Original Article

European Journal of Clinical Nutrition (2011) 65, 26–31; doi:10.1038/ejcn.2010.197; published online 29 September 2010

Gender inequality in food intake and nutritional status of children under 5 years old in rural Eastern Kenya

M Ndiku1,2, K Jaceldo-Siegl1, P Singh1,3 and J Sabaté1,2

  1. 1Department of Nutrition, School of Public Health, Loma Linda University, Loma Linda, CA, USA
  2. 2Department of Family and Consumer Science, School of Sciences and Technology, University of Eastern Africa, Baraton, Eldoret, Kenya
  3. 3Department of Epidemiology and Biostatistics, School of Public Health, Loma Linda University, Loma Linda, CA, USA

Correspondence: Dr J Sabaté, Department of Nutrition, NH 1102, School of Public Health, Loma Linda University, Loma Linda, CA 92350, USA. E-mail: jsabate@llu.edu

Received 13 November 2009; Revised 24 May 2010; Accepted 24 May 2010; Published online 29 September 2010.





Although gender inequality in nutritional status has been consistently reported in several parts of South Asia, in sub-Sahara Africa there is a paucity of data and conflicting conclusions. We conducted a study to assess if gender inequality in food intake and nutritional status is present in rural Eastern Kenya.



This was a descriptive cross sectional study conducted in the Mwingi and Makueni districts of Ukambani region in Eastern Kenya, two rural districts where grains are the main contributor of energy intake. There were 629 children aged <60 months, randomly selected for participation in the study.



Boys consistently had higher energy intakes than girls (P=0.005). More girls were stunted, underweight and wasted 51.7%, (49.9–53.5), 32.1%, (30.4–33.7), 4.6%, (3.9–5.4) than boys 35.9% (34.2–37.7), 14.6% (13.4–15.9) and 1.2% (0.8–1.6), respectively, P<0.001. Of the total, 24.6% (23.1–26.2) of the girls were severely stunted compared with boys 16.3% (15.0–17.7). Boys had higher Z-score indices (height-for-age (HAZ)=−1.33±1.86, weight-for-age (WAZ)=−0.60±1.53 and weight-for-height (WHZ)=0.25±1.23) than girls (HAZ=−2.02±1.94, WAZ=−1.37±1.27 and WHZ=−0.10±1.49), all P<0.001.



The prevalence of malnutrition among children in rural Eastern Kenya is sizable. However, girls were more stunted, underweight and wasted than boys at all age categories due to their consistent lower food intake. Further research is needed to expose the social and cultural determinants underlying gender discrimination in intra-household allocation of food.


stunting; underweight; Ukambani; gender; nutritional assessment; prevalence



Under-nutrition characterized by lack of sufficient food and variety in food choices continues to be a major public health concern in most developing nations. It is exacerbated by diseases, poverty, hostile climates and lack of nutritional knowledge in mothers or caretakers (Bridge et al., 2006a; Bridge et al., 2006b). The main determinants of a child's nutritional status are the child's dietary intake and all the other elements that contribute to the overall health status. At the household level, these two factors are influenced by food security, adequate care for mothers and children and a proper health environment (Smith and Haddad, 2000).

Gender inequalities in quantity and quality of food intake may contribute to under-nutrition mainly in settings where the girl child is still considered less important than the boy child (World Bank, 2006; Dey and Chaudhuri, 2008). Gender inequality in nutritional status due to intra household allocation of resources has been consistently reported in several parts of South Asia (Pal, 1999; Sen, 2001; Moestue et al., 2004; Dey and Chaudhuri, 2008; Moestue, 2008; Dancer et al., 2008). In sub-Sahara Africa, however, there is a paucity of data and conflicting conclusions (Klasen, 1996; KDHS, 2003; Gillet and Perry, 2005; Hadley et al., 2007; Wamani et al., 2007).

We assumed that if boys and girls under 5 years old were cared for and nourished in the same way, their anthropometric status should be similar. Differences in their physical development can be attributed to inequalities in diet and health care. Thus, the aim of this study was to assess if gender inequalities in food intake and nutritional status is present in rural Eastern Kenya. This was part of a study that looked at dietary patterns of under 5 years old children from two rural regions with different dietary patterns (Ndiku et al., 2009).


