Contents - Previous - Next

This is the old United Nations University website. Visit the new site at

The introduction of semi-solid and solid foods to feeding infants

Priyani Soysa


A workshop was convened during the International Union of Nutritional Sciences (IUNS) meeting in Brighton, England, in early 1986 to discuss the introduction of semi-solid and solid foods during infancy and the early preschool years.

There is much confusion about the meaning of the word "weaning," although the definition of "wean" given in the Concise Oxford Dictionary, to "accustom to food other than milk," was quoted. In the context of the present need to promote breast-feeding for as long as possible, the term "complementary food" was recently defined by the ACC Sub-committee on Nutrition Consultative Task Force on Maternal and Young Child Nutrition [1, 2]. This emphasizes that foods should be used in addition to breast milk and not as mere replacements for it.

It has not been necessary in most parts of the world to advise mothers to add formula to the infant's diet when their milk appeared inadequate. Most mothers do this on their own, quite often even when it is unnecessary. There was no intention, therefore, to discuss the use of formulae.

In many countries, however, the introduction of semi-solids and solids to infants' diets is unsatisfactory in timing, and the quality and quantity of the foods are insufficient. The effect on infants is growth faltering, which incipiently worsens from around six months of age and results in malnutrition in later months and years.


Global review of infant-feeding practices and maternal attitudes

Developed countries

Feeding patterns in the United States, the United Kingdom, and Scandinavia are somewhat similar. In Scandinavia [Y. Hofvander, personal communication], mothers generally know the current recommendations for introducing semi-solids and solids around the age of four months (sometimes postponed to five months). This usually starts with purees of legumes and/or fruits. By the time the infant is six months old, at least one meal is a full meal with fish and/or eggs and/or meat as well as cow's milk, based on advice given at child health centres. Growth curves show a rate of weight gain in excess of the current normal standards during the first three to four months, after which they fall off a little. There was general agreement in the group that the growth curves used presently are based on data that refer mainly to bottle-fed infants. It was observed that when infants are fed breast milk and the current recommendation to introduce semi-solids by about the fourth month is observed, the curves are different: growth "appears to falter" This faltering is probably due to the use of inappropriately high "normal" weight-for-age values in later infancy.

Approximately 60% of all infants in the United States today begin breast-feeding at birth [ B. Nichols, personal communication]. By two months of age, 61% of all infants are receiving prepared formula. When they are three to four months old, formula usage peaks at approximately 67%, then falls to 64% at five and six months, 41% at eight months, 30% at ten months, and 14% at twelve months.

By two months of age, approximately 15% of the caloric intake is derived from prepared baby foods or pureed home food items. The proportion of calories from these sources increases to approximately 25% at six months of age, and is associated with a decrease in caloric intake from milk from 100% at birth to 60% at six months of age. At twelve months of age, approximately 35% of energy intake is derived from milk and approximately 12% of total calories is derived from prepared baby foods. Between six and twelve months, energy intake from table foods increases from 5% to almost 50%.

Developing countries

The position is different in the developing countries.

In many African countries, some premasticated food such as rice or a tuber is introduced to infants very early. In rural Gambia [3, 4], semi-solids are introduced around the age of four months, although the amounts given tend to be very small. They consist largely of rice, millet, and sorghum, and do not contain more than 30 kcal per 100 g. Deficiencies in vitamins and minerals are even more marked, but this is not the main problem. The foods contain potentially hazardous micro-organisms that are little affected by the rather perfunctory boiling the gruels undergo in the course of preparation. If they are cooked too much, the gruels become over-thick for younger infants' consumption.

In Thailand [5] premasticated rice and banana are usually offered by the end of an infant's first month. No other food is introduced before 20 weeks of age, and then only sugar-based snacks are added by six months. The mothers do not perceive that rice is needed to increase the amount of calories and nutrients for increased needs of growth. Thus the weaning diet through infancy is limited in quantity and variety.

