This dataset is a more different and reliable version to KumarRajarshi's Life Expectancy (WHO) dataset - where some of his values and methods can be questioned.

Context All of the data in this dataset is compiled and downloaded from the Global Health Observatory (GHO) – which is a public health data repository established by the World Health Organisation (WHO). This makes the dataset very reliable and valid.

Challenges - Perform EDA to explore factors that affect life expectancy? - Produce a model to predict life expectancy?

Dataset Contents Life Expectancy from birth: - https://www.who.int/data/gho/data/indicators/indicator-details/GHO/life-expectancy-at-birth-(years)

Mean BMI (kg/m²) (crude estimate): - https://www.who.int/data/gho/data/indicators/indicator-details/GHO/mean-bmi-(kg-m-)-(crude-estimate)

Alcohol, total per capita (15+) consumption (in litres of pure alcohol): - https://www.who.int/data/gho/data/indicators/indicator-details/GHO/total-(recorded-unrecorded)-alcohol-per-capita-(15-)-consumption

The rest of the factors: - https://www.who.int/data/gho/data/themes/mortality-and-global-health-estimates/ghe-leading-causes-of-death (BY COUNTRY, Summary tables of mortality estimates by cause, age and sex, by country, 2000–2019, Number of Deaths [2000, 2010, 2015, 2019]). All of the values are crude estimates number of deaths per 1000.

I did this so you don't have to!

Data Collected: March 2023

Life expectancy worldwide has seen significant improvements over the past three decades, with notable variations across regions. In 2021, a child born in the Americas could expect to live an average of **** years, compared to ** years in 1990. However, the COVID-19 pandemic caused a universal decline in life expectancy from 2019 to 2021, affecting all World Health Organization regions. Regional disparities and global trends While global life expectancy has generally increased over time, stark regional differences persist. ****** consistently reports the lowest life expectancy, with **** years in 2021. In fact, the twenty countries with the lowest life expectancy in the world are all found in ******, with **** and ******* reporting the lowest life expectancies at just ** years. In contrast, the *************** now has the highest life expectancy, reaching **** years in 2021. These disparities reflect broader socioeconomic factors, with low-income countries facing challenges such as limited healthcare access and higher rates of infectious diseases. Impact of health issues on life expectancy Various health issues contribute to differences in life expectancy across countries and regions. Mental health has emerged as a significant concern, with a survey of 31 countries identifying it as the biggest health problem facing people in these countries in 2024. The COVID-19 pandemic not only directly impacted life expectancy but also exacerbated mental health issues worldwide. Additionally, non-communicable diseases play a crucial role in determining life expectancy. In 2021, ********************** was the leading cause of death globally, highlighting the importance of addressing chronic health conditions to improve overall life expectancy.

The World Health Organization provides data on Life Expectancy, Healthy Life Expectancy, and Healthy Life Years Lost.

Life expectancy and Healthy life expectancy data by country and wealth group from the World Health Organization (WHO). Historical data starts from 2000 and latest data points are from 2019.

Life Expectancy of the World Population

The dataset from Worldometer provides a ranked list of countries based on life expectancy at birth, which represents the average number of years a newborn is expected to live under current mortality rates. It includes global, regional, and country-specific life expectancy figures, with separate data for males and females. The dataset highlights disparities in longevity across nations, with countries like Hong Kong, Japan, and South Korea having the highest life expectancies. This data serves as a key indicator of public health, quality of life, and healthcare effectiveness, offering valuable insights for policymakers, researchers, and global health organizations.

Data Analysis & Machine Learning Approaches for Life Expectancy Data

Data Analysis Approaches Life expectancy data can be analyzed using descriptive statistics (mean, variance, distribution) and correlation analysis to identify relationships with factors like GDP, healthcare, and education. Time series analysis helps track longevity trends over time, while clustering techniques (e.g., K-Means) group countries with similar patterns. Additionally, geospatial analysis can visualize regional disparities in life expectancy.

Machine Learning Models For prediction, linear and multiple regression models estimate life expectancy based on socioeconomic indicators, while polynomial regression captures non-linear trends. Decision trees and Random Forests classify countries into high- and low-life expectancy groups. Deep learning techniques like neural networks (ANNs) can model complex relationships, while LSTMs are useful for time-series forecasting.

