A programme of multiple-antigen childhood immunization in Yaoundé, Cameroon: first-year evaluation, 1975-1976

Bulletin of the World Health Organization, 55 (5): 633-642 (1977) A programme ofmultiple-antigen childhood immunization in Yaoundd, Cameroon: first-year evaluation, 1975-1976 BERNARD GUYER 1 & SIMON ATANGANA2 The Yaounde multiple-antigen childhood immunization programme began in November 1975, making it one of the first expanded programmes on immunization operational in Africa. During the first 9 months, more than 22 000 children were immunized against poliomyelitis, measles, tuberculosis, smallpox, whooping cough, tetanus, and diphtheria. Evaluation of the programme showed the following rates of immunization coverage in the target population- 30%. for DPT (one dose or more), 27 % for poliomyelitis (one dose or more), 27 % for BCG, 33 % for measles, and 20 % for smallpox. Eighty per cent ofchildren received the correct vaccinesfor their age and vaccination status. Seroconversion to measles vaccine was 89 % in those over 12 months of age but only 50 % in those between 6 and 11 months ofage. The major factor in low immunization coverage was felt to be inadequate publicity. The cost of the programme was estimated to be US $10 920. The cost of immunizing a childcompletely was estimatedat US $1.90. Some logisticproblems encountered during this initial year of operation are discussed. The Twenty-seventh World Health Assembly in 1974 resolved that the World Health Organization should stimulate a major effort to expand childhood immunization programmes, especially in developing countries (1). In another resolution in 1976, the Twenty-ninth World Health Assembly noted that many problems remained to be solved in these expanded immunization programmes, including de- termination of the best immunization strategies, assurance of adequate vaccination coverage and efficacy, and maintenance of the " cold chain ", the system that ensures vaccines are kept cold between storage and use (2). The Yaounde multiple-antigen childhood immu- nization programme began in November 1975, mak- ing it one of the first " expanded " programmes on immunization operational in Africa. After the first 9 months of operation, from November 1975 to July 1 Medical Epidemiologist, Bureau of Smallpox Eradica- tion, Center for Disease Control, Atlanta, GA 30333, USA assigned to the Organisation de Coordination pour la Lutte contre les End6mies en Afrique Centrale (OCEAC), B.P. 288, Yaound6, United Republic of Cameroon. Present address: Children's Hospital Medical Center, 300 Longwood Avenue, Boston, MA 02115, USA. ' Sous-Directeur de la M6decine pr6ventive et de l'Hy- giene publique, Ministere de la Sant6 publique, Yaound6, United Republic of Cameroon. 1976, the Yaound6 programme was evaluated in order to: (a) assess vaccination coverage and efficacy; (b) control the quality of operations, including the selection of patients and vaccine wastage; (c) identify the characteristics of the population being reached by the programme publicity; (d) determine the ability of the programme to prevent diseases, especially measles; and (e) estimate the cost of the operation. The results of this evaluation have been used to modify the Yaounde programme and will be valu- able to other urban African centres planning an expanded immunization programme. BACKGROUND Yaound6 is the capital and administrative centre of the United Republic of Cameroon. The popula- tion has grown from 166 000 at the time of the 1968 census to an estimated 250 000 in 1975. The birthrate is estimated at 45 per 1000 or about 1000 births per month. The population under 5 years of age is estimated at 20.7% or 51 750 children. The popula- tion is ethnically mixed. Migration into the city has been a major factor in its growth, and the rate of movement between the city and the countryside remains high. 3648 -633 B. GUYER & S. ATANGANA Health care in Yaounde is delivered through a group of five government dispensaries, three civil service and military facilities, a large child welfare clinic of the Service de la Protection Maternelle et Infantile (PMI Centrale), four smaller political party (OFUNC)-sponsored maternal and child health cen- tres, four private mission dispensaries, a large central hospital, two community clinics affiliated with the University Centre for Health Sciences, and numer- ous traditional healers. The Yaound6 immunization programme was de- veloped by the Sous-Direction de la Medecine pre- ventive et de l'Hygiene publique of the Ministry of Health with the technical assistance of the Organisa- tion de Coordination pour la Lutte contre les Ende- mies en Afrique