Materials and methods

Study area and population

Out of the four main geographical regions of Eastern Kenya (Ukambani) we randomly selected two districts. In all, 4 of the 27 divisions were also randomly selected from these two districts, and within the divisions a total of 16 villages were surveyed. All the households in these villages were visited and were eligible for data collection if they included children under 5 years of age. Households where the biological mothers were not present at the homestead at the time of data collection were excluded. A total of 403 households were surveyed and interviews were conducted with the mothers. In all, 201 Mwingi and 202 Makueni mothers were interviewed, and these yielded 629 children: 325 girls and 306 boys. This sample was drawn on the assumption of a two-tailed α of 0.05 and no matching of any variables with a power of 80% to detect a 10% in the prevalence of malnutrition between the two gender groups (boys and girls).

Data collection procedure

The research team in the field consisted of a leader (MN) and four trained research assistants. Data collection proceeded after permission was granted from the district commissioner's office. Data were collected in two phases: January–March 2008 (rainy season), and June–September 2008 (dry/harvest season). At the village level, the team was led by a village elder from house to house during the home visitations. This research was approved by Loma Linda University Institutional Review Board and Ministry of Higher Education, Kenya Government

The data collection instruments included a mother and child section that was a modified rapid, knowledge, practice and coverage survey questionnaire (KPC, 2000) and a 24-h dietary recall of the mother on the child's intake. The instrument was piloted and revised accordingly. A verbal consent was obtained from the mother after a consent statement had been read to her in the local language (Kikamba) before proceeding with data collection. Data collected included demographics, anthropometrics, and biomedical and clinical manifestations of malnutrition, and dietary habits. Age for the children was recorded as reported by the mother and verified with the maternal and child health cards. Where there was discrepancy, the age on the maternal and child health card prevailed.

Anthropometric and clinical data

For children older than 2 years of age weight was measured using an electronic floor scale (Scale-Troni-x-5125, NY, USA), whereas height was measured using a stadiometer (Health scale RGX-120, North Shore Health Supplies, Northbrook, IL, USA) as described by Nieman and Lee, (2006). Height was reported to the nearest 0.1cm and weight was reported to the nearest 0.1kg. For children <2 years old, length was measured using a recumbent length board (Crown, London, UK) and weight was measured using an infant beam balance (Crown) as described by Gibson (2005). Children wore minimal clothing while being weighed.

A hypolet automatic pricking device (Auto-Lancet, Palco Laboratories, Scotts Valley, CA, USA) and sterile hypoguard disposable lancets (TechLite Lancet, Arkray Factory, Minneapolis, MN, USA) were used to make a finger prick. Hemo-control microcuvettes (EKF Diagnostics, Magdeburg, Germany) were used to collect the blood and a hemo-control analyzer (EKF Diagnostics) was used to measure hemoglobin levels on site.

Clinical manifestations of malnutrition in the children were observed and documented as wasting, brown hair, dry skin and edema. Diseases suffered within 1 month before data collection and hospital visitations and hospitalization encountered were documented.

24-h diet recall

Dietary intake of each child was assessed by a face-to-face 24-h dietary recall interview with the mother. An accurate and complete listing of all food and drink consumed by each child within the last 24h was recorded including quantity and time. Preparation methods and how the food was served were noted (details given elsewhere; Ndiku et al., 2010).

Children who were exclusively breastfed and were less than or equal to6 months were estimated to take an average of 25 fluid oz (750ml) of breast milk per day. Therefore, estimating that the babies were fed six times within 24h, an average of ½ cup of breast milk was estimated for every feed in the 24-h recall record. After introduction of solid foods, the milk was adjusted accordingly to accommodate the use of solid foods (Steenbergen and Kusin 1981KDHS, 2003; Kelly, 2008).

Data analysis

Dietary intake was assessed in terms of food groups and food nutrients and was analyzed using NDS-R 2008 (The Nutrition Coordinating Center, Minneapolis, MN, USA; details given elsewhere; Ndiku et al., 2010). Height, weight and age were used to compute the malnutrition indices (Z-scores) using EPI INFO 2000 (NCHS, 2000).

Statistical analysis was performed using SAS version 9.1 (SAS Institute Inc., Cary, NC, USA). Results are reported as mean±s.d. unless otherwise indicated. The nutritional data were normalized using log transformations where appropriate. Descriptive analyses were done, using student's t-test for continuous variables and Fisher's exact test and Pearson χ2-test for categorical variables, with significance assumed at an α of 0.05.



A total of 403 mothers were interviewed for 24-h diet surrogate recalls and 629 diet recalls for their children were collected (Table 1). Of the 629 children, 49% were boys and 51% were girls. The average age for the interviewed mothers was 30±7 years and the mean age for the children was 26.5±17 months. Table 1 presents the distribution of the under 5 years children according to age and gender. There was a progressive decrease in the number of children as age increased.

