Evaluation of a poliomyelitis immunization campaign in Madras city.

Bulletin of the World Health Organization, 64 (6): 861-865 (1986) © World Health Organization 1986 Evaluation of a poliomyelitis immunization campaign in Madras city* V. BALRAJ' & T. J. JOHN2 A city-wide oral polio vaccine (OPV) immunization campaign, aimed at the 3-36- month age group, was carried out in Madras city during the first 3 months of 1985 as a joint effort by government and voluntary agencies. Wide publicity was given to the campaign. An evaluation of the results was conducted using the 30-cluster sample survey method. The campaign succeeded in rapidly increasing the vaccine coverage to 94%, 88% and 72% for, respectively, the first, second, and third doses of OPV. Although a variety of publicity methods were used, nearly half the parents were reached through word-of- mouth information. It is proposed that similar campaigns for immunization should be carried out every year till the health care system develops the capability ofattaining a high level ofpopulation coverage. As immunization coverage by the health care system is relatively low in most parts of India (1), the incidence of preventable diseases like paralytic poliomyelitis remains high (2). In such a situation the strategy of 'annual pulse immunization', in which governmental and voluntary agencies together con- duct annual immunization campaigns in geographi- cally defined units of the population (using a broad range of the target age group of eligible children), has been recommended for disease control until the primary health care system becomes capable of achieving near-universal immunization coverage (3). This strategy to control poliomyelitis was adopted by the city of Madras in the State of Tamil Nadu. A pilot project for the campaign was conducted in two areas of the city on a population of more than 20 000 and was evaluated by the 30-cluster sample survey method (4). A city-wide campaign using oral polio vaccine was planned and conducted by a committee (composed of representatives from the city corpor- ation and state health department, the Integrated Child Development Scheme,' voluntary organiz- ations, and private companies) under the guidance of Impact India Foundation. b * From the Department of Virology, Christian Medical College and Hospital, Vellore 632 004, Tamil Nadu, India. ' Lecturer in Community Health, Department of Virology. 2 Professor and Head, Department of Virology. Requests for reprints should be sent to this author. a The ICDS is a national programme of nutrition supplemen- tation and immunization of selected population groups, e.g., city slum dwellers and the rural poor. b Impact is an international initiative against avoidable disablement and is jointly sponsored by the UNDP, WHO and UNICEF. Impact India Foundation was inaugurated in October 1983 and is a registered public trust. Wide publicity was given to the campaign through the mass media (television, radio, newspapers) for several days. In addition, in many areas the city corporation's health workers, 'balwadi ayahs' (women workers in day-care centres for pre-school children), and volunteers visited households and informed parents by word of mouth. Wall posters, descriptive handbills, and slides in cinemas were also used to disseminate information. In many areas announcements were made through mobile loud- speakers. The target age group for immunization was 3 to 36 months. The centrally stored oral polio vaccine (OPV) was moved to 450 immunization booths via two dis- tribution centres early in the morning on the days of the campaign (i.e., on 4 days each in January, February and March 1985). The cold chain remained intact until the vaccine was taken out of the boxes immediately before administration by health workers and volunteers. A total of 240 100 doses of OPV were utilized in the campaign. Immunization cards were distributed in many but not all booths. After the campaign was over, an independent evaluation was conducted and the results are reported he-re. EVALUATION METHODOLOGY The objectives of the evaluation were to determine the vaccine coverage, the relative success of various publicity methods, and the reasons for non-response. A WHO-recommended cluster-sample technique was used to choose 30 among the total of 150 adminis- trative divisions of Madras city (4). In every chosen 4733 -861- V. BALRAJ & T. J. JOHN Table 1. Oral poliovirus vaccine (OPV) coverage according to the age of children and number of doses No. of children who received OPV: Age in No. of months respondents 1 dose 2 doses 3 doses 4 doses 5 doses 6 doses 7 doses 6-12 308 281(91.2)° 256(83.1) 200(64.9) 54(17.5) 27(8.8) 