Centrale (OCEAC). The vaccination team was composed of Ministry of Health employees. From November 1975 to July 1976, the seven- member Yaounde vaccination team held a total of 96 vaccination sessions at 11 dispensaries and child health centres in Yaounde. The team visited each centre on a fixed monthly basis, e.g., the first Monday of the month at centre A. The team was responsible for all technical aspects of vaccine deliv- ery, and maintaining the cold chain presented no major problems. The vaccines were transported in commercial cold boxes using frozen " cold-packs ". The following age schedule of immunization was used for the selection of patients. The chief of the team and two other team members were responsible for selection. Age Immunizations 6-8 weeks DPT and poliomyelitis No. 1 + BCG (marker) 1 month later DPT and poliomyelitis No. 2 1 month later DPT and poliomyelitis No. 3 9-23 months Measles and smallpox (marker) Vaccinations for older children were decided indi- vidually, based on vaccination and disease histories and vaccination scars. Children over 4 years of age received tetanus vaccine rather than DPT. The following commercial vaccines were used: DPT (diphtheria-pertussis-tetanus): 50-dose vials for intramuscular injection. Poliomyelitis (trivalent oral poliomyelitis vaccine, Sabin): 100-dose vials. BCG (lyophilized): 50-dose vials for intradermal injection. Measles (live attenuated, Schwarz strain): 10- and 50-dose vials for delivery by jet injector (Ped-O-Jet). Smallpox (lyophilized): 100-dose vials for admin- istration by bifurcated needle. Tetanus (60 Lf, absorbed): 50-dose vials. The target group for the programme comprised all Yaounde children under 2 years of age. However, older children and nonresidents were always vac- cinated. Programme publicity was effected mainly by means of the personnel of the vaccination centres, although some radio announcements were made. The team kept records of the total number of vaccine doses given. During the 9-month period, the team was the major source of childhood immunization in Yaounde, although some vaccines were available for purchase in local pharmacies. All immunizations by the team were done free of charge. METHODS Four evaluation methods were used: (1) Throughout the 9-month period, team mem- bers recorded the vaccinations given. (2) During August 1976, visits were made by an assessment team to randomly selected households in Yaounde to assess vaccination coverage. (3) During July 1976, an independent assessment team visited 10 vaccination sessions to evaluate team operations and the population presenting themselves for immunization. (4) Continuous active surveillance of measles was maintained. Vaccinations At all vaccination sessions, the selection personnel kept records of the number of new vaccinees, return- ing vaccinees, number of doses of each vaccine given, and (for DPT and poliomyelitis vaccines) whether this was the first, second, or third dose. Evaluation of vaccination coverage A two-stage cluster sample, based on the 1969 census figures for Yaounde, was used to select children included in the study; this was the method described by Henderson (3), and previously used in Yaounde by McBean (4). Thirty neighbourhoods were visited by an assessment team and in each a sample of houses was randomly selected. The first 32 children aged between 8 weeks and 35 months encountered were selected. Identifying information was taken on each child and all vaccination records were examined. The child 634 CHILDHOOD IMMUNIZATION PROGRAMME was examined for evidence of scars from previous smallpox and BCG vaccinations. Only documented records were accepted as evidence of immunization. Evaluation ofprogramme operations including selec- tion An independent assessment team was trained. They visited 10 vaccination sessions and interviewed every tenth mother leaving the vaccination centre. They recorded the age and vaccination status of the child and the vaccinations actually received during that particular session on a standard form. Later, the correctness of selection was determined by the OCEAC epidemiologist. Selection criteria. Selection of immunizations for each child was based on age and vaccination status. Correct selection decisions were as follows: (1) Between 1 and 8 months of age, a child should have started his DPT and poliomyelitis series, with BCG vaccine given at the time of the first dose. The series of three DPT-poliomyelitis vaccine doses should have been completed during this period. (2) Between 9 and 23 months of age, measles vaccine (together with smallpox vaccine) should have been given, even if this necessitated interrupting the