Reducing childhood mortality is an ambitious public health project of India. To achieve the expected results in the public health policies requires a deep understanding on the issues and strong implementation strategies (Joseph L Mathew, et al. 2011a). According to the India Human Development Survey (IHDS), analysis of the household data provides a broad overview of the implementation and success of the public policies and the future policy planning and the challenges involved in it. Assets and amenities provided for a household reflects the quality of life. While the availability and number of services/amenities and assets vary from every region. Sandip Mondal (2020) focused the spatial pattern of Household Quality of Living (HOL) in West Bengal corresponding

Arjun Kumar (2015) stated access to basic amenities, like health and family welfare centres, is imperative for a good quality of life. These basic amenities are universally standard and playing a major role in the formulation of policies and programmes in India. In India, rural households have witnessed a gradual improvement in access to health centres; however, they experience high level of service. Access to such amenities in the rural India has been insufficient, which is reflected in the poor health. Providing an equal health service across the country is a challenging task, however, to provide such a service would be an ideal solution to reduce the disparity. Increase in the accessibility and availability of quality health services is shown to be an effective method to serve the needy and to improve the health system (Emily Oster, 2009).

Fabio Boncinelli et al. (2015) identified and evaluated the most important territorial services, infrastructure, and geographic factors in order to measure several dimensions of quality of life in rural areas. The study evaluates how the availability of services supplied in a rural area is able to influence local residents’ quality of life.

More than half of deaths in children (age less than five years) occur in the neonatal period; infections (including sepsis, pneumonia, diarrhoea, and tetanus), prematurity, and birth asphyxia are the three major causes of death in this period. (Vinod Kumar Paul, et al. 2011; Chandrakant Lahariya, et al. 2010; Marion Krämer, et al. 2021; Dheeraj Shah, et al. 2012). Climate change, among the other factors, affects child health through direct and indirect pathways, with implications for determinants of child health as well as morbidity and mortality from a range of diseases related to vector borne, water and food borne, respiratory, pathogens etc. (Daniel Helldén, et al. 2021). Respiratory related diseases are one among the chief causes of child mortality in India (Joseph L Mathew, et al. 2011b). Despite the plans and policies implemented by the government of India, pre-natal and post-natal health care was not delivered to the fullest across all the parts of the country, especially those are poor and weaker sections of the society.

The geographical accessibily is another impediment for the mother and the child to get the health care service on time. stratification and clustering are an effective method to group the variables and infer valid inputs to draw effective strategies to strengthen the health care systems (Yuvaraj Krishnamoorthy, et al. 2020; Siddhartha Gogia. et al. 2011).

Despite the vast establishments of the health care units, home deliveries are high in India. The important limiting factors for home-based deliveries were myths and cultural beliefs, lack of awareness of mother and families regarding danger and postnatal complications, and hesitancy to contact health workers. On the other side the health workers, at times, lack clarity about job responsibilities, poor quality of training, skills building and supervision of front-line workers (Vaishali Deshmukh, et al. 2020).

Among the many intricate methods to understand health care systems and the associated pros and cons is the cross-sectional profile on how the health care systems like maternal, early child care Centres, family planning etc. were utilized by the locals around it (Shankar Prinja, et al. 2019; Shubhranshu Kumar Upadhyay, et al. 2020). Associating with the child mortality, breastfeeding is an important factor, especially in the India context, is attributed to the gender biased socio-cultural beliefs; in certain cases, which keeps the mothers to wean on feeding girl child and promotes to get a boy child (Seema Jayachandran, et al. 2011). This also affects the reproductivity cycles and the mother’s health (Abhishek Singh, et al. 2012).

Health Management Information System (HMIS) and Rural Health Survey (RHS) provide wide range of reports on health-related information up to village level across India. The data are valuable in evaluating the health condition and health care developments in providing the service with expected quality.

For the present study, child mortality, up to the age of five, has been aimed to study associating with number of other factors. The study has been carried out with the integration of spatial and statistical methods. Conventionally, the statistical methods have lacked the ability to visualize the results, while with the integration of these methods into the geospatial applications provides higher level of understanding about the problem.

