Nutritional status of preschool children in Togo, 1976-77*

Bulletin of the World Health Organization, 58 (6): 889-895 (1980) Nutritional status of preschool children in Togo, 1976-77* H. C. STETLER,1 A. AYEBOUA,2 E. W. BRINK,3 A. N. AGLE,4 N. W. STAEHLING,5 &J. M. LANE6 Survey teams visited 163 rural villages and41 urban quarters in Togo and collected data on weight, height, arm circumference, age, andpedal oedemafrom 6120 children aged 6-72 months. Haemoglobin levels were determinedfor one-fifth of the survey children. When 80% ofthe reference median weight-for-height was used as the cut-offpoint, theprevalence ofacute undernutrition was 2.0%. When 90% of the reference median height-for-age was used as the cut-off point, the prevalence of chronic undernutrition was 19.1%. The prevalence of both types of undernutrition was significantly higher in the northern rural regions, with the lowest prevalence in the urban areas. The prevalence of anaemia was 58.6%, with the highestprevalence in the northern regions. Anthropometric data on socio- economically privilegedpreschool children were collected and compared with thosefor the survey children and the reference population. The socioeconomically privilegedpreschool group of Togolese children had a nutritional status substantially better than that of the survey children and nearly equivalent to that of the American reference population. The survey method was economical in terms ofmoney, time, andpersonnel. The methodology employed rapidly provided objective data on the extent and distribution ofprotein-energy undernutrition and anaemia in the preschool children of Togo. Togo is located in West Africa on the Gulf of Guinea between Ghana and Benin. It is only 149 km * This study was conducted under the auspices of the Ministry of Rural Development of the Government ofTogo and was supported by the United States Agency for International Development and the Center for Disease Control, Public Health Service, Department of Health and Human Services. The opinions expressed herein are those of the authors and do not necessarily reflect the official views of the sponsoring authorities and organizations. I Medical Officer, Bureau of Smallpox Eradication, Center for Disease Control, Public Health Service, Department of Health and Human Services, Atlanta, GA 30333, USA. 2 Chef de la Division de la Nutrition appliquee et de l'Economie alimentaire, Ministere du Developpement rural, Lome, Togo. 3 Medical Officer, Bureau of Smallpox Eradication, Center for Disease Control, Atlanta, GA 30333, USA. 4 Public Health Advisor, Bureau of Smallpox Eradication, Center for Disease Control, Atlanta, GA 30333, USA. 5 Statistician, Bureau of Smallpox Eradication, Center for Disease Control, Atlanta, GA 30333, USA. ' Director, Bureau of Smallpox Eradication, Center for Disease Control, Atlanta, GA 30333, USA. across at its widest point, but it extends 576 km from the Gulf to the border of Upper Volta. The basic food sources are cereals in the north and tubers in the south. Protein-energy undernutrition is recognized as a problem in Togo. In 1964-65, a national study on the food consumption of Togolese families identified 23.2% of rural households as having energy intakes that were less than 80% of the recommended amount and 35.2% as having protein intakes that were simi- larly deficient (1). Urban households showed lesser deficits in energy and protein intakes. In 1974, Ayeva noted that severe clinical protein-energy under- nutrition accounted for 208 of 1766 paediatric admissions (11.8%) to the University Hospital of Lome (2). The objectives of the 1976-77 National Nutrition Survey were to identify the prevalences of protein-energy undernutrition and anaemia in Togo- lese children by age group and by geographical area. 4015 889- H. C. STETLER ET AL. METHODS The entire population of Togo (1.9 million) was considered as the survey population. The rural inhabitants of each of the five administrative regions formed the five rural groups of the survey. Lome, the capital of Togo, and six other cities composed the urban region (Fig. 1). To permit statistical inferences on individual regions and interregion comparisons, we selected at least 30 sample sites in each of the six regions, employing the 1970 Togo census (3) and population proportionate sampling. In each rural sample site the survey team sub- grouped the households geographically and then UPPER VOLTA REGION 5: SAVANNA REGION BENIN 4;' BASSAR SOKODE REGION 3: GHANA CENTRAL REGION 2; PLATEAU 0 ATAKPAME {@PALIME J ARITIMEi ;EVIEO AN