DPT-poliomyelitis series. (At one centre, the pediatrician in charge insisted on beginning measles vaccination at 6 months of age.) (3) After the age of 24 months, selection was based on individual vaccination requirements but measles vaccine should not have been administered. After 4 years of age, tetanus vaccine was substituted for DPT. Measles immunity and seroconversion. From each child who received measles vaccine, finger-prick blood specimens were collected on to filter paper as described by Mathews (5). One month later, an effort was made to revisit the homes of these children to obtain a second blood specimen and to assess scar formation following smallpox or BCG vaccination. Measles haemagglutination inhibition (HI) antibody assays were performed at the Center for Disease Control (CDC), Atlanta, GA, USA by the method reported by Hierholzer et al. (6, 7), using antigen prepared by the Norrby method (8). The titre of the vaccine in representative measles vaccine vials was measured quantitatively to deter- mine the attenuated virus content at the beginning of the programme. During four sessions, partially used diluted vials were collected and sent for titration to the CDC. In addition, at each centre, the number of measles vaccine vials diluted for use and the total number of children immunized was counted. Characteristics of parents seeking vaccinations. A questionnaire was delivered to mothers to determine how they had learned of the programme, the distance travelled to the centre, the reason for choosing that particular centre, and the means of transport used. Measles surveillance Since 1975, cases of measles have been reported by 11 dispensaries around Yaounde to the OCEAC epidemiologist on a weekly basis by name, age, sex, place of residence, and vaccination status. Before 1975, reporting from the PMI Centrale, the largest single child health facility in Yaounde, was moni- tored at OCEAC. As measles is a common, easily recognized disease, clinical diagnoses are accepted without laboratory confirmation. RESULTS Vaccinations given During the 9-month period between November 1975 and July 1976, the Yaounde vaccination team held 96 vaccination sessions at the 11 regular vac- cination centres. Table 1 shows the monthly tally of vaccinations given. An average of 1270 new children were registered on the programme each month, and the number of children returning for vaccination increased dramatically over the year. Overall, 51 522 vaccine doses (all vaccines) were given during 22 736 child-visits, giving an average of 2.3 vaccine doses per child-visit. An average of 4.5 doses of all vaccines were given to each child registered on the pro- gramme over the first 9 months. (A completely vaccinated child would have received a total of 9 doses of vaccine.) Only 5890 measles vaccinations were given. How- ever, during the month of February, a special publi- city effort increased measles vaccinations for that month to more than 2000. Vaccination coverage Complete vaccination histories were obtained from a random sample of 596 Yaounde children (Table 2). Overall, 61 % of these had never received any immu- nization. Of the target group aged 1-8 months, only 22.6% and 21.3 % had received one dose or more of DPT and poliomyelitis vaccine, respectively. This level of vaccination coverage is supported by the finding of BCG vaccination scars in 20.9% of the age group (BCG vaccine was given to children as a marker of their first DPT-poliomyelitis immuniza- 635 B. GUYER & S. ATANGANA Table 1. Monthly vaccinations given by the Yaound6 vaccination team between November 1975 and July 1976 DPT dose Poliomyelitis dose Month New Returning BCG Ploy tsds Measles Smallpox Tetanusvaccinees vaccinees 1st 2nd 3rd 1st 2nd 3rd November 1975 1112 131 663 575 49 82 642 36 79 389 449 December 1164 259 745 355 94 22 741 190 69 357 419 - January 1976 1306 382 603 701 241 75 668 263 119 524 605 39 February 2867 2398 1695 1732 456 265 1971 479 270 2071 1730 525 March 1282 1212 829 832 509 261 903 524 275 615 754 224 April 1129 1703 900 922 648 409 1036 642 424 516 634 306 May 945 1622 839 819 632 436 879 634 463 510 567 268 June 991 1961 978 818 735 598 1018 800 588 470 619 386 July 664 1608 715 470 526 492 620 632 529 438 543 350 Total 11 460 11 276 7967 7224 3890 2640 8478 4200 2816 5890 6320 2097 Table 2. Random sample survey of vaccination coverage, Yaound6, August 1976 Vaccination coverage (%) Age group No. of children (months) studied Never DPT: 1 dose Poliomyelitis: BCG: vaccina- Smallpox: vac- Measles vaccinated or more 1 dose or more tion scar cination scar 1-8 230 76.5 22.6 21.3 20.9 0.4 0.4 9-23 239 45.5 37.2 33.9 33.1 25.9 