Figure 1 represents the contribution of foods and food groups to daily energy intake in children by gender according to different age categories. The contribution of breast milk to energy intake tapered off and ceased to be a major contributor after 18 months. There was a consistent increase in energy intake for boys as age increased. The girls showed inconsistent increases of energy intake with age, with two main drops in energy at age category 25–36 months and age category 49–60 months. Overall, boys had a 12% higher mean energy intake 4688±2378kJ than girls 4169±2231kJ, P=0.005. Boys had significantly higher energy intake from grains 3533±2240kJ than girls 3018±2093kJ, P=0.021. Intakes from all the other food groups were not significantly different (data not shown).

Figure 1.
Figure 1 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

Contribution of foods and food groups to daily energy intake of under 5 years children in rural Eastern Kenya by gender according to different age categories. Others include: legumes, fruits/vegetables, fats and meat/eggs.

Full figure and legend (167K)

The prevalence of malnutrition was significantly greater in girls than boys as seen in all the anthropometric indicators (Table 2). After 6 months of age, anthropometric indices of malnutrition (mean Z-scores) for all the children dropped below the reference line (Z-score of zero) with height-for-age (HAZ) mean Z-scores decreasing below the moderate malnutrition cut point (−2 s.d.) at 13–18 months of age (Figure 2). HAZ and weight-for-age (WAZ) mean Z-scores were significantly higher for boys than girls in all age categories (P<0.001). Weight-for-height (WHZ) Z-score means were significantly higher in boys than girls only at (7–12, 25–36, 37–48 and 49–60 months; Figure 3).

Figure 2.
Figure 2 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

Mean Z-score values for three anthropometric indicators height-for-age (HAZ), weight-for-age (WAZ) and weight-for-height (WHZ) for <5 years in rural Eastern Kenya (n=629). The horizontal line at the Z-score value of zero represents the median Z-score of the reference population. Z-score of –2 s.d. represents moderate malnutrition and Z-score of –3 s.d. represent severe malnutrition. The National Center for Health Statistics/World Health Organization reference was used to calculate the Z-scores (NCHS, 2000). Cases were weighted by village size.

Full figure and legend (63K)

Figure 3.
Figure 3 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

Height-for-age, weight-for-age and weight-for-height for under 5 years children in rural Eastern Kenya by gender. The horizontal line at the Z-score value of zero represents the median Z-score of the reference population. Z-score of –2 s.d. represents moderate malnutrition and Z-score of –3 s.d. represent severe malnutrition. The National Center for Health Statistics/World Health Organization reference was used to calculate the Z-scores (NCHS, 2000). Cases were weighted by village size.

Full figure and legend (83K)

Overall, boys had significantly (P<0.001) higher mean Z-scores for HAZ, WAZ and WHZ (−1.33, −0.60, 0.25, respectively) compared with girls (−2.02 HAZ, −1.37 WAZ and −0.10 WHZ; data not shown).

Preventive-measures differences were not significant except for deworming, which was favorable for girls. Apart from diarrhea, girls were prone to have slightly more infectious diseases than boys. Except for edema and hemoglobin all other clinical signs of malnutrition were more prevalent in girls than boys (Table 3).



The salient findings from this study were that boys overall had higher energy intakes than girls, and this difference appeared to be primarily from a greater intake of grains among the boys. Girls at all age categories had greater prevalence of malnutrition than boys. There were no major gender differences in preventive measures; however, the girls had experienced slightly more infectious diseases than boys.

Energy and protein requirement seem to be no different for boys and girls <5 years old as there are no energy and protein intake recommendations until >8 years (WHO, 1985; IOM (Institute of Medicine) of the National Academies, 2006). Thus, it is likely that differences found in anthropometric indicators of malnutrition among the girls in our study were due to lower energy intake combined with greater prevalence of infectious disease. As there was no difference in preventive measures, it is suggested that the greater prevalence of infectious disease in the girls was a consequence rather than a cause of their poorer nutritional status. Differences in food intake between genders because of intra-household allocation of resources most probably led to these marked gender differences in anthropometric indicators of malnutrition.

Findings of this study, that in general, boys’ energy intake is higher than girls, agree with a study in Ethiopia among adolescents. In that study despite no differences in their households food insecurity status, girls reported being food insecure themselves (Hadley et al., 2007). Our findings are also supported by anthropometric, mortality and population data analyzed by Klasen, who found evidence of a slight and rising anti-female bias in sub-Saharan Africa, which was particularly apparent in mortality and population indicators (Klasen, 1996).