In Cuba [6] fruit juices and purees are popular and are fed to infants before two months of age. Tubers (molonga) are introduced before three months of age. Besides their nutritional value, mothers consider them to be good for digestive function. Although precooked cereals are available, mothers prefer to cook rice and mix it with meat or other foods. Bread is also offered at this age, before the fifth month by 50% of mothers. Legumes, eggs, yellow vegetables, and leafy foods are introduced at about the same time.

In most areas in India [7] mothers are totally unaware of children's nutrition needs. Breast-feeding continues for over a year. Fruits and green leafy vegetables are rarely fed. At about one year some family food is given but is very short of children's requirements. There is a belief that children should not be given solids until they "cut some teeth."

In Sri Lanka [8] about 25% of mothers in urban areas commence semi-solid feeding after the fourth month versus 6% of those in rural areas. Solids such as rusks and biscuits are offered in significant amounts only after the child is six months old. It takes twelve months for 80% to 100% of mothers to give solids. Faltering growth (in comparison with NCHS standards) may be observed around the fourth month in urban areas and earlier in rural areas. This relates to the late complementary feeding in Sri Lanka. Mothers are reluctant to feed earlier because of a traditional rice-eating ceremony around the end of the year or until teeth have erupted. Furthermore, there is a real fear of diarrhoea. This illustrates the observation by Mata and Behar [9] that children in the developing world are constantly exposed to infection and the frequency of diarrhoea increases after complementary feeding commences.

Thus in the developing world, the quality and quantity of the weaning diet require, in addition to more energy and protein, more iron, calcium, vitamins A and D, and trace elements. Also, the diet is monotonous.

Summarizing maternal attitudes, many sociocultural considerations, specific to the different regions, alter the timing of introduction of semi-solids and solids into infants' diets. The privileged populations follow the so-called Western system of feeding. Thus economic background appears to govern the timing and the quality of complementary feeding. The variety of foods offered is limited by the fuel available to cook them and the employment status of mothers. Certainly, the more literate or educated mothers introduce semi-solids and solids earlier than the others.



It is necessary to consider the appropriate timing and quality of complementary foods (semi-solids and solids) because of the prevalence of under-nutrition beginning in late infancy. The strategy for improving the nutrition of young children in underprivileged populations during infancy would be to continue breast-feeding as long as possible, to promote the use of high-quality home-prepared weaning or complementary foods as well as village-prepared weaning foods, and to increase the availability of low-cost, indigenous, centrally processed foods marketed through commercial channels.


Weaning foods

Home-prepared weaning foods

In a series of experiments in India [7] it was shown that simple mixtures of malted cereals, pulses, and oilseeds (in proportions of 4:1:1) are acceptable to mothers and children. Malting consists of steeping for 12 hours, germinating for 24 to 72 hours, sun drying, roasting, and milling. The process is time-consuming and requires space, however. The viscosity of the malted mix is far less than that of roasted mixes (this viscosity is contributed by the malted cereal rather than the pulses).

The most recent experiment was to add a small amount of amylase-rich food (ARF) to a large amount of raw cereal flour. A cooked hot paste of this is thinner than one without ARF. Sorghum germinates well. Its malted germinate is used as the catalytic material. A gruel of 100 g cereal powder.

10 g malt. and 200 ml water is boiled. The cost of the ARF is infinitesimally small compared to that of Taka-diastase, a pure amylase enzyme. It is suggested that rice ARF would be best used for rice, wheat ARF for wheat, maize ARF for maize, etc.

Village-prepared weaning foods

In Thailand, formulation of foods prepared at the home or village level is based on locally available raw ingredients. A rice-legume mixture has been developed that can be roasted simply in a household pan. Proportions are determined by weighing scales or by volumes using a typical village bowl. Manually packed bags may be kept for six to eight weeks.