32.6 24-35 127 30.0 30.7 25.2 26.8 41.7 48.0 Total 596 60.6 30.2 27.2 27.0 19.6 37.9 a a Coverage for measles vaccination in those aged 9-35 months only. Table 3. Random sample survey of DPT vaccination coverage by number of doses, Yaound6, August 1976 No. of DPT vaccination coverage (%) (months) children at least at least 3 doses(month)studied 1 dose 2 doses or more 1-8 230 22.6 13.9 5.6 9-23 239 37.2 22.8 13.4 24-35 127 30.7 21.3 15.0 Total 596 30.2 20.0 11.9 tion). One-third of the target 9- to 23-month-old children had received measles vaccine. Only one- third of children beginning their DPT-poliomyelitis immunizations completed the 3-dose series during these 9 months (Table 3). In Table 4, the actual number of immunizations given by the Yaounde team is compared to a calculation of the projected number of children immunized, derived by applying the percentage cov- erage figures to the estimated child population of Yaounde. Evaluation ofprogramme operations including selec- tion Selection. Of 196 child records examined, selection decisions were correctly made in 156 cases (80 %). Most selection errors, that is, errors in deciding the correct vaccinations for a particular child at the session, involved the use of measles vaccine or the failure to associate measles and smallpox vaccines. Of all measles doses given, 6% went to children under 9 months of age and 12% to those over 24 months. Marker vaccines. BCG vaccine left a scar in 76 of 636 CHILDHOOD IMMUNIZATION PROGRAMME Table 4. Number of vaccinations given in Yaound6 compared to projected number of children immunized Age Percentage Estimated Projected Actual Vaccine group immunized Yaound6 number doses given (months) (coverage survey) population immunized by team DPT 1-23 30.1, 1 8.3, 9.6 a 22 000 13 642 13 754 BCG 1-23 27.1 22 000 6 306 7 967 Measles 9-23 32.6 14 000 4 564 5 890 a Coverage for DPT vaccine includes those children given one, two, and three or more doses. 84 children vaccinated (91 %). Of 35 children receiv- ing their DPT-poliomyelitis vaccination, 33 (94%) correctly received simultaneous BCG vaccine. Small- pox vaccination left a scar in 26 of 27 children (97 %). Of 77 children receiving measles vaccination, 72 (94%) received simultaneous smallpox vaccina- tion. Vaccine usage. During 5 of the vaccination ses- sions evaluated, three 50-dose measles vaccine vials and nine 10-dose vials were opened, representing 240 measles vaccine doses. With these vials, 188 children were vaccinated, resulting in an average use of 78% of vaccine doses. Work load. An average of 240 children were vaccinated at a session. However, at some all-day sessions, more than 400 children were vaccinated over a period of about 7 h. Measles immunity and seroconversion During the 10 sessions evaluated, blood specimens were taken from 45 measles vaccinees, of whom 76% were seronegative at the time of selection for measles vaccination (Table 5). Of 21 children located for a Table 5. Measles immune status and seroconversion to measles vaccine in children selected for measles immunization Children selected Seronegative children for measles immunization followed-up for Age seroconversion group seronegative(months) (Hi titre61 :10) Seroconverted No. No. (%) No. No. (%) 6-11 26 22 (85) 12 6 (50) a 12-35 19 12 (63) 9 8 (89) a Total 45 34 (76) 21 14 (67) a Difference significant, P=0.001, by Fischer's Exact Test. follow-up serological examination, 67 % showed seroconversion. However, seroconversion in children less than 12 months of age was significantly lower; these children had been vaccinated at the same sessions with the same vaccine. The children had not had clinical measles during the study interval. Measles vaccine titre, as determined on represen- tative vials at the beginning of the programme, was 2.5 TCID50 logs per 0.5-ml dose. However, partially used, diluted vaccine vials collected during 4 sessions had titres ranging from 0.5 to 1.5 logs. Although the numbers tested were small, seroconversion did not appear to correlate with these low vaccine titres or with specific vaccination sessions. Population reached by the programme During July, 198 persons accompanying children to vaccination sessions were questioned (Table 6). At the 10 centres, an average of 87% of all children were brought for vaccination by their own mothers. Over 40% of parents chose to attend a centre more distant than the one in their area, mainly because they knew the personnel or also went there with the child for curative treatment. Only 18% learned of the programme through radio or poster publicity. More than half of all parents arrived at the centres by some public or private vehicle, despite the high cost. Of these July vaccinees, 10% were children who had arrived in Yaounde during the previous month. Measles morbidity During 1976, 2019 cases of measles were reported in Yaounde residents through active surveillance at 11 Yaounde dispensaries. This measles incidence represents a 25% reduction compared to 1975 (2639 cases) and a 30% reduction from the average annual number of cases reported for the years 1973-1975 (3230 cases, see Fig. 1). 