Our study findings that girls had higher prevalence rates for both moderate (−2 s.d.) and severe malnutrition (−3 s.d.), concur with a study in India reporting 55.9%, 51.4 and 42.3% of the girls underweight, stunted and wasted, respectively, compared with 46.6, 40.5 and 35.3% of the boys (Dey and Chaudhuri, 2008). Underlying reasons for the differences were suggested to be that girls have less access to nutrition, physical and mental health care, and education (Dey and Chaudhuri, 2008).

Our findings that boys (compared with girls) had more favorable mean Z-scores for the three anthropometric indices (that is, HAZ, WAZ and height-for-weight), agree with those of the KDHS. The 2003 KDHS reports higher Z-scores for HAZ, WAZ and WHZ in boys (1.4, 1.0 and 0.3, respectively) compared with girls (1.1, 0.9 and 0.2, respectively; KDHS, 2003).

Moestue et al. (2004) and Moestue (2008), two studies conducted in Bangladesh, caution against the use of anthropometrics to measure gender nutritional disparity because of the different conclusions drawn when the three known references are used, that is, National Center for Health Statistics (NCHS/WHO, 2006), Center for Disease Control (CDC, 2000) and British growth references 1990. Our study used the NCHS as reference. While this caution is an important consideration, the difference we found between genders in anthropometric measures seems to be supported by the differences in energy intakes and lends strength to the study.

Inequality between girls and boys can take many different forms. One of those could be neglect of health, nutrition and other needs of girls that influence survival (Sen, 2001; Osman and Sen, 2003). The girl child in most cases is disadvantaged from birth and throughout her entire life (Dey and Chaudhuri, 2008). Women's deprivation in terms of nutrition and healthcare rebounds on society as a whole in the form of ill-health not only for themselves including reduced productivity, but also for their offspring (males and females alike—both as children and as adults; Osman and Sen, 2003; World Bank, 2006).

Malnutrition causes growth retardation, a physiologically and economically costly human condition (Leenstra et al., 2004). It retards children's physical and cognitive development and increases susceptibility to disease (World Bank, 2006). Under-nutrition of children erodes human capital, as it negatively influences the chances that a child will go to school, stay in school and perform well (World Bank, 2003). Malnutrition is a violation of a child's human right (Smith and Haddad, 2000). The World Health Organization recognizes the importance of investing in nutrition as a critical component to achieving the millennium development goals (World Bank, 2006).

Our study illustrates the complex causes of malnutrition. Future research should ascertain if this pattern is the same in other parts of rural Africa. If confirmed as a generalized problem, it warrants further study to explore the possible determinants of malnutrition at the household level such as the reasons for intra-household discrimination in allocation of food. Results from our study may be used to support nutritional intervention policy in Eastern Kenya to mitigate malnutrition in young children.


Conflict of interest

The authors declare no conflict of interest.