For pattern detection, K-Means clustering groups countries based on life expectancy trends, and DBSCAN identifies anomalies. Principal Component Analysis (PCA) helps in feature selection, improving model efficiency. These methods provide insights into longevity trends, helping policymakers and researchers improve public health strategies.

Life expectancy at birth. Data based on the latest United Nations Population Division estimates.

Source: https://www.worldometers.info/demographics/life-expectancy/#countries-ranked-by-life-expectancy

Healthy life expectancy (HALE) at birth (years)

The data is collected as part of a project to predict the life expectancy. The data is loaded in the notebook that can be found in https://github.com/MarouaJaoua/DataStewardship1 . The data is from different sources which are Kaggle, World Bank data and World Health Organisation.

Healthy life expectancy (HALE) at age 60 (years)

Additional file 1: Table A1. Age- and cause-specific contributions to gap in life expectancy between Iran and the neighbour countries in males. Table A2. Age- and cause-specific contributions to gap in life expectancy between Iran and the neighbour countries in females. Table A3. Age- and cause-specific contributions to gap in lifespan inequality between Iran and the neighbour countries in males. Table A4. Age- and cause-specific contributions to gap in lifespan inequality between Iran and the neighbour countries in females. Table A5. Age- and cause-specific contributions to gap in life expectancy and lifespan inequality between Iran and the neighbour countries in males based on the life table from the institute for health metrics and evaluation. Table A6. Age- and cause-specific contributions to gap in life expectancy and lifespan inequality between Iran and the neighbour countries in females based on the life table from the institute for health metrics and evaluation. Table A7. Age- and cause-specific contributions to gap in life expectancy and lifespan inequality between Iran and the neighbour countries in males based on the life table from the United Nations. Table A8. Age- and cause-specific contributions to gap in life expectancy and lifespan inequality between Iran and the neighbour countries in females based on the life table from the United Nations. Table A9. Age- and cause-specific contributions to gap in life expectancy and lifespan inequality between Iran and the neighbour countries in males based on the life table from the world health organization. Table A10. Age- and cause-specific contributions to gap in life expectancy and lifespan inequality between Iran and the neighbour countries in females based on the life table from the world health organization.

Description:

This dataset explores the factors influencing life expectancy across various countries and years, aiming to uncover patterns and disparities in health outcomes based on geographic locations. By examining key features such as adult mortality, alcohol consumption, healthcare expenditures, and socioeconomic indicators, this dataset provides insights into the complex interplay of factors shaping life expectancy worldwide.

Feature Description:

Source:

World Health Organization (WHO), United Nations (UN), World Bank, etc.

Context

I am developing my data science skills in areas outside of my previous work. An interesting problem for me was to identify which factors influence life expectancy on a national level. There is an existing Kaggle data set that explored this, but that information was corrupted. Part of the problem solving process is to step back periodically and ask "does this make sense?" Without reasonable data, it is harder to notice mistakes in my analysis code (as opposed to unusual behavior due to the data itself). I wanted to make a similar data set, but with reliable information.

This is my first time exploring life expectancy, so I had to guess which features might be of interest when making the data set. Some were included for comparison with the other Kaggle data set. A number of potentially interesting features (like air pollution) were left off due to limited year or country coverage. Since the data was collected from more than one server, some features are present more than once, to explore the differences.

Content

A goal of the World Health Organization (WHO) is to ensure that a billion more people are protected from health emergencies, and provided better health and well-being. They provide public data collected from many sources to identify and monitor factors that are important to reach this goal. This set was primarily made using GHO (Global Health Observatory) and UNESCO (United Nations Educational Scientific and Culture Organization) information. The set covers the years 2000-2016 for 183 countries, in a single CSV file. Missing data is left in place, for the user to decide how to deal with it.

Three notebooks are provided for my cursory analysis, a comparison with the other Kaggle set, and a template for creating this data set.

