The Importance of Investing in the First 1000 Days of Life: Evidence and Policy Options

Abstract

The first 1000 days of life starts from conception to a child’s second birthday. Research suggests that the period is critical for cognitive, physical, and emotional development. Investments in maternal and child healthcare during this period have a profound impact on long-term health, educational attainment, and economic productivity. This study examined the impact of such investments on child health outcomes in Kenya, using data from the 2015/2016 Kenya Integrated Household Budget Survey (KIHBS). Key areas of focus included maternal healthcare, early antenatal care, skilled delivery, exclusive breastfeeding, proper weaning practices, immunization, and the timely treatment of childhood illnesses. Using the Cox regression hazard model, the study revealed that twins faced a higher risk of mortality compared to single births, while firstborns were less likely to die before their fifth birthday; larger household sizes were associated with reduced child mortality, and children in female-headed households had a lower likelihood of dying, likely due to better adherence to proper health and nutritional practices. Maternal health conditions, the place of delivery, and assistance during childbirth significantly influenced survival, with government health facility deliveries yielding better outcomes than homebirths. This study emphasizes the importance of educating pregnant women and mothers on health risks and public health protocols during this critical period. Strengthening healthcare systems and promoting equitable access to essential services during the first 1000 days could improve child survival rates and enhance long-term economic productivity.

1. Introduction

The first 1000 days of life refers to the period from conception to the second birthday in a child’s life, that is, gestation, infancy, and toddlerhood. This is an important period since the body is undergoing significant brain, physical, and immunity development, as well as neurodevelopment (Sanefuji et al., 2021). The period is, therefore, important for public health interventions that will support child development during this critical period, preventing deficiencies, such as malnutrition and infections, that become counterproductive to the child’s health and result in negative economic impacts. The first 1000 days of life also play a crucial role in human capital formation, a fundamental driver of economic development (Martorell, 2018). This critical period has lasting effects on a child’s future health, development, and productivity in adulthood. The South African Child Gauge highlights how early life experiences serve as the foundation for human capital development (Jamieson et al., 2017). A healthy pregnancy, followed by adequate child healthcare and supportive parenting, positively influences a child’s educational attainment and smooth transition into the workforce, ultimately contributing to sustained economic growth and development.

Human capital constitutes the impact of health and education on productivity in the workplace. With a primary focus on human capital formation, it is important to recognize that human capital begins from the time of conception, through pregnancy, to the delivery of a child and their development to adulthood. A study published by the International Institute for Applied Systems Analysis (IIASA) in 2008 compared demographic and human capital trends in Eastern Europe and in Sub-Saharan Africa. The world was forecasted to have two major demographic occurrences in two decades: rapid population growth and rapid population ageing (Lutz et al., 2008). From the study, Sub-Saharan Africa was predicted to experience a population explosion that would boost the workforce, while Eastern Europe would experience a population shrink. It, however, also brought up an inverse correlation between population growth and education attainment, predicting higher per capita productivity in Eastern Europe owing to increased education attainment among the adult working population as compared to Sub-Saharan Africa. The Human Capital Index (HCI) comprises six inputs in health and education. The health inputs are the probability of a child surviving to the age of 5 years, the adult survival rate, and the proportion of children without stunted growth. The education inputs are the expected years of schooling, harmonized test scores, and the learning-adjusted years of school. The HCI for Kenya in 2018 was 52%, relative to a global average of 56%. This implies that a child born in Kenya today will have a productivity of 52%, relative to what it could be if the child enjoyed full health and completed their education. The Seychelles and Mauritius are the best performing among the Sub-Saharan African countries, with HCIs of 68% and 63%, respectively. The main factors explaining the level of the HCI include health outcomes since the productivity of individuals is highly dependent on the quality of their health and early investment in the education they receive. Sub-Saharan Africa was noted to have a low HCI of 0.57 in 2020 as it has higher mortality rates from childhood to adulthood, lower education levels amongst the population, and higher rates of poverty than do developed nations (Shobowale et al., 2024).

The child mortality rate in children under 5 years old in Kenya was 43.2 deaths per 1000 live births in 2021, having improved from 52 per 1000 live births (KNBS, 2015). The rates have been improving over the years, but the country has not reached the set target. The SDG target 3.2 aims to reduce the under-5 mortality rate to less than 25 deaths per 1000 live births and the neonatal mortality rate to less than 12 deaths per 1000 live births. The country can only reach this target once there is adequate treatment and prevention of the common causes of child mortality, which are pneumonia, diarrhea, malaria, and malnutrition. The morbidities account for a third of the causes of child morbidities.

Low investment in a child’s early days of life has been associated with stunting, becoming underweight, and wasting, which have long-term implications on human capital accumulation. According to the KNBS (2015), the prevalence of stunting was estimated at 26%, wasting at 4%, and being underweight at 11% in 2014. The levels improved in 2022 to the prevalence of child stunting being at 18%, wasting at 5%, and being underweight at 10% (KNBS & ICF, 2023). Stunting refers to an individual having a low height for their age and is a marker of chronic malnutrition and wasting, which refers to an individual having a low weight for their height and is a marker of acute malnutrition: examples include marasmus and kwashiorkor. Being underweight refers to an individual having a low weight for their height and is a combination of both stunting and wasting. Understanding and preventing malnutrition is crucial because it not only exposes the child to opportunistic infections but also has negative effects on the physical and mental development of the child.

Maternal healthcare has remained a major public health concern across the globe and more so in low- and middle-income countries. Specifically, in Kenya, in 2022, the maternal mortality ratio (MMR) was 362 maternal deaths per 100,000 live births (translating to about 6000 maternal deaths per year), and the stillbirth rate was 23 deaths per 1000 live births (translating to 35,000 stillbirths per year), far below the target of 147 maternal mortalities per 100,000 live births and 12 stillbirths per 1000 live births by 2030 (KNBS & ICF, 2023).

The current levels of child undernutrition illustrate the continuing challenges for the reduction of child hunger. Statistics show that stunting affects 2 in every 10 children under 5 years old (KNBS & ICF, 2023). This is way below the Sustainable Development Goal (SDG) target of reducing the percentage of stunted children to 14.7% by 2030 and has long-term effects on human capital development in the country.