GJF OF G/J/NEA Fig. 1. Map of Togo showing the five rural survey regions and the urban centres. chose one of these subgroups by random selection. The team then proceeded to the centre of the chosen subunit and randomly selected a compass direction. The first household encountered in this direction became the initial household. All children in the 6-72-month age group from this household were enrolled and the same procedure was followed among the next closest houses until 30 children had been included. In urban sample sites, the teams randomly selected the initial household from malaria household lists or from urban census lists. The survey teams also measured a special group of 450 Lome preschool children from private day nurseries and from the families of high-ranking military and police officers to estimate the nutritional status of socioeconomically privileged children. Seven two-man teams were trained. We followed a standardization procedure involving duplicate measurements and inter-team measurement com- parisons to estimate and improve the accuracy and reproducibility of the anthropometric measurements. Supervisors monitored the collection of data in the field and edited the survey forms. The field teams measured stature in children 2 years of age and older and length in children less than 2 years of age on a portable measuring board (4). Weight was determined with a portable Salter hanging scale (5). Arm circumference was measured with a Zerfas slot- ted tape (6). The presence or absence of pedal oedema was determined by firm thumb pressure bilaterally for 3 seconds on the dorsal surface of the foot. Capillary blood was collected by finger prick from every fifth child in the survey, and haemoglobin determinations were done each evening using the cyanmethaemo- globin method and a Bausch & Lomb Minispect 20 TM spectrophotometer. The recently developed National Center for Health Statistics/Center for Disease Control (NCHS/CDC) height and weight reference data (7) were used for evaluation of the Togo survey data. The NCHS/CDC reference population is larger and more representative than the commonly used Stuart-Meredith reference population (8). These factors result in improved statistical precision of outlying percentiles and a statistically more acceptable basis for analysis of anthropometric survey data. The analyses of anthropometric and haemoglobin data require reliable age information. The field workers sought a verifiable record, such as a birth registration document, for each child. If this was not available, the field workers employed historical calen- dars (prepared for each region by the local social workers) to locate the year and month of the child's birth. Children without a verifiable record were excluded from the survey only when the field workers could not estimate the child's age to within 3 months by the historical calendar method. 890 NUTRITIONAL STATUS OF TOGOLESE CHILDREN RESULTS Complete sets of data were obtained for 6094 chil- dren, whose age and sex distributions are shown in Table 1. Because many children in Togo begin school at 5 years of age and were not at home at the time of the survey team's visit, the 60-72-month age group is underrepresented. Table 1. Sample distribution by age and sex Males Females Total Age (months) No." %b No. % No. % 6-11 335 5.6 373 6.3 708 11.8 12-23 703 11.6 641 10.7 1344 22.3 24-35 600 10.0 572 9.3 1172 19.3 36-47 601 9.7 551 9.0 1152 18.7 48-59 535 8.6 483 7.9 1018 16.6 60-71 385 6.2 315 5.1 700 11.3 Total 3159 51.7 2935 48.3 6094 100.0 a Actual number of children in age group. b Weighted percentages to correct for the influence of under- sampling in the Maritime and Plateau regions. Weight-for-height is a good index of wasting of body mass, and children who are below 80% of the reference median may be considered as acutely under- nourished. Table 2 presents the percentage distri- bution of the sampled children by region according to the NCHS/CDC reference weight-for-height median. The prevalence of acute undernutrition is significantly higher among the children of the most northern Table 2. Percentage distribution of survey population by NCHS/ CDC reference weight-for-height mediana % of NCHS/CDC reference median Region < 80.0 80.0-89.9 90.0-99.9 > 100.0 Savanna 5.1 29.3 44.3 21.4 Kara 1.7 20.6 43.1 34.6 Central 2.0 19.7 48.9 29.4 Plateau 1.7 20.4 48.4 29.6 Maritime 1.8 21.6 49.0 27.6 Urban 0.8 18.3 48.1 32.8 Average (Togo) 2.0 21.4 47.4 29.1 region, the Savanna, than among those of the other regions (t-test, P<0.001). There is little