637 638 Table 6. Responses to a questionnaire: sample of parents leaving vaccination centres 1. Who brought the child to be vaccinated ? (a) Mother (b) Other relative (c) Other person 172 (87 %) 24 (12 %) 2 (1 %) 198 2. Why was the child brought to this centre for vaccination ? (a) Closest vaccination centre to home 109 (59 %) (b) Personnel known or centre used for health care 51 (27 %) (c) Other 25 (13 %) 185 3. How did the parent learn of the vaccination programme? (a) From the personnel of the centre (b) From a friend, neighbour, or relative (c) From publicity (radio, posters) 112 (58 %) 38 (19 %) 35 (18 %) 195 4. How did you travel to the center for vaccination today? (a) By foot 89 (47 %) (b) By taxi (CFA frs. 100) 69 (36 %) (c) By bus (CFA frs. 50) 11 (6 %) (d) By private car 21 (11 %) 190 600 500- 00 E 2 = <t 400- 111a.C (n < <3 300 LU uacowc < X a: 300- 100 1973-1975 J F M I A M J i A S N D MONTH MOIS Fig. 1. Reported cases of measles in Yaound6: monthly averages for 1973-1975 and 1976. The number of reported cases during the period were: 1973, 4439 cases; 1974, 2660 cases; 1975, 2639 cases; average for 1973-1975, 3230 cases; 1976, 2019 cases. Table 7. Estimate of overall costs for the Yaound6 vaccination programme for 9 months November 1975- July 1976 costs Item Per- CFA frs. US $ b centage of total 1. Vaccines 1 289 500a 5200 47 2. Personnel (1 full-time, 6 part-time) 687 500 2752 25 3. Materials and transport 549 000 2200 20 4. Surveillance and evaluation 192 000 768 7 Total 2 718 000 10 920 a This estimates the value of the vaccines if purchased as: Measles 5 500 doses at CFA frs. 37.5 per dose Smallpox 6600 doses ar CFA frs. 5 per dose DPT 15 000 doses at CFA frs. 17.5 per dose Poliomyelitis 15 000 doses at CFA frs. 47 per dose BCG 8 800 doses at CFA frs. 6 per dose Tetanus 2 200 doses at CFA frs. 15 per dose However, of these the measles, smallpox, poliomyelitis, and BCG vaccines were donated by foreign governments of international agencies and their cost was not directly incurred by the programme. b Calculated at an exchange rate of CFA francs 250 to US $1.00. Programme costs An attempt was made to estimate the overall cost of the Yaounde programme, despite the fact that some of these costs were not actually incurred by the programme (cost of donated vaccines, for example). In Table 7, the overall cost for 9 months has been estimated at CFA francs 2 718 000 or US $10 900. The itemized table shows the vaccine to be the major cost (47 %) with personnel salaries second at 25 %. In Table 8, the estimated costs per unit pro- gramme activity have been calculated. The average cost of a dose of vaccine (CFA frs. 24) is the total cost of all vaccines used, i.e., CFA frs. 1 289 000, Table 8. Estimated programme costs CFAfrs. USS Average cost of a dose of vaccine 24 0.10 Average administrative cost per dose 28 0.11 Total cost per dose administered 52 0.21 Average cost per child-visit 120 0.48 Average cost per child registered 237 0.95 Projected cost per child fully vaccinated 470 1.90 B. GUYER & S. ATANGANA CHILDHOOD IMMUNIZATION PROGRAMME 639 divided by the number of doses given, i.e., 51 522. The average administrative cost per dose (CFA frs. 28), is calculated as all programme expenses except vaccine cost, i.e., CFA frs. 1 428 000, divided by the number of doses given. The cost per child- visit and child registered are simply the total pro- gramme cost divided by the number of children in each of these categories. It is projected that to vaccinate a child fully in this programme would cost CFA frs. 470 or US $1.90. DISCUSSION This evaluation provides some data on a multiple- antigen immunization programme in urban Africa and demonstrates the problems to be overcome in the expansion of immunization services in devel- oping countries. Comparisons will be limited to the few published evaluations of mass vaccination pro- grammes and to limited data available from other programmes. Vaccination coverage The overall 30% coverage of the target group 1-23 months of age achieved in 9 months in Yaounde is disappointing. In 1972, a biennial mass measles immunization campaign achieved 78% coverage of the 6- to 36-month-old target population over a 3- month