Inequities in the immunization coverage between populations in Kenya have persisted, even with the devolution of primary healthcare in the country. In 2014, there was a 17.7 percentage point difference in the DPT3 immunization coverage between the highest coverage in Central Province and the lowest in North Eastern Province (WHO, 2018). With almost 1.5 million children born each year in Kenya, relatively small proportional differences in the immunization coverage between subgroups translates into large absolute numbers of under-immunized and zero-dose children (UNICEF, 2019). In 2022, the vaccination rate was 76% for children aged 12–23 months and 61% for children aged 24–35 months (KNBS & ICF, 2023).

In addition, despite government efforts to improve the lives of pregnant mothers and unborn children through the introduction of various health programs (such as free maternity services being introduced in 2013 in all public health institutions across the country), the number of mothers utilizing the services is still below the targets. Moreover, only about 61.8% of deliveries are attended by skilled providers (Ministry of Health, 2018). The low utilization of free maternity services was linked to patients’ satisfaction in previous deliveries, which was influenced by the healthcare worker’s activities and actions towards women in labor. Other factors included the mother’s level of education, which played a role in demystifying poor-delivery-outcome-related beliefs and increased parity (Ngesa et al., 2021).

If these issues are left unresolved, related illnesses would reduce productivity and increase health expenditure, burdening Kenya’s economy, while addressing the issues would lower under-five morbidity and mortality. Therefore, it is important to address the issues that affect children in their early stages in life for improved health outcomes in the future and increased productivity later in life, especially in the labor market. This study, therefore, assessed the link between investing in the first 1000 days of life and health outcomes in Kenya using KDHS 2022 data (KNBS & ICF, 2023).

This study focused on the effects of investing in the first 1000 days of life, consequently giving each child an opportunity to survive and realize and maximize their potential in adulthood, which is important for human capital development. The basic needs and services during the first 1000 days of life are maternal healthcare; child healthcare; the mobilization of mothers to seek early antenatal care; increasing hospital deliveries; enhancing exclusive breastfeeding for the first 6 months of life; increasing knowledge on the proper weaning diet; immunization; and the early diagnosis and treatment of common childhood illnesses.

To explore the effects associated with investment in the first 1000 days of life on child outcomes in Kenya, this study focused on three questions. These included (i) assessing the status of the essential needs of children and households with children under 2 years old; (ii) estimating the effect of access to critical services on child health and mortality outcomes; and (iii) to provide policy recommendations to enhance investment in children’s health and reduce child mortalities.

2. The Theoretical and Empirical Literature

This study is grounded in the Human Capital Theory, which posits that investments in health, education, and training enhance individual productivity and income potential (Marginson, 2019). The theory, stemming from Smith (1776)’s The Wealth of Nations, emphasizes education as a capital good that is crucial for economic transformation and skill development. Investments in child education and healthcare are vital for improving human capital and economic output (Becker, 1993). Research indicates that enhancing children’s health significantly contributes to better educational outcomes and productivity in adulthood, making it a valuable economic investment.

The period between conception to a child’s second birthday is widely recognized as a critical period for growth and development. Research highlights that investments in maternal and child healthcare during this period significantly impact child survival, cognitive development, and long-term well-being (M. M. Black et al., 2017). This section reviews key research on maternal health, child healthcare, and essential interventions that shape outcomes for children under five years old. A comparative study (Timaeus & Lush, 1995) of urban areas of Ghana, Egypt, Brazil, and Thailand with an objective of assessing the intra-urban differentials in child health for health transition clearly indicated that children’s health is affected by environmental conditions and the economic status of the household. Children from better-off households had lower diarrhea morbidity and mortality in Egypt, Thailand, and Brazil. The differentials in diarrheal diseases by household economic status were due to differences in childcare practices, such as the preparation of weaning foods and personal hygiene (Timaeus & Lush, 1995). Jacoby and Wang (2003) examined the linkages between child mortality–morbidity and the quality of the household–community environment in rural and urban China using a competing risks approach. The study established that access to immunization reduces the diarrhea incidence in rural areas, and access to modern sanitation facilities like flush toilets reduces the diarrhea prevalence in urban areas.

Maternal health plays a pivotal role in determining child survival and development. According to the World Health Organization (WHO, 2021), complications during pregnancy and childbirth contribute to a high burden of neonatal mortality, with 289,000 maternal deaths, 2.6 million stillbirths, and 2.7 million neonatal deaths occurring globally each year (Lang’at et al., 2019). Evidence suggests that early antenatal care (ANC), skilled delivery, and postpartum support reduce maternal and infant mortality rates (Islam & Tabassum, 2021). In low- and middle-income countries (LMICs), access to quality maternal healthcare remains a major challenge, contributing to adverse birth outcomes, such as low birth weights, preterm births, and perinatal asphyxia (Salam et al., 2023).

Childhood mortality remains a major concern, particularly in LMICs where under-five deaths are often linked to preventable infectious diseases and malnutrition (WHO, 2021). Studies show that over 45% of under-five deaths are nutrition-related (UNICEF, 2019). The high mortality rates in the first five years of life are primarily due to pneumonia, diarrhea, malaria, and neonatal complications, all of which can be mitigated through timely immunization, proper nutrition, and improved healthcare access (Victora et al., 2016).

Research suggests that key interventions for child survival and development include access to early antenatal care, increased skilled assistance with births, hospital deliveries, exclusive breastfeeding, immunization, proper weaning, and nutritional support. Antenatal care is essential in reducing pregnancy-related complications. Community health workers (CHWs) play a critical role in mobilizing mothers for ANC visits, providing nutritional counseling, micronutrient supplementation, and early screening for maternal conditions (Olaniran et al., 2019). ANC reduces the risk of preterm birth, malnutrition, and iodine-deficiency-related cognitive impairment (M. M. Black et al., 2017).

Skilled birth attendance significantly reduces perinatal asphyxia, stillbirths, and postpartum complications (WHO, 2021). Hospital deliveries lower maternal mortality rates, particularly from postpartum hemorrhage and obstructed labor (Campbell & Graham, 2006). Evidence suggests that facility-based deliveries with skilled personnel could prevent up to 60% of maternal deaths in LMICs (WHO, 2021).