difference in prevalence among the remaining regions. In the whole of Togo, the distribution of acute undernutrition by age groups (Table 3) indicates that children aged 12-24 months are most frequently affected. Table 3. Percentage of children suffering from acute under- nutrition" by age Age (months) % acutely undernourished 6-11 3.2 12-23 4.5 24-35 1.8 36-47 0.8 48-59 0.5 60-71 0.7 Average (Togo) 2.0 B Weight-for-height < 80% of reference median; percentages based on weighted data. Table 4 presents data on the height-for-age of the sample children. Height-for-age is a good index of linear body growth, and children who are below 90%/ of the reference median may be considered to be chronically undernourished. The total survey preva- lence of stunting is high, but significant variation exists among the six regions (F-test, P<0.001). The prevalence of chronic undernutrition is highest in the northern rural regions and lowest in the urban areas. The overall prevalence of stunting increases with age through the first 3 years of life and remains at approximately 25% from 3 to 6 years of age (Table 5). Table 4. Percentage distributiona of survey population by NCHS/CDC reference height-for-age median % of NCHS/CDC reference median Region < 90.0 90.0-94.9 > 95.0 Savanna 28.2 40.0 31.4 Kara 26.1 38.6 35.3 Central 20.3 41.2 38.5 Plateau 16.2 42.5 41.3 Maritime 17.8 39.7 42.5 Urban 11.4 37.3 51.3 Average (Togo) 19.1 40.1 40.8 8 Percentages based on weighted data. 891 11 Percentages based on weighted data. H. C. STETLER ET AL. Table 5. Percentage of children suffering from chronic undernutrition8 by age Age (months) % chronically undernourished 6-11 6.5 12-23 15.1 24-35 18.6 36-47 24.4 48-59 23.9 60-71 24.8 Average (Togo) 19.1 8 Height-for-age < 90% of reference median; percentages based on weighted data. Table 6 presents the basic anthropometric data for the entire survey population and for each region using the Waterlow categories (9). Children were classified as "stunted only" if they were below 90% of the refer- ence height-for-age median but above 80% of the reference weight-for-height median. Conversely, they are considered "wasted only" if they were below 80% of the weight-for-height median, but were over 90% of the height-for-age median. Children were con- sidered "normal" if they were above and "wasted and stunted" if they were below both of these cut-off points. The severest form of undernutrition, com- bined wasting and stunting, is most prevalent in the Savanna (t-test, P<0.001). There are more normal children in the urban region than in the rural regions (t-test, P<0.001) and there tend to be fewer normal children the further north one goes. Table 6. Percentage distribution' of children 6-71 months of age by Waterlow categories b Wasted Region Stunted Wasted and Normal only only stunted Savanna 69.4 25.5 2.4 2.6 Kara 72.7 25.6 1.3 0.4 Central 77.9 20.1 1.8 0.2 Plateau 82.4 15.9 1.3 0.3 Maritime 81.0 17.3 1.2 0.6 Urban 88.1 11.1 0.5 0.3 Average (Togo) 79.6 18.4 1.3 0.7 a Percentages based on weighted data. b See ref. 9. Weight-for-age (Table 7) is a commonly used anthropometric index of undernutrition, but it does not separate the acute and chronic components. Arm circumference-for-height has been considered a possible substitute index for weight-for-height in the identification of acute undernutrition. Both indi- cators-arm circumference-for-height and weight- for-height-identified the Savanna region as the nutritionally worst region. However, arm circum- ference-for-height as an index of acute under- nutrition, failed to rank the remaining regions in the same order as the weight-for-height index. Table 7. Percentage distribution° of survey children by G6mez classification of undernutritionb Region Normal 1st degree 2nd degree 3rd degreee > 9o.oc 75.0-89.9 60.0-74.9 < 60.0 Savanna 21.8 51.1 24.5 2.6 Kara 30.8 50.4 17.8 0.9 Central 30.5 54.0 14.8 0.7 Plateau 32.9 53.4 13.3 0.3 Maritime 30.8 56.1 12.5 0.6 Urban 40.3 50.8 8.3 0.6 Average (Togo) 31.6 53.1 14.5 0.8 B Percentages based on weighted data. b See ref. 11. c Percentage of NCHS/CDC reference median. Anthropometric and age data were collected on 450 children aged 6-72 months in several private nursery schools and in the families of civil servants within the capital city of Lome. These children are from the highest socioeconomic level in Togo; their nutritional status is substantially better than that of the survey children and almost equivalent to that of the American reference population (Table 8). Pedal oedema of nutritional origin is diagnostic