ICTIwfifirw III! III!!! IIII. I'M IIIIIIII I'n'I /!!!I .' IIIL 'I'.‘f"I !I III IIIIIII WK! In!!!l;! IdIIIlI'!!!I'!!I|! ' "fl ! ' ENE 'II .I .. IMWI! .I! III"1 II"! !II!II.I» I!!! IIIIIJ! I II I "III!!! "I “ III! . . i. I!" 'IIIHHIII qt!!! I!!! III M W I !! II! I! 1!!!! !IIIIIIII5 I III!! I!!II!III! I I. ’ If!" “I... ! I!!! I!!! I!!! I! ' !.!'!I!' II!!! M 'It’! I'I!!I!I!! II: IIIIII l!!!’ IIIIIIIII !I II ' !|!!!!I ! I'II !I!!II., II I ‘I. -. I”!Wl‘yh I, !W I h!!!! . 7.. ‘ I! ' I . I! . AIIJWIMI WIIIH n!!!!WII!I!IHM! I IIIIIII IIIII III!IIIIIIIIIIII ! I!IIII III! III! !MIhI . at 6,14%”) III!) .. $§Wuf§w . b "IIIEL v.4 "55' ’5‘ mi! 4»; 'I“ III ,.:.II, " III§£95dEI . f..I!!.!!.I'L.!~.:!;1:’! Inn-“b: "$19.1: ' J Iifi. Ifi!€3§”fi“g&%§! ffifil;§§%éfigflf = If?! “WI“1!!!3§!%iv‘!i§!§’z Si ' I 233‘ . ! , !. h; fHIIb!’ : «15‘; = . IIIIIIi-III. ”I???" “‘ Auufif I i! I? 2.13% ifi '. I to; LhHM JR wh‘.‘ m4 —- :I‘IIIH Q M I!!! “III!!!“I!!! W! MI I'HW _. . I... IIIII. i!! ! "I be'a eil ' J i?!“ fi'.‘ #‘ ”I! 'g ,?"I“” ILISIIIEVL 'r" 'fi fi I'EI‘I' 53' ”MI - o o— '.-' .' . . .- ’10 .."” ' .f0 .If .v ‘ .. p a I ‘ -‘ II II n . .,II"'I! IIHI | 'Il!!! I!I!!I II “HI I I:I «IQ! I!!!!..I,t\ -: l‘. I III IIII ! .III III ! -! I I! .I I I! !!!.|!!! II!. III!!! III .!!!!I!! III !! !I!!!I II II! .I-III “III J!!! !!!!!I 1.3....3. l!” ”1!!!!I !!!!I!I!!IH' ! !!I I!!! !!! “3!! II‘IIIEIjIJ !!I 5.1! fiIm !r!'!Ir.::. %: ~ .IVII‘II III‘I. l I! II ! I ”III!!!” III II !I!II!‘I h!!! !' I!I III; 'IIIII II I! III! I!!:!I!!I‘I IN“ I!” IIIIJ! “I L 3 v . . II!!!I.II‘I‘« II!!!I !!II!III IIIIIIIII. III‘III; II I III I‘I‘IIIII . ' .I'Iri":"!|!! ,!'!III;.I IIII!!II.II"III IIII'II'!III I II I-.: . .II .I, IHIII-I I!!! III II'I'II III !I!I| I'III!III II II!I! II III.I ITJI’IIUI II !I!II IIIII IIII IIIIII II!IIIIIIIII I III . IIIIII IIIIII IIIIIII-nI IIIII !IIIIII! IIIIIII! !!I MiSu' ‘ ’Q‘. — v... — ., , On .4. III II I . .. IfI. I! .I!l I" I!!! I. ”II! I!' !!‘!I II“ M ' II 5:"?!- IIIInIIH” “H“ ! I”! W!"h‘%”ImUM I.” !!!Iflm!!bfl!w!%!'fi . :‘:_ .~ IIIII .I! " “I ‘I!i“‘. HIIII‘” IIIII !! IIII!I!.I”~"!!III! I!!I'!III!I . I!HI. IMMH ,IIquM I .!III . , . : . I! II'”! “M .III WI IIIIIP III III‘! !I' L I II!I !!I I! I!!! !!!I -! I ! !l:!!‘ IIII I III !I!| .IIIII II I'II'I'.“ II III IIIIIIIIII !,!II!I!I . 0" IIIIII III !!!III IIIII !!!! I!!!II!! !!I.' III III III: ' 'i‘ w ! “II ."I "I!!! I!!! I!!!” I!'..I."!III!‘ III!!! ! !!I! ! _——.__._____;‘.__:2_ -: v I. w ! IHIw ‘IM. HI! . .:‘I“ '! HII‘III% II!IW!I “W‘ d'! IIIHIHII WI :"'MI I!!! III!!!” IW I H!MX I '!!'!!!E!"!li! ‘13:!!! I!!! ‘ ’ «1!!!! I25“! .II!!,!!!!!!I!!! I!!! I!!!" “'7'“ III I 11"'.'!'!!Ii !:!I"I ll ! I I ' |"3! II "~ ' I. ‘! w! ! II‘ .II , 992:1! ~.'.._ ! h_-!~!II!I!!~!LAU.J.AUM .!I!LJ I!u!!rM!mIII!II!I. !!!!!!!! I!n!!!!!!!. .m!! llllllll llllllllllllll\l . “my S Michigan Sac g University This is to certify that the thesis entitled A COMPARISON OF NUTRITIONAL FACTORS IN SELECTED CHILDREN WITH AND WITHOUT IRON DEFICIENCY ANEMIA presented by Helen Dersch has been accepted towards fulfillment of the requirements for .Ma_S_t;er;S_degree in_Human_Nutrition 1/ éé/ 41151:” M Major professor Date /[’(/'77 0-7639 OVERDUE FINES ARE 25¢ PER DAY PER ITEM Return to book drop to remove this checkout from your record. A COMPARISON OF NUTRITIONAL FACTORS IN SELECTED CHILDREN WITH AND WITHOUT IRON DEFICIENCY ANEMIA By Helen Dersch A THESIS Submitted to Michigan State university in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Food Science and Human Nutrition 1979 ABSTRACT A COMPARISON OF NUTRITIONAL FACTORS IN SELECTED CHILDREN WITH AND WITHOUT IRON DEFICIENCY ANEMIA BY Helen Dersch The work described in this thesis was a part of an interdisciplinary study investigating some of the physio- logical, sociological, psychological, and nutritional factors which may contribute to the occurrence of iron deficiency anemia. Emphasis in this thesis was on nutritional aSpects. Sixteen infants 6 to 24 months of age from low income families were included in the thesis group. They were placed in iron deficiency anemic or non—anemic control groups on the basis of hematological values. Data collect- ion was conducted during the fall of 1971 and winter of 1972. Results indicated generally adequate diets in all nutrients except iron. Iron intake was significantly higher for the control group but both groups fell well below the recommended dietary allowance for iron. Mothers of control children had significantly greater nutritional knowledge. Families of anemic children had significantly lower per capita income and larger number of children per family. ACKNOWLEDGMENTS The author wishes to express her thanks to her committee chairman, Dr. Dorice Narins, Department of Human Development, and committee members, Dr. Dena Cedarquist and Dr. Dale Romsos, Department of Food Science and Human Nutrition, for their guidance throughbut‘ this study, to the mothers and children interviewed for their willing participation, and to her husband, Eckhart, without whose patient understanding this thesis could not have been completed. ii TABLE OF CONTENTS INTRODUCTION Purpose of Study REVIEW OF LITERATURE Iron Metabolism Iron Coumpounds in the Body Absorption of Iron Development of Iron Deficiency Anemia Standards for Determining Iron Deficiency Anemia Effects of Iron Deficiency Prevalence of Iron Deficiency and Iron Deficiency Anemia Physiological Factors Sociological Factors Psychological Factors Summary EXPERIMENTAL PROCEDURE Selection of the Sample Medical and Laboratory Evaluation Home Interview Statistical Analysis Informed Consent FINDINGS Laboratory Data Demographic Data Nutritional Data Diet History Food Attitudes Nutrition Knowledge 24-Hour Dietary Intake Records Shopping Patterns Parent Attitude and Expectations DISCUSSION AND RECOMMENDATIONS Laboratory Data Demographic Data Race Sex Socio-Economic Level Psychological Factors Nutritional Data Diet History Nutrition Knowledge iii Page Table of Contents Page Nutritional Data (con'd) Food Attitudes 62 Shopping Patterns 63 Intake Data Compared to RDA 64 Food Sources of Iron, Fortification, and 67 Utilization Summary 69 LIST OF REFERENCES 70 APPENDICES Appendix.A: Iron Deficiency Anemia Study 75 Physical Examination Form Appendix B: Iron Deficiency Anemic Questionnaire 76 Appendix C: Nutrition Knowledge Questionnaire 87 Appendix D: Food Attitude Questionnaire 93 Appendix E: Parent Attitude Questionnaire 97 Appendix F: Parental Expectations Questionnaire 101 Appendix G: Consent Forms 103 Appendix H: Text of Iron Deficiency Pamphlet 105 iv LIST OF TABLES Table Page 1. Distribution of Iron in the Body 3 2. Stages of Iron Deficiency 15 3. Physiological Characteristics 36 4. Familial Characteristics 37 5. Diet History 40 6. Food Frequency- Percent of Children Served Each 42 Food Category per Week 7. Mothers' Favorite Meal- Food Categories Included 43 8. Nutrition Knowledge Scores- Number Correct 45 9. Nutrition Knowledge Scores- Right Minum %-Wrong 45 10. Percentage of Correct Answers to Grouped 46 Nutrition Questions 11. Mean Daily Nutrient Intake— Percent of RDA 47 12. Sources of Iron- Mean Percent of Total Daily 48 Iron Intake 13. Milk Consumption- Mean Daily Intake 48 14. Combined Responses to Parent Attitude Categories- 52 Percent Agree and Disagree Responses INTRODUCTION Purpose of the Study Iron deficiency anemia is the most prevalent nutri- tional deficiency in the United States and continues to be as common today as 30 years ago (1). While iron deficiency and iron deficiency anemia may occur in any segment of the population, certain groups seem to be more at risk than others. In these high risk groups, iron deficiency does not occur alone but as part of a constellation of medical, social, and economic factors (2). Although much research has been done on iron deficiency anemia, relatively little of that research has investigated the social and ecological factors which lead to iron deficiency in some children with- in an environment where others within the same family are not iron deficient (2). Because of the complex etiology of iron deficiency anemia, a multidisciplinary project was undertaken to investigate some of the medical, physiological, nutritional, and socio—psychological factors which may contribute to the occurrence of iron deficiency and iron deficiency anemia.1 The age group of children 6 months to 2 years of age was chosen because of the high reported incidence of iron deficiency anemia in this age group (4, 5, 6, 7). The 1 Principle investigators of the project, of which this thesis was a part, were Theresa B. Haddy, M.D., David J. Kallen, Ph.D., and Dorice C. Narins, Ph.D., of the Department of Human Development, College of Medicine, Michigan State University. A report of the larger study has been made (3). 2 study included physical examinations of the children, collect- ion and evaluation of medical and hematological data for in- cidence of iron deficiency anemia, and appropriate treatment of those children with iron deficiency. Families were inter- viewed for demographic and sociological data, and for the mother's attitudes concerning homemaking and child rearing. The nutritional information included collection and evalua- tion of data regarding past and present nutritional status of the children and the nutritional practices of their families. The emphasis of this thesis was on the nutritional aspects of iron deficiency and iron deficiency anemia, although some other data were also included. The objectives of this thesis were to investigate and compare: (1) the nutritional adequacy of diets of children with and without iron deficiency anemia; (2) the food patterns and habits of the children and their families; (3) the level of nutrition knowledge of the mothers; and (4) some of the parental attitudes toward child-rearing. ggyrsw OF LITERATURE Iron Metabolism Igon Compounds in the_§ody Due to its ability to undergo reversible oxidation and reduction, iron functions in the body in oxygen tranSport and cellular respiration (8). There are a number of iron compounds in the body. Hemoglobin is the most prevalent and, with myoglobin, a similar compound present in smaller quantities, account for 70-75% of total body iron. The storage forms or iron, ferritin and hemosiderin, account for about 15—25% of body iron. Iron is also found in serum transferrin and as a component of a number of cytochromes and tissue enzymes (8, 9, 10). Table 1 shows the distribu- tion of iron by approximate percentages of total body iron. Table 1: Distribution of Iron in the Body(10) Iron Compound Percentage of Total Body Iron Hemoglobin 67 Storage Iron 27 (Ferritin, Hemosiderin) Myoglobin 3.5 Labile Iron Pool 2.2 Enzyme Iron 0.2 Transport Iron 0.08 The exact amounts of iron in the body flucuate with storage iron subject to the greatest flucuation (10). The largest proportion of body iron is in hemoglobin. The hemoglobin molecule consists of 4 polypeptide chains each with an attached iron-porphyrin (heme) group. The iron 3 4 content of hemoglobin is constant; iron deficiency results in decreased hemoglobin synthesis. The formation of red blood cells (erythropoiesis) is controlled by the hormone erythropoietin. Changes in hemoglobin mass such as would occur with transfusions, hemorrhage, or erythrocyte destruc- tion, results in a compensatory change in erythrocyte syn— thesis in the bone marrow (10). Most of the iron in hemo- globin is conserved when the red blood cells are catabolized and the iron is then available for reutilization (10). This reutilization of iron accounts for more than 90% of the daily bone marrow iron requirement (11). Myoglobin, an iron-containing heme compound similar to hemoglobin, is found in muscle tissue where it apparently functions to provide oxygen during muscle contraction. The myoglobin content is greatest in muscles Specialized for sustained contraction, when blood flow is most dimished. However, the contribution that oxygen from myoglobin makes to total oxygen consumption is relatively small (12). Iron is stored primarily in the liver, Spleen and bone marrow. There are two main forms of storage iron, ferritin and hemosiderin. Some researchers have suggested that iron may be stored first as ferritin and as the level of iron increases to a critical value, further storage is primarily as hemosiderin (13). Both ferritin and hemosiderin seem to be readily available for use in the body. It has been suggested that ferritin also has important regulatory and transport functions in the mucosa of the small intestine and possibly in the placenta (10). 5 Iron is transported by the iron-binding plasma protein transferrin, whose major function appears to be the transport of iron to the bone marrow (10). The amount of iron—binding sites which contain iron are measured as percent transferrin saturation. In the normal adult, approximately one third of the iron-binding sites of transferrin contain iron, giving a normal transferrin saturation of about 32% (10). In iron deficiency, the amount of transferrin-bound iron is usually diminished (10). Iron is also important in tissue enzymes, although the amount of total body iron represented by enzymes is small. There are 3 types of iron enzymes: (1) enzymes and cytochromes which are heme—proteins; (2) iron-flav0proteins; and (3) enzymes requiring iron as a cofactor. In the heme enzymes and iron-containing cytochromes, iron is a part of the prosthetic group at the active site of the molecule. In other iron enzymes, the function of iron is less clear (10). Absogption of Iron Absorption of dietary iron is influenced by a number of factors. These include the form of iron present in food, the presence of other food substances which can solubilize, chelate, or precipitate iron, the quantity of iron in the diet, and the body's need for iron. Iron is present in foods in two basic forms, ferric- iron complexes and heme compounds. These are absorbed by different mechanisms and affected differently by various other factors within the gastrointestinal tract (14). 