The data available in the reports are to be generalized, inverted, and calculated into ratio before it can be used in the analysis. From the reports, 32 variables were selected for Hot-spot analysis and 20 variables were selected for regression analysis (Table 1 and Table 2). For regression, child death has been selected as the dependent variable and rest are treated as independent variables. Independent variables are further grouped into five categories namely, new born health condition, number of health care Centres (SC, PHC, CHC), grade of the health Centres, doctors, and Staffs available at the health Centres, and population characteristics.

Clustering and linear regression methods were implemented in the study. To cluster the values, Getis-Ord Gi method, and Ordinary Least Square method for finding the regression were adopted; both the methods were proved to be impactful.

Getis-Ord Gi was given as,

\begin{aligned} &\Gamma^n W_{i, j} X_{i, j}-X \sum^n W_{i, i}\\ &\boldsymbol{G i} \boldsymbol{i}^*=\frac{j=1}{s \overline{\left[n \sum^n \quad W^2 \equiv \Sigma^n \quad v v i_j^2\right.}} \end{aligned}

Where, xj the attribute is value for feature j;

wi,j is the spatial weight between feature I and j; n is equal to the total number of features

\begin{aligned} &\sum^n x_i\\ &\boldsymbol{X} \underset{n}{i} \quad \mathbf{S}=\sqrt{\frac{i}{n}}-(X)^2 \end{aligned}

The Gi* statistic is a z-score, so no further calculations are required.

Regression analysis is probably the most commonly used statistic in the social sciences. Regression is used to evaluate relationships between two or more feature attributes. Identifying and measuring relationships lets you better understand what's going on in a place, predict where something is likely to occur, or begin to examine causes of why things occur where they do.

Both the Multiple R-Squared and Adjusted R-Squared values are measures of model performance. Possible values range from 0.0 to 1.0. The Adjusted R-Squared value is always a bit lower than the Multiple R-Squared value, because it reflects model complexity (the number of variables) as it relates to the data and is consequently a more accurate measure of model performance. Adding an additional explanatory variable to the model will likely increase the Multiple R-Squared value but may decrease the Adjusted R-Squared value.

The coefficient for each explanatory variable reflects both the strength and type of relationship the explanatory variable has to the dependent variable. When the sign associated with the coefficient is negative, the relationship is negative

Ordinary Least Square method was given as,

=\mathrm{Q}_0+\mathrm{Q}_1 \times{ }_1+\mathrm{Q}_2 \times 2+\cdots \mathrm{Q}_n \times_n+\mathrm{s}

In the formula,

(y): dependent variable;

(x): independent/Explanatory variables;

(β): regression coefficients; \varepsilon: random error term.

Factors of infant deaths below five years of age are categorized into 5 groups for regression analysis (Table 2).

Hot-spot and regression analysis are powerful methods to derive realistic results; implementing them in the spatial analysis reveals intricate information about the pattern of things which are analyzed. The greatest advantage of these methods incorporated in spatial analysis is the capability of the software to display the numerical information as maps. The visual perception increases the ability to understand the problems effectively.

The Universal Health Care (UHC) resolution in the United Nations highlighted the need for national strategic direction on Quality of Care (QoC) and made a strong call to expand the provision and prioritization of primary health Care (PHC) for more sustainable, people-centred and integrated health systems. The resolution reaffirmed the commitment to the 2030 Agenda for Sustainable Development and the achievement of UHC by 2030.

With reference to the Sustainable Development Goals 2015, India targets to reduce the neonatal mortality to as low as 12/1000 live births and under five years mortality to 25/1000 live births. To attain the target, major improvisations and intensification of the efforts are needed at the root level.

As part of evaluation of the efforts in the implementation of the plans, the sub-district level flow of data at yearly interval through the Health Information Management System (HMIS) is a valuable data resource to provide insightful information to the policy makers and health workers.

The tools are very effective in manipulating the data and the results are realistic, which can be verified with further round of validation with a select sampling method. The detailed report from the study can be very insightful to the policy makers and implementing agencies for effective outcome.

The study has vast scope for further investigation while the major limitation is data. Currently the data is available at state level, is insufficient to carry out the research deeper like at district level.