period using intensive publicity (4). Wide coverage, however, was not a primary goal in the early months of programme operation. Since the team was inexperienced and many logistic prob- lems needed to be solved, small vaccination sessions were desirable. Health centre personnel were the main means of disseminating information since they were in contact with the parents who brought their children for baby care or medical consultation. Although attendance did increase in subsequent months, coverage never reached the target; several factors were thought to have contributed to this low vaccination coverage. First, mass publicity efforts, radio, and posters were inadequate to inform the majority of the community of the new vaccination services. Radio broadcasts were not made at the peak morning listening hours, nor were they made in local African languages. However, it may be that, in general, mothers of young African children do not listen to the radio. The large turnout of children in February, when loudspeaker trucks circulated in the town, indicated that the community was ready to respond when the techniques of mass immunization cam- paigns were used. However, the Yaounde pro- gramme was never intended to be a mass (attack phase) campaign, and these methods were not felt to be appropriate on a continuing basis. The diversity of ethnic groups and languages made publicity even more difficult. Second, these newly available immunization ser- vices attracted a large number of older children, outside the target age group, who had never received immunizations. This diverted both vaccines and team time away from the target groups. Finally, as the response to the programme in- creased, the size of vaccination sessions reached the upper limit of 350-400 children per day. In these large, chaotic vaccination sessions, women some- times waited 3-4 hours without reaching the front of the crowd. These women became discouraged and did not return to seek immunization for their children. Thus, in community-based multiple-antigen im- munization programmes, vaccination coverage is linked to publicity and to the efficiency with which the programme can meet the demand for immuniza- tion services. The Yaounde programme has been reorganized to simplify the selection of patients, to improve the flow of patients to vaccination stations, and to deliver DPT vaccine by jet injector rather than by syringe and needle. While pre-programme immunization levels were not documented in Yaound6, a baseline immuniza- tion survey in Douala a in September 1976, using the same methods as in Yaounde, provided some useful comparative data.b In Douala, a random sample of 337 children aged 1-23 months was evaluated. Only 19.2% had had any kind of immunization compared to 39.4% in Yaound6. In Douala, only 7.4% of the target 1- to 8-month-old children had received at least one DPT-poliomyelitis dose, while in Yaound6 22.6% of this age group had begun the course. Finally, only 9.6% of Douala children age 9 to 23 months had been immunized against measles, com- pared to 32.6% of Yaound6 children. Completion of the DPT-poliomyelitis course Only one-third of children beginning the DPT series completed the third dose during the 9-month a Douala is the major port and commercial city of the United Republic of Cameroon with a population of about 400 000, and is situated about 375 km west of Yaound6 on the Atlantic coast. In Douala, vaccines for children could be purchased in private pharmacies and at some child health centres in an uncoordinated programme such as in Yaounde prior to 1975. A mobile vaccination team visited the city every 2 years for the mass measles, smallpox, BCG, and yellow fever campaigns. b OCEAC. Evaluation of vaccination services in Douala, Cameroon. Unpublished OCEAC report, 1976. 640 B. GUYER & S. ATANGANA period. While this low figure is somewhat biased by the short period under study, the return rate is low and disappointing. A special health education effort directed at women first presenting their babies for DPT-poliomyelitis vaccination will need to ensure they understand that the full course comprises three doses. The programme may adopt a more widely spaced two-dose schedule of DPT-poliomyelitis vac- cination if and when it is shown to be as effective as the three-dose course. Programme operations Selection ofpatients. Selection decisions were cor- rectly made according to programme criteria in 80% of cases. Of children selected for measles immuniza- tion, 76% were seronegative compared to the 1972 mass measles compaign when only 49% of vaccinees were susceptible (4). The limitation of measles im- munization to the 9- to 