6 In the stomach, hydrochloric acid helps keep ionic iron in solution; iron absorption decreases if there is a lack of gastric acid as in the case of achlorhydria. Muco- protein substances help form soluble iron chelates which remain in solution as food passes into the alkaline medium of the small intestine (10, 14). The exact mechanism by which iron is transported across the intestinal mucosa is not completely clear despite inten- sive investigation (10). Iron is absorbed primarily in the duodenum and upper jejunum. The duodenum is the only portion of the intestine able to transport ferrous iron against a concentration gradient. This portion of the intestinal mucosa contains the most ferritin, although the role of ferritin in the tranSport process has not been definitely determined. Iron absorption is a Z-stage process. In the first, relatively rapid step, iron enters the mucosal cell. There the iron accumulates and is either distributed through- out the cytoplasm or incorporated in ferritin. The second, slower step, is the release of iron into the blood stream. This transfer from mucosal cell to plasma is regulated by the body need for iron. Iron is carried by the plasma globulin transferrin (15). Because of the short life-Span (2-3 days) of mucosal cells, iron remaining in the cells as they are sloughed is lost (15). Plasma iron can be also taken up by the mucosal cell, retained, and ultimately sloughed into the intestinal lumen. This suggests an excretory role of the mucosa, but 7 the capacity of this mechanism is limited (10). In absorption of heme iron, the metalloporphyrin por- tion is split from the protein by digestion and is absorbed directly into the mucosal cells. Substances which affect absorption of inorganic iron do not seem to change the absorption of heme iron which is absorbed best in an alkaline medium (14). However, some researchers have stated that heme iron (or other complexed iron) may not be as efficient- ly used in hemoglobin synthesis as ionizable iron (16). Observations support the concept of some kind of mucosal regulation of the rate of iron absorption. Addition- ally, the level or iron stores in the body influences the rate of iron absorption, and it has been suggested that the level of saturation of iron-binding proteins also plays a role in iron absorption rate (10). There may also be a re- lationship between iron absorption and rate of erythropoiesis, due either to the effect of erythropoiesis alone or the ac- tion of erythropoietin which stimulates erythropoietic activ- ity (10). In general, iron absorption is increased in iron deficiency anemia. The total quantity of iron in the diet is the most im— portant single factor in determining the amount of iron absorbed. The greater the quantity of iron in the diet, the greater the amount which is absorbed (up to toxic amounts) although the percentage of dietary iron absorbed decreases as quantity increases (8). Both the form of iron in various foods and other sub- stances within foods consumed at the same time affect iron 8 absorption. Heme iron is better absorbed than inorganic iron, and ferrous iron better than ferric (14). Dietary constituents which solubilize iron increase iron absorption (ascorbic acid, sugars, amino acids); compounds which cause molecular aggregation or precipitation of iron decrease absorption. Carbonates, oxalates, phytates, and phosphates all can combine with iron to form insoluble or poorly absorb~ ed cemplexes (8). Thus there is considerable variation in availability of iron depending on the composition of the total meal with which it is consummed as well as the type of iron—containing food. Iron absorption is decreased with increased bulk of the diet (17). Interaction between foods modify the absorp- tion of iron compared to iron absorption from individual foods (18). Phosphates and phytates decrease absorption of inorganic iron by precipitation (14). Ascorbic acid increases iron absorption (17); orange juice will increase absorption of non-heme iron such as from eggs and cereals. But egg, (considered a good source of iron in diets of infants and children) decreases absorption of non-heme iron from other foods (8, 16). Addition of animal foods to a meal may in- crease absorption of iron from vegetables sources (14). Meat increases iron absorption of some foods, with cysteine shown to be of importance in the enhancing effect (19). Calcium may interfere with iron absorption since iron absorp- tion increases if milk is absent or reduced in the diet, according to one report (16). And a study with rats indicated that excess calcium may inhibit iron absorption but the 9 authors cautioned that it remains to be seen if the rat is a suitable model for absorption studies (20). Another re- port suggests that wide variations found in the availability of iron from the same sources but between laboratories may be due in part to assay methods rather than true differences in availability of the iron (21). Absorption of dietary iron is generally believed to be about 10% for individuals with normal iron status and 20% for iron deficient subjects. However, some researchers suggest that these figures are too high and that absorption may be generally less than 10% for normal persons and only somewhat over 10% for iron deficient patients (8). Studies of iron absorption from single foods gave values of less than 10% from vegetable sources and less than 13% for most animal products. Using a mixed "Western" diet, iron absorp- tion values were 6% for males and 14% for females with normal iron status and 20% for iron deficient subjects, but less even for the iron deficient subjects when a predominately cereal diet was used (19). There is also question about the effective absorption of iron from fortified foods. This may be particularly true in infants. There is some evidence that during the first 4—5 months of life only a small amount of dietary iron is incorporated into hemoglobin and has little effect on hemoglobin level of full term, normal birth-weight infants (22). (In the case of premature and low birthweight infants, the benefit of early iron supplementation is more widely recognized.) Little is known about iron absorption in 10 infants of either sodium iron pyrophOSphate or metallic iron, forms commonly used for fortification, according to one study (23). Another report states that some of the com- monly used salts are poorly absorbed, although ferrous sul— fate is better absorbed than ferric orthophOSphate and so- dium ferric phyrophosphate (19). Other studies of iron absorption from cereal products indicate that natural iron in whole meal flours may be absorbed somewhat better than iron added to fortify white flour (15). Development ogfiIron Defigienpy Anemia Iron deficiency anemia is generally considered to be a nutritional deficiency where dietary intake of iron is insuf- ficient to meet body needs. Prolonged dietary intake below the level necessary for replacement of iron losses may lead to eventual iron deficiency in any individual, but both in- creased need and excessive loss of iron put some people at greater risk than others. Increased need is generally the result of increased growth rate, and occurs during infancy, especially the period of 6 - 24 months of age, adolescense, and pregnancy. The pregnant adolescent is at particular risk due to demands for her own growth in addition to that of the fetus and maternal tissues associated with pregnancy. Pregnancy increases need for iron for both maternal and fetal tissues. Total iron demand during pregnancy is about 1000 mg (8). Iron deficiency anemia is the most common complication of pregnancy and is frequently due to inadequate dietary iron intake. Routine iron supplementation is 11 generally recommended, particularly during the second and third trimesters (24). However, iron deficiency anemia of the mother, unless very severe, does not seem to produce anemia of the infant (25). There is no difference in frequency of iron deficiency anemia among infants born to iron deficient mothers compared to those born to non-anemic mothers (10). Infants of iron deficient mothers show normal hemoglobin levels at birth, although evidence of low iron storage levels may become apparent later in the first year of life (25, 26). Maternal hemoglobin concentration does not affect the infant's iron stores unless the mother's hemoglobin level falls below 9 gm/dl (27). Iron supplementation during pregnancy does not seem to influence hemoglobin levels in the infant (25). Iron transport across the placenta to the fetus appears to be an active process against a concentration gradient. The exact transport mechanism is not clearly understood, but there is some indication that ferritin may play a role (10). Iron transfer to the fetus does not take place at a uniform rate throughout the gestation period, but increases during the latter half of pregnancy and particularly in the third trimester (8, 25). At birth the normal, full term infant has about 300 mg of iron, the majority as circulating hemoglobin. Normal hemoglobin levels range from 13-20 gm/dl with the mean 16-17 gm/dl. An average of about 70% of this is fetal hemoglobin and 30% adult hemoglobin. During the first few weeks of life, hemoglobin level drops, reaching its lowest level of about 12 10—11 gm/dl at 6-10 weeks of age (25). This decrease in hemoglobin level, termed the physiologic anemia of the new- born, is due to the Shorter life Span of fetal red blood cells, about 2/3 that of adult RBC, a slightly increased rate of RBC hemolysis, and Stationary or decreased rate of erythro— poiesis. The resulting fall in hemoglobin and RBC levels allows adjustment to the increased oxygen saturation of blood with lung rather than placental oxygenation (25). The physiologic anemia is self-limited and does not respond to iron or other nutritional supplementation (28). When hemoglobin concentration falls to about 11 gm/dl, erythropoiesis begins to increase. The iron need for this rise in hemoglobin synthesis is met by the iron stores including prenatal accumulation and iron from RBC degrada- tion during the first few weeks of life (25). Iron stores of the full term infant are usually sufficient for normal erythrOpoieSiS for 4-6 months without dietary supplementation (28). Individual variability of infants in growth and iron stores determine how quickly iron is depleted, but by the age of 6 months the infant is usually dependent on dietary intake to meet iron needs. Hemoglobin level normally remains fairly constant about 10—12 gm/dl during the first two years of life (25). Since the greatest quantity of iron is stored by the fetus during the last months of pregnancy, premature birth results in smaller iron stores for the infant. These stores are rapidly depleted and the premature infant is therefore 13 at greater risk of developing anemia, which is common and may be severe (10). Low birth weight infants also have in- creased risk of developing iron deficiency anemia because of smaller iron stores and rapid growth (25). Hemoglobin level is lower in the premature than the full term infant and fetal red blood cells have an even Shorter life Span (28). In addition, the bone marrow of the premature infant is im- mature, and the capacity of the hematopoietic system is in- adequate for the requirements of rapid growth. The degree of immaturity of bone marrow is related to the degree of prematurity of the infant (29). The decrease in hemoglobin level resulting from physiologic anemia is more pronounced in premature than full term infants (28). Supplemental iron therapy is generally successful in treating iron deficiency in premature and low birth weight infants(10). Iron is well conserved by the body and daily loss of iron is relatively Small, averaging 0.5-2 mg/day (27). Iron is normally lost primarily through desquamation of gastrointestinal mucosa and some excretion of iron in bile. Urinary loss of iron is usually less than 0.1 mg/day, and loss through sweat and sloughed Skin cells is not likely to exceed 0.15 mg/day under normal circumstances. Adolescent girls and women of reproductive age have the additional loss due to menstruation of an average of 0.5-1 mg/day. Children have smaller iron losses in feces, urine and sweat than do adults (27). Greater iron losses, however, may lead to development 14 of iron deficiency anemia. ExcessiVe losses of iron result most often from blood loss such as hemorrhage or heavy men- struation. Intestinal bleeding is one avenue of blood loss which may result in chronic loss of iron. It has been sug- gested by some that intestinal blood loss may be a major factor in iron deficiency anemia in infants (10), but the Committee on Nutrition reported that such blood loss in in- fants was negligible and may be a secondary gastrointestinal manifestation of iron deficiency rather than a cause (21). Cow's milk has been Shown to contain a substance which in- duces gastrointestinal bleeding in infants, but the substance is thermolabile and thus heat processed milk and formulas should not cause bleeding (30). The development of iron deficiency anemia is a progres- Sive decrease in total body iron. Storage forms are depleted, evidenced by the absense of bone marrow iron. Then follows a gradual decline in serum iron concentration and concomitant rise in serum iron binding capacity. Transferrin saturation of 16% or less is characteristic of iron deficiency. Red blood cell changes characteristic of iron deficiency occur with the appearance of hypochromic microcytosis (28). Hemo- globin synthesis and erythropoiesis are decreased in iron deficiency and the hemoglobin concentration of erythrocytes also decreases (10). Table 2 shows the stages of deficiency in development of anemia. .mOGH Cfi COfiHMflUOmmaQ HNUHUQZ CMUflHmEAN 15 one no :ofipfluusz pcm poom no Aflocsoo one an pmnmoum mos cofiDMUSMflmmmao mass m Ufisounu leaks on Ufleounu mmoa Ho Xmfi 30H :OEuo: haemSms Seaman: ucmmnm mucmfiufimmp couH mmwa no Xbfi 30H wcoc SHHMSmS haamsms uzmmnm hocwfluflmmp