Inspiration

There is a lot to explore, if the user is interested. The GHO server alone has over 2000 "indicators". - How are the GHO and UNESCO life expectancies calculated, and what is causing the difference? That could also be asked for Gross National Income (GNI) and mortality features. - How does the life expectancy after age 60 compare to the life expectancy at birth? Is the relationship with the features in this data set different for those two targets? - What other indicators on the servers might be interesting to use? Some of the GHO indicators are different studies with different coverage. Can they be combined to make a more useful and robust data feature? - Unraveling the correlations between the features would take significant work.

In the World Health Organization (WHO)-coordinated Cardiovascular Disease and Alimentary Comparison Study, isoflavones (I; biomarker for dietary soy) and taurine (T; biomarker for dietary fish) in 24-hour—urine (24U) were inversely related to coronary heart disease (CHD) mortality. High levels of these biomarkers are found in Japanese people, whose CHD mortality is lowest among developed countries. We analyzed the association of these biomarkers with cardiovascular disease risk in the Japanese to know their health effects within one ethnic population. First, to compare the Japanese intake of I and T with international intakes, the ratios of 24UI and 24UT to creatinine from the WHO Study were divided into quintiles for analysis. The ratio for the Japanese was the highest in the highest quintiles for both I and T, reaching 88.1%, far higher than the average ratio for the Japanese (26.3%) in the total study population. Second, 553 inhabitants of Hyogo Prefecture, Japan, aged 30 to 79 years underwent 24-U collection and blood analyses. The 24UT and 24UI were divided into tertiles and adjusted for age and sex. The highest T tertile, compared with the lowest tertile, showed significantly higher levels of high-density lipoprotein-cholesterol (HDL-C), total cholesterol, 24U sodium (Na) and potassium (K). The highest I tertile showed significantly higher folate, 24UNa and 24UK compared with the lowest tertile. The highest tertile of both T and I showed significantly higher HDL-C, folate, and 24UNa and 24UK compared with the lowest tertile. Thus, greater consumption of fish and soy were significantly associated with higher HDL-C and folate levels, possibly a contributor to Japan having the lowest CHD mortality and longest life expectancy among developed countries. As these intakes were also associated with a high intake of salt, a low-salt intake of fish and soy should be recommended for healthy life expectancy.

Japan stands out as one of the countries with the highest population longevity, from a global perspective 🌏, having the highest estimated life expectancy at birth of 84.26 years. The longevity of Japanese women is notable, ranking first worldwide with a life expectancy of 86.94 years, while Japanese men rank second with 81.49 years (World Health Organization, 2020). Japan's high life expectancy can be attributed to various factors. Technological progress, especially in the medical field, along with the country's accelerated economic development, in recent decades, have inevitably led to an increase in the average life expectancy of the population.

The dataset contains information about life expectancy and economic&social variables for Japan's prefectures as of 2020. - Life expectancy data source: Ministry of Health, Labour and Welfare, Japan - Independent variables data source: Japanese Government Statistics - Geospatial prefecture data: GitHub

The life expectancy experiences significant growth in all gender groups in 2023. As part of the positive trend, the life expectancy reaches the maximum value for the different genders at the end of the comparison period. Particularly noteworthy is the life expectancy of women at birth, which has the highest value of 69.08 years. Life expectancy at birth refers to the number of years the average newborn is expected to live, providing that mortality patterns at the time of birth do not change thereafter.Find further similar statistics for other countries or regions like Suriname and Saint Vincent and the Grenadines.

Life expectancy at age 65 for Quebec, Canada, 2012 to 2050 (years for the baseline; additional years compared to baseline for other scenarios).

Context Although there have been lot of studies undertaken in the past on factors affecting life expectancy considering demographic variables, income composition and mortality rates. It was found that affect of immunization and human development index was not taken into account in the past. Also, some of the past research was done considering multiple linear regression based on data set of one year for all the countries. Hence, this gives motivation to resolve both the factors stated previously by formulating a regression model based on mixed effects model and multiple linear regression while considering data from a period of 2000 to 2015 for all the countries. Important immunization like Hepatitis B, Polio and Diphtheria will also be considered. In a nutshell, this study will focus on immunization factors, mortality factors, economic factors, social factors and other health related factors as well. Since the observations this dataset are based on different countries, it will be easier for a country to determine the predicting factor which is contributing to lower value of life expectancy. This will help in suggesting a country which area should be given importance in order to efficiently improve the life expectancy of its population.