Commercially prepared weaning foods

Several countries have projects to produce low-cost weaning foods as an intervention for childhood malnutrition. This is especially useful for working mothers. In India several mixtures have been developed, including Indian multipurpose food and corn-soy mixture (CSM). Other mixtures have been developed, for example, in the Sri Avnashilingam Home Science College in Coimbatore.

In Sri Lanka the Food and Nutrition Policy Planning Division has promoted a low-cost weaning food produced by extrusion and consisting of rice flour, soya flour, and green gram flour [7] that is now commercially marketed. Its approximate composition is protein 21%, carbohydrate 62%, and fat 8% (2.5% linoleic acid); its NPU is 79. A vitamin-mineral premix must be imported, but its cost per unit of product is very small.

The choice of strategy

There is no universal way to wean. Each country has its own situation, and within a country there are variations with respect to the privileged and the underprivileged, working and non-working mothers, and urban and rural settings.

The following steps are suggested as a problem-solving strategy.

1. Define the dimensions of the problem:

- the percentage of infants at risk for malnutrition if adequate food is not given at the appropriate time;
- the availability of nutritious foods.

2. Establish the etiology of the problem:

(a) the situation

- the mother's lack of time to prepare and resources (in terms of health and economics) to purchase and prepare foods and fuel;
- household access to water, etc.:
- how foods are obtained, stored, prepared, and fed to children; the manner of feeding: quantity, quality, and variety (descriptive).

(b) attitudes

- beliefs, fears (diarrhoea, etc.), and values:
- intrafamilial relationships and the mother-child relationship;
- the mother's access to household income;
- organization;

(c) access to nutritious food.

3. Determine locally oriented interventions:

- education of the community in general and parents in particular;
- communication of information relevant to the promotion of good weaning practices, such as food hygiene and sanitation, signals of the need to introduce foods (timing), e.g., growth faltering, crying due to hunger;
- development of new foods.



  1. Consultative group on maternal and young child nutrition. Food Nutr Bull 1979;1(3):20-22.
  2. Hofvander Y. Maternal and young child nutrition. Unesco education series no. 3. Paris: Unesco, 1983.
  3. Whitehead RG. Infant feeding practices and the development of malnutrition in rural Gambia Food Nutr Bull 1979;1(4):36-41.
  4. Rowland MGM, Whitehead RG. The epidemiology of protein-energy malnutrition in children in a west African village community. A summary of the work of the Protein-Energy Malnutrition Group of the MRC Dunn Nutrition Unit, Cambridge, UK, and Gambia, 1974-78. Cambridge, UK: MRC Dunn Nutrition Unit, 1979.
  5. Tontisirin K, Moaleekoonpairoj B, Dhanamitta S. Formulation of supplementary foods for infants. In: Proceedings of workshop on breast feeding and supplementary foods. Bangkok: United Production Press, 1980.
  6. Cordova L, Amador M. Prevencion y eradicacion de la male nutrición en Cuba. Rev Cubana Pediat 1978;50: 171.
  7. Gopaldas T. Malted versus roasted weaning mixes: development, storage, acceptability and growth trials. In: Achaya KT, ed. Interfaces between agriculture. nutrition, and food science. Tokyo: United Nations University, 1984.
  8. Soysa P, Senanayake M. The introduction of a low-cost weaning food, its acceptability and effectiveness in a well-baby clinic. Ceylon J Child Hlth 1985.14(1):21-26.
  9. Mata LJ, Behar M. Malnutrition and infection in a typical rural Guatemalan village: lessons for the planning of preventive measures. Epol Food Nutr 1975:4:41-47.

Nutrition and diarrhoeal diseases

Necla Çevik, Benal Büyükgebiz, Atilla Büyükgebiz, and Namik Çevik


Diarrhoea is a major problem of child health in Turkey as in other developing countries [1-3]. It is especially hazardous in children because it may result either in death due to dehydration and electrolyte imbalances or in malnutrition from repeated attacks [2, 3]

The major reasons for malnutrition occurring during and after diarrhoea are restriction of food intake; loss of fluids, nitrogen, and other nutrients; and alterations in digestion, absorption, and metabolism. In addition, parents adhere to superstitions and traditions, and accept incorrect facts with respect to nutrition during diarrhoea [1, 4-6]. Recently developed oral rehydration therapy (ORT), using a sugar-salt solution combined with continued regular feeding, has proved effective in maintaining the nutritional status of children with diarrhoea [7-10].