couH Dammnm no one: HmEuo: Hmsuoc pmmmmnomp soapoaowp couH masons :oflumusumm Sfluummmcmue couH Esnmm monoum SouH mufiumfinmuumumno hocwflowmwn mo wmmum AoHv mNocmfioflmmn conH mo mommum .m wanna 16 Aberrations in absorption or utilization of iron may also lead to iron deficiency. The emphasis in this study however, was on iron deficiency caused by low dietary iron intake coupled with increased iron demand for growth. Standards for Determiningigon Defigiencv Anemia Although iron deficiency is considered by many to be a major public health problem, there are differences of opinion as to what hematological levels constitute iron deficiency and iron deficiency anemia. The Ten-State Nutrition Survey set hemoglobin levels for infants 6-23 months of age as less than 9 gm/dl "deficient"; 9.0-9.9 gm/dl "low"; and above 9.9 gm/dl "acceptable" (31). Other studies have considered hemoglobin of less than 10 ngl (32) or 10.5 gm/dl (33) as evidence of iron deficiency anemia, and still others regard less than 11 gm/dl as "anemic" (6). Likewise there is variation regarding transferrin levels. The Committee on Nutrition considered that trans- ferrin saturation of less than 15% as indicative of body iron depletion (22). One author stated that erythropoiesis is impaired when transferrin saturation is less than 18% (18), while another researcher used a 20% saturation level as in- dicating low body iron stores (33). Elwood stated that ideal values are unknown and the accepted "normal" values are either empirically determined or based on findings in selected groups of individuals (16). Since it was found that some apparently normal infants showed a substantial in- crease in hemoglobin level in reSponse to iron therapy (34), some researchers have suggested that such a response may be 17 a better indication of iron deficiency than any Single laboratory value (35, 36). Effects of iron Deficiency Common clinical manifestations of iron deficiency and iron deficiency anemia include pallor, fatigue, and increased irritability as well as characteristic hematological changes such as decreased hemoglobin, hematocrit, mean corpuscular volume, Serum iron, transferrin saturation, and increased total iron binding capacity (17). There has been relatively little information about the pathologic changes which occur in most tissues of the body. Greatest attention has been given to changes in bone marrow Since this tissue is readily available for study through biopsy. Although there is usually increased erythroid activity in the bone marrow of iron deficient patients, this is not always the case and the correlation between severity of anemia and degree of erthyroid hyperplasia is poor. The morphology of bone marrow of iron deficient individuals is not sufficiently distinctive to be of diagnostic value (10). Gastric achlorhydria is frequently associated with iron deficiency as are mucosal abnormalities of the stomach and epithelial changes of the tongue and esophagus (10, 13). Some authors have suggested that the gastric atr0phic changes occurring with iron deficiency are of significance in the origin of the deficiency state while others take the view that gastric atrophy is a result of prolonged iron deficiency rather than its cause (10). Some studies have found no correlation between the severity of the anemia and the 18 presense of changes in gastric mucosa. The results of iron therapy to patients with mucosal atrophy is mixed in terms of improvement of histological appearance of gastric mucosa or ability to secrete gastric acid (10). The presense and intensity of various symptoms often associated with iron deficiency anemia are not directly correlated with the severity of the anemia (10). Symptoms of anemia such as pallor, fatigue and weakness depend mainly on the rate of development of the anemia rather than its severity. In slowly developing anemia the body may gradually become adapted to anemia Showing fewer symptoms, whereas a relatively sudden onset of anemia may produce more definite symptoms (11). Particularly in the case of slowly developing mild anemia, compensatory processes may lead to more efficient release of oxygen from hemoglobin, with few noticeable symptoms (14). In slowly developing anemia, symptoms may be due to adaptation to the decreased oxygen-carrying capac- ity of the blood, such as acceleration of heart and respir- atory rate and increased cardiac output. Symptoms of fatigue may be related to decreased oxygen transport, pallor to de- creased blood volume and hypochromic blood, and palpitation to increased cardiac output (37). The decreased hemoglobin levels may play a part in producing Symptoms common to all anemias such as weakness, pallor and dizziness (10). In infants, the early stages of development of iron deficiency are usually unaccompanied by clearly recognizable symptoms. Even low hemoglobin levels are tolerated if development is Slow (38). 19 Although it was once thought that enzyme iron was not reduced significantly in iron deficiency, careful studies have Shown that the activities of several iron enzymes are definitely affected by body iron depletion. Iron deficient individuals have been Shown to have a deficiency of the iron enzyme cytochrome oxidase and a rapid return to normal activ— ity of this enzyme followed iron therapy (10). Other enzymes shown to have diminished activity in iron deficient persons are aconitase and xanthine oxidase (10). Decreases in cyto- chrome c in heart, kidney and muscle tissues in iron defi- cient rats (39), cytochrome c and catalase in rats and pigs, and cytochrome oxidase, aconitase and succinic dehydrogenase in rats have been Shown by several studies (10, 40, 41). Those enzymes which show such sensitivity to iron de- ficiency may decrease in activity before the characteristic hematologic changes of anemia are evident, as well as in cases of iron deficiency without anemia (10, 40). One study found a decrease in cytochrome oxidase level in one half of patients with iron deficiency without anemia and in three- fourths of patients with iron deficiency anemia (42). The consequences of changes in tissue enzymes are not yet clear, but iron has an important role in cellular metab- olism: nearly half the enzymes and cofactors in the Kreb's tricarboxylic acid cycle either contain iron as an integral part of their molecular structure or require its presense (10). The influence of iron deficiency on tissue enzymes may cause symptoms such as headache, fatigue, palpitation, and changes in fingernail and epithelial tissues (10, 17). 20 Epithelial changes such as atrophy of the papillae of the tongue and angular stomatitis may be caused, at least in part, by diminished function of iron containing respiratory enzymes (14). Researchers do not completely agree on possible lasting effects of iron deficiency or iron deficiency anemia, eSpe- cially in cases of mild deficiencies which are most pre- valent. (See section on prevalence.) One author Stated that "no deleterious effect of mild iron deficiency has... been demonstrated" (16). Another study found that infants with iron deficiency anemia Showed no decrease in either weight gain or linear growth when compared to normal infants (1), but other reseachers found that iron deficient children tended to be underweight and Slightly shorter in linear growth, and that normal growth patterns returned following adequate iron therapy (10). Another study found that those children ages 1-6 years who were below the 25th percentile in height had lower hemoglobin and transferrin saturation levels than did children whose heights were above the 25th percentile, and suggested that iron deficiency may be asso- ciated with measureable uncerachievement in somatic growth (43). Since this latter study included children already past the time of highest incidence of iron deficiency (6-24 months), these findings may reflect effects of a longer period of iron deficiency than studies limited to infants. Research has also been conducted on possible effects of iron deficiency anemia, as well as malnutrition generally, 21 on mental development. One study of children 4—6 years of age with mild iron deficiency anemia found that attention Span and ability to concentrate on mental tasks were lower in anemic than non—anemic children, although intellectual potential as measured by I.Q. tests was not affected (44). Another study reported Significant differences between child— ren with low hemoglobin and those with normal hemoglobin levels on some intelligence tests and associative reaction time tests. It was not determined whether this lower level of achievement of the iron deficient children resulted from lower learning ability, from lower or less sustained level of motivation, or greater fatigue as the testing progressed (45). Another study of children in which diet history records in infancy were taken, found that even mild deficiencies in nutritional adequacy during infancy may be related to dif- ferences in mental abilities on subsequent mental tests taken at beginning school age (46). Prevaience of Iron Deficiency and Iron Deficiency Anemia Many studies have investigated the prevalence of iron deficiency and iron deficiency anemia in various segments of the population. Although Specific figures vary, there are similarities in the findings. Physiolpgical Factors Growth A number of studies have found that a higher incidence of iron deficiency and iron deficiency anemia occurs during the rapid growth period of about 6-24 months of age than in other children, until the adolescent growth period is reached 22 when the incidence of iron deficiency and iron deficiency anemia again increases (4, 5, 6, 7). One researcher, Speak- ing of malnutrition generally (of which iron deficiency is the most common type among infants and children under age 2), stated that growth rate was a major factor, either in precipitation of malnutrition by rapid growth or in pro- tection against malnutrition by slower growth. Further, borderline nutritional status converted to manifest mal- nutrition with the added stress of rapid growth (7). The incidence of iron deficiency, determined by low transferrin saturation levels, was found to be considerably higher than the incidence of iron deficiency anemia, deter- mined by low hemoglobin values. One Study reported a 5% incidence of anemia and a 35% incidence of low transferrin saturation (33); another Study found anemia in 6% and low plasma iron in 50% of children studied (47); atthird found that the incidence of low transferrin saturation was more than 1.5 times as frequent as identifiable anemia (18); and yet another suggested that iron deficiency may exist as a "subclinical" or marginal deficiency state in Spite of "normal" hemoglobin concentration and diets apparently adequate in iron (5). .1322: Racial differences in incidence of iron deficiency also seem to exist. The Michigan nutrition survey showed a high- er incidence of low hemoglobin values for blacks in the 0-6 year age group as well as for all ages (children and adults) combined (31). Mickelsen et al found somewhat lower hemoglobin 23 and hematocrit values in black Head Start children than in their white classmates. Additionally, he cites other studies where Similar racial differences have been found and quotes one researcher that "it would appear that iron deficiency is at least twice as prevalent" among black as among white children. Mickelsen states that while racial differences seem to be consistently observed "causal factors remain un- explained" (6).1 However, Garn suggested that there may be genetic differences in hemoglobin levels between blacks and whites such that comparing blacks to the "white" standard for hemoglobin may result in a higher apparent incidence of iron deficiency anemia in blacks (48). Sociological Factors Economic Levei Several studies Showed that differences in the incidence of iron deficiency anemia were found to correSpond to econom- ic level. For example, one study found a 6% incidence of anemia among children 12-23 months of age whose families rep- resented an economic cross-section, but a 20% incidence when only those children in the study from families economically classified as below the "poverty" level were considered (47). Other studies have cited the incidence of anemia as high as 56% (7) and 64% (18) among disadvantaged children. However, one study of 9 month old patients of private physicians, pre- sumably from middle class families, revealed that 14% of the 1 Racial differences may be related to socio-economic factors which are important in prevalence of iron deficiency anemia. (See section on sociological factors.) 