of kwashiorkor. In the Togo survey, only 26 children had bilateral oedema (i.e., 4.0 per 1000 children examined). The prevalence of bilateral pedal oedema was 12-fold higher in children with stunting and/or wasting than in children with normal anthropometry. The survey was carried out from October through December, the season of greatest food availability and lowest diarrhoea prevalence, and thus the season of the lowest expected prevalence of kwashiorkor. A haemoglobin determination was done on every fifth child in the survey. Anaemia is defined as a haemoglobin level below 100 g/litre if the child is under 2 years of age and below 110 g/litre if the child 892 NUTRITIONAL STATUS OF TOGOLESE CHILDREN Table 8. Percentage distribution by Waterlow categories' of the survey children, special group, and the NAS refer- ence population b Wasting Normal Wasting Stunting and Sample only only stunting sizes Total Togo (weighted) 79.6 1.3 18.4 0.7 6094 Rural (weighted) 78.1 1.5 19.7 0.7 4868 Urban 88.1 0.5 11.1 0.3 1226 Special group (6-71 months) 98.3 0.4 1.3 - 450 NAS reference population 98.7 0.8 0.5 - 6195 B See ref. 9. b An American reference population recommended by the National Academy of Sciences, developed by CDC, and similar to the NCHS/CDC reference population. is 2 years of age or older (10). Table 9 presents the haemoglobin distribution and percentage of anaemic children by region. Haemoglobin levels tend to be lower in the northern rural areas. The rate of anaemia in the northern regions of Savanna, Kara, and Central is significantly higher than in the remaining regions (x2, P<0.005). Severe anaemia, identified as a haemoglobin level below 80 g/litre and requiring urgent clinical attention, has a countrywide prevalence of 10.1%. Overall, there is a very high prevalence of anaemia in Togo. Table 10 shows by age group the percentage of Togolese preschool children who are anaemic. There is a high prevalence in all age groups, but with some Table 9. Percentage distribution' of survey children by haemoglobin levels Haemoglobin values (g/litre) Region anaemic < 80 80-89 90-109 110 Savanna 68.9 13.8 15.3 46.4 24.5 Kara 65.6 17.2 6.1 48.9 27.8 Central 65.0 11.8 16.7 46.3 25.2 Plateau 56.0 6.1 11.7 43.6 38.6 Maritime 55.7 8.3 13.1 47.6 31.0 Urban 50.3 8.6 10.4 44.3 36.7 Total (Togo) 58.6 10.1 12.3 46.0 31.6 8 Percentages based on weighted data. Table 10. Percentage of children with anaemia by age groups Age (months) % anaemic8 6-11 61.2 12-23 49.3 24-35 70.7 36-47 65.7 48-59 53.4 60-71 49.2 Average (Togo) 58.6 a Percentages based on weighted data. decrease in prevalence in older children. The apparent rise at 2 years of age is accounted for by the change in the cut-off point for anaemia from <100 g/litre to <110 g/litre. For a better analysis of haemoglobin data it would be necessary to make a comparison with a curve showing a continuous distribution of haemo- globin by age in a reference population. Such a ref- erence curve is not yet available. In the survey, 68% of the children had their age determined from an official record and the remaining 32% by the historical calendar method. Table 11 contrasts age-dependent measurements (height-for- age) obtained using official records of ages with those obtained using historical calendar ages. There is no statistical difference in chronic undernutrition as calculated by the two methods ofage distribution, and the historical calendar method ofage determination of preschool children was found to have sufficient accuracy. Table 11. Percentage distributionI of survey children with chronic undernutritionb by age determination method Verifiable Historical Region record calendar Total Savanna 28.5 25.6 28.2 Kara 23.8 30.2 26.1 Central 18.9 22.1 20.3 Plateau 16.8 15.2 16.2 Maritime 18.3 16.8 17.8 Urban 11.6 11.0 11.4 Average (Togo) 19.3 18.5 19.1 8 Percentages based on weighted data. b Height-for-age < 90% of reference median. 893 H. C. STETLER ET AL. DISCUSSION A Joint FAO/WHO Expert Committee on Nu- trition (12) in 1971 emphasized the need for simplified techniques to assess and compare the nutritional status of countries. Cross-sectional survey collection of data by selected nutrition indicators provides useful information for establishing nutrition priorities and targets for remedial programmes. This survey has utilized the CDC methodology (13) of selected anthropometric indices, clinical signs, and haemoglobin determinations within a framework of representative sampling, measurement standardiz- ation, and close field supervision to characterize the nutritional status of Togo. Comparable surveys have been completed in Nepal (14) and Sri Lanka (15). Except for the Savanna region, acute protein- energy undernutrition is not a problem in Togo. Anec- dotal information from the Savanna region relating drought conditions and subsequent poor harvests during the year preceding the study reinforces the survey findings of significantly higher rates of acute undernutrition in the Savanna. High prevalences of chronic undernutrition occurred in all regions (19. 