1(0.3) 0 13-24 301 291(96.7) 267(88.7) 226(75.1) 97(32.2) 58(19.3) 11(3.7) 2(0.66) 25-36 308 287(93.2) 280(90.9) 238(77.3) 103(33.4) 62(20.1) 18(5.8) 3(0.97) Total 917 859(93.7) 803(87.6) 664(72.4) 254(27.7) 147(16.0) 30(3.3) 5(0.5) 0 Figures in parentheses are percentages. division, a sample survey was conducted until 10 children each from the age groups of 6-12 months, 13-24 months, and 25-36 months were identified, following the procedures recommended by Hender- son & Sundaresan (4), and their parents interviewed. The information obtained at the interview included how they came to hear about the campaign, history of previous OPV administration, receipt of OPV or the reasons for not receiving one or more doses of OPV at the campaign, and whether an immunization card was issued. The survey was conducted by 5 trained interviewers in April 1985, over a period of 5 days; a supervisor checked and verified a 10% random sample of the answers. Seven samples of OPV were drawn from the cold storage and transported in ice to the virology laboratory in Vellore where their potency was assayed by infectivity titration in primary monkey kidney cell culture. Serial tenfold dilutions of the samples were inoculated in 6 tubes of cell culture, cytopathic changes were observed till the seventh day, and the vaccine virus titres were then calculated using standard formulae and expressed as the median tissue culture infectious dose (TCIDso). RESULTS A total of 917 children were identified in the 30- cluster samples in the three age groups and their parents were interviewed. The vaccine coverage according to the age of the children is summarized in Table 1. The levels of coverage were generally higher in the older age groups than among infants. The Table also shows the decline in coverage as the number of doses increased; the gradual decline from the first to the third dose became steep thereafter (Fig. 1). 100 0-o-O Prior to campaign 90 ~ . - After campaign Coverage achieved 80 _ by campaign 70 460- 0 0 - O - 840 20 1 2 3 4 5 6 7 NUMBER OF DOSES OF OPV Fig. 1. Percentage immunization coverage according to the number of doses of oral polio vaccine (OPV) received before and during the Madras campaign in 1985. The campaign accounted for 27% of all the oral polio vaccines the children had ever received. In children aged 6-12, 13-24 and 25-36 months the campaign delivered 19.4%, 31.4% and 24.6%, respectively, of all the OPV received prior to the evaluation; among the parents of children in these age groups, 13.9%, 11.5% and 9.8% respectively had 862 POLIOMYELITIS IMMUNIZATION CAMPAIGN Table 2. Percentage frequency of various channels by which the parents came to know about the immunization campaign Word of mouth: from health workers or volunteers 30.8 Word of mouth: from 'balwadi ayahs' or others 16.6 Television 10.2 Radio 7.3 Newspapers 4.0 Mobile loudspeakers 2.1 Handbills, posters or slides in cinemas 1.0 No information received 28.0 retained their immunization cards. Based on the immunization data, the estimated child population, and a vaccine wastage of up to 10%, the total number of doses of OPV used during the campaign was calculated to be between 197 722 and 241 659 (95% confidence interval). As the campaign workers had distributed 240 100 doses of OPV (including wastage), there is close agreement with the information collected through the 30-cluster sample survey. Table 2 indicates the responses given by the parents of 874 children regarding how they came to hear about the campaign. Health workers, volunteers, the 'balwadi ayahs', neighbours and relatives were the source of word-of-mouth information for 47.4% of the parents while 28% denied receiving any infor- Table 3. Percentage distribution of respondents according to the reasons given for not taking their children to one or more sessions of the immunization campaign No information received, or received too late Considered that 3 doses of vaccine given before the campaign were adequate Not in the city at the time of the campaign Child was sick and could not be taken for immunization Preferred to be immunized by private practitioners Was not informed about all 3 sessions of the campaign Was informed by health workers that 3 doses of vaccine given before the campaign were adequate Expected the immunization would take place at home Immunization booth was located too far away Various reasons: advised against immunization, forgot, or did not like the mass campaign Immunization booth was not open on the scheduled day Afraid to immunize the child Could give no definite reason 38.1 22.2 7.7 7.4 4.6 3.4 1.4 1.3 0.9 0.9 0.5 0.4 11.2 mation; the mass media all together had reached only 