24 infants had iron deficiency anemia (32). Owen summarized his study by stating that iron deficiency was fairly common among preschool children regardless of socio-economic status, although the incidence was probably greater among the poor, and that additionally, it appeared that the preschool child without iron deficiency anemia was frequently iron deficient (47). I Education and Nutrition Knowledge Several studies have found correlations between the nutritional status of children and the general education level or the nutritional knowledge of the mother. Filer and Martinez found that in general, infants consumed less of several nutrients, including iron, when the average education of the parents was lower (5). Garn Stated that low hemo- globin concentration was related to the educational level of parents and to the mothers' knowledge of food (48). Other studies (49, 50) found that, although health and nutrition reasons were often given as a basis for food choices by school children and adults, there was a "disturbing reliance on erroneous nutrition information" (49). Wang, using a nutrition knowledge questionnaire, found that higher levels of education and greater varities of experience seemed to increase knowledge of nutrition and that the poor knew less about nutrition and the nutritional value of foods (51). Food Preferences Studies of food preferences of children have supported theories that early home environment is instrumental in 25 development of food attitudes. A child's attitudes toward food may be formed at an early age and continue to influence his food choices in later life. The range of the child's experience is limited by parental food attitudes, particular- ly those of the mother who is most influential in development of the child's food habits and behavior (49). PsycholggicaiFactors Parental Attitudes toward Child Rearipg It has long been recognized that infants who lack ade- quate maternal nurturing may fail to grow and develop normal- ly. This failure has been attributed to lack of social, psychological and sensory contact and Stimulation (52). Hepner, in his study relating quality of "mothering" to malnutrition, found a very strong correlation between serum vitamin A and/or hemoglobin and mean corpuscular hemoglobin concentration values, and the quality of mothering. He con- cluded that adequate mothering was protective to a child even under the combined stress of rapid growth and low nu- trient intake, but inadequate mothering precipitated malnu- tritioniJithe rapidly growing child even with a more adequate and balanced diet. He stated that despite a mother's best intention and desire to perform adequately, social and eco- nomic situations beyond her control might interfere with the maternal-child relationship needed for successful child de- velopment (7). Whitten and Fischoff, however, have suggested that some problems of growth failure in infants are due not so much to the socio-psychological effects of lack of "mothering" but 26 to simple undernourishment; that is, failure on the part of the mother to provide the child with adequate food (52, 53, 54). Summary Clearly there are many factors which contribute to the development of iron deficiency and iron deficiency anemia, but a pattern begins to emerge in its etiology; in young children it occurs most frequently during the infant-toddler period of rapid growth, and particularly in the poor and the black child (although the latter may be due in part to their higher representation among the "poor and disadvantaged"). The child under age two is of interest not only because of his high risk for develOpment of iron deficiency but for several other reasons as well. He experiences an extensive change in diet from a total (or nearly so) milk preparation diet as a newborn to essentially an "adult" diet in type and variety of food as he reaches the age of one to two years. His eating pattern changes from one which may be unique to him as an infant (such as eating every few hours) to one which probably corresponds to the existing family eating pattern, whatever that may be. He is dependent upon an adult for the quality and quantity of food provided, but, as Dwyer points out, "nutritional considerations...have never played the major factor in determining the dietary habits of the American people" (50). Perhaps the most interesting point is the fact that only some children become iron deficient within an environment where other children do not. EXPERIMENTAL—PROC EDURE Selectipn of the Sampie The children selected for this study were between the ages of 6 and 24 months of age at the time of their initial participation in the program in the fall of 1971 and winter of 1972. Contact with the families was made through the Ingham County child health clinics where the children were seen for routine health care. These clinics serve a general— ly low income population of families who do not have other access to medical care. Racially the population seen at the clinics is about equally distributed among black, Spanish-American, and white families. All the children in— cluded in this study were considered to be in generally good health without congenital abnormalities, chronic illnesses or obvious neurological defects. A total of sixteen child— ren were included in the study group for this thesis. Those children with hemoglobin values below 11 gm/dl as tested at the Ingham County clinic were referred to the Michigan State University health clinic for further medical and laboratory study. Those with hemoglobin values above 11 gm/dl were contacted as "control" children. Medical and Laboratory Evaluation Of the Sixteen children, ten (8 anemic and 2 control) participated in the medical and laboratory evaluations at the Michigan State University health clinic. They were given complete physical examinations by Dr. Theresa Haddy.1 l See.Appendix A for medical examination form. 27 28 Diet history and 24—hour dietary recall information were obtained at that time. Laboratory tests included: hemoglobin, hematocrit, red blood cell count, serum iron, total serum iron binding capacity, transferrin saturation, routine urin- alysis and determination of occult blood in the stool. In addition, hemoglobin electrophoresis and a glucose-6-phos- phate dehydrogenase (G6PD) screening test were carried out as a service to the black patients. Hemoglobin was deter- mined by the cyanmethemoglobin method; erythrocyte counts, mean corpuscular erythrocyte volumes, and leukocyte counts were done on the Coulter S model counter. Hematocrits were done in duplicate by the microhematocrit method. Serum iron and iron binding capacity levels were measured by a one-tube colorimetric method and percent saturation of trans- ferrin was determined by dividing the serum iron by the iron binding capacity and multiplying the result by 100. Stools were checked for blood with a tablet containing ortholioin. Diagnosis of iron deficiency anemia was based on the presense of 9 of the following 11 conditions: hemoglobin less than 11 gm/dl hematocrit less than 35 % ' mean corpuscular volume less than 82 cuu mean corpuscular hemoglobin less than 27 umcg mean corpuscular hemoglobin concentration less than 32% hypochromia present microcytosis present poikilocytosis present serum iron less than 65 ugm/dl total serum iron binding capacity greater than 350 ugm/dl transferrin saturation less than 17% Oral ferrous sulfate in doses of 5 mg per kg per day was prescribed for those children with iron deficiency anemia. Medication continued until normal hemoglobin, hematocrit, 29 and transferrin saturation levels were achieved. The six children for whom home interview but no medical of laboratory data was obtained were included in the non-anemic (control) group on the basis of hemoglobin values of 11 gm/dl or above tested at the county health clinic. Home Interview The interviews were conducted by appointment in the homes of the families. This was done for several reasons. First, it was generally more convenient for the mothers, many of whom did not have transportation of their own. Secondly, it eliminated the problem of mothers bringing one or more children with her for the interview. Third, it was felt that the mother would be more relaxed in her own home and more at ease answering questions. In addition, it provided the opportunity to observe the home environment and the child in his family situation. All the mothers interviewed were willing to do so; those who did not wish were not inter- viewed. Several mothers expressed appreciation for the interest taken in them and their children. Questions were read to the mothers and the answers written on the interview form by the interviewer to eliminate error which might be introduced by a mother's inability to read the questionnaire accurately. The interview questions covered a range of topics.1 Demographic data included the number of persons in the 1 See Appendices B-F for complete interview forms. 30 family, their ages and relationships, education, occupation, income, marital status and housing situation. A substantial portion of the interview dealt with nutrition practices. Since iron deficiency is a condition that develOps over a period of time, a diet history for the child was taken as well as a 24-hour recall of diet for the day preceeding the interview. In some cases additional in— formation on 24—hour intakes were obtained later by mail. Information concerning diet history and recall was also obtained by the doctor during clinic visits for those children seen at the university health clinic. Dietary intake for calories, protein, vitamin A and C, and iron was calculated from the 24-hour recall record and compared to recommended dietary allowances (RDA) for each nutrient. Further, total iron intake was divided according to the food source of iron. The nutrition knowledge section of the interview was given to determine whether or not there was a Significant difference in knowledge between mothers of anemic and control children. Later it was decided to determine in which areas of nutrition the mothers had the most and the least knowledge. This questionnaire contained 67 questions all of a True—False form.1 Questions were obtained from three sources: twenty-three were adapted from Lund and Burk (55) and 24 questions from Wang (51); the others were 1 In the early part of the study only the first 29 ques- tionS were used. This was determined to be inadequate and the other questions were added making the total 67. 31 added by those working on this study. General questions on foods and eating habits, and more Specific questions concern- ing the food sources and functions of various nutrients and the nutritional needs of infants and children were included. Nearly half of the questions were concerned with the food sources of nutrients. Mothers were given the option of replying "I don't know" to the questions. This choice was provided to eliminate a forced guess and to help keep the interview on a relaxed and informal level by indicating that the mothers were not necessarily expected to know all the answers. It also gave some idea of what they knew correctly or incorrectly and also what was unfamiliar to them. Answer Sheets containing a brief explanation of the nutritional facts covered by each question were provided at the end of the interview or by mail. Two different food attitude questionnaires were used. The first was an adaptation of the list used by Lund and Burk (55), listing various meats, fruits, and vegetables. Interviewees were asked whether they liked, did not like, had no feeling, or had never eaten each of the foods. Reasons for disliking a food were also asked. Although many foods were reportedly liked, food intake records and conversations with the mothers seemed to indicate that the variety of foods served to the family, at least on a regular basis, was not as great as the number of foods which the mothers said they liked. A new questionnaire was prepared listing various categories of foods and the frequency with 32 which these were served to the child in the study. Another section of the interview included open-ended questions on food shopping such as frequency of shopping, the type of food store, expenditures for food, purchase of "convenience" and "health" foods, use of shopping lists, advertized "Specials" and discount coupons. A parent attitude scale was given to determine whether or not there were significant differences between mothers of anemic and control children in their attitude toward the role of "mother." Items on the parent attitude scale were adapt- ed from Schaefer and Bell (56). Thirty-five items were selected reflecting a variety of parental characteristics such as abdication of parental role, rejection of homemaking role, martyrdom and irritability, and such attitudes toward children as showing affection, encouragement of development, and punishment. The items were read to the mother who re- sponded with the extent to which she agreed or disagreed with the statements on a 4-part scale: strongly agree, agree, disagree or strongly disagree. A short section of the interview included questions concerning parental expectations for their children. There were two parts to these questions: the first was a list of activities to which parents gave the age at which they would expect a boy or girl to perform the activity; the second part consisted of open-ended questions about what the parents felt were the most important things they could teach their children in life, what they hoped their children 33 would become as adults, and whether luck or effort would play the greatest part in their child's future. This part of the questionnaire was completed by the parent. Statistical Analysis Because of the Small numbers involved the children were placed in two groups,"control" and "anemic" on the basis of hematological data, that is presense of 9 of the 11 conditions listed on page 28. Student's t-test was used to determine Significant differences between the two groups for each section of the data (57). Informed Consent The study was explained to the mother and if they wished to participate with their children, an informed con- sent form was signed by each child's parent or legal guardian at the time of the initial interview. Two forms were used, one for control and one for anemic children, the difference being a statement regarding treatment of the anemic children.1 1 See Appendix G for consent forms. W1 Labgiatory Data Mean hemoglobin values were 12.5 gm/dl for the controls and 9.7 gm/dl for the children with anemia. Among the anemic children, those having hemoglobin values between 10.0 and 10.9 gm/dl were considered to be mildly anemic while those with hemoglobin values below 10.0 gm/dl were considered to be moderately or severely anemic. Sixty-three percent of the anemic children fell into the mildly anemic category. Additional hematological data was available on all children with anemia but only three of the control children. However, in the larger study of which this thesis is a part, laboratory data indicated that the control children can be placed into two different groups on the basis of transferrin saturation. Those children with a transferrin saturation of less than 17% were considered to be iron deficient despite a hemoglobin concentration of 11 gm/dl or above. On this basis it was found that among control infants and children up to 24 months of age, nearly 42% were iron deficient. Dempgraphig_9ata The demographic data are summarized in Tables 3 and 4. Seventy-three percent of the control children were only children while only one (11%) of the anemic children was an only child. On the other hand, seven (77%) of the anemic 1 Throughout this report, several items have been omitted from discussion Since responses were extremely variable and no trends were discernable. Since this study con- tinued beyond data collected for this thesis, certain items may become Significant to the findings when the number of respondents is increased. All major findings and trends are reported here from data collected for this thesis. 