectiveness Efficiency Relevance Feasibility Reproducibility Participation of the partners ANALYSIS BY CRITERION - Strong mobilization of the parents of children targeted at the time of the passage of the teams in the villages CONCLUSION Satisfactory - Better knowledge of the populations of the campaign schedule, of the strategy of progression of the teams and of the campaign’s target disease - Reduction of the number of people reluctant to vaccinate Satisfactory - Improvement of the vaccine coverage in the localities benefiting from the support of mobilizers - Facilitate the acceptance of vaccination by the populations in the urban zones - Valid for all the vaccination campaigns even the JNV polio - Implementation in the country’s 3 largest cities and in 12 other communes of the country - Activities mainly undertaken by the volunteers of the Red Cross, the Community and members of the Church of Jesus Christ of Latter Day Saints Satisfactory

WHAT THE DATA SHOWS § § § Large chunks of the urban population, and even of many rural populations, are accessible through mass media approaches. However, we are unlikely to achieve herd immunity in campaigns without house to house mobilization. In addition to campaigns, intercampaign house canvassing, 1 ½ years after the SIA, is a promising possibility to reduce the risk of outbreaks between campaigns.

METHODS FOR EVALUATING COMPARATIVE PERFORMANCE OF H 2 H AND CONVENTIONAL APPROACHES BEST OF ALL POSSIBLE WORLDS PLANET EARTH § 30 cluster surveys, intervention and nonintervention areas §Yes; §Admin §Yes, §Spot coverage estimates surveys at vaccination sites to ascertain mothers’ source of info. Cheap and easy; permits assessment of comparative role of H 2 H and other info sources so far, only in mainland Tanzania, with results ranging from 72 to 100 percent in areas with house visiting. but check your denominators. With data retention and/or recording errors, check your numerators as well. §Spot surveys at vaccination sites to ascertain mothers’ source of info. Cheap and easy; permits assessment of comparative role of H 2 Hand other info sources §Compare to IM data where available.

WHY WE NEED MORE SPOT SURVEYS § § § Cluster coverage surveys, with more scientific rigor, are not always done, and rarely permit comparison between areas with and without house visiting. Admin coverage data are based on high side population figures (Eritrea) or low side population figures (Uganda). >>100% coverage = high degree of flakiness; true of ½ of all districts in Uganda’s 2012 measles SIA. Data retention by health workers (Senegal, Kampala) makes it impossible to calculate SIA admin coverage.

COSTING OF HOUSE TO HOUSE MOBILIZATION § § Vitamin m, the indispensable micronutrient Single partner funding by American Red Cross is not a viable option for H 2 H mobilization to go to scale.

ADDED COST PER BENEFICIARY, H 2 H STRATEGY, FIVE MOST RECENT NATIONAL CAMPAIGNS, AVERAGE $0. 32. UNIT COSTS VARY. TANZANIA FINANCED DAR ES SALAAM, WITH LOW UNIT COSTS. NAMIBIA FINANCED RURAL AREAS. BENIN, 2011 BURUNDI, 2012 NAMIBIA, 2012 TANZANIA, 2011 UGANDA, 2012 BUDGET FOR HOUSE VISITING USD 99, 233 USD 154, 546 USD 95, 759 USD 272, 957 (exclusive of UNICEF funding) USD 272, 957 BENEFICIARIES 322, 572 473, 890 166, 750 1, 687, 000 1, 300, 000 COST PER BENEFICIARY USD 0. 31 USD 0. 33 USD 0. 57 USD 0. 16 USD 0. 21 VOLUNTEERS WORKING ON CAMPAIGN 685 3100 1450 2679 2911

CONCLUSIONS § § § In areas with H 2 H mobilization, measles SIA cost per child rises from ~$1 to ~$1. 32 or more. We need to be selective in choosing areas for H 2 H. Selection criteria used by American Red Cross and, in some countries, UNICEF: § Underserved populations, especially slums § Areas with low coverage and/or high cases based on case based surveillance § Geographically remote areas

CONCLUSIONS (CONT. ) § § Some countries (Kenya, Burkina Faso) have widespread viral seeding from town to country. There, it may be necessary there to target whole cities, not just slums. In some settings, the dollar goes farther in urban H 2 H mobilization (Tanzania vs. Namibia). Where funding is short, first priority goes to underserved urban and periurban areas.