24.6% of the parents. The reasons given by the parents of 789 children who did not attend one or more of the three im- munization sessions are presented in Table 3. As many as 38% did not know about the campaign at all, or about any one of the immunization sessions; another 22% reported that their children had already received 3 doses of OPV from various sources prior to the campaign, and believed that no further doses were required. About 11I% could give no definite reasons. The remaining 29% gave a variety of reasons. Of the 7 samples of OPV tested for potency, 6 had infectivity titres of 105 74 to 106 06(satisfactory range) and one had a titre of 105.60 (marginally low) TCID5o per dose. DISCUSSION The population of Madras city (area, 174 kmi2) was estimated to be 3 698 377, based on the 1981 census and the annual growth rate between 1971 and 1981. Although many institutions and practitioners offer immunization either free of charge or on payment, there is no system available for assessing the coverage or for locating unimmunized infants. As the paediatric hospitals in the city have not registered any decline in the numbers of children with acute paralytic poliomyelitis in recent years, it would appear that polio immunization coverage is inad- equate (5). The annual pulse immunization strategy is designed to rapidly increase vaccine coverage and thereby reduce the incidence of disease (6). In Madras city, in the first application of this strategy, only oral polio vaccine was used; in subsequent years immunization against other diseases will be included until all vaccines in the Expanded Programme on Im- munization (EPI) are offered during the annual pulse campaigns. The use of a single vaccine is justified in the first year so that health staff, volunteers and the community could get used to the concept. Oral polio vaccine was found to be ideal on account of the simplicity of its administration and the need to combat the high incidence of poliomyelitis. Our results show that the Madras campaign was successful because it raised the immunization coverage very rapidly, from pre-campaign rates for the first, second and third doses of OPV of 66%, 62% and 52%, respectively, to 94%, 88% and 72%. Vaccine coverages for the fourth and subsequent doses were very low. This was because some of the campaign organizers and the parents of 22% of the children had understood that 3 doses of OPV were sufficient for every child. The Indian Academy of 863 V. BALRAJ & T. J. JOHN Pediatrics recommends 5 doses, WHO recommends 4 doses, and the Indian EPI recommends 3 doses of OPV for primary immunization plus one or two more doses for reinforcement (7-9). The immunogenic efficacy of 3 doses of OPV in India is in the range of only 75-80% while that of 5 doses is in the range of 90-95% (10). Thus, in poliomyelitis-endemic countries, children should receive 5-7 doses of OPV. Fortunately, this vaccine costs less than US$ 0.03 per dose. We therefore recommend a liberal policy in pulse campaigns, offering 3 doses of OPV to all children of the target age group, regardless of the number of doses that may have been given previously. Recently in Brazil annual nationwide pulse immunizations were carried out, 2 doses of OPV being offered to all children below 5 years, irrespective of the number of previous doses (11); in this country, 3 doses for primary immunization plus 2 doses annually for 5 years could add up to 13 doses per child (C. de Quadros, personal com- munication, 1985). The success of immunization campaigns depends largely on three factors: information announcing the campaign reaching the parent, their compliance, and an effective vaccine delivery system. In the Madras campaign, OPV was efficiently delivered through the immunization booths and the vaccine's potency was safeguarded by proper cold storage and handling. However, 28/o of parents did not hear about the campaign at all, and another 10% had not heard about one or two sessions despite the use of a variety of publicity channels. As nearly half of all the parents came to know about the campaign by word of mouth, this appears to be the most effective channel of communication. The mass media were not as effective as the messages from person to person although it is possible that some persons may have obtained their information through the mass media. Any immunization programme that is to continue will have to be based on accurate census data and periodic updates on new births and on the immunization status of infants; in addition, the programme should keep parents informed about facilities and the need for immunizing their children, thereby gaining their compliance, and should maintain