34 35 children were youngest children in their families while only two (18%) of the control children were the youngest child. The average number of children per family was 1.8 for con- trol and 3.2 for families of anemic children.1 This differ- ence in number of children is reflected in number of persons in the household, 3.8 and 5.1 for control and anemic child- ren reSpectively. DeSpite the larger Size of families of anemic children, they lived in slightly smaller housing measured by number of rooms and of bedrooms. Race and sex distributions of the children showed a tendency for the anemic child to be black and male. The control children were about equally distributed among races and between sexes. More mothers of control children were married with an adult male present in the home. Total income was Slightly higher for families of anemic children, but because of the larger family Size, per capita income was less by nearly 25%. Educational level was similar for all parents, with control mothers attaining the highest level, 12.2 years. Fathers of control children averaged just about high school graduate level, followed by fathers and then mothers of anemic child- ren. The differences in educational level between mothers and between fathers of control and anemic children was not Significant by the t-test. 1 Significant @ t( 1) 36 Table 3: Physiological Charactegisticsa Control Anemic Mean age of children 16.4 mos. 13.2 mos. Mean birth weight 6.9 lbs. 6.7 lbs. Birth position: oldest/only 8 (73%) 1 (11%) middle 1 (9%) 1 (11%) youngest 2 (18%) 7 (78%) Child's race: white 5 (50%) 1 (10%) black 2 (20%) 7 (70%) Spanish-American 2 (20%) 2 (20%) mixed 1 (10%) Child's sex: male 7 (54%) 9 (82%) female 6 (46%) 2 (18%) Mean age of mother 22.2 yrs. 24.3 yrs. Mean time between pregnancies 13.3 mos. 17.7 mos. a Information on some items was not available for all families, therefore total numbers vary from item to item. Percentage figures are for each item separately. 37 Table 4: Familial Characteristicsa Control Anemic Mean number of children/family 1.8b 3.2b Mean number of persons/family ' 3.8 5.1 Marital Status of mother: married 10 (83%) 3 (33%) unmarried 1 ( 8%) 4 (45%) separated, divorced 0 2 (22%) unknown 1 ( 8%) Mother's employment: housewife 11 (83%) 6 (55%) employed 0 1 (9%) unemployed 0 1 (9%) student 2 (15%) 3 (27%) Adult male in the home: yes 10 (83%) 5 (50%) no 2 (13%) 5 (50%) Father's employment: employed 5 (56%) 2 (6T%) unemployed 4 (44%) 1 (33%) Major source of income: job 6 (55%) 4 (40%) public assistance 3 (36%) 5 (50%) other 1 ( 9%) l (10%) Income: family/year $4488.50 $4554.67 person/year $1180.75C $ 899.47C Mean educational level attained mother 12.3 yrs. 10.8 yrs. father 11.9 yrs. 11.0 yrs. a Information on some items was not available for all families, therefore total numbers vary from item to item. Percentage figures are for each item separately. b Significant @ t( 1) c Significant @ t( 05) 38 Nutritionai Data Diet History There are many Similarities in the diet histories of control and anemic children. Most children were bottle fed, 83% and 64% of control and anemic children reSpectively. A higher percentage of anemic children were breast fed (36% vs. 18%) or if bottle fed, received iron—fortified formula (7r% vs. 50%); however, anemic children were changed to whole milk at a slightly earlier age. Consumption of cereals and strained baby foods started earlier and table foods later for the control children than for the anemic children. In both cases fruits and vegetables were started before meats, but by 6-8 months of age all infants were receiving meat. weaning from the bottle to cup occurred at an average age of one year for those children who were weaned at the time of the interview. Of the control children, 3 still drank from a bottle at ages 7, 11, and 12 months reSpectively; two anemic children ages 12 and 23 months, were still taking bottles. Several other children, drinking from a cup at mealtime, received bottles at bedtime (3 control and 5 anemic children). More mothers of anemic children reported having "feeding problems" with their children and described the child's appetite as "poor" than did mothers of control children. The average age at which the child sat up to the table with the family at mealtime was 7.0 and 8.2 months for con- trol and anemic children respectively. Whether or not the child was fed at that time was not asked. One anemic child 39 was, at 14 months of age, still not at the table during family meals. The average age at which table foods were started is almost exactly the reverse of the age of sitting at the table with the family. Table foods were started at 7.1 and 8.0 months of age, the anemic children having the younger age. Table 5 Shows the diet history data. Food Attitudes Results of the food attitude questionnaire Showed little difference between the two groups of mothers. Many foods were reportedly liked by both groups, but dietary recall re— cordsanuiconversations with the mothers during the interview seemed to indicate that the variety of food likes were not closely related to the variety of foods served to the family, at least on a regular basis. Some mothers commented that although they liked a particular food, they had eaten it only a few times. The reasons why so many foods were liked but seldom served was not asked directly but conversations indicated that high price, unusual preparation and seasonal availability of a food were factors. Favorite foods of the children included some predicat- able choices. Fruits were generally well liked. A number of meats and vegetables were also listed as favorite foods, particularly chicken, hot dogs, mashed potatoes, corn and green beans. Vegetables, however, were the most frequently mentioned foods which the child reportedly did not like. The same pattern was found in the foods given as favorite or disliked by the mother and the family as a whole. Responses to the food frequency questionnaire, the 40 Table 5: Diet Historya Control Anemic Breast fed as newborn number, percentage 2 (16%) 4 (36%) average length of time 4.5 mos. 3.8 mos. Bottle fed as newborn number, percentage 9 (82%) 7 (64%) bottle fed later in infancy 2 4 Type of furmula: (number, percentage) b iron fortified 6 (46%) 6.5b (72%) non-fortified 7 (54%) 2.5 (26%) Mean age changed to regular milk 5.4 mos. 4.7 mos. Mean age baby cereal started 1.6 mos. 1.9 mos. Mean age strained foods started 2.7 mos. 3.0 mos. meat started later, latest age 6-7 mos. 7-8 mos. Mean age table foods started 8.0 mos. 7.1 mos. Mean age most of milk from cup 12.0 mos. 11.5 mos. number still on bottle 3 2 ages of those still on bottle 7, 11, & 12, 23 mos. 12, mos. Receive bottle at bedtime: yes 6 7 no 3 1 Mean age child first present at 7.0 mos. 8.2 mos. table for family meal Mother's assessment of child's appetite: (number, percentage) good 9 (82%) 3 (37.5%) fair 1 ( 9%) 2 (25%) poor 1 ( 9%) 3 (37.5%) "Feeding problems" reported by mother (number, percentage) yes 1 (12.5%) 6 (86%) no 7 (87%0 1 (14%) a Information on some items was not available for all families, therefore total numbers vary from item to item. Percentages are for each item separately. b One child received both. 41 second version of the food attitude questionnaire, are again very similar between the anemic and control groups, and sug- gest that both receive a fairly well balanced diet. All the children reportedly received meat, fish or poultry, as well as milk or other dairy product, and bread and/or cereal pro- ducts every day. Citrus fruits or juices were served several times a week, and in some cases every day. Eggs and vege- tables were each eaten several times per week. No signif- icant differences between the two groups of children were found in frequency of various foods eaten. The mothers were asked to describe their favorite meal to give some idea of whether or not such a meal would be balanced in terms of food groups, including a main meat dish, vegetables, fruits, cereal products and beverage, or at least most of these. Every favorite meal contained a meat dish, and all but one included a vegetable. Many included two or more vegetables. Fruits and milk were included least often. Nutrition Knowledge Results of the nutrition knowledge questionnaire for mothers of anemic and control children were compared accord— ing to: (a) total number of correct answers; and (b) right minus %-wrong answers. The "don't know" reSponses did not enter into the scoring. The scores of the shorter questionnaire (the early form) were corrected mathematically for comparison with the longer form. 42 Table 6: Food Frequencv- Percent of Children Served Each Food Category:per Week Food Categorya Daily 1-4 Sb c Eggs C 0 100 A 20 80 Meat, fish, poultry C 100 A 100 Milk products C 100 A 100 Bread, cereal products C 100 A 100 Yellow vegetables C 0 100 A 25 75 Other vegetables C 50 50 A 20 80 Citrus fruits, C 75 25 tomatoes A 60 40 Other fruits C 75 25 A 0 75 25 Candy, desserts C 50 50 A 20 60 20 a Some categories have been omitted from the table. b Daily = everyday; 1-4 combines 3-4 and 1-2 times per week frequencies; S = seldom. c C = control; A = anemic. 43 Table 7: Mothers' Favorite Meal- Food Categories Included Control Anemic Main meat dish 100% 100% Vegetables 100% 83% Bread and /or cereal product 38% 50% Fruits 25% 0% Milk 63% 17% Beverage (including milk) 75% 33% Dessert 25% 50% 44 Table 8 shows the total number correct and mean correct scores. A Significant difference was found between the mean scores of the two groups of mothers. AS can be seen, the mean score, 46, achieved by the mothers of control children is higher than all but one of the scores of mothers of anemic children, while the latter's mean score of 34 is lower than any score of the control mothers. Calculating the scores by right minus %-wrong to correct for guessing which might falsely increase the number of correct answers gave very Similar results (Table 9). The 67 questions were grouped into four major sections: (1) food sources of nutrients; (2) function of nutrients; (3) food needs of infants and children; and (4) eating pat- terns.1 The number of correct reSponses of the two groups of mothers was compared within each section. Results were calculated as the percentage of correct responses (100% being the number of questions in each section times the number of mothers answering each question). Results are shown in Table 10. AS can be seen, mothers of control children correctly answered more questions in each section and Scored Significantly higher in 3 of the 4 sections. It is interest- ing to note that the highest number of correct answers for both groups was in the section on food needs of infants and children. This was the only section in which mothers of anemic children scored higher than about 50% correct, and the only section where the differences between scores of the 1 See Appendix C for the nutrition knowledge questionnaire and grouping of the questions, questions answered correct- ly by most and by less than half the respondents. 45 Table 8: Nutrition Knowledge Scores- Numbe; Correcta Control Anemic 59 49b 53 41 49 36 a c 49 a c 34 (mean = 34 ’ ) 47 (mean = 46 ’ ) 33 44 30 39 28 36 25 35b a nearest whole number of 67 total b Shorter form corrected to compare with longer form c significant @ t( 005) Table 9: Nutrition Knowledge Scores- Right Minus fi-wrong Control Anemic 56.5 39.1a 50 36 b c 42.5 26 (mean = 25 ’ ) 40.5 b C 24.5 40 (mean = 38 ’ ) 24.5 36.5 19 28 18 25 16 23.1a a shorter form corrected to compare with longer form b nearest whole number c Significant @ t( 025) 46 two groups of mothers was not statistically significant. Table 10: Percentage of Correct Answers to Grouped Nutrition Questionsa Opestion Group Control Anemic I Food Sources of Nutrients 54b 40b II Function of Nutrients 79C 53C III Food Needs of Infants 81 75 and Children . d d IV Eating Patterns 76 50 a nearest 1% c Significant @ t(.025) 0.. b Significant @ t( 025) Significant @ t( 005) Individual questions were also considered in analyzing the results. No Single item distinguished between the two groups of mothers. Fifteen questions were answered correct— ly by nearly everyone, with approximately equal numbers of these questions in each of the four sections. Twelve items were answered correctly by less than half of all mothers.1 All but one of these was in the section on food sources of nutrients; the other was in the section on function of nutrients. 24-Hour Dietary Intake Recgggg Food intake was calculated from the recall records and intake of several nutrients compared to the recommended dietary allowance (RDA). Table 11 Shows the mean intake of each nutrient as a percentage of the RDA for each group of 1 See Appendix C for questions answered correctly the most and the least frequently. 