In this study we investigated a community's level of knowledge and concepts of nutrition during diarrhoea.

A questionnaire containing 15 questions about nutrition was used. Between 1 May and 1 June 1986, 1,680 parents who applied to Hacettepe Children's Hospital and Gulveren and Yunus Emre health centres were questioned by well-trained interns, nurses, and midwives. Results are shown in tables 1-6.



Acute diarrhoeal disease are one of the leading causes of childhood mortality and morbidity in the developing countries, and a major contributor to malnutrition [4 6]. Repeated attacks lead to malnutrition and growth retardation because of anorexia and malabsorption and the associated food restriction instituted by mothers as therapy. A number of studies have revealed that acute diarrhoeal diseases were most severe in undernourished infants and young children [11, 12]. Field studies in developing countries have disclosed a strong correlation between the prevalence of diarrhoea and growth retardation [6, 12]. In addition, the mortality rate is relatively increased in patients with marasmus and kwashiorkor [6, 12].

TABLE 1. Reasons given for stopping feeding during diarrhoea




Feeding increases diarrhoea



Loss of appetite












TABLE 2. Methods of resting the intestine during diarrhoea




Give diarrhoea diet



Give solid food



Stop giving milk



No change in diet









TABLE 3. Reasons given for feeding liquid diet during diarrhoea




Loss of water from the body






To rest the bowel









TABLE 4. Reasons given for not feeding liquid diet during diarrhoea




Increases diarrhoea



Induces vomiting



Is not nutritious



May cause loss of appetite






TABLE 5. Reasons given for stopping breast-feeding during diarrhoea




Breast-feeding increases diarrhoea



Breast-feeding causes diarrhoea



No answer






TABLE 6. Reasons given for diluting cow's milk during diarrhoea




Undiluted milk increases diarrhoea



Undiluted milk is harmful to the bowel



Undiluted milk cannot be absorbed from the bowel



No answer






Withholding food from children with diarrhoea is one of the therapies most frequently employed by parents [13]. The theory is that feeding could enhance the passage of frequent watery stools, thus increasing the severity and prolonging the duration of diarrhoea. Today, however, it is strongly recommended that breast-feeding or any kind of usual feeding be continued. A recent intensive, village-level study of diarrhoeal diseases in Bangladesh documented an annual incidence of 6.8 episodes and an annual prevalence of 15% (55 cases per year) for children under three years of age [10 13, 14]. If these children had to fast for the duration of symptomatic diarrhoea, they would forfeit a sizeable proportion of their potential annual nutrient intake during those years.

It is obvious that after recurrent diarrhoea, patient's nutritional status is affected negatively and existing malnutrition becomes worse. Greater weight gain, however, has been documented in infants given a liberal diet during diarrhoea than in others receiving a more restricted diet [7, 12, 15]. It is well known that even during acute disease 60% or more of nutrients are absorbed normally [7, 9]. Thus it is not necessary to interfere with the regular feeding of children with diarrhoea. In spite of this, according to the answers to our questionnaire, 56.5% of the parents believed that normal feeding should be avoided in cases of diarrhoea.

It is known that lactose carbohydrates, fat, and protein are malabsorbed during diarrhoea [13, 16, 17]. Impaired digestion and absorption in infectious diarrhoea result in a small part from the accelerated transit of food through the intestine, but mostly from the direct action of agents or their products on mucosal function. Some agents multiply to live in the intestinal lumen without causing overt lesions, but they liberate toxins, enzymes, or products that impair digestion and either diminish absorption or increase secretion into the lumen [4, 5]. Also, unabsorbed carbohydrates may give rise to systemic acidosis, and high-molecular-weight proteins absorbed from the damaged intestine may cause food allergy [18].