Content The project relies on accuracy of data. The Global Health Observatory (GHO) data repository under World Health Organization (WHO) keeps track of the health status as well as many other related factors for all countries The data-sets are made available to public for the purpose of health data analysis. The data-set related to life expectancy, health factors for 193 countries has been collected from the same WHO data repository website and its corresponding economic data was collected from United Nation website. Among all categories of health-related factors only those critical factors were chosen which are more representative. It has been observed that in the past 15 years , there has been a huge development in health sector resulting in improvement of human mortality rates especially in the developing nations in comparison to the past 30 years. Therefore, in this project we have considered data from year 2000-2015 for 193 countries for further analysis. The individual data files have been merged together into a single data-set. On initial visual inspection of the data showed some missing values. As the data-sets were from WHO, we found no evident errors. Missing data was handled in R software by using Missmap command. The result indicated that most of the missing data was for population, Hepatitis B and GDP. The missing data were from less known countries like Vanuatu, Tonga, Togo, Cabo Verde etc. Finding all data for these countries was difficult and hence, it was decided that we exclude these countries from the final model data-set. The final merged file(final dataset) consists of 22 Columns and 2938 rows which meant 20 predicting variables. All predicting variables was then divided into several broad categories:​Immunization related factors, Mortality factors, Economical factors and Social factors.

Acknowledgements The data was collected from WHO and United Nations website with the help of Deeksha Russell and Duan Wang.

Association of tertiles of biomarkers of fish and soy intakes (24UT, 24UI) with cardiovascular risks, fasting blood, and 24U in Japanese, Hyogo inhabitants aged 30 to 79 Years.

Abstract

The Afghanistan Mortality Survey (AMS) 2010 was designed to measure mortality levels and causes of death, with a special focus on maternal mortality. In addition, the data obtained in the survey can be used to derive mortality trends by age and sex as well as sub-national estimates. The study also provides current data on fertility and family planning behavior and on the utilization of maternal and child health services.

OBJECTIVES

The specific objectives of the survey include the following: - National estimates of maternal mortality; causes and determinants of mortality for adults, children, and infants by age, sex, and wealth status; and other key socioeconomic background variables; - Estimates of indicators for the country as a whole, for the urban and the rural areas separately, and for each of the three survey domains of North, Central, and South, which were created by regrouping the eight geographic regions; - Information on determinants of maternal health; - Other demographic indicators, including life expectancy, crude birth and death rates, and fertility rates.

ORGANIZATION OF THE SURVEY

The AMS 2010 was carried out by the Afghan Public Health Institute (APHI) of the Ministry of Public Health (MoPH) and the Central Statistics Organization (CSO) Afghanistan. Technical assistance for the survey was provided by ICF Macro, the Indian Institute of Health Management Research (IIHMR) and the World Health Organization Regional Office for the Eastern Mediterranean (WHO/EMRO). The AMS 2010 is part of the worldwide MEASURE DHS project that assists countries in the collection of data to monitor and evaluate population, health, and nutrition programs. Financial support for the survey was received from USAID, and the United Nations Children’s Fund (UNICEF). WHO/EMRO’s contribution to the survey was supported with funds from USAID and the UK Department for International Development and the Health Metrics Network (DFID/HMN). Ethical approval for the survey was obtained from the institutional review boards at the MoPH, ICF Macro, IIHMR, and the WHO.

A steering committee was formed to coordinate, oversee, advise, and make decisions on all major aspects of the survey. The steering committee comprised representatives from various ministries and key stakeholders, including MoPH, CSO, USAID, ICF Macro, IIHMR, UNICEF, UNFPA, WHO, and local and international NGOs. A technical advisory group (TAG) made up of experts in the field of mortality and health was also formed to provide technical guidance throughout the survey, including reviewing the questionnaires, the tabulation plan for this final report, the final report, and the results of the survey.