Exclusive breastfeeding is crucial for optimal infant nutrition, immune system development, and cognitive growth (Victora et al., 2016). The WHO’s Baby-Friendly Hospital Initiative (BFHI) promotes breastfeeding in hospitals; however, post-discharge support is limited, leading to early weaning and increased malnutrition risks (Victora et al., 2016). There is a need for community-based breastfeeding education programs to improve adherence.

The weaning period (6–24 months) is a vulnerable phase, often characterized by poor dietary diversity and malnutrition (Pinchoff et al., 2021). Studies indicate that many children in LMICs receive starch-heavy diets with limited protein and micronutrient intakes, leading to stunting, marasmus, and kwashiorkor (M. M. Black et al., 2017). Addressing food security, maternal education, and nutrition-sensitive interventions are critical for improved health outcomes (GNR, 2020).

Vaccination remains one of the most cost-effective public health interventions. According to the WHO (2021), immunization prevents 2–3 million child deaths annually. Expanding access to vaccines against pneumonia, diarrhea, and measles could significantly reduce child mortality rates (Maruta & Afoakwah, 2023). Additionally, the early diagnosis and treatment of childhood illnesses, such as pneumonia and gastroenteritis, are crucial in reducing under-five mortality (Maruta & Afoakwah, 2023).

However, the available studies were limited in scope, with none focused on children aged 2 years and below. Therefore, our study will focus on the impact of investing in the first 1000 days of life on the health outcomes of children under the age of 5 years and, by extension, on human capital development.

3. Methodology

This section provides a summary of the theoretical framework adopted, the research design used, the data type and source, and the definition and measurement of the variables pertaining to children under two years old, focusing on their demographics, health, and nutrition in the household.

3.1. Research Design

A quantitative research design was employed to investigate the relationship between investment in the first 1000 days of life and child health outcomes in Kenya. This involved a review of the literature and descriptive statistics on basic child needs, including maternal and child healthcare, exclusive breastfeeding, immunization, and the treatment of common childhood illnesses. This study assessed the relationship between access to these basic services and child mortality.

3.2. Study Data

This study used data from the 2015/16 Kenya Integrated Household Budget Survey (KIHBS), covering 24,000 households across Kenya. The data relevant to children under two years were extracted, resulting in a sample size of 12,630 observations to aid in examining demographics, health, and nutrition through household data.

3.3. Empirical Model

This study employed the Cox Proportional Hazards model to evaluate the impact of access to basic health services on infant and child mortality rates in Kenya. This semi-parametric regression model, established by Cox (1972), is widely utilized in medical research to analyze the relationship between survival time and predictor variables.

The Cox model is appropriate for this research as it accommodates both quantitative and categorical predictor variables, allowing for the assessment of multiple risk factors simultaneously. The model is semi-parametric in nature and allows for analysis without assuming a specific distribution for survival times, making it flexible and robust for real-world data. Further, the model allows for the inclusion of multiple covariates, enabling this study to assess factors such as breastfeeding, maternal education, and healthcare access on child mortality. The model provides hazard ratios (HRs), offering intuitive interpretations for study. It also assumes proportional hazards, allowing for stable risk estimations over time. Its applicability to both large and small datasets makes it useful in this context. The model is expressed mathematically as follows:

$$h\left(t\right)=h_0\left(t\right) \ast \mathrm{e}\mathrm{x}\mathrm{p}(b_1{x}_1+b_2,\dots ,+b_p{x}_p)$$

(1)

where $h\left(t\right)$ is the hazard function, $h_0\left(t\right)$ is the baseline hazard, and b₁, $b_2$ …, and $b_p$ are the estimated coefficients indicating the effect sizes of the covariates. Hazard ratios (HRs), defined as exp(b_i), provide insights into the relationship between the covariates and the risk of death.

The interpretation of the HR values is as follows:

• HR = 1: no effect.
• HR < 1: reduced hazard.
• HR > 1: increased hazard.

The model aims to estimate the multivariate Cox regression as follows:

$$h\left(t\right)=h_0\left(t\right) \ast \mathrm{e}\mathrm{x}\mathrm{p}(b_1{x}_1+b_2{x}_2+\dots +b_p{x}_p)$$

(2)

The relative risk of death will be estimated by

$${\displaystyle \frac{h_i\left(t_i;x_i\right)}{h_0\left(t_i\right)}}={exp}^{b^i{x}_i}$$

(3)

The decision rules are based on the HR relative to the comparison group (counties). The model assumes that regular clinical interventions, such as antenatal and postnatal care visits, take precedence over maternal medical conditions, affecting the probability of child mortality.

The dependent variable, childhood mortality, is analyzed in two age intervals:

• Infant mortality (birth to 12 months).
• Child mortality (12 months to 5 years).

The independent variables include the following:

• Maternal and child healthcare indicators (e.g., the place of birth, assisted delivery, antenatal care visits).
• Basic healthcare inputs (e.g., exclusive breastfeeding, immunization).
• Nutrition-related outcomes (e.g., stunting, wasting, being underweight).
• Health morbidity factors (e.g., maternal morbidity, gestational diabetes).

The dependent variables comprise the following:

• Infant mortality: the probability of dying between birth and the first birthday.
• Child mortality: the probability of dying between ages one and five.
• Child morbidity: the probability of a child experiencing diarrhea from birth to age two.

The analysis focuses on households with children born alive but who died within the period 2015–2016. The KIHBS 2015–2016 data were utilized to assess mortality rates across the specified age groups. The STATA software (version 18) facilitated the estimation of the coefficients and the hazard ratios, with the individual level as the unit of analysis. The results were presented using appropriate tables and figures.

Additionally, this study employed a Probit regression model to examine the determinants of child morbidity, focusing on the probability of children aged 0 to 2 years falling sick from diarrhea. The Probit model is appropriate for estimating binary dependent variables, as it assumes a normal distribution of the error term, allowing for a more accurate representation of the probability of illness occurrence.

