The primary goal and responsibility of a data scientist is to solve problems. Identifying datasets for testing and training models is likely the initial step before commencing model construction. Furthermore, gaining access to a dataset can provide opportunities for creativity and inspire the development of a machine-learning model by utilizing the data.  

The Indian government has recently made many health and family welfare statistics accessible to the public. The government likely anticipates that developers and scientists will be tasked with resolving these issues. Addressing any issue for India, based on Indian data, is a challenging undertaking that also carries significant accountability.

The following is the collection of freely accessible datasets related to Indian health and family welfare. These datasets can be used for data analysis and developing machine learning models.  

Global Youth Tobacco Survey (GYTS-4)

The International Institute for Population Sciences (IIPS) under the Ministry of Health and Family Welfare performed GYTS-4 in 2019. The survey estimated nationwide tobacco usage among 13-15-year-old schoolchildren in states and UT by sex, school location (rural-urban), and school administration (public-private). 

The first three GYTS rounds were in 2003, 2006, and 2009. The survey included 97,302 students from 987 schools (544 public, 443 private). The report included 80,772 13-15-year-olds. The poll covered tobacco use, cessation, second-hand smoke, access and availability, anti-tobacco information, tobacco marketing awareness and receptivity, knowledge, and attitudes.

National Family Health Survey (NFHS) - 5

The fifth in the NFHS series, the National Family Health Survey 2019-2021, provides population, health, and nutrition data for all states and union territories of India. NFHS-5, similar to NFHS-4, offers estimates at the district level for many significant indicators. Similarities exist between NFHS-5 and NFHS-4, so comparisons over time are possible. NFHS-5 incorporates several novel subjects, including primary education, disability, toilet facility accessibility, death registration, menstrual bathing practices, and abortion methods and justifications. 

Additionally, waist and hip circumferences are now included in the scope of clinical, anthropometric, and biochemical testing, and the age range for blood pressure and glucose measurements has been broadened. Nevertheless, HIV testing has ceased. The NFHS-5 sample has been specifically crafted to furnish estimates of various indicators addressed in the survey at the national, state, union territory, and district levels. However, data on domestic violence, knowledge, attitudes, and behaviour regarding HIV and AIDS, indicators of sexual behaviour, the background of the spouse and the woman's occupation, and domestic violence are only accessible at the state, union territory, and national levels.

Monthly Maximum and Minimum Performing Public Health Facilities Across States

Obtain statistics on the monthly highest and lowest performing public health facilities on specified data items across states. Deliveries made at the facility, C-section deliveries made at the facility, IPD attendance, OPD attendance (all), operation major (general and spinal anaesthesia), and operation are small (no or local anaesthesia). The data includes the overall number of facilities, the number of facilities reporting no performance, the details of the best and worst performing facilities (name, sub-district, value reported), and so on.

HMIS district level Cumulative comparison of various indicators of M/O Railways

Obtain data on the Health Management Information System (HMIS) for the State M/O Railways, item-by-item and month-by-month cumulative comparison of various indicators. Data on various parameters at the district level have been provided here. Primary indicators consist of antenatal care services. Pregnancy outcomes and newborn weight, number of Caesarean C-section deliveries performed, complications during pregnancy, postnatal care, medical termination of pregnancy (MTP), cases of recurrent and sexually transmitted infections (RTIs/STIs), family planning, child immunization, quantity of vitamin A doses administered, number of childhood disease cases reported, blindness control programme, patient services, laboratory testing, stock position, and details of reported deaths with probable causes. 

The National Rural Health Mission (NRHM) was initiated by the Ministry of Health and Family Welfare to implement essential structural improvements in the system of fundamental healthcare provision. Constant access to high-quality data regarding inputs, outputs, and outcome indicators enables effective oversight of NHM's objectives. The Ministry of Health & Family Welfare, Government of India, is implementing various mechanisms to enhance the monitoring and evaluation systems to oversee the quality and performance of health services provided under the NRHM. These mechanisms include performance statistics, surveys, community monitoring, and quality assurance. HMIS is defined as an instrument that facilitates the collection, aggregation, analysis, and subsequent application of the resulting data to enhance the performance of health systems.

Image source: Unsplash