47 children. Of greatest importance is the fact that the difference in iron intake between the control and anemic children is significant at the .025 level by the t-test. Additionally, it is the only Significant difference in in- take of the nutrients analyzed. Although the percentage- point difference in RDA is greater for some other nutrients, the variance for those nutrients was also greater, resulting in an insignificant difference by the t-test. Of particular interest, however, is the fact that not even the control children averaged 50% of the RDA level for iron intake. Table 11: Mean Daily Nutrientintake- Percent of RDAa Control W Calories 112 101 Protein 218 195 Vitamin A 230 136 Vitamin C 194 129 Iron 47b 30b a nearest 1% b Significant @ t(.025) To determine the major sources of iron in the child's diet, food sources were categorized into 6 groups, and the mean percentage of iron coming from each food group was calculated (Table 12). These figures indicate that the diets of the two groups of children were very similar as far as the source of iron is concerned and there are no statistical- ly significant differences. 48 Table 12: Soupce of Iron- Mean Percent of Total Daily iron Intakea Food Group Control Anemic Eggs 15.4 21.8 Meat 22.3 26.1 Breads and Cereals 32.2 29.7 Fruits 18.1 12.8 Vegetables 18.2 17.9 Milk and Milk Productsb 5.0 5.7 a Numbers are the mean % of total daily iron intake con- tributed by each food group. Although figures for each individual child totaled 100%, percentages listed here are the means for the group as a whole, and therefore, totals do not equal 100%. Milk consumption was nearly the same for the two groups, not only in the percentage of daily iron it provided, but in total milk intake and percent of daily calories as well. Differences were not significant (Table 13). Table 13: Milk Consumption- Mean Daily Intakea Control Anemic Mean Milk Consumption (ounces/day) 18.8 19.1 Percent of Total Daily Calories 27.1 34.8 a Includes infant formulas and cow's milk. 49 Shopping Patterns DeSpite the fact that some families lived in the "inner" city and did not have cars, they shopped at suburban super- markets, sometimes located miles from their homes. Their reasons for Shopping at supermarkets were most often the lower prices, and better quality and selection of foods. While the homemakers generally went to those stores where they felt the prices were lowest, not all took advantage of newSpaper discount coupons or items advertized at Special prices. Several families did not take a newSpaper because of its cost. Most of the families ShOpped every two weeks which was the pay interval of many jobs as well as the schedule of welfare payments. Three shoppers mentioned making daily or bi-weekly trips to the store for a few items such as bread or milk, in addition to larger trips made less frequently. Two mothers said they made a major Shopping trip every month or two months, with smaller trips in between on occasion. More Shoppers than not took a list, at least some of the time, although a few made such comments as ~"I know what to get" or "I get what I always do," indicating a somewhat repeti- tious nature and limited variety of food preferences and intake. It would appear that most of the mothers are fairly organized in their Shopping, buying groceries for a two-week or greater time period and ShOpping where they felt the over- all prices were lowest. Five women mentioned that they might go to more than one Store. There were no Significant dif- ferences between mothers of control and anemic children in 50 Shopping practices. The families spent an average of $24 and $26 per week for food, families of anemic children having the higher figure. When calculated on a per-person-per-day basis, the food cost figures are $.80 and $.73, control families having the greater amount. These figures are not significantly different.1 Neither group used "convenience", prepared or semi- prepared foods to any great extent. Cake mixes were the convenience food most frequently mentioned. Two homemakers used so-called "health" foods occasionally. Most of the mothers liked to cook and used recipes occasionally for Special dishes and for variety. Most learned to cook from their mothers and had had some kind of home economics training in school or through clubs such as Girl Scouts or 4-H. Some women felt that the public was getting confusing nutrition information from mass media sources. When asked what sources they would use for nutrition information, most seemed hesitant in deciding.. A variety of sources were given, some potentially knowledgeable such as doctors, dietitians, health clinic personnel, and nutrition books. Other sources mentioned were of more questionable reliability such as other family members or store clerks.2 1 Cost figures are for 1971-1972. 2 Data collection was conducted in 1971 and 1972. Responses to questions on use of convenience foods and sources of nutrition information might be quite different if the study were repeated due to increases in types of conven- ience foods and in public awarneSS of nutrition concerns. 51 Parent Attitudes andygxpectations Two points stand out on the parent attitude question- naire when reSponseS are compared: (1) the extent of Similar— ity between the two groups of mothers; and (2) the extent to which mothers within each group were divided between agree- ment and disagreement to a given item. ReSponseS to some items, including autonomy of the child (24)} non-punishment (22), avoidance of communication (17), avoidance of tender- ness (23), expressing love and affection (29), and irritabil- ity (6) were nearly identical between the two groups. Close agreement in reSponse was given to items on encouraging verbalization (l, 11), breaking the will (3), harsh punish- ment (31), rejection of homemaking role (14), comradship and sharing (20), and equalitarianism (25), and infantilization (30). Items where both groups were divided between agreement and disagreement to a given Statement included acceleration of development (21, 26), suppression of agression (9), martydom (4), abdication of parental role (35), irritability (12), excluding outside influences (7, 13), ascendency of mother (18), ignoring the child and baby (28, 32, 34), and avoidance of tenderness (27). No item Showed a clear-cut division between the two groups of women. Items Showing some degree of difference between the groups included approval of activity (16), strictness (5), deification of mother (8), rejection of the 1 Numbers refer to item number on the parent attitude questionnaire. See Appendix E for complete question- naire and grouping of the items. 52 mother role (10), abdication of parental role (33), equal- itarianism (15) and intrusiveness (19). However, the dif— ferences between the two groups were not statistically significant on any item. The items were also grouped into four general categories: (1) mother role; (2) mother-child relationship; (3) develop- ment of the child; and (4) discipline. ReSponses are shown in Table 14. There were no Significant differences between reSponseS of the two groups of mothers. Within each category there were items to which the two groups of mothers gave some nearly identical responses, some divided within their own group, and some which showed a degree of difference be- tween groups. Table 14: Combined ReSponses to Parent Attitude Categories- Percent Aggee and Disagree Responses ReSponse Control Anemic I Mother Role Aa 57b 54 D 43 46 II Mother-Child A 52 46 Relationship D 48 54 III Development of Child A 68 76 D 32 24 IV Discipline A 36 34 D 64 66 a A combines "strongly agree" and "agree" answers; D combines "Strongly disagree" and "disagree" answers. b nearest 1% 53 The mothers were about equally divided between those who did and those who did not like housekeeping. There seem- ed to be no correlation between the homemaker's answer to that question and the condition of the home at the time of the interview. The most poorly-kept home (with broken glass on the kitchen floor and the children barefoot) was that of a mother who said She liked to keep house, while the homes of mothers who reportedly didn't like housekeeping were at least reasonably neat and clean for the interview. This may have been related to concern for the interviewer's expectations. The questionnaire concerning the ages at which parents expect children to be capable of certain tasks indicated that mothers of anemic children generally expected perform— ance at an earlier age on such tasks as toilet training, dressing himself, and taking care of younger children. An important exception from a nutritional standpoint was the age at which mothers expected a child to eat at the table with adults. Control mothers gave an average age of 10 months while mothers of anemic children cited 18 months. However, this difference is greater than that indicated in the diet histories of the ages at which the children in the study began to Sit at the table with the family at mealtime, although in that case also the age for anemic children was the greater (7.0 vs. 8.2 months). In contrast, mothers of anemic children responded with 10-11 months of age for a child to begin eating adult food rather than baby food, while control mothers gave 15-18 months as an average age. 54 Again the difference between the two groups is greater than that indicated by diet histories as the actual practice, but the order is the same (7.1 vs. 8.0 months for anemic and control children reSpectively). The questions concerning the most important thing a mother teaches her child and how She does this were broad, general questions and elicited broad, general answers. How- ever, some interesting trends do appear, particularly regard- ing what is most important to teach a girl versus a boy. Cleanliness, neatness, and being a "young lady" were cited most often for a girl. A greater variety of things was given for boys with no one thing predominant. Things cited gpiy for boys included reSponsibility, self-assurance, and self-reliance. Things listed equally for boys and girls in— cluded self-respect and teaching the "facts of life". The most frequent answer to the question of how one Should teach the most important thing were by "example" and "talking and explaining". It seemed more difficult for the mothers to answer these general questions than the other more Specific questions in the rest of the interview. DIS CUSS I ON AND REC OMMENDATI ONS Laborappgy Data Laboratory data collected in this study and the larger study of which it was a part indicated that iron deficiency anemia was most often mild, with the children having hemo- globin levels between 10.0-—10.9 gm/dl. Additionally, iron deficiency without anemia was more prevalent among the child- ren in the larger study. It was not possible to predict the incidence of iron deficiency without anemia on the basis of hemoglobin value. This would indicate that testing of hemo- globin alone is not adequate to detect those children with marginal iron status. Dempgraphingata Bees A greater proportion of black than white or Spanish- American children were found to have iron deficiency anemia in this study, as has been the case in some other studies (6, 31). Since the children in this Study were all from "low income" families, the higher incidence of anemia among black children cannot be dismissed Simply on the basis of socio-economic level. AS was already mentioned, the "normal" hemoglobin values for blacks may be slightly lower than for whites so that comparison of blacks to standard hemoglobin levels for whites may not be appropri- ate (48). §_e_x In this study more children with iron deficiency anemia were male than female. This is consistent with some other 55 56 studies (8, 25). The RDA recommendations for iron, as well as other nutrients and calories, are the same for bOth sexes during infancy and childhood, as are standard hematological values.1 Such lack of distinction between sexes at this age would suggest that biological parameters and nutrient needs are Similar. Socio-Economic Level The childmen.in this study were all from families whose incomes were generally low, with the average annual family income approximately $4500. A number of family characteristics indicate trends which may become more apparent when data from the entire study are assembled.2 Eighty-three percent of mothers of control children were married compared to 33% for mothers of anemic children. This difference may be an important factor in the somewhat greater reliance of families of anemic children on public assistance as the major source of income (50% vs. 36%). Per capita income was Significantly greater for families of control children, but this difference was not reflected in the food budgets where average expenditures per-person-per- day for food were nearly identical for the two groups of families. Educational level of parents was similar, but mothers of control children had completed the most Schooling on the 1 Although Standard hemoglobin values for adults are dif- ferent for each sex, this difference becomes apparent during puberty and seems to be the result of a higher androgen level which increases the hemoglobin level in males (13). 2 A report of the larger study has been made (3). 