The empirical model specification is presented below, where the probability that a child, $i$, experiences diarrhea ($Y_i$ = 1) is modeled as a function of various socioeconomic, environmental, and health-related factors. The Probit model is expressed as follows:

$$P\left(Y_i=1/X_i\right)=\varnothing (X_i\beta +{\epsilon }_i)$$

(4)

where the

• $P\left(Y_i=1/X_I\right)$ represents the probability that child $i$ suffers from diarrhea.
• $\varnothing$ is the cumulative distribution function of the standard normal distribution, ensuring that the estimated probabilities lie between 0 and 1.
• $X_i$ is a vector of explanatory variables influencing child morbidity.
• $\beta$ represents the coefficients to be estimated.
• ${\epsilon }_i$ is the normally distributed error term.

This empirical approach allows for a quantitative assessment of key factors influencing child morbidity, informing targeted interventions to reduce childhood diarrhea in Kenya.

3.4. Variables and Measurement

Households were deemed to have access to basic needs and services if they had access to key maternal and child health indicators, as indicated by the predictor values below. The dependent variables included infant mortality (deaths from birth to 12 months), child mortality (deaths from 12 months to 60 months), and child morbidity (diarrhea incidence). The predictor variables included indicators related to maternal healthcare, such as the place of birth, assisted delivery, the number of antenatal care visits, and household income, and child health indicators, such as exclusive breastfeeding, immunization, and maternal participation in community nutrition programs.

3.5. Data Sources

The main source of data was the Kenya Integrated Household Budget Survey (KIHBS) 2015–2016, conducted by the Kenya National Bureau of Statistics (KNBS), which provided comprehensive information on health, education, and socioeconomic indicators for households with children under five years old. These data were essential for analyzing the determinants of child health and mortality across various households.

4. Results

This section highlights the descriptive statistics results, including the results on the status of the main needs of children and households with children in Kenya. In addition, it presents the results on the access to critical services on children health mortality outcomes in Kenya, and lastly, it presents a discussion on the empirical results

4.1. Descriptive Statistics

This study utilized the KIHBS 2015/16 dataset, which included 92,846 individuals, of whom, 16.71% (15,560) were under 5 years old. Among these, 81.56% (12,690) were aged less than 1 year.

Table 1. Average age of children in months.

Age in Months	Frequency	Percent
0	1573	12.4
1	1067	8.41
2	1110	8.75
3	1061	8.36
4	1042	8.21
5	1035	8.16
6	1082	8.53
7	1020	8.04
8	960	7.57
9	954	7.52
10	905	7.13
11	881	6.94
	12,690	100

Source of data: KIHBS, 2015/16.

summarizes the age distribution of these children.

Table 2. Summary statistics of continuous variables by area of residence.

Variable	Individuals	Rural	Urban	Pooled	t-Test
Underweight children (% underweight)	10,320	14.45 (8.41)	13.52 (13.24)	14.14 (8.33)	5.31 ***
Stunted children (% stunted)	12,630	28.60 (8.73)	26.51 (7.91)	27.93 (8.53)	12.95 ***
Wasted children (% wasted)	9797	8.92 (11.75)	7.75 (9.63)	8.54 (11.12)	4.90 ***
Household size (number)	12,630	6.20 (2.38)	5.52 (2.35)	5.98 (2.38)	15.02 ***
Age of children (months)	12,690	5.08 (3.52)	5.08 (3.45)	5.08 (3.50)	−0.07

Note: asterisks (*) indicate the level of statistical significance *** (p < 0.001).

shows that approximately 14% of children were underweight, with higher rates in rural areas. The t-statistic (5.31 ***) suggests that this difference is statistically significant at the 1% level, meaning there is a very low probability that this difference occurred by chance. Stunting affected about 28% of children, with rural areas again being most affected. The t-statistic (12.95 ***) confirms that this difference is highly significant, indicating that children in rural areas are more likely to experience chronic malnutrition and stunted growth, possibly due to poorer nutrition and limited access to healthcare. About 9% of children experienced wasting, which was also higher in rural areas. This is attributed to the increased reliance on staple food like maize with a limited consumption of proteins and vegetables and a lack of dietary diversity in rural areas, as compared to urban areas. The average household size was six, being significantly larger in rural compared to urban areas. The average age of children was 5 months. The t-statistic (−0.07) shows no statistically significant difference, confirming that the sample is evenly distributed in terms of child age across rural and urban settings.

Approximately 8% of the respondents reported diarrhea in the last two weeks prior to the survey in 2015/16, which is almost in range with the KDHS 2022, which estimated the diarrhea prevalence at 14%. Child mortality varied, with significant differences between urban and rural areas. Most mothers (61%) delivered in health facilities, with higher rates in urban areas (81% vs. 52% rural). About 64% of mothers who delivered in health facilities were assisted by health professionals, with urban mothers having better access (83% vs. 55% rural). This reflects better access to healthcare facilities in the urban areas, as compared to rural areas, where patients have to travel longer distances to get to the healthcare facilities. Also, there is preference for homebirths in rural areas, as compared to urban areas. Nearly 90% of mothers exclusively breastfed for 6 months, with a higher percentage in rural areas. However, 39% supplemented breastfeeding with milk other than breast milk. Participation in community nutrition programs was similar in rural and urban areas (about 31%).

Significant disparities in access to improved water, sanitation, hygiene, and other health services were observed between urban and rural areas. The findings indicate that many households face economic challenges, with approximately 28% reporting no income and over 80% relying on out-of-pocket payments for healthcare services, emphasizing the need for affordable and accessible health services for all households (

Table 3. Summary statistics of categorical variables by area of residence.

Variable	Individuals	Rural	Urban	Pooled	Pearson Chi2 Test
Diarrhea incidence (1 = yes)	12,582	7.98	8.51	8.15	1.02
Child mortality (month age groups)
0–30 days		1.73	7.69	3.36
1–12 months	238	35.84	36.92	36.13	5.39 *
13–60 months		62.43	55.38	60.50
Child supplements (1 = milk other than breastfeeding)	12,630	41.12	35.08	39.17	42.37 ***
Birth assistance (1 = health professional)	12,561	55.12	83.21	64.21	943.38 ***
Place of delivery (1 = health facility)	12,630	52.09	80.64	61.32	949.64 ***
Child exclusively breastfeeding (1 = 0 to 6 months)	12,630	91.09	89.01	90.42	13.88 ***
Mother participation in community nutrition program (1 = yes)	12,529	30.59	30.31	30.50	0.10
Vaccination card (1 = availability of health card in the last 12 months)	12,510	90.38	92.99	91.22	23.40 ***
Access to information (1 = yes)	5114	58.62	74.13	63.51	114.56 ***
Access to water (1 = improved)	12,490	52.68	74.74	59.81	552.67 ***
Access to sanitation (1 = improved)	12,602	69.35	91.40	76.47	745.09 ***
Access to hygiene (1 = improved)	12,170	8.51	18.44	11.82	255.44 ***
Absolute poverty (1 = poor households)	12,630	42.70	36.99	40.86	37.35 ***
Food poor households (1 = yes)	12,630	13.14	6.07	10.86	142.81 ***
Number of births (count)	12,630	1.97	1.98	1.98	0.18
Age (years)	12,630	2.06	1.99	2.04	2.71 ***
Gender (1 = male)	12,630	48.70	49.67	49.01	1.05
Household assets	12,630	7.13	17.04	10.33	293.02 ***