THE MAGIC FORMULA § § There is no magic formula for targeting areas to cover through H 2 H. Where viral seeding is well documented, target the source of the viral seeding. Where coverage data are reliable, target areas with low coverage. Where case based surveillance is good, target areas with cases (Burundi: 4/17 regions were home to 29/30 confirmed measles cases).

URBAN PARTICULARITIES § § § Weekend SIA start is preferable; no traffic jams to tie up logistics; pulpit announcements on Fridays and Sundays Multilingual house visitors and vaccinators are needed – Dakar, Nairobi, for example. Mapping of neighborhoods with many migrants and floating populations, for special emphasis by gov’t, RC and other partners

H 2 H CANVASSING FOR ROUTINE IMMUNIZATION? § § § American Red Cross & partners need to consult on how best to apply lessons learned from SIAs to routine immunization. A network of volunteers already exists to sensitize the community. Possible modalities: birth registration and follow-up; periodic village canvasses; linkages to health facilities for defaulter follow-up

SO WHERE IS H 2 H GOING? § § § Depends on decisions made by the MRI, as by the GPEI in 2001, when the polio initiative opted for H 2 H OPV SIAs. If H 2 H mobilization goes global with measles, as with polio, then more resources and partners will be needed. You can’t go global on a shoestring, and you can’t do it with 1 or 2 partners, as at present. Decision whether to go global with H 2 H should predate any WHA resolution. No 1988 -2001 gap as with GPEI, SVP!

THANK YOU/ASANTE SANA/ AMESEGNALEHU/SIYABONGA/ MERCI/OBRIGADO/MUCHAS GRACIAS § § § § § § § § AMESEGNALEHU ANSAKUSU ARIGATO ASANTE SANA BAIE DANKIE BARKA BEDANKT BINOBONDI DEUS PAGARAPUSUNKI DHANJABHAT DJERE DIEUF DIOKO NDIAL DYARAAMA/FOOFU DZIĘKUJEMY DUNABAT EFHARISTO ESE GELETOMA GRAMACI (PROVENÇAL) HAKHENTSA HARTELIJK DANK! HSEHSE HVALA INITCHIE INKOSI KAMSAHAMNIDA KANIMAMBO KEAITUMETSE KEA LEBOHA KHOBKHUN MAG KIITOS PALION (FINNISH) § § § § § § § KÖSZÖNÖN LONGONIA MINGI LOSAKA MADLOBT (DIDI MADLOVA) MAHADSANIT MANAM MAZVITA MEDASE MERCI MILLE GRAZIE MIRISE MISAOTRA MUCHAS GRACIAS MURAKOZE CYANE MWASHUMA MPUSIYA NAGODE NAMVERA NAPANDULA NASOM NATONDI NDAU YA NDUNA NDA BOKA NDIYABULELA NGASSAKIDILA § § § § § § § OKUHEPA PANDU RE A LEBOGA RO LIVHUWA SALAMAT SHNORHAKALOUTYOUN SINGUILA MINGUI SHUKRAN SIYABONGA SOSONGO SPASIBA TATENDA TERIMAKASIH TEŞEKKŰRLER TODAH RABAH TSE ZU TIN BA DEH TVASAKIOILA TWATASHA TWATOTELA VIELEN DANK VILLMOLS MERCI WAKOOZECHANE WEBAALE NNYO YABONGA YAQENYILEY ZIKOMO

Anne Ray Charitable Trust