Geographic coverage

National

Kind of data

Sample survey data [ssd]

Sampling procedure

The AMS 2010 is the first nationwide survey of its kind. A nationally representative sample of 24,032 households was selected. All women age 12-49 who were usual residents of the selected households or who slept in the households the night before the survey were eligible for the survey. The survey was designed to produce representative estimates of indicators for the country as a whole, for the urban and the rural areas separately, and for each of the three survey domains, which are regroupings of the eight geographical regions. The compositions of the domains are given below: - The North, which combines the Northern region and the North Eastern region, consists of nine provinces: Badakhshan, Baghlan, Balkh, Faryab, Jawzjan, Kunduz, Samangan, Sari Pul, and Takhar. - The Central, which combines the Western region, the Central Highland region, and the Capital region, consists of 12 provinces: Badghis, Bamyan, Daykundi, Farah, Ghor, Hirat, Kabul, Kapisa, Logar, Panjsher, Parwan, and Maydan Wardak. - The South, which combines the Southern region, the South Eastern region, and the Eastern region, consists of 13 provinces: Ghazni, Hilmand, Kandahar, Khost, Kunar, Laghman, Nangarhar, Nimroz, Nuristan, Paktika, Paktya, Uruzgan, and Zabul.

The sample for the AMS 2010 is a stratified sample selected in two stages from the 2011 Population and Housing Census (PHC) preparatory frame obtained from the Central Statistics Organization (CSO). Stratification was achieved by separating each domain into urban and rural areas. Because of the low urban proportion for most of the provinces, the combined urban areas of each domain form a single sampling stratum, which is the urban stratum of the domain. On the other hand, the rural areas of each domain are further split into strata according to province; that is, the rural areas of each province form a sampling stratum. In total, 34 sampling strata have been created after excluding the rural areas of Hilmand, Kandahar, and Zabul from the domain of the south. Among the 34 sampling strata, 3 are urban strata, and the remaining 31 are rural strata, which correspond with the total number of provinces and their rural areas. Samples were selected independently in each sampling stratum by a twostage selection process. Implicit stratification and proportional allocation were achieved at each of the lower administrative levels within a sampling stratum, by sorting the sampling frame according to administrative units at different levels within each stratum, and by using a probability proportional to size selection at the first stage of sampling.

The primary sampling unit was the enumeration area (EA). After selection of the EA and before the main fieldwork, a household listing operation was carried out in the selected EAs to provide the most updated sampling frame for the selection of households in the second stage. The household listing operation consisted of (1) visiting each of the 751 selected EAs, (2) drawing a location map and a detailed sketch, and (3) recording on the household listing forms all structures found in the EA and all households residing in the structure with the address and the name of the household head. The resulting lists of households serve as the sampling frame for the selection of households at the second stage of sampling. In the second stage of sampling, a fixed number of 32 households was selected randomly in each selected cluster by an equal probability systematic sampling technique. The household selection procedure was carried out at the IIHMR office in Kabul prior to the start of fieldwork. An Excel spreadsheet prepared by ICF Macro to facilitate the household selection was used. A level of non response, or refusals on the part of households and individuals, had already been taken into consideration in the sample design and sample calculation.

The survey interviewers interviewed only pre-selected households, and no replacements of pre-selected households were made during the fieldwork, thus maintaining the representativeness of the final results from the survey for the country. Interviewers were also trained to optimize their effort to identify selected households and to ensure that individuals cooperated to minimize non-response. It is important to note here that interviewers in the AMS were not remunerated according to the number of questionnaires completed but given a daily per diem for the number of days they spent in the field; in addition, it is also important to note that respondents were neither compensated in any way for agreeing to be interviewed nor coerced into completing an interview.

For security reasons, the rural areas of Kandahar, Hilmand, and Zabul, which constitute less than 9 percent of the population, were excluded during sample design from the sample selection; however, the urban areas of these provinces were included. Of the 751 EAs that were included in the sample, 34 EAs (5 urban and 29 rural) were not surveyed. Six of the selected EAs in Ghazni, 16 in Paktika, 1 in Uruzgan, 3 in Kandahar, 3 in Daykundi, and 2 in Faryab were not surveyed because of the security situation. In addition, two EAs from Badakshan and one from Takhar were not surveyed because base maps from the CSO were unavailable. The non-surveyed EAs-which were primarily in rural areas-represent 4 percent of the total population of the country,

Table 1.1 - Sample coverage (Percentage of the population represented by the sample surveyed in the Afghanistan Mortality Survey, Afghanistan 2010) Region / Urban / Rural / Total North / 97 / 98 / 98 Central / 100 / 98 / 99 South / 94 / 63 / 66 Total / 98 / 84 / 87

Overall, approximately 13 percent of the country was not surveyed; most of these areas were in the South zone. As shown in Table 1.1, the survey covered only 66 percent of the population in the South zone. Sample weights were adjusted accordingly to take into account those EAs that were selected but not completed for security or other reasons.