1% average prevalence), with significantly higher levels in the northern region, and the lowest prevalence (11.4%) in the urban region. The highest prevalence of acute undernutrition occurs during the weaning period of 6-23 months of age, reflecting traditional inadequate food intake patterns during the transition from breast-feeding to the household diet. Chronic undernutrition begins in the first 6 months of life with the prevalence of stunting increasing up to the end of the 4th year of life. During the 5th and 6th years of life, the rate of stunting remains constant. This is probably an indi- cation that these age groups are receiving sufficient nutrition to prevent new cases of stunting. It is unlikely that the diet has improved sufficiently to cure any stunting that has already occurred. Concurrent wasting and stunting identifies the most seriously undernourished group of children. Children in this category are in urgent need of remedial measures (9) and 3449 children (0.7% of Togo's 492 729 children 6-71 months of age) come into this priority category. The socioeconomically privileged group of pre- school children from Lome were found to have a better nutritional status than the survey children, and in fact, closely resembled the NAS (National Academy of Sciences) reference population. The latter finding supports Habicht's suggestion (17) that ethnic background has little effect on the attainable stature of preschool children. Theoretically, the nutritional status of the privileged group should be attainable by the entire preschool population ofTogo. The 58.6% prevalence of anaemia in Togolese pre- school children is high, but is similar to the 62.1% prevalence found in Liberian preschool children (16). Severe anaemia (a haemoglobin level below 80 g/litre) was found in 1O.1I% of the survey children. Extra- polating this figure to the total population, it would appear, therefore, that nearly 50 000 Togolese preschool children are in urgent need of treatment for anaemia. In the US ten-state nutrition survey, black pre- school children in low-income states had a 34.3% prevalence of anaemia (18). Almost all the cases were caused by deficient iron intakes. In Togo, the high rates of anaemia are probably caused by a combi- nation of dietary lack of iron, malaria, intestinal parasites, and hereditary haemoglobinopathies. The Malaria Bureau's statistics show malaria to be highly endemic throughout Togo, and there is no reason to suspect regional differences in infestation with intestinal parasites or in the prevalence of hereditary haemoglobinopathies. Although anaemia prevalence is high throughout Togo, it is highest in the northern regions where the highest prevalences of protein- energy undernutrition were also recorded. Further studies are indicated to clarify the contribution of low dietary intake of iron to increased anaemia preva- lences in certain regions. The differences in nutritional status between the urban and rural poor need to be better quantified to improve programme planning. Previous CDC nutrition status surveys did not attempt to assess urban populations because of sampling difficulties. In Togo, the 1970 census and malaria campaign records provided sufficient demographic data to permit urban sampling (3). The survey results show that the urban preschool population is significantly better off by all anthropometric indices. This indicates a much greater need for nutrition intervention in rural areas. This survey has defined the distribution and esti- mated the prevalence of protein-energy under- nutrition and anaemia. Ecological studies are now needed to define the causal factors of these nutritional problems. Government intervention programmes can then be designed rationally and aimed at the high-risk age groups and priority geographical areas. Finally, further evaluations are needed, either by the use of limited surveys or by the establishment of a national nutrition status surveillance system. 