Because of the potentially sizeable deficit of nutrients induced by fasting during diarrhoea, many public health planners encourage continued feeding throughout the episode. Emphasis on malabsorption overlooks the degree of absorption that does occur when nutrients are offered. Continued feeding during episodes has also been promoted because of the observation that intestinal mucosal disaccharidase levels are dramatically reduced during fasting, and the decrease in intestinal enzyme levels is merely a temporary adaptive response to feeding.

Pancreatic secretions are potent stimulus for intestinal mucosal hyperplasia. Minimal feeding might induce pancreatic enzyme secretion sufficient to maintain a near normal mucosa [5]. Rehydration therapy stops and reverses the dangerous dehydration caused by diarrhoea [13]. Several studies are being performed to reduce the volume and duration of diarrhoea by adding water-soluble, organic molecules such as D-hexoses, neutral amino acids, and dipeptides of neutral amino acids to WHO-formulated ORT solution.

A large percentage of parents in our study (78.8%) fasted their babies because they believed that nutrition would increase the severity and duration of the disease (table 1). Other reasons given were children's loss of appetite (13.2%) and vomiting (4.7%). The fact is, however, that ORT corrects acidosis, hypotension, and hypokalaemia, and when the normal stomach pH is restored, vomiting ceases and appetite returns to normal [13, 19].

The community erroneously believed that it is advisable to rest the bowel during diarrhoea: 65.5% of the parents answering our questionnaire subscribed to this concept, although it has no physiological basis [17, 18, 20] A majority of parents (63.9%) believed in feeding children a liquid diet to rest the bowel and in restricting various nutrients (table 2). although it is now recommended that a fibreless diet. rich in calories and potassium, should be administered during diarrhoeal diseases [7, 12]. We found that 17.3% of parents tended to give solid food to rest the bowel, although this in fact increases dehydration. The ORT or homemade solutions are the preparations of choice for patients with diarrhoea [7, 8,12].

Plain water or similar drinks containing only a small amount of salt are not recomended for dehydrated children, except where sugar and salt are not available.

In such extreme circumstances, any drink available should be used to treat a dehydrated child. In our study 78.2% of the parents gave as their reason for giving water or other solutions during diarrhoea the need to replace the excess loss of water from the body (table 3). This indicates the awareness of parents of some of the consequences of diarrhoea.

Their concern, however, that restricting liquids would provoke vomiting and loss of appetite, leading to malnutrition, is unfounded except when excessive amounts of sugar are added to the solution. Another erroneous belief is that it is advisable to stop breast-feeding or giving cow's milk to children with diarrhoea. In fact, breast-feeding should not be discontinued, and cow's milk diluted by one-half should be given [7, 8, 12] Continuing breast-feeding, or simply normal feeding, is one of the principles of ORT [1,9,10,13,20].

In a study reported from Bangladesh, one group of children with diarrhoea stopped receiving breast milk for 8 to 24 hours and a second group continued to be breast-fed. In the second group, the need for ORT, the time required to achieve rehydration, and the severity of diarrhoea were less than in the first group [6]. The percentage of parents who stopped breast-feeding was 16.4% in our study. The investigators in Bangladesh revealed that the percentage of parents withholding other foods was 61%, while only 12% restricted breast-feeding.

Cow's milk should be diluted during diarrhoea and can be given undiluted when the episode is over. Diluting the milk decreases the lactose concentration and reduces carbohydrate malabsorption; some, however, claim that cow's milk may be given undiluted to patients with diarrhoea [4] In our study, 59% of parents gave diluted cow's milk to their children, while 41% did not.