Source: researchers’ computations. Note: asterisks (*) indicate the level of statistical significance * (p < 0.05), and *** (p < 0.001).

).

4.2. Establishing the Status of the Main Needs of Children and Households with Children

Children’s well-being is crucial for their development, and their household’s ability to meet their basic needs significantly affects their growth and productivity. This study examined the status of key needs, including parental presence, healthcare access, nutrition, and income among children in Kenya.

4.2.1. Parental Presence

Research shows that the involvement of both parents positively impacts children’s social, psychological, and health outcomes (Van Voorhis et al., 2013). Data from the Kenya Integrated Household Budget Survey (KIHBS) 2015/16 revealed that approximately 61.3% of families had both biological parents present, while one in two children lacked a father figure (

Table 4. Presence of biological parents in the life of a child.

	Frequency		Percent
	Father	Mother	Father	Mother	Average
Presence in the household	34,204	46,964	51.72	71.01	61.37
Absent in the household	24,158	15,877	36.53	24.01	30.27
Deceased	7023	3247	10.62	4.91	7.77
Do not know	748	45	1.13	0.07	0.6
Total	66,133	66,133	100	100	100

Source of data: KIHBS, 2015/16.

). The absence of either parent, especially mothers, negatively impacts child health and nutrition, leading to increased risks of infections and long-term health issues (Abdulla et al., 2022).

4.2.2. Maternal and Child Healthcare

Maternal and child health services are crucial for ensuring the well-being of mothers and infants during pregnancy, labor, and postpartum. Following the abolishment of user fees for maternity care in public facilities in 2013, Kenya saw improvements in facility births, increasing from 39.1% in 2005/06 to 65.3% in 2015/16. Out of 12,612 women surveyed, 41.67% delivered in hospitals, with the assistance from healthcare professionals varying (Figure 1). However, 38.19% still delivered at home and were hence unable to benefit from skilled birth attendants, increasing the risk of maternal and neonatal complications. Most healthcare facilities, especially in rural areas, are centrally located, and there is low motivation for healthcare workers to work in remote areas, hence reducing the number of hospital deliveries (Samuel et al., 2021). This indicates gaps in healthcare accessibility.

4.2.3. Nutrition

Proper nutrition is vital for children’s health. Despite interventions to improve child nutrition, many infants lack dietary diversity (WHO, 2021). In Kenya, while 99.4% of children were ever breastfed, only 92% were exclusively breastfed for the first six months. Many children received other supplements, including porridge and commercial infant foods (Figure 2 and Figure 3). Despite high breastfeeding rates, malnutrition remains a challenge, with stunting affecting 28% of children, being underweight affecting 14%, and wasting affecting 8.5% (

Table 5. Undernutrition-related consequences.

Continuous Variables	Mean	Std Dev.	Min	Max
Stunted children (sample = 12,630)	27.93	8.53	10.8	58.1
Underweight children (sample = 10,320)	14.14	8.33	2.9	41.2
Wasted children (sample = 9797)	8.54	11.12	1.4	50.9
Household size (number) (12,630)	5.98	2.39	2	28
Child under 1 year (months) (12,616)	5.11	3.48	1	11

Source: KIHBS and computed statistics.

).

Further, at the global level, child undernutrition in terms of stunting, wasting, and being underweight contributes to about 2.2 million deaths and 21 per cent of disability-adjusted life years for children under 5 years old (R. E. Black et al., 2008). The early detection and management of cases associated with malnutrition among children is necessary in the prevention of such related deaths. In Kenya, the proportion of stunted children stood at 28 per cent in 2016, with some households recording a high rate of 58 per cent. Underweight and wasted children were recorded at 14 per cent and 8.5 per cent on average, respectively, during the same period (Table 5). Failure to address these consequences at an early age may result in more deaths and disability-adjusted life years for children below 5 years.

4.3. Impact of Access to Essential Services on Child Mortality in Kenya

4.3.1. Means to Facilitate Access to Childcare Needs

Access to the basic infant needs in terms of child basic healthcare services, maternal and childcare services, and nutrition-related services is dependent on other factors, including the financial ability of the family to access the services, availability, and affordability. In most cases, the financial status of households determines whether some health services are affordable or not.

Nearly one-third of households reported no income, and 71% earned less than KES 100,000 annually (

Table 6. Household income.

Monthly Income (Kenyan Shillings)	Freq.	Percent
0	10,992	28.42
1–100,000	27,556	71.24
100,001–200,000	108	0.28
200,001–300,000	17	0.04
300,001–400,000	4	0.01
400,001–600,000	2	0.01
Total	38,679	100

Source of data: KIHBS, 2015/16.

). This implies that at least one in every three households might be facing challenges in affording the requisite infant health services, such as access to skilled delivery services and proper nutrition, among other services.

Despite the government championing the need for citizens to enroll into the National Hospital Insurance Fund, which has a mandate of providing accessible, affordable, sustainable, and quality health insurance for all Kenyan citizens, only 15.36% of individuals had health insurance, leaving the majority reliant on out-of-pocket payments, exacerbating the financial strain on households and limiting access to essential child healthcare services (

Table 7. Insurance coverage.

Health Insurance Coverage in the Last 12 Months	Freq.	Percent
Yes	14,252	15.36
No	78,507	84.64
Total	92,759	100

Source of data: KIHBS, 2015/16.

).