  1. Bridge A, Kipp W, Jhangri GS, Laing L, Konde-Lule J (2006a). Nutritional status of young children in AIDS-affected households and controls in Uganda. Am J Trop Med Hyg 74, 926–931.
  2. Bridge AK, Raine W, Konde-Lule J (2006b). Nutritional status and food consumption patterns of young children living in Western Uganda. East Afr Med J 83, 619–625.
  3. CDC (2000). Clinical Growth Charts. Available online at http://www.cdc.gov/growthcharts/
    . Accessed 12 June 2008.
  4. Dancer D, Rammohan A, Murray SD (2008). Infant mortality and child nutrition in Bangladesh. Health Econ 17, 1015–1035. | Article | PubMed
  5. Dey I, Chaudhuri RN (2008). Gender inequalities in nutritional status among under five children in a village in Hooghly district, West Bengal. Indian J Public Health 52, 218–220. | PubMed |
  6. Gibson RS (2005). Principles of Nutritional Assessment, 2nd ed. Oxford University Press: New York.
  7. Gillet R, Perry A (2005). Gender and nutritional status at the household level among Gwembe valley Tonga children, 0–10 years. Am J Hum Biol 17, 372–375. | Article | PubMed
  8. Hadley C, Lindstron D, Tessema F, Belachew T (2007). Gender bias in the food insecurity experience of Ethiopian adolescents. Soc Sci Med 66, 427–438. | Article | PubMed
  9. IOM (Institute of Medicine) of the National Academies (2006). Dietary Reference Intakes. The National Academies Press: Washington, DC.
  10. KDHS (2003). Kenya demographic health survey. Calverton National Council of Populations and Development: Claverton, MD, 153–168.
  11. Kelly B (2008). How much milk do babies need: Kellymom breastfeeding and parenting. Available online at http://www.kellymom.com/bf/pumping/milkcalc.html. Accessed 12 December 2008.
  12. Klasen S (1996). Nutrition health and mortality in sub-Sahara Africa: is there a gender bias? J Dev Stud 32, 913–932. | Article
  13. KPC (2000). Rapid knowledge, practices and coverage (KPC) survey module 3: growth monitoring and maternal/child anthropometry. Available online at http://www.childsurvival.com/kpc2000/mod3.pdf. Accessed 15 March 2007.
  14. Leenstra T, Kariuki SK, Kurtis JD, Oloo AJ, Kager PA, ter Kuile FO (2004). Prevalence and severity of anemia and iron deficiency: cross-sectional studies in adolescent schoolgirls in western Kenya. Eur J Clin Nutr 58, 681–691. | Article | PubMed | ISI | ChemPort |
  15. Moestue H, de Pee S, Hall A, Hye A, Sultana N, Ishtiaque MZ et al. (2004). Conclusions about differences in linear growth between Bangladeshi boys and girls depend on the growth reference used. Eur J Clin Nutr 58, 725–731. | Article | PubMed
  16. Moestue H (2008). Can anthropometry measure gender discrimination? An analysis using WHO standards to assess the growth of Bangladeshi children. Public Health Nutr 12, 1085–1091. | Article | PubMed
  17. NCHS (2000). CDC Growth Charts: United States. Available online at http://www.cdc.gov/growthcharts/. Accessed 12 June 2008.
  18. NCHS/WHO (2006). WHO Child Growth Standards. Available online at http://www.cdc.gov/growthcharts/who_charts.htm. Accessed 12 June 2008.
  19. Ndiku M, Jaceldo-Siegl K, Singh P, Sabaté J (2010). Dietary patterns of children <5 years old in mwingi and makueni districts of Ukambani region, Eastern Kenya. Afr J Food Agric Nutr Dev 10, 2804–2817.
  20. Ndiku M, Sabaté J, Jaceldo-Siegl K (2009). Dietary patterns of children 0–59 months in Ukambani region of Kenya. FASEB J 23, 916 (abstract).
  21. Nieman D, Lee R (2006). Nutritional Assessment, 3rd ed. Library of Congress Cataloging-Publication Data: New York.
  22. Osman S, Sen A (2003). The hidden penalties of gender inequality: fetal origins of ill-health. Econ Hum Biol 1, 105–121. | Article | PubMed
  23. Pal S (1999). An analysis of childhood malnutrition in rural India: role of gender, income and other household characteristics. World Dev 27, 1151–1171. | Article
  24. Sen A (2001). Many faces of gender inequality. The Hindu: India's National Magazine 18, Available online at http://www.hinduonnet.com/fline/
    . Accessed 10 November 2009.
  25. Smith LC, Haddad L (2000). Overcoming child malnutrition in developing countries: Past achievements and future choices A 2020 vision for food, agriculture, and the environment. International Food Policy Research Institute (IFPRI): Washington, DC. pp 1–2.
  26. Steenbergen VM, Kusin JA (1981). Lactation performance of Akamba mothers, Kenya. Breast feeding behavior, breast milk yield and composition. J Trop Pediatrics 27, 155–161.
  27. Wamani H, Astrom AN, Peterson S, Tumwine J, Tylleskar T (2007). Boys are more stunted than girls in sub-Saharan Africa: a meta-analysis of 16 demographic and health surveys. BMC Pediatr 7, 1–10. | Article | PubMed
  28. World Bank (2003). Gender equality and the millennium development goals. Gender and Development Group. Available online at http://siteresources.worldbank.org/INTGENDER/Publications/
    . Accessed 10 September 2009.
  29. World Bank (2006). Repositioning nutrition as central to development a strategy for large-scale action. Washington DC: USA. Available online at http://siteresources.worldbank.org/NUTRITION/Resources/
    . Accessed 15 September 2009.
  30. WHO (1985). Energy and protein requirements. Report of a Joint FAO/WHO/UNA expert consultation WHO Technical Report Series No. 724. Geneva.


We thank the Federico Foundation, Switzerland for funding this project, Dr Patricia Johnson for revising a previous version of this manuscript, and the field research team for invaluable assistance during data collection. We also acknowledge the time given by the subjects and mothers who participated in the study.

Extra navigation