Education and training should not only transfer facts but also stimulate the imagination. Scientific studies may support the efficacy of oral rehydration, but the most convincing evidence will come when educated health personnel successfully pass their knowledge to the community [21-23] We believe that we will reach the aim of decreased deaths from diarrhoea in our country by implementing an intensive education programme.



  1. WHO-UNICEF. The management of diarrhoea and use of oral rehydration therapy. A joint WHO/UNICEF statement. Geneva: WHO, 1985.
  2. Santos Ocampo PD, Carandang-Bravo L. Oral rehydration therapy in the Philippines: scenario and challenges. Presented at the IPA/UNICEF/WHO symposium on child survival strategies, 18th International Congress of Pediatrics, Honolulu, Hawaii. USA, 7-12 July 1986.
  3. Bertan M. Control of diarrhoeal diseases and oral rehydration therapy in Turkey. Presented at the IPA/UNICEF/WHO workshop on pediatrics in the tropics, Cairo, Egypt, 14-15 Dec 1986.
  4. Mata L. How harmful is diarrhoea? World Health 14. 1986; Apr: 5-7.
  5. Brown KH, Maclean WC. Nutritional management of acute diarrhea: an appraisal of the alternatives. Pediatrics 1984;73(2):119-125.
  6. Recent advances and researches on feeding during and after acute diarrhea. Report of the scientific working group on drug development and management of acute diarrhoea. (CDD/DDM/85.2) Geneva: WHO. 1984.
  7. Santosham M, Reid R. Diarrhoea management. World Health 1986; Apr:8-9.
  8. Merson MH. Diarrhea and oral rehydration. Bull Int Pediatr Assoc 1985;6(1):47-52. normal feeding, is one of the principles of ORT [1, 9, 10, 13, 20].
  9. Hirschorn M. Treatment of acute diarrhea in children: an historical and physiological perspective. Am J Clin Nutr 1980;33(3):637-663.
  10. Çevik N. The importance of nutrition in diarrhea! diseases. Presented at the Nutricia Conference series, Ankara, Turkey, 1986.
  11. Nabarro D. Diarrhea and growth. Dialogue on Diarrhea 1985;23:4.
  12. Fourth programme report 1983-1984. (WHO/CDD/ 85.13) Geneva: WHO, 1985.
  13. Hirschorn B. Successful ORT. Dialogue on Diarrhea 1985 ;22:6.
  14. Black RE, Brown KH, Becker S, et al. Longitudinal studies of infectious diseases and physical growth of children in rural Bangladesh. I. Patterns of morbidity. Am J Epidemiol 1982 ;115: 305-314.
  15. Black RE, Brown KH, Becker S, et al. Longitudinal studies of infectious diseases and physical growth of children in rural Bangladesh. II. Incidence of diarrhea and association with known pathogens. Am J Epidemiol 1982;115:315-324.
  16. Brown KH, Black RE, Parry L. The effect of acute diarrhea on the incidence of lactose malabsorption among Bangladesh children. Am J Clin Nutr 1980;33:2226-2227.
  17. Jonas A, Avigal S, Diver-Haber A. Disturbed fat absorption following infectious gastroenteritis in children. J Pediatr 1979;95:366-372.
  18. Walker-Smith JA. Cow's milk intolerance as a cause of post-enteritis diarrhea. J Pediatr Gastroenterol Nutr 1982;1:163-173.
  19. Hirschorn N. Oral rehydration therapy for diarrhea in children: a basic primer. Nutr Rev 1982;40(4):97-104.
  20. A manual for the treatment of acute diarrhoea. (WHO/CDD/SER/80.2 Rev. 1) Geneva: WHO, 1984.
  21. Cutting W. Convincing the doctors. Diarrhea Dialogue 1983; 13:5.
  22. Fendall NRE, Shattock FM. Health education for diarrhea. Diarrhea Dialogue 1980;3:4-5.
  23. Smith W. Promoting ORT: integrating mass media. print and visual aids. Delivering the goods. Diarrhea Dialogue 1983;14:4-5.

Contents - Previous - Next