4.3.2. Rural vs. Urban Access

Access disparities exist between rural and urban areas. While urban areas generally have better access to essential services, disparities exist within marginalized urban settlements, sometimes leading to worse child health outcomes than in rural areas. (UNICEF, 2019). The data findings indicate that there is an improved infrastructure and healthcare access in urban areas. The urban households had better access to water (74.74% vs. 52.68%), sanitation (91.40% vs. 69.35%), and hygiene services compared to rural households. These factors contribute to lower infection rates, better nutritional support, and improved maternal and child health outcomes in urban settings. The KIHBS data indicate improved access to water and sanitation in urban settings (

Table 8. Categorical variable in percentages.

Variable	Total Sample	Rural	Urban	Pooled
Access to Water (% improved)	12,490	52.68	74.74	59.81
Access to Sanitation (% improved)	12,602	69.35	91.40	76.47
Access to Hygiene (% improved)	12,170	8.51	18.44	11.82
Child Mortality (0–30 days, 1–12 months, 13–59 months)	238	1.73	35.84	62.43
Household Being Food Poor (1 = yes)	12,630	13.14	6.07	10.86
Household Being Absolutely Poor (1 = yes)	12,630	42.70	36.99	40.86
Diarrheal Incidence (1 = yes)	12,582	7.98	8.51	8.15
Mother in Community Nutrition Program (1 = yes)	12,529	30.59	30.31	30.50

Source of data: KIHBS, 2015/16.

). However, child mortality rates for children under 12 months were higher in urban areas, likely due to factors such as poor breastfeeding practices and a higher diarrheal incidence. This may be linked to overcrowding, poor waste disposal, and exposure to contaminated food and water in informal urban settlements. Further, close living quarters in urban areas facilitate the faster transmission of respiratory and gastrointestinal infections, particularly in low-income neighborhoods.

4.3.3. The Effect of Access to Critical Services on Child Health Mortality Outcomes in Kenya

Child mortality rates were examined in relation to various independent variables. Of 29,148 children born, 82.45% of males and 80.49% of females were born alive. Among those born alive, mortality rates were 8.75% for males and 7.90% for females (

Table 9. Children who were born alive and children who were born alive but died.

	NO	YES	TOTAL
Male Children	Frequency	2558	12,017
	Percent	17.55	82.45
Female Children	Frequency	2843	11,730
	Percent	19.51	80.49
Male Mortality	Frequency	13,294	1275
	Percent	91.25	8.75
Female Mortality	Frequency	13,414	1151
	Percent	92.10	7.90

Source of data: KIHBS, 2015/16.

).

The World Health Organization estimates that diarrheal diseases claim the lives of nearly 525,000 children under five years old annually. In 2018, Kenya reported 15,450 deaths from diarrheal diseases among children under five years (WHO, 2021). The KIHBS data show that only 8.15% of the surveyed children experienced diarrhea in the preceding two weeks (

Table 10. The proportion of children experiencing diarrhea in the last 2 weeks.

Diarrhea in Last 2 Weeks?	Frequency	Percent
Yes	1026	8.15
No	11,556	91.85
Total	12,582	100

Source of data: KIHBS, 2015/16.

), reflecting a decline from earlier reports such as the KDHS 2022, which was at 14%.

Generally, this study showed the key factors influencing child health outcomes, including the household size, nutritional status, birth assistance, and place of delivery, highlighting critical areas for intervention in Kenya’s child health landscape.

5. Discussion

This section provides a discussion on the incidence of children between the age of 0 and 2 years falling sick from diarrhea. Further it also indicates the policy implication of the empirical results of this study.

5.1. Incidence of Child Diarrhea in Children Aged 0–2 Years

This study employed a Probit model to estimate the probability of children aged 0 to 2 falling sick from diarrhea, highlighting significant contributors to child morbidity. The findings indicate that the household poverty status increases the risk of diarrhea in this age group. Children from poor households had poor access to water and sanitation, hence increasing their exposure and susceptibility to infections, including diarrhea (Demissie et al., 2021). Additionally, access to improved water, sanitation, and hygiene negatively correlates with diarrhea incidence, underscoring the importance of enhancing basic services, like clean water and proper sanitation(

Table 11. The probability of a child aged between 0 and 2 years falling sick from diarrhea (child morbidity).

Variable	Coefficient	Robust Std Error	P > |Z|
Access to water (1 = improved)	−0.0486	0.0371	0.189
Access to sanitation (1 = improved)	−0.0113	0.0529	0.831
Access to hygiene (1 = improved)	−0.0944	0.0570	0.095
Age (years)	−0.1524	0.0133	0.000
Gender (1 = male)	−0.0578	0.0341	0.090
Household size (number)	−0.0263	0.0083	0.002
Household assets (1 = yes)	−0.0719	0.0699	0.303
Household being absolutely poor (1 = yes)	0.0050	0.0389	0.897
Birth assistance (1 = health professional)	−0.1781	0.0819	0.030
Place of delivery (1 = health facility)	−0.1922	0.0807	0.017
Child supplements (1 = milk other than breast)	0.0162	0.0381	0.671
Residence (1 = rural)	−0.0384	0.0395	0.331
Regional Bloc (NOREB = base category)
LREB	0.0585	0.0519	0.260
PWANI	0.2546	0.0587	0.000
CEKEB	0.0141	0.0628	0.823
FCDC	−0.4423	0.0761	0.000
SEKEB	0.0877	0.0800	0.273
NAKAEB	0.0805	0.0835	0.335
NAIROBI	0.1786	0.1313	0.174
Constant	−0.6535	0.1219	0.000
Observation = 11,938
Wald chi2 (21) = 279.390
Prob > chi2 = 0.0000
Pseudo R2 = 0.0438
Log pseudo likelihood = 3212.2238
AIC = 6464.448
BIC = 6612.197

Source of data: researcher’s computation.

).

Access to birth assistance positively affects child nutrition and significantly reduces the risk of diarrhea by reducing the risk of neonatal infections. The household size significantly reduces the probability of a child under five reporting diarrhea (

Table 12. The maximum likelihood estimates of Probit model on the drivers of diarrhea prevalence among children under 5 years of age (child morbidity) in Kenya.

