Measles surveys*

BRIEF COMMUNICATIONS 673 Measles surveys* A. N. D. NANAVATI 1 & NASEEMA A. MEHTA 2 Abstract The authors present background data indicating that there is a close correlation between serological findings and children's measles histories obtained from parents. They conclude that interrogative surveys can yield reliable epidemiological information even though the sample of the survey population checked serologically may be small. It has been stated in previous communications (Mehta et al., 1972; Shah et al., 1972) that subclinical Table 1. Location, number, and age distribution of children serologically assessed for measles in India no.of Plaeaea Category No. of Age in yearsSo rof lace Year (seetexteX) children *1 Aei erstudy (setx) cide %-1 1-3 3-5 >5 1 Bombay 1967 A 200 a 50 50 50 50 2 Bombay 1969 B 56 b 17 27 12 - 3 Aurangabad 1969 A 79b 31 42 6 4 Aurangabad 1970 A 61 c 11 32 18 - 5 Indore 1969 A 348b 140 176 20 12 6 Indore 1971 B 106 15 66 25 - 7 Poona 1970 B 40 - 1 11 28 total 890 264 394 142 90 a Data of Mehta et al. (1972). b Prevaccination serum samples collected during a measles vaccine trial conducted by the Indian Council of Medical Research. c Data of Shah et al. (1 972). infections of measles do not occur in the nonimmune child to an appreciable extent. This view was based on the close correlation found, in tests of over 600 sera obtained from different sources, between the history given by the parents and the serological findings in the child. It was therefore suggested that information on the epidemiology of the dis- ease might be obtained by interrogative surveys, and a test survey of this nature including a serologi- * From the Haffkine Institute, Bombay, India. 1Assistant Director, and Head of the Department of Virology. ' Senior Research Fellow, Department of Virology. 2950 cal check on 8% of the survey population has recently been reported (Shah et al., 1972). It should be noted that certain prerequisites set forth in that communication must be satisfied if such surveys are to be reliable. If these are neglected, the infor- mation obtained is likely to be erroneous and mis- leading. For the information of interested workers, the background data on which the study of Shah and associates was planned, with additional material ob- tained subsequently, is presented here in tabular form. It will be seen from Tables 1-3 that the evidence is convincing in spite of certain obvious defects. Many of the sera were collected for a vaccine trial, so that there was a selection for children without a history of measles. Further, the necessary pre- conditions and planning are known to have been adhered to only in the studies in category A- studies 1, 3, 4, and 5-whereas in the remainder, in category B, we have no knowledge of the condi- tions. It may be a coincidence that these are the groups where the greatest discrepancy between his- tories and serological findings was noted. It should be emphasized that to avoid any possi- 674 BRIEF CONMMUNICATIONS Table 2. Correlation of serological results and history taken from parents No. of No. of children with history No. of children without history Serial children of measles of measles no. of serologi- Seropositive a Seronegative Seropositive a Seronegativestudy cally __________________________________ assessed No. % No. % No. % No. % 1 200 68 34.0 - - 3 1.5 129 64.5 2 56 - - - - 4 7.1 52 92.9 3 79 2 2.5 - - - - 77 97.5 4 61 26 42.6 1 1.6 1 1.6 33 54.2 5 348 6 1.7 - - - - 342 98.3 6 106 48 45.3 2 1.9 1 5 14.1 41 38.7 7 40 30 75.0 - - - - 10 25.0 total 890 180 20.2 3 0.3 23 2.6 684 76.9 a Sera with an HI titre of 1: 20 or above were taken as positive (Mehta et al., 1972). Table 3. Correspondence between history and sero- logical results H istory History did not No. of corresponded correspond with Serial no. of children with serum serum antibody study serologically antibody results results assessed No. % No. % 1 200 197 98.5 3 1.5 2 56 52 92.9 4 7.1 3 79 79 100.0 - - 4 61 59 96.7 2 3.3 5 348 348 100.0 - - 6 106 89 84.0 17 16.0 7 40 40 100.0 - - total (A studies only) 688 684 99.4 4 0.6 total 890 865 97.2 25 2.8 bility of bias (other than stated above) the serological tests in each study were performed without any know- ledge of the survey data, and the correlation was made only after the tests had been completed. These data are presented to indicate the back- ground information for the hypothesis that such surveys may be useful, and also to sound a warning against undertaking them without the necessary planning and preparation. The findings do not rule out the possibility of subclinical infection in measles (Stocks & Karn, 1928; Bech, 1960; Snyder et al., 1962; Krugman et al., 1965). However, this appears to be a phenomenon of older children who are partially immune or with waning immunity, as suggested by Mehta,' and does not seem to be of relevance in infants or young children at first exposure to the disease. ACKNOWLEDGEMENTS The authors express their thanks to the following persons and institutions for their help in the collection of data and serum samples: Dr S. M. Merchant, Bai Jerbai Wadia Hospital for Children, Bombay; Dr P. M. Udani, Pediatric Centre, Sir J. J. Group of Hospitals, Bombay; Dr H. I. Jhala, Dr V. N. Ingle, and Dr Usha Shah, Medical College, Aurangabad; Dr J. N. Pohowalla and Dr S. D. Singh, Mahatma Gandhi Medical College, Indore; Dr N. S. Doodhar, B. J. Medical College, Poona, and the Director-General, Indian Council of Medical Research, New Delhi. REFERENCES Bech, V. (1960) Acta path. microbiol. scand., 50, 322-330 Krugman, S. et al. (1965) J. Pediat., 66, 471-488 Mehta, N. A. et al. (1971) Ind. J. med. Res., 60, 661-669 Shah, U. et al. (1972) Bull. Wld Hlth Org., 46, 130-13 8 Snyder, M. J. et al. (1962) Amer. J. Dis. Child., 103, 250-251 Stocks, P. & Karn, M. V. (1928) Ann. Eugen. (Lond.), 3, 361-398 1 Mehta, N. A. (1970) Seroepidemiology of measles, Ph.D. Thesis, University of Bombay, India.