894 NUTRITIONAL STATUS OF TOGOLESE CHILDREN 895 R.SUME ETAT NUTRITIONNEL DES ENFANTS D'AGE PRESCOLAIRE AU TOGO (1976-1977) Des equipes d'enquete se sont rendues au Togo dans 163 agglomerations rurales et dans 41 secteurs urbains pour recueillir des donnees concernant le poids, la taille, la circon- ference du bras, I'age et la presence d'ordeme chez 6120 enfants de 6 mois A 6 ans. On a d6termin6 le taux d'hemo- globine chez un cinquieme d'entre eux. Si l'on prend comme point limite 80% de la valeur mediane de reference du poids- pour-la-taille, la prtvalence de la denutrition aigue etait de 2,0%. Si l'on prend 90/o de la valeur mediane de reference de la taille-pour-I'age comme point limite, la prevalence de la denutrition chronique etait de 19,1 %. La prevalence des deux types de dtnutrition etait significativement plus elevte dans les regions rurales du nord alors qu'elle etait la plus faible dans les zones urbaines. La prevalence de l'anemie etait de 58,6% avec un maximum dans les r6gions du nord. Des donn6es anthropo- metriques relatives A la population prescolaire togolaise socio-economiquement privilegiee ont e reunies et comparees avec celles des enfants soumis A l'enquete et celles de la population americaine de reference. L'etat nutritionnel du groupe d'enfants privilegies etait nettement meilleur que celui des enfants soumis a l'enquete et atteignait pratique- ment celui de la population de reference. La technique d'enquete adoptee s'est revelee economique du point de vue argent, temps et personnel. GrAce A la methodologie suivie, on a rapidement obtenu des donnees objectives sur l'etendue et la repartition de la d6nutrition proteino-energetique et de l'anemie chez les enfants d'Age prescolaire au Togo. REFERENCES 1. Enqugte sur les budgets familiaux et la consommation des m6nages au Togo, Paris, Societe d'Etudes pour le Developpement Economique et Social, 1967. 2. AYEVA, A. Etude de la malnutrition proteo-calorique chez l'enfant au Centre Hospitalier Universitaire de Lome, Lome, Ecole des Assistants Medicaux, Univer- sit6 du Benin, 1975 (Memoire No. 8). 3. Recensementgeneralde lapopulation du Togo, ler mars au 30 avril 1970, Lome, Direction de la Statistique, Ministere du Plan, 1974. 4. BRINK, E. W. ET AL. Improvised measuring board, suitable for the field. Tropical doctor, 7: 96 (1977). 5. MORLEY, D. Paediatric priorities in the developing world. London, Butterworths, 1973, pp. 140-141. 6. ZERFAS, A. J. The insertion tape: a new circumference tape for use in nutritional assessment. Americanjournal of clinical nutrition, 28: 782-787 (1975). 7. NATIONAL CENTER FOR HEALTH STATISTICS. NCHS growth charts. Monthly vital statistics report, vol. 25, No. 3, suppl., Rockville, US Department of Health, Education, and Welfare, 1976 (DHEW publication No. HRA 76-1120). 8. VAUGHAN, V. C. & McKAY, R. J., ed., Nelson's text- book of pediatrics, 10th ed., Philadelphia, Saunders, 1975, pp. 39-48. 9. WATERLOW, J. C. & RUTISHAUSER, I. H. Malnutrition in man. In: Cravioto, J. et al., ed., Early malnutrition and mental development. Proceedings of the sym- posium of the Swedish Nutrition Foundation, Sweden, Almquist & Wiksell, 1974, pp. 13-26. 10. WHO Technical Report Series, No. 405, 1968 (Nutritional anaemias: report of a WHO Scientific Group). 11. GOMEZ, F. ET AL. Mortality in second and third degree malnutrition. Journal of tropical pediatrics, 2: 77 (1956). 12. WHO Technical Report Series, No. 477, 1971 (Food fortification-protein-calorie malnutrition: eighth report of the Joint FAO/WHO Expert Committee on Nutrition). 13. MILLER, D. C. ET AL. Simplified field assessment of nutritional status in early childhood: practical suggestions for developing countries. Bulletin of the World Health Organization, 55: 79 (1977). 14. BRINK, E. W. ET AL. Nutritional status of children in Nepal, 1975. Bulletin of the World Health Organ- ization, 54: 311 (1976). 15. BRINK, E. W. ET AL. Sri Lanka Nutrition Status Survey, 1975. Internationaljournal ofepidemiology, 7: 41 (1978). 16. Liberia National Nutrition Survey (December 1975- March 1976), UCLA Nutrition Assessment Unit, Division of Population, Family, and International Health, School of Public Health, University of California, Los Angeles, in cooperation with the US Agency for International Development and the Government of Liberia, 1976. 17. HABICHT, J. P. ET AL. Height and weight standards for preschool children: how relevant are ethnic differences in growth potential? Lancet, 1: 611-614 (1974). 18. CENTER FOR DISEASE CONTROL. Ten-state nutrition survey reports, IV, Atlanta, US Department of Health, Education, and Welfare, 1972 (Publication No. (HSM) 72-8130).