Variable	dydx	Delta Method- Std Error	P > |Z|
Access to water (1 = improved)	−0.0070 ***	0.0053	0.189
Access to sanitation (1 = improved)	−0.0016 ***	0.0076	0.831
Access to hygiene (1 = improved)	−0.0136 ***	0.0082	0.095
Age (years)	−0.0220 ***	0.0019	0.000
Gender (1 = male)	−0.0083 ***	0.0049	0.091
Household size (number)	−0.0038 ***	0.0012	0.002
Household assets (1 = yes)	−0.0104 ***	0.0101	0.303
Household being absolutely poor (1 = yes)	0.0007 ***	0.0056	0.897
Birth assistance (1 = health professional)	−0.0257 ***	0.0118	0.030
Place of delivery (1 = health facility)	−0.0277 ***	0.0116	0.017
Child supplements (1 = milk other than breast)	0.0023 ***	0.0055	0.671
Residence (1 = rural)	−0.0055 ***	0.0057	0.331
Regional Bloc (NOREB = base category)
LREB	0.0086 ***	0.0076	0.257
PWANI	0.0426 ***	0.0101	0.000
CEKEB	0.0020 ***	0.0090	0.823
FCDC	−0.0455 ***	0.0072	0.000
SEKEB	0.0131 ***	0.0124	0.288
NAKAEB	0.0120 ***	0.0129	0.350
NAIROBI	0.0284 ***	0.0229	0.215
Constant	−0.6581	0.1161	0.000
Observation = 11,938
Wald chi2 (21) = 279.390
Prob > chi2 = 0.0000
Pseudo R2 = 0.0438
Log pseudo likelihood = 3212.2238

Notes.1 Source of data: researcher’s computation. Note: asterisks (*) indicate the level of statistical significance *** (p < 0.001).

). Previous studies indicate that a larger household size is positively correlated with a higher incidence of diarrhea. Our findings suggest the opposite, which may be attributed to the increased supervision of younger children by more household members.

The place of delivery significantly impacts diarrhea prevalence, as delivering in health facilities reduces the probability of diarrhea-related mortality in children under five by 2.7%. This finding aligns with Bitew et al. (2023), indicating hygiene is crucial in preventing childhood diarrhea. Conversely, the lack of skilled birth assistance increases the risk of diarrhea, suggesting unsupervised deliveries lead to poorer hygiene practices and greater vulnerability to infections.

Using NOREB as the base category, the spatial assessment indicated that children in various economic blocs, especially Jumuia ya Kaunti za Pwani, Nairobi, and SEKEB, experience a higher diarrhea incidence than those in NOREB. In contrast, the FCDC demonstrated reduced diarrhea incidences, possibly due to varying levels of water contamination and sanitation practices.

5.2. Cox Proportional Hazard Regression on Drivers of Child Mortality

The Cox proportional hazard regression analysis using the Breslow method revealed a strong fit with significant Wald chi² (20), of 295.26, and Prob > chi² (0.0000).

Table 13. The Cox proportional hazard regression on the drivers of child mortality (the probability of a child dying between birth and the first year).

Variables	Hazard Ratio (Robust Std Error)	Relative Hazard Ratio dy/dx (Delta-Method Std Error)
Access to water (% improved)	1.0006 (0.0185)	0.0005 (0.0159)
Access to sanitation (% improved)	0.9307 (0.0244) ***	−0.0615 (0.0215) ***
Access to hygiene (1 = improved)	1.0317 (0.0282)	0.0267 (0.0234)
Household size (number)	0.9866 (0.0041) ***	−0.0116 (0.0035) ***
Age (years)	1.0197 (0.0068) ***	0.0167 (0.0060) ***
Gender (1 = male)	0.9975 (0.0169)	−0.0021 (0.0145)
Household assets	1.0011 (0.0311)	0.0010 (0.0266)
Number of births (number)	1.0178 (0.0063)	0.0167 (0.0061)
Exclusively breastfeeding	0.6710 (0.0931) ***	−0.0518 (0.0318) ***
Child supplements (1 = milk other than breast)	1.0137 (0.0185)	0.0126 (0.0187)
Birth assistance (1 = health professional)	0.7881 (0.1072) ***	−0.0260 (0.0611)
Place of delivery (1 = health facility)	0.6774 (0.0630) ***	−0.0165 (0.0368)
Residence (1 = rural)	1.1053 (0.0865) ***	0.0603 (0.0428) ***
Age of mother (years)	1.0231 (0.0091)	0.0220 (0.0083)
Maternal education (1 = secondary school)	0.7943 (0.1048) ***	−0.0629 (0.0608) ***
Maternal participation in community program	0.6418 (0.0962) ***	−0.0512 (0.0272) ***
Stunting (1 = yes)	1.7166 (0.1902) ***	0.1984 (0.0545) ***
Wasting (1 = yes)	1.2135 (0.2118)	0.1837 (0.0890)
Underweight (1 = yes)	1.3503 (0.2471) ***	0.1568 (0.0852) ***

Number of observations: 6774; number of events: 474; Wald chi² (15) = 158.850; Prob > chi² = 0.0000. Source of data: researcher’s computation. Note: asterisks (*) indicate the level of statistical significance *** (p < 0.001).

shows the Cox proportional hazard regression results, which indicate that stunting, wasting, an underweight status, the area of residence, child supplementation, exclusive breastfeeding, and maternal participation in community nutrition programs significantly affect child survival rates.

This study found that supplementation with milk other than breast milk had an adverse effect on the child mortality hazard risk. Specifically, mothers who supplemented their children reduced the risk of child mortality by 7.3% compared to those who did not use any supplements. This suggests that alternative milk sources can provide essential nutrients that contribute to child survival, particularly in cases where exclusive breastfeeding may not be feasible due to maternal health issues, insufficient breast milk production, or other socioeconomic factors. However, this finding does not contradict the well-established benefits of exclusive breastfeeding. At the same time, this study confirmed that exclusive breastfeeding significantly reduces the child mortality risk. An increase in the exclusive breastfeeding duration correlated with a 1.4 times lower risk of child mortality compared to children who were never breastfed. This aligns with previous research (Rahman, 2008) that demonstrated how prolonged exclusive breastfeeding enhances infant immunity, prevents infections, and improves overall survival rates.