23-month-old target group focused immunization on the high-risk group. Effective vaccine administration. The overall 67% seroconversion to measles vaccine in susceptible children was an improvement over the 40% found by McBean during the 1972 mass campaign (4). As expected, seroconversion was age-related: although the numbers tested were small, seroconversion in children under 12 months of age was noticeably lower. Breman (9) also found reduced seroconversion in children 6-8 months of age (84.3% compared to 94.7% in children 9-23 months of age). This finding reinforces the need to delay measles vaccination to reduce apparent vaccine failure. In this programme, 9 months was taken as the age at which seroconver- sion rates were considered adequate and when vac- cination would result in at least half of the important risk group under 1 year of age being immunized. Seroconversion rates did not seem to vary by session, and the low vaccine titres discovered in used vials may have resulted from problems in transport- ing the vaccine to the laboratory. Smallpox and BCG efficacy were adequate at 97% and 91% scar rates, respectively. Logistics. In brief, good organization and super- vision of a complex operation such as a multiple- antigen immunization programme is crucial. At the time of the vaccination session, all operational ele- ments from transport to supplies, personnel, and organization must be well coordinated if children are to be effectively immunized. In developing countries, where health facilities frequently function with major elements lacking, logistic aspects must be stressed. Further, the publication of a year-long vaccination schedule should be seen as a commitment to provid- ing immunization services at a given place and time. Success in the operation will depend, above all, on an operations officer who has the technical and fiscal authority to make the programme work. Civil Ser- vice policies need to be changed to provide incentives to immunization workers. Measles surveillance The ultimate evaluation of an immunization pro- gramme is its ability to prevent disease. Although measles incidence in Yaounde was reduced in 1976, it would be difficult to argue that the level of vaccination coverage achieved by the team was responsible for this effect. Nevertheless, the focusing of vaccination on a precise age group and the timing of the February campaign may have played a role in modifying the pattern of the annual epidemic. Programme costs No previous reports exist with which to compare cost figures for a multiple-antigen immunization programme. In Yaounde, an effort was made to keep costs low by using available materials, vaccines, and personnel. However, the costs shown in Table 7 were estimated as though incurred in the programme. In the 5-year smallpox eradication and measles control programme in West and Central Africa (1967-1971), about US $31 000 000 was spent to vaccinate about 122 000 000 persons against small- pox and 20 000 000 children against measles. The average cost of the programme was approximately $0.27 per person. Measles and smallpox vaccine purchases accounted for 25% and 10% of the budget, respectively.a Clearly, the Yaound6 multiple-antigen programme is more expensive per person vaccinated and the cost of the vaccines represents virtually half of the total budget. The US $1.90 estimated to vaccinate a child fully is more than the annual per caput expenditure on health of many African states (10). In Yaounde, to immunize fully the estimated 12 000 infants born every year would cost about US $24 000. Therefore, multiple-antigen immunization is expensive and means of reducing overall costs must be found, otherwise countries undertaking programmes with foreign assistance will later have to abandon them when faced with bearing the entire cost. a BUCK, A. A. ET AL. Evaluation of the smallpox eradica- tion and measles control program, Central and West Africa. APHA/USAID report, 1971. CHILDHOOD IMMUNIZATION PROGRAMME 641 ACKNOWLEDGEMENTS The authors wish to acknowledge the excellent pioneering work of the Yaounde vaccination team and its chief, Mr Paul Djogdom. Others have assisted the team in their work, including Mr Simon Ndengue, Mr Engelbert Binyet, Mr Sylvestre Mbarga, and Mr Gabriel Effila. We thank Dr J. Nakano, Chief, Viral Exanthems Branch, Center for Disease Control, Atlanta, for performing the measles HI tests and for determining measles vaccine titres. We also thank Dr Bernard Durand (OCEAC Secretary-General), Mr Robert C. Hogan, Dr J. Michael Lane, Dr W. Schaffner, and Dr S. 0. Foster for their encouragement and assistance. RtSUMt PROGRAMME D'IMMUNISATION PAR ANTIGENES