Overall, the AMS 2010 covered 87 percent of the population of the country, 98 percent of the urban population and 84 percent of the rural population. Nevertheless, the lack of total coverage and the disproportionate exclusion of areas in the South, and particularly the rural South, should be taken into consideration when interpreting national level estimates of key demographic indicators and estimates for the South zone and regions within. For this reason key indicators will be presented for all Afghanistan and Afghanistan excluding the South zone. Despite these exclusions, the AMS is the most comprehensive mortality survey conducted in Afghanistan in the last few decades in terms of geographic coverage of the country.

Throughout this report, numbers in the tables reflect weighted numbers unless indicated otherwise. In most cases, percentages based on 25-49 cases are shown in parentheses and percentages based on fewer than 25 unweighted cases are suppressed and replaced with an asterisk, to caution readers when interpreting data that a percentage may not

The English Longitudinal Study of Ageing (ELSA) Covid-19 study can be seen as a follow-up study based on the sample of the regular ELSA study (held under SN 5050). ELSA was launched in 2002 with the primary objective of exploring ageing in England through the operationalisation of a longitudinal design, where repeated measures are taken over time from the same sample of study participants, composed of people aged 50 or above.

After the beginning of the Coronavirus Disease 2019 (COVID-19) outbreak at the end of 2019, its classification as global pandemic by the World Health Organisation in March 2020 and the gradual escalation of protective measures in the UK, culminating with the enforcement of a nation-wide lockdown in late March, the ELSA research team identified the need to carry out a new ad-hoc study that measures the socio-economic effects/psychological impact of the lockdown on the aged 50+ population of England.

Acknowledgment statement:

The ELSA COVID-19 Substudy was funded through the Economic and Social Research Council via the UK Research and Innovation Covid-19 Rapid Response call. Funding has also been received from the National Institute of Aging in the US, and a consortium of UK government departments coordinated by the National Institute for Health Research.

Further information can be found on the http://www.elsa-project.ac.uk/covid-19" style="background-color: rgb(255, 255, 255);">ELSA COVID-19 Study webpage.

ELSA COVID-19 study: End User Licence and Special Licence data

The main data and documentation for the ELSA COVID-19 study are available under SN 8688, subject to standard End User Licence conditions. This study (SN 8918) contains only interview week variables, which are subject to stringent Special Licence conditions. Users should obtain SN 8688 and check whether it is suitable for their needs before considering an application for this study (SN 8918).

Data on causes of death (COD) provide information on mortality patterns and form a major element of public health information.

The COD data refer to the underlying cause which - according to the World Health Organisation (WHO) - is "the disease or injury which initiated the train of morbid events leading directly to death, or the circumstances of the accident or violence which produced the fatal injury".

The data are derived from the medical certificate of death, which is obligatory in the Member States. The information recorded in the death certificate is according to the rules specified by the WHO.

Data published in Eurostat's dissemination database are broken down by sex, 5-year age groups, cause of death and by residency and country of occurrence. For stillbirths and neonatal deaths additional breakdowns might include age of mother and parity.

Data are available for Member States, Iceland, Norway, Liechtenstein, Switzerland, United Kingdom, Serbia, Turkey, North Macedonia and Albania. Regional data (NUTS level 2) are available for all of the countries having NUTS2 regions except Albania.

Annual national data are available in Eurostat's dissemination database in absolute number, crude death rates and standardised death rates. At regional level the same is provided in form of 3-years averages (the average of year, year -1 and year -2). Annual crude and standardised death rates are also available at NUTS2 level. Monthly national data are available for 21 EU Member States from reference year 2019 and in 24 Member States from reference year 2022 in absolute numbers and standardised death rates.