57 average (12.3 yrs.) and the mothers of anemic children the least (10.8 yrs.). This difference was not significant for this sample, but as the number of families studied is increased a similar finding might be significant. Seventy-three percent of the control children were the oldest or only children, while 77% of anemic children were the youngest in their families. The mean number of children per family was significantly larger for anemic than control children. Smith reported similar results, stating that iron deficiency was less likely to occur in first born than in later children (25). While each of these factors may play a minor role in the etiology of anemia, their total effect may be of im- portance. A pattern which seems to emerge is that, in comparison to her counterpart in the control family, the mother of the anemic child has more children, less dispos- able income and less education. The absence of a husband in the home places greater reSponsibility on her for the care of the children and it is the youngest child who is most likely to develop anemia. It is possible that older children were also anemic during the vunerable period of infancy, but Specific data on older siblings were not available. Psycholggical Factogg The data regarding parental attitudes toward child rearing were very Similar between the two groups of mothers. On many items mothers of both control and anemic children were divided as to agreement or disagreement to a given 58 statement. Only a very few items showed differences be- tween the two groups and these differences were not signif- icant. The parental expectations questions indicated a general expectation of task performance at a slightly younger age for anemic than control children on at least some tasks. This may be related to the fact that a greater number of anemic children were the youngest in a family of several children. Included among the tasks were several which would be expected of children older than those in the study group. More mothers of anemic than control children had the benefit of experience with older siblings in deter- mining ages of expected performance of these tasks. Npppitional Data The major part of this thesis was concerned with the nutritional status of the child, the differences between the knowledge of mothers of control and anemic children regarding nutrition, and possible ways to educate the moth- ers in nutrition. Diet History Diet histories of the two groups of children were similar. However, some differences in consumption of baby cereals and meats, important sources of iron in infant diets, may play a part in distinguishing between control and anemic children. (See Table 5). The number of iron-rich foods available to infants is somewhat limited, fortified cereals and meat being two of the most frequently consummed (58). The control children were started earlier on baby cereal, 59 as well as other baby foods (fruits, vegetables, meats) and continued on them longer before changing to adult foods. The anemic children were started on solid foods later and fed both baby cereals and meats for a shorter average length of time. Baby food cereal was generally discontinued before the age of greatest risk of iron deficiency, but meat was started about that same time (6-8 months of age). While the differences in consumption of cereal and meat between control and anemic children were slight and not Significant Statistically, in combination they may work to the advantage of the control child. Many adult type cereals do not contain as much iron as do baby cereals (about 1.5-2.0 mg vs. 14 mg per average serving) and thus the amount of iron consumed by an infant eating adult cereal would likely be consider— ably reduced (59). In addition, infants may have greater difficulty in eating adult meat unless it is ground or cut into small bites. The quantity of regular meat eaten by a child at one meal or in one day might therefore be less than the quantity of prepared baby food meat eaten. Iron-fortified formulas did not seem to play an impor- tant role in this study. In fact, more anemic than control infants were given fortified formula (68% vs. 42% of all anemic and control children), but in both groups formula was no longer used by an average age of less than 6 months. 1 Unfortified adult cereals contain about 1.5 mg of iron per serving while fortified adult cereals contain 8-10 mg per serving. Therefore, careful selection of forti- fied adult cereals could contribute a Sizeable percent- age of the RDA for infants if a full adult serving were consumed. 6O Nutpition Epowiedge Results of the nutrition knowledge questionnaire indi- cate that mothers of control children had greater knowledge in all four sections of that questionnaire: food sources of nutrients, function of nutrients, food needs of infants and children, and eating patterns. However, the section in which the mothers of anemic children had their highest scores (and the only section in which the differences in scores between the two groups of mothers was not significant) was that on food needs of children. This is interesting Since children's nutrition was the area of particular con- cern in this study, and the mothers were grouped according to the nutritional status of their children in regard to iron. There are several possible reasons for these results. The questions about food needs of children may have been too few and too general so that a minimal knowledge was sufficient to answer them correctly. The low score (mean 40%) of mothers of anemic children in the section on food sources of nutrients may have been an important part of the reason for the apparent discrepancy between their knowledge of food needs of children as indicated by their relatively high score on that section of the questionnaire, and the nutritional status of their children. The control mothers' higher scores on the sections on food sources and functions of nutrients may reflect a greater overall nutrition know- ledge enabling them to better understand and put into practice the general knowledge of children's nutritional needs. There may also be a discrepancy between what the 61 mother knows or believes to be good nutrition for her family and her actual feeding practices. One study found evidence of this, reporting that 72% of mothers questioned believed that vegetables, and 86% believed that fruits were important in a child's diet, yet these foods were rarely served to their children. It was concluded that nutrition knowledge and Opinion could not be equated with feeding practices (60). One additional fact stands out; neither group of mothers knew very much about the food sources of nutrients. That section of the questionnaire had the lowest scores for both groups. While the control mothers scored significantly higher, their mean score of 54% correct (compared to 40% for mothers of anemic children) can hardly be considered a high score. It would seem that this is an area where efforts in nutrition education for mothers of small children could focus attention. During the course of the interviews it became apparent through conversations with the mothers that they lacked knowledge about iron deficiency anemia. There was confusion among black mothers concerning the differences between iron deficiency anemia and sickel cell anemia. Some mothers showed a lack of concern about iron deficiency, particularly if they or other members of their family had had it previous- ly. One mother commented that she didn't realize that it could be serious enough to require medication. Perhaps the ready availability of iron supplementation in over-the-count- er preparations such as multivitamin and iron tablets and 62 "tonics" suggests that medical consultation is not necessary. A short educational pamphlet was developed as an aid in helping mothers better understand iron deficiency anemia.1 A brief explanation of iron deficiency was included as well as a list of nutritional risk factors which contribute to its development, such as rapid growth during the period of 6-24 months of age, high milk consumption and low meat con- sumption. A list of iron-rich foods was included so that the mother could include these in the child's diet. A sample pamphlet was distributed by the Ingham County child health clinic to a small number of mothers whose children were 6-24 months of age and had hemoglobin levels close to 11.0 gm/dl. The pamphlet was explained to the mothers by clinic personnel. Telephone calls were made to some of these mothers for their comments about the pamphlet. The mothers found the list of iron-rich foods the most help- ful part of the pamphlet. Additional copies of the pamphlet were not made, but it might be used for distribution to mothers of infants through nutrition education programs, health clinics, or social service programs. Food Attitudes The food attitude questionnaire was an attempt to elicit a more complete response conderning overall diet than was obtained by the 24-hour recall records. The questionnaire itself was in the process of development during the time data for this thesis was collected, and, in fact, two 1 See Appendix H for text of pamphlet. 63 versions were used in this study. The first version, a list of Specific foods to which the mothers indicated the degree that they liked each food, did not indicate the frequency with which the food was eaten. This was considered a ser- ious omission, and the second version was develOped listing categories of foods according to essential nutrients pro- vided. The frequency with which each group was served to the study child was asked. Because of the small number of mothers given each version it is difficult to draw conclu- sions. The food frequency reSponses indicated a fairly well balanced diet in terms of the number of time per week important food sources of essential nutrients were served. However, the quantity of the various foods actually eaten by the children appeared to be a more important factor in the incidnece of iron deficiency. Quantities of foods usually eaten by the child might be incorporated into the food frequency chart. In addition, the frequency question- naire did not indicate the variety of foods within each category and might be expanded to include a section on the number of different fruits, meats, vegetables and so on served each week, or the Specific foods. Further attempts should be made to design better instruments for evaluating dietary intakes both on a short term and long term basis. Shopping Patterns Some studies have indicated that the urban poor pay more for food because they cannot shop at suburban chain supermarkets where prices are generally somewhat lower (61); this was not true in this study. Families of both control 64 and anemic children shopped at the suburban markets to take advantage of lower prices. Other studies have also found this to be the case (62). iptakeyData Cgmpared to RDA Although a Significant difference between the control and anemic children was found in intake of iron compared to the RDA for that nutrient, the striking fact was that not even the control children averaged an iron consumption of 50% of the RDA. Of all intake records of control children, only three were above 60% of the RDA for iron, with the mean intake being 47%. It would appear that daily iron intake is marginal at best. This is not different from the findings of a number of other studies. One study of infants found that iron intake expressed as percent RDA drOpped steadily from close to 100% at 6 months of age to about 50% at 13 months (62). .Another study found that less than 26% of infants met the recommendation for iron suggested by the Committee on Nutrition (5). A study of diets of preschool children found that only 20% met the recommendations for all major nutrients and that iron was the nutrient least well supplied. Fewer children had iron intakes of 100% of the RDA and more had iron intakes below 67% than for any other major nutrient (63). Other investigations have found that iron intake was below the RDA despite adequate protein and calorie intake in preschool children (33) and that among school children, 84% had iron intakes of less than 66% of the RDA (7). 65 Uhder such circumstances the question arises why more of the children were not anemic. It may be that the RDA is unnecessarily high or that the present standards for diag— nosis of iron deficiency are too low. These values have been the subject of controversy as was already discussed, and have been revised from time to time reflecting current research findings and interpretations. Beal and Meyers did not find a clear relationship between iron intake and hemo- globin, hematocrit or mean corpuscular hemoglobin concentra- tion levels in their longitudinal study (64). It was found that while iron intakes seldom reached the RDA recommenda- tion, hematological levels were satisfactory. Further, Beal, in another study, pointed out that little is known about possible adaptation of individuals to their levels of iron intake (65). The accuracy of intake records is, of course, an important factor in RDA calculations and one which is difficult to assess in the 24-hour recall record. In this study, the recall was taken for the day prior to the inter- view, and it is unlikely that the mother made any particular note of the child's intake during that day. In obtaining the intake records, mothers were more hesitent about the quantities eaten than about which foods were served, and were particularly hesitent if the child left some food on his plate or was given second helpings. Whitten et al found that in cases of growth failure due to undernourishment, parental dietary recalls were unreliable (54). Beal found the preschool period one of unevenness of intake and observed 66 100% variations between children in intake of calories, carbohydrates and protein, and even greater variations in ranges of some other nutrients. Daily variations for a given child were not so great (65). The use of Single 24-hour recall records poses the possibility of unrepresentative intake compared to a child's "usual" diet. However, in those cases in this study where 2-3 recalls were obtained, differences in intake were not so great as to change the statistical findings. The range of intake for iron was less than for any other nutrient evaluated, and deSpite the inherent problems of 24-hour recall data, the findings of low iron with generally adequate intakes of other nutrients is consistent with other studies. Further, there is vari- ation in iron content of different samples of a given food, and the published data on nutrient content of various foods used to calculate dietary intakes may contain errors. Although complete hematological data were obtained on too few control children in this study to make a meaning- ful comparison of iron intake expressed as percent RDA be- tween iron deficient children and non-deficient control children, preliminary data from the larger study indicated that there may be a difference between these two groups in average iron intake, with the non-deficient children having higher iron intakes than those with iron deficiency, and iron intake level of deficient children higher than that of anemic children. This would indicate some correlation be- tween iron intake and hematological data, contrary to the findings of Beal and Meyers (64) already cited. 