MULTIPLES AU COURS DE L'ENFANCE A YAOUNDE, CAMEROUN: EVALUATION DE LA PREMIERE ANNAE, 1975-1976 Un programme d'immunisation par antigenes multiples au cours de l'enfance a dtd entrepris a Yaoundd, Repu- blique-Unie du Cameroun, en novembre 1975; c'est donc un des premiers programmes elargis de vaccination qui soit operationnel en Afrique. Apres 9 mois, on a evalue le programme afin de d6terminer la couverture et l'efficacite de la vaccination, de controler la qualite des operations, y compris la s6lection des sujets (triage), et le gaspillage de vaccins, d'identifier les caracteristiques de la population touchee par la publicite relative au programme et d'estimer le couit de ce dernier. Entre novembre 1975 et juillet 1976, les 7 membres de l'equipe de vaccination de Yaound6 ont procede 'a 96 s6ances, selon un calendrier fixe a l'avance, dans 11 dispensaires urbains de Yaounde. En tout, 51 522 doses de vaccins (de toutes les categories) ont et6 administrees au cours de 22 736 visites-enfants a raison d'une moyenne de 2,3 doses de vaccin par visite-enfant. Une moyenne de 4,5 doses de tous les vaccins a 6ete administree a chaque enfant enregistre. L'evaluation de la couverture en ce qui concerne la vaccination a ete effectuee sur un echantillon statistique d'enfants de Yaounde et a montr6 que, parmi le groupe d'Age cible (1 a 8 mois), seules des proportions de 22,6% et 21,3% respectivement ont recu au moins une dose de vaccin DTC ou de vaccin antipoliomyelitique buccal. Une cicatrice de BCG a ete d6couverte chez 20,9% des enfants. En ce qui concerne le vaccin antirougeoleux, un tiers du groupe d'age cible (9 a 23 mois) en avait requ. Parmi les enfants qui avaient commence a recevoir leur vaccination DTC-polio, un tiers seulement a recu la totalite des trois doses. L'equipe de vaccination a pris des decisions de s6lec- tion correcte en fonction de l'age et de l'etat d'immunisa- tion chez 80% de ces enfants. En moyenne, 240 enfants etaient vaccines au cours d'une seance. Parmi ceux qui ont ete retenus pour la vaccination antirougeoleuse, 76% etaient seronegatifs: il y a eu seroconversion, apres la vaccination antirougeoleuse, chez 89% des enfants ages de 12 a 35 mois, mais 50% seulement ont ete pro- teges parmi ceux de moins de 12 mois. Une 6valuation de la population touchee par le pro- gramme a montre que 87% des enfants ont et6 amenes par leur mere. Plus de 40% des parents ont choisi un centre plus eloigne et non celui qui 6tait situe dans leur voisinage. Seulement 18% des personnes ont 6te infor- mees du programme par la radio ou des affiches, le reste en avait entendu parler directement par des personnes connues et surtout par des membres du personnel sani- taire. I1 semble que la publicite de masse n'ait pas donne un resultat satisfaisant. L'incidence de la rougeole au cours de la periode d'evaluation a 6te reduite de 25% par rapport a l'ann6e precedente. Le coiut global du programme a ete de 2 718 000 francs CFA (US $10 900), le co(ut du vaccin representant 40% des d6penses et les salaires du personnel 25%. II a 6te calcule que l'immunisation complete d'un enfant cofiterait 470 francs CFA (US $1,90). REFERENCES 1. WHO Official Records, No. 217, 1974, p. 28. 2. WHO Chronicle, 30: 339-350 (1976). 3. HENDERSON, R. H. ET AL. Assessment of vaccination coverage, vaccination scar rates, and smallpox scar- ring in five areas of West Africa. Bulletin of the World Health Organization, 48: 183-194 (1973). 4. McBEAN, A. M. ET AL. Evaluation of a mass measles immunization campaign in Yaounde, Cameroon. 642 B. GUYER & S. ATANGANA Transactions of the Royal Society of Tropical Medi- cine and Hygiene, 70: 206-212 (1976). 5. MATHEWS, H. M. ET AL. Persistence of malaria antibody in Tobago, West Indies following eradica- tion as measured by the indirect hemagglutination test. American journal of tropical medicine and hy- giene, 19: 581-585 (1970). 6. HIERHOLZER, J. C. & SUGGS, M. T. Standardization of viral hemagglutination and hemagglutination inhibition tests. I. Standardization of erythrocyte suspensions. Applied microbiology, 18: 816-823 (1969). 7. HIERHOLZER, J. C. ET AL. Standardization of viral hemagglutination and hemagglutination inhibition tests. II. Description and statistical evaluation. Ap- plied microbiology, 18: 824-833 (1969). 8. NORRBY, E. Hemagglutination by measles virus. Proceedings of the Society for Experimental Biology and Medicine, 11: 814-818 (1962). 9. BREMAN, J. G. ET AL. Evaluation of a measles- smallpox campaign by a sero-epidemiological me- thod. American journal ofepidemiology, 102: 564-571 (1975). 10. WORLD BANK. Health sector policy paper. Washing- ton, DC, World Bank, 1975, Annex 3, p. 74.