an efficient vaccine delivery system. Until these requirements can be met, immunization cam- paigns, like the one described here, offer an alternative strategy so that high immunization coverage and reduced disease incidence can be achieved. ACKNOWLEDGEMENTS This evaluation was conducted on behalf of the Polio Foundation of India and was supported by the Impact India Foundation. We thank the campaign coordination committee, the Madras city corporation officials, and the Secretary and the Minister for Health and Family Welfare of Tamil Nadu State for their help and encouragement. We also thank R. Albert, D. Michavel, B. Paranthaman, G. Stalin and K. Lakshmipathy for interviewing the parents. RESUME tVALUATION D'UNE CAMPAGNE DE VACCINATION CONTRE LA POLIOMYELITE A MADRAS Une campagne de vaccination contre la poliomyelite par administration de trois doses de vaccin buccal a et entre- prise a Madras (population estimee: 3 698 377 habitants) en janvier, fevrier et mars 1985. Cette campagne, qui avait ete precedee par une etude pilote, 6tait organisee par un comite representant divers organismes gouvernementaux et b6n& voles sous la direction de l'Impact India Foundation. Elle avait pour cible les enfants ages de 3 a 36 mois. Le present rapport resume les resultats d'une 6valuation de la campagne selon la methode de sondage par 30 grappes recommand&e par I'OMS. La couverture vaccinale, qui etait de 66%, 62% et 520/ pour les premiere, deuxieme et troisieme doses de vaccin avant la campagne, est passee a 94%, 88% et 72% respectivement apres celle-ci. Vingt-sept pour cent de toutes les doses de vaccin buccal reques par les enfants depuis leur naissance ont e administrees dans le cadre de la campagne au cours de laquelle 241 000 doses ont ete effectivement utilisees, alors que les resultats de l'evaluation donnent un chiffre compris entre 197 722 et 241 659 doses (intervalle de confiance a 95%). Parmi les parents, 47% ont declare qu'ils avaient ete informes verbale- ment de la campagne, 25% ont appris son existence par les medias et 28% n'ont requ aucune information. Parmi les parents qui n'ont pas amene leurs enfants a une ou plusieurs 864 POLIOMYELITIS IMMUNIZATION CAMPAIGN 865 des seances de vaccination, 38% n'avaient pas et suffisam- ment inform6s et 22% croyaient que leurs enfants n'avaient plus besoin d'etre vaccines puisqu'ils avaient deja recu 3 doses precedemment. Notre evaluation montre qu'une campagne de vaccination bien conduite peut accroitre rapidement la couverture de la population; les ressources de plusieurs organismes peuvent etre mises en commun avec profit et l'information verbale est le moyen le plus efficace d'annoncer la campagne. En attendant que l'infrastructure des services de sante permette d'entreprendre un programme permanent et efficace d'immunisation, des campagnes annuelles contribueront i atteindre un niveau eleve de couverture, a condition que 3 doses de vaccin buccal soient offertes au groupe cible, independamment du nombre de doses reques precedemment. En temps utile, ces campagnes pourraient egalement comprendre d'autres vaccinations, jusqu'a ce que soient incluses les six maladies visees par le programme elargi de vaccination. REFERENCES 1. BHARGAVA, I. & S. SOHKEY, J. Immunization pro- gramme in India. Indian pediatrics, 22: 95-104 (1985). 2. BASU, R. N. Magnitude of the problem of poliomyelitis in India. Indian pediatrics, 18: 507-511 (1981). 3. JOHN, T. J. & STEINHOFF, M. C. Appropriate strategy for immunization of children in India. 3. Community- based annual pulse (cluster) immunisation. Indian journal ofpediatrics, 48: 677-683 (1981). 4. HENDERSON, R. H. & SUNDARESAN, T. Cluster sampling to assess immunization coverage: a review of experience with a simplified sampling method. Bulletin of the World Health Organization, 60: 253-260 (1982). 5. SUNDARAVALLI, N. ET AL. Spurt in poliomyelitis in Madras. Indian pediatrics, 18: 539-543 (1981). 6. JOHN, T. J. ET AL. Control of poliomyelitis by pulse immunization in Vellore, India. British medical journal, 286: 31-32 (1983). 7. INDIAN ACADEMY OF PEDIATRICS. Workshop in Immunisation. Academy today, 1: 1-6 (1983). 8. KEJA, K. & HENDERSON, R. H. Expanded Programme on Immunization: the continuing role of the European Region. WHO Chronicle, 39: 92-94 (1985). 9. BASU, R. N. & SOHKEY, J. The expanded programme on immunization. A review. New Delhi, Directorate- General of Health Services, 1982. 10. JOHN, T. J. Antibody response of infants in the tropics to five doses of oral poliovaccine. British medical journal, 1: 811-812 (1976). 11. RISI, J. B. The control of poliomyelitis in Brazil. Reviews of infectious diseases (Suppl.2): S400-S403 (1984).