67 Food Soupces of Iron, Fortification, and Utilization The incidence of iron deficiency and iron deficiency anemia and the low intake levels of iron have led to much debate concerning both iron fortification of various foods and general feeding practices of children. A number of questions arise, including the kinds of foods which could or Should be fortified, the type and amount of iron supple— mentation, the effectiveness of such foods in preventing iron deficiency and anemia, and the increased use of iron fortified foods by individuals with higher risk of develop- ment of iron deficiency. Analysis of the source of iron in diets of children in this study showed a Similarity between control and anemic children. The difference in percent of the RDA for iron intake seems to be due to greater quantity of food consumed by the control children on a daily basis rather than to higher quality in terms of iron-rich foods. For all child— ren, the greatest percentage of iron came from the bread and cereal food group, a finding that would support the proponents of continued or expanded iron fortification of this food group in view of the persistence of iron deficiency in children. While fortification of commercial infant foods, particularly formulas and cereals, is common, these foods are frequently discontinued at an early age. Filer found that formulas were continued until about 4 months of age, and by 6 months of age 75% of infants studied were receiving regular milk rather than formula (66). Purvis found that while over 90% of infants studied consumed baby cereal at 68 age 4 months, this number drOpped to less than 50% by 12 months of age, 10% by 18 months, and essentially 0% by age 24 months. Further, that study found that consumption of iron by infants was closely related to consumption of fortified cereal, the average iron intake falling below recommendations when these cereals ceased to be a major part of the child's diet (58). Another study found that only about 66% of 6 month old infants were receiving fortified baby cereals and that there were wide variations in iron intake depending upon the amount of cereal consumed (5). There is also some question as to the effectiveness of iron fortification in the early months of life in full term, normal birth weight infants. As has already been mentioned, incorporation of dietary iron into hemoglobin may be low, and Since formulas and cereals are so often discontiued at an early age, their role in prevention of iron deficiency is somewhat diminished. However, the RDA for iron during this period is so seldom met by the usual diet which supplies adequate amounts of other nutrients and calories, and the Committee on Nutrition suggests that fortification is prob— ably needed (22). AS mentioned earlier, there is some ques- tion of the availability of iron used for fortification. One author stated that the pyrophOSphate in cereals has a very limited absorption (1). Filer, in referring to forti- fied formulas stated that the utilization of iron was greater when protein content of the formula was less than 28 gm/100 ml (66). 69 It seems apparent that while fortification is recommend- ed on the one hand, there remains much research yet to be done on the source of iron and the foods to be fortified, and on the interaction on iron absorption of various foods as combined in a meal. Summary This thesis investigated some of the sociological, psychological and nutritional aSpectS of anemic and control children 6-24 months of age, from low income families. However, particular emphasis was on nutritional factors. Results indicated that the diets of all children were generally adequate except for iron, and, although there was a significant difference in iron intake between control and anemic children, the iron intake of all children was well below the RDA for this nutrient. There were also signifi- cant differences in family per capita income and mother's knowledge of nutrition, in both cases the families of con- trol children having the higher values, and in number of children per family with the families of anemic children having the higher number. The anemic child was more often the youngest in his family of several children than was the control child. Black males seemed more likely to become anemic than black females or SpanisheAmerican or white children of either sex. Other variables which were consider- ed did not Show Significant differences between groups, al- though some trends were observed and these are more fully analyzed in the final report of the entire interdisciplinary study (3). LIST OF REFERENCES 10. 11. 12. 13. LIST OF REFERENCES Levine, R.H., Iron deficiency anemia in the pediatrics patient, J.Am. Phar. Assoc. NSll:670-76, 1971 Read, M.S., Nutrition and ecology: crossroads for re— search. Bibliotheca Nutritio et Deita, No 14, pp 202-18, 1970 Haddy, T., C. Jurowski, H. Brody, D. Kallen, D. Narins, Iron deficiency with and without anemia in infants and children, Am J. Dis. Child. 128:787, 1974 Young, H.B., Socio-economic factors in child development, Bibliotheca Nutritio et Dieta, No 14, pp 43-63, 1970 Filer, K.J., jr., and G.A. Martinez, Intake of selected nutrients by infants in the United States, Clin. Ped., 3:6333-45, 1964 Mickelsen, 0., L.S. Sims, R.P. Boger, E. Earhart, The prevalence of anemia in Head Start children, Mich. Med. 69:569-75, 1970 Hepner, R., and N.C. Maiden, Growth rate, nutrient in- take and "mothering" as determinants of malnutrition in disadvantaged children, Nut. Rev. 29:219-23, 1971 Hallberg, L., and H.G. Tarwerth, A. Vannotti (eds), Iron Deficiency. New Yerk: Academic Press, 1970 Underwood, E.J., Trace Elements in Human and Animal Nutrition, (3rd ed). New York: Academic Press, 1971 Fairbanks, V.F., and J.L. Fahey, E. Beutler, Clinical Disorders of Iron Metabolism. New York: Grune & Stratton, 1971 Kelemen, E., PhysiOpathology and Therapy of Human Blood Diseases. New York: Pergamon Press, 1969 Ganong, wm.F., Review of Medical Physiolggy. Los Altos, Calif: Lange Medical Publications, 1969 Gross, F. (ed). IEOn Metabolism. Berlin: Springer- Verlag, 1964 70 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 71 Howard, A.N. and I.M. Baird, (eds), Nutritional Defi- ciences in Modern Society. London: Newman Books Ltd, 1973 Texter, E.C. jr., C.-C. Chow, H.C. Laureta, G.R. Van- trappen, Physiolggy of the Gastrointestinal Tract. St. Louis: The C.V. Mosby Company, 1968 Elwood, P.C., Utilization of food iron- an epidemiol- ogist's view, Nutr. et Dieta 8:210-25, 1966 Wohl, M.G. and R.S. Goodhart (eds), Modern Nutrition in Health and Disease. Phil: Lea & Febige, 1968 Finch, C.A., Iron deficiency anemia, Am. J. Ciin. Nut. 22:512-7, 1969 Nutritional Anaemias, World Health Organization Tech- nical Report No. 503, Genewa 1972 Absorption of iron in rats, Nut. Rev. 30:47-9, 1972 Nutrition evaluation of preschool children, Nut. Rev. 30:34-7, 1972 Committee on Nutrition, Iron balance and requirements in infancy, Ped. 43:134-42, 1969 Anderson, T.A., S.J. Fomon, Commercially prepared strained junior foods for infants, J. Am; Diet. Assoc. 58:520-7,1971 Shank, R.E., The role of nutrition in the course of human pregnancy, Nut. News, Vol.33, No.3, p11, Oct. 1970 Smith, C.H., Blood Diseases of infancy and Childhood, 3rd. ed.). St. Louis: The C.V. Mosby Company, 1972 Montagu, M.F.A., Prenatal Influences. Springfield, Ill: Charles C. Thomas, publisher, 1962 Witts, L.J., Hypochromic Anaemia. Phila: F.A. Davis Co., 1969 Lascari, A.D. (ed), The Pediatric Clinics of North.Ameg- ica, vol.19, no.4, Phila: W.B. Saunders Company, 1972 Barr, H.S., S. Barr, E. Stransky, K.B. Rogers, Disorders of Bipod and Blood-forming Organs in Childhood, New YOrk: S. Karger, 1962 Wilson, J.F., D.C. Heiner, M.E. Lakey, Milk-induced gastro- intestinal bleeding in infants with hypochromic microcytic anemia, JAMA 189:568-72, 1964 Ten State Nutrition Survey, USDEW publication 72-8134, 1972 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 72 Fuerth, J.H., Incidence of anemia in full-term infants seen in private practice, J. Ped. 79:560-2, 1971 Sandstead, H.H., J.P. Carter, F.R. House, F. McConnell, K.B. Horton, R.VanderZwaag, Nutritional deficiencies in disadvantaged preschool children, Am. J. Dis. Child. 121:455-63, 1971 Sturgeon, P, Studies of iron requirements in infants and children, II, Ped. 17:341, 1956 Marner, T, Haemoglobin, erythrocytes and serum iron values in normal children 3-6 years of age, Acta Paed. Scand 58:363, 1969 Brigerty, R.E., H.A. Pearson, Effects of dietary and iron supplementation on hematocrit levels of preschool child- ren, J. Ped. 76:757, 1970 Beck, W.S.(ed), Hematolpgy.Cambridge,Mass: MIT Press, 1973 Mauer, A.M. Pediatgic Hematology. New York: The Blakeston Division, McGraw-Hill Book Company, 1969 Dallman, R R., H.C. Schwartz, Cytochrome c and myoglobin in iron-deficient rats, Pediatrics 35:677, 1965 SanPietro, A. (ed), Non-Heme Iron Proteins: Role in Energy Conversion. Yellow Springs, Ohio, The Antioch Press, 1965 Roehm, R.R., H.L. Mayfield, Interrelationship between the biological oxidation mechanism, serum lipid, and serum iron transport system in the rat, J.Nutr. 87:322-30, 1965 Marrow, J.J., J.H. Dagg, A. Goldgerg, A controlled trial of iron therapy in Sideropenia, Scot Med. J. 13:78-83,1968 Owen, G.M., A.H. Lubin, P.J. Garry, Preschool children in the United States: Who has iron deficy, J.Ped. 79:563-8,l971 Beller, E.K., D.A. Howell, A study of anemia and mental functioning in underprivileged children. International Congress of Pediatrics, pp. 109-15, 1971 Kallen, D.J. (ed), Nutrition, development and social be- havior. DHEW Publication No (NIH)73-242, 1970 Brown, A.M., A.P. Matheny, Feeding problems and preschool intelligence scores: a study using the co-twim method, J. Clin. Nutr. 24:1207-9, 1971 Owen, G.M., C.E. Nelson, P.J. Garry, Nutritional status of preschool children: hemoglobin, hematocrit and plasma iron values, J. Ped. 76:761-3, 1970 48. 49. 50. 51. 52. 53. 54. 55. 56. 57. 58. 59. 60. 61. 62. 63. 73 Garn, S., Malnutrition Conference, Merrill-Palmer In- stitute, Detroit, Nov. 16, 1971 Litman, T.J, J.P. Cooney, R.Stief, The views of Minnesota school children on food, J.Am. Diet.Assoc. 45:433-40, 1964 Dwyer, J, J.J. Feldman, J.Mayer, Nutritional Literacy of high school students, J. Nutr. Ed., pp 59-66, Fall 1970 Wang, V.L., Food Information of homemakers and 4-H youths, J. Am. Diet. Assoc. 58:215-9, 1971 Whitten, C.F., TLC and the hungary child, Nut. Today, pp 10-14, Jan-Feb 1972 Fischhoff, J., C.F. Whitten, M.G. Pettit, A psychiatric study of mothers of infants with growth failure secondary to maternal deprivation, J. Ped. 79:209-15, 1971 Whitten, C.F., M.G. Pettit, J. Frischhoff, Evidence that growth failure from maternal deprivation is secondary to undereating, JAMA 209:1675-82, 1969 Lund, L.A., M.C. Burk, Amultidisciplinary analysis of children's food comsumption behavior, Technical Bulletin 265, Agricultural Experiment Station, U. Minn, 1969 Schaefer, E.S., R.Q. Bell, Development of a parental attitude instrument, Child. Dev. 29:339-61, 1958 Snedecor, G.W., W.G. Cochran, Statistical Methods, 6th ed. Ames, Iowa: Iowa State University Press, 1967 Purvis, G.A., R.D. Wallace, J.R. Lovasz, J.W. Harper, R.A. Stewar, The role of supplementary foods- the nutrition of US infants, Gerber Research Center, Fremont, Mich, 1972 Home and Garden Bulletin No. 72, USDA, Wash. D.C., 1970 Sanjur, D., E. Romero, M. Kira, Milk consumption patterns of Puerto Rican preschool children in rural New York, Am. J. Clin. Nut. 24:1320-26, 1971 Captain, O.B., M.S. McIntire, Cost and quality of food in poverty and non-poverty urban areas, J. Am. Diet. Assoc. 55:569-71, 1969 Coltrin, D.M., R.E.Bradfield, Food buying practices of urban low-income consumers- a review, J. Nutr. Ed. pp 16-7,‘Winter 1970 Metheny, N.Y., F.E. Hunt, M.E. Patton, H. Heye, The diets of preschool children, J. H. Ec. 54:297-308, 1962 64. 65. 66. 74 Beal, V.A., A.J. Meyers, R.W. McCammon, Iron intake, hemoglobin and physical growth during lst and 2nd year of life, Pediatrics 30:518, 1962 Beal, V.A., Dietary intake of individuals followed through infancy and childhood, AJPH 51:1107-17, 1961 Filer, L.J., Infant feeding, CMD 555-565, 1970 A PPENDI C ES APPENDIX A IRON DEFICIENCY ANEMIA STUDY PHYSICAL EXAMINATION FORM Weight - lbs, oz Weight - kg Weight - Zile Height - in Height - cm Height - Zile B P P R Child's Name Date Head circumference - cm (L) Mid-upper arm - cm Chest - cm Abdomen - cm I General (physical development, developmental items, disposition, a1ertness:) Head: Eyes: Ears: Nose: Mouth and pharynx: Teeth: Neck: Lymphatics: Chest: Lungs: Heart: 75 APPENDIX B IRON DEFICIENCY ANEMIA QUESTIONNAIRE Child's Name: Sex: Birth Date: Mother's Name: Father's Name: Address: Telephone Number: Interview-Appointment Information Section Date Place Time Interviewer Introduction: We are trying to learn more about the growth and eating habits of young children. In order to learn not only what children eat, but why they eat it, we need to know certain things about the children's families. You, as a parent of a young child, can help to supply the information we need. There are no "right" or "wrong" answers to the questions you will be asked. 76 .mucwomao mo Sofiumumomm musumswuo one mucwomao one mo unmeoomao Hmeuocnm .mcfipwman cu mocwuwmmu Hmwowom m mwcwm: so» mum .mcm we .Howucoo nuuwn we oocuwe use: zuuwm m>wa 6 sec H wcmnoawno once has w>mc cu wcwccmao so» mw< SHH.um o wwmwpumomwz n hocmcwowo mo :oHumcweuwu mo wumn m 77 mucmesoo Spo>wamo Hosea socmcwwum cowumummu- ucwwoz 6A om oz: mouse Socmcwwwm mo numama sauna wmcoaumowdmaoo mw< ucmmwwm m.ewcuoz owoowm hocmqmohm 78 mac: cw wcw>wa mw>wumamulcoz xwmz\owxuoz pwuwaoeoo Sufism: meson wwnesz coHumosooo Hoocom .mnh mo mumum xwm wumocuuwm aficmcowumawm mmedz .oc ummcwam muwnewzlwwwemm spaocwmson use» as m>HH mcomumo some so: mUHHmHmmHo