Furthermore, the area of residence significantly influenced the child mortality risk. This study found that mothers living in urban areas had a 3.5% lower likelihood of experiencing child mortality hazards compared to those in rural areas. This supports Ayele et al. (2017), who identified urban–rural disparities in child mortality due to better healthcare access, improved sanitation, and higher maternal education levels in urban settings.

This study also considered the probability of a child dying between birth and 12 months of age. Using the STCox proportional hazards model, this study found a weak significant relationship between nutritional factors (proportions of stunted, wasted, and underweight children) and infant mortality (deaths between birth and 12 months of age) (

Table 14. Probability of a child dying between birth and 12 months of age (infant mortality).

Variables	Coefficients with Std Errors in Parentheses	p Values
Access to water (% improved)	1.0000 (0.0182)	0.998
Access to sanitation (% improved)	0.9346 (0.0241) ***	0.001
Access to hygiene (1 = improved)	1.0253 (0.0275)	0.351
Household size (number)	0.9859 (0.0040) ***	0.000
Age (years)	1.0168 (0.0067) ***	0.011
Gender (1 = male)	1.0011 (0.0167)	0.949
Household assets	0.9986 (0.0305)	0.962
Number of births (number)	1.0247 (0.0082) ***	0.002
Household being absolutely poor (1 = yes)	1.0372 (0.0196) **	0.053
Residence (1 = rural)	0.9817 (0.0193)	0.347
Diarrhea incidence (1 = yes)	0.9974 (0.0297)	0.930
Place of child delivery (1 = health facility)	0.9597 (0.0376)	0.0293
Birth assistance (% health professional)	1.0253 (0.0404)	0.526
Child supplements (1 = milk other than breast)	0.9293 (0.0173) ***	0.000
Child exclusively breastfeeding (1 = 0 to 6 months)	1.0016 (0.0301)	0.957
Observation	10,550
Wald chi2 (20)	204.040
Prob > chi2	0.0000
Log pseudo likelihood	−89,059.797
Failure_-d	Age in months
Analysis time—t	Age in months
Number of subjects	10,550
Number of failures	10,550
Time at risk	61,259

Source of data: researcher’s computation. Note: asterisks (*) indicate the level of statistical significance ** (p < 0.01), and *** (p < 0.001).

). This suggests that the effects of poor nutrition may manifest more significantly later in a child’s life. Additionally, both the household size and financial status were positively correlated with infant mortality rates. Larger household sizes and poorer financial conditions were associated with an increased likelihood of children dying before reaching their first birthday. This is likely linked to challenges in accessing healthcare services due to financial constraints and competing needs arising from larger families.

5.3. Conclusions and Implications for Policy

This study identified several key factors influencing child health in Kenya, particularly focusing on the prevalence of diarrhea and child mortality among children under five. The findings revealed that larger household sizes significantly reduce the likelihood of diarrhea, while nutritional factors, such as being underweight, stunting, and wasting, are critical contributors to child mortality risks. The overall prevalence of diarrhea in children under five in Kenya was reported at 8%, slightly below the global average of 9%. Access to improved water, sanitation, hygiene, household assets, and skilled birth assistance positively affects child nutrition and health outcomes. This study highlights the importance of investment in the first 1000 days of life, particularly through maternal healthcare, improving access to water and sanitation, and improving infant feeding practices. Consequently, this study strongly advocates for targeted intervention programs aimed at reducing childhood diarrhea and improving child healthcare and nutrition.

5.3.1. Policy Recommendations

• This study recommends enhancing community understanding regarding the significance of attending antenatal healthcare visits and opting for deliveries at healthcare facilities, rather than at home. This can be achieved by training and equipping Community Health Promoters and workers to identify and support expectant mothers and ensuring early hospital linkages and postnatal follow-ups.
• The findings indicate the need for intervention programs that are focused on preventing childhood diarrhea and promoting better nutrition. The government, private sector, and civil society should collaborate to expand the WASH infrastructure in both rural and urban areas.
• Further, the findings suggest the need for a legislative review. This study calls for action to the government to conduct timely reviews of legislation and policies to enhance child nutrition and healthcare, particularly regarding interventions related to breastfeeding, nutritional supplements, and education for new mothers.
• Targeted health interventions should also be employed. There is a need to guide national and county-level health interventions to address higher child mortality hazard risks, ensuring that resources and policies prioritize areas with the greatest need.

5.3.2. Limitations of the Study and Future Prospect for Research

Despite providing valuable insights into the significance of investing in the first 1000 days of life for improved child health outcomes, this study has certain limitations that should be considered when interpreting its findings.

• Data constraints were the main challenge. This study relied on secondary data from the Kenya Integrated Household Budget Survey (KIHBS) 2015/16 and Kenya Demographic and Health Survey (KDHS). The KIHBS 2015/16 data were the most recent data at the time of the study. The use of cross-sectional data limits the ability to establish causal relationships over time. A longitudinal study would provide a more comprehensive understanding of the long-term effects of early childhood investments.
• Geographical and socioeconomic variability also constraints the findings of the study: the study recognizes regional disparities in child health outcomes but does not comprehensively analyze the role of specific cultural, economic, and environmental factors across different counties. A more detailed county-level analysis could help tailor interventions to specific local contexts.
• The exclusion of certain health determinants was another factor: while the study focuses on key indicators such as maternal healthcare, exclusive breastfeeding, and immunization, it does not extensively cover other determinants like parental education, household income dynamics, and access to healthcare infrastructure, which could significantly impact child health outcomes.

5.3.3. Future Prospects

To address these limitations and enhance research in this area, future studies should

• Incorporate longitudinal data: conducting longitudinal studies would provide deeper insights into how investments in the first 1000 days impact long-term health and human capital development outcomes.
• Expand the scope of analysis: future research should explore the influence of socioeconomic factors, gender disparities, and community health programs on child health and nutrition outcomes.
• Leverage advanced analytical approaches: using geospatial and econometric modeling can help identify high-risk regions and tailor targeted interventions.

6. Conclusions

In conclusion, these findings emphasize the critical role of improving water and sanitation access, enhancing maternal education, and addressing malnutrition in reducing child morbidity and mortality in Kenya. Additionally, this study provides the measures necessary to support the government effort of reducing child mortality rates and enhancing the overall health of children in Kenya.