Difference between revisions of "Human development/Description"

Jump to: navigation, search
(Replaced content with "{{ComponentSubDescriptionTemplate}}")
Line 1: Line 1:
|Description=In the GISMO model the impacts of global environmental change on human development are  modeled by considering impacts on human health, either directly through for example the impact of climate change on malaria, or indirectly through for example the impact of climate change on food availability. Next to environmental factors, human health is also driven by socio-economic factors, including income levels and educational attainment. To take account of these different factors and their interrelation, the GISMO model consists of three modules that address human health, poverty and education (Figure 7.6.1). The modules are linked through a cohort component population model, including endogenous fertility and mortality (for details see Hilderink, 2000). Fertility levels are modeled using a convergence level, determined by female education levels, and a speed of convergence, determined by the HDI (see below). Mortality rates are determined by the health module, which is discussed more in-depth in the remainder of this Section. Future trends in migration, including urbanization, are taken exogenously to the model (for details see Hilderink, 2000)
The Human Development Index (HDI), introduced in the human development report 1990 to rank development achievement, is a composite index of life expectancy, education, and income indices (UNDP, 1990, 2010). The underlying indicators have been refined several time during the years, while the three elements have remained the same. The index links to the three GISMO model components.
==GISMO health model==
The GISMO health model describes the causal chains between health-risk factors and health outcomes, (morbidity and mortality) taking into account the effect of health services. The mortality rate is modelled through a risk factor-attributable component and a non-attributable component. Historically, the non-attributable component represents the remainder of mortality not covered by the risk-factors included. For future projections this component is assumed to reduce by regional average historic rates of reduction.
The risk factor-attributable component is based on a multi-state approach, distinguishing exposure, disease and death (Cairncross and Valdmanis, 2006; WHO, 2002). This implies that for various health risk factors, incidence and case fatality rates (i.e. the ratio of the number of deaths caused by a specific disease over the number of diagnosed cases of that disease) are taken into account. The level of health services modifies case fatality rates. This methodology is used for malaria, diarrhoea and pneumonia; the approach for protein deficiency is discussed below. The method for projecting mortality due to other causes of death (non-communicable chronic diseases, other communicable diseases and injuries) follow the global burden of disease (GBD) approach that uses a parsimonious regression technique to related mortality rates of ten major disease clusters with GDP, smoking behavior and human capital  (Mathers and Loncar (2006). This methodology is used for the causes of death related to urban air pollution.
The GISMO health model considers mortality due to (i) malaria; (ii) communicable and infectious diseases associated with undernourishment, poor access to safe drinking water and basic sanitation and poor indoor air quality; (iii) diseases caused by poor outdoor air quality; (iv) HIV-AIDS; and (v) chronic diseases including high blood pressure and obesity. Here, we only discuss the first three causes of mortality as they are linked to environmental factors. The mortality rate due to a specific disease is a multiplication of the incidence rate (fraction of the population with the specific disease) and the case fatality rate (fraction of the people who die from the specific disease), distinguishing for the two sexes and 5-year age cohorts. These mortality rates can then be used to calculated age-specific life expectancy (for details see Hilderink, 2000).
*Malaria risk. In GISMO, incidence rates of malaria are determined by the areas which are suitable for the malaria mosquito based on the monthly climatic factors temperature and precipitation ([[Atmospheric composition and climate]]) (Craig et al., 1999). The incidence rates are decreased by the level of insecticide treated bed nets and indoor residual spraying, modeled separately as potential policy options. The case fatality rate of malaria is increased by underweight levels and decreased by case management, i.e. treatment).
*Access to food, water and energy. GISMO relates incidence and case fatality rates of major communicable (infectious) diseases to access to food, water and energy (Table 7.6.2), with access defined by per capita food availability, access to safe drinking water and improved sanitation, and access to modern energy sources for cooking and heating. The per capita food availability (Kcal/cap/day) is obtained from the IMAGE model (chapter 4.2).  The levels of access to safe drinking water and improved sanitation are modeled separately by applying linear regression. The explanatory variables include GDP per capita, urbanization rate and population density. Developments in water supply are assumed to be implemented ahead of sanitation. As such, developments in access follow a pathway from no sustainable access to safe drinking-water and basic sanitation, to improved water supply only, to improved water supply and sanitation, to a household connection for water supply only, to a household connection for water supply and sanitation. The level of access to modern energy sources for cooking and heating distinguishes between the use of 1) traditional biomass and coal on traditional stoves; 2) traditional biomass and coal on improved stoves; and 3) the use of modern fuels (electricity, natural gas, LPG, kerosene, modern biofuels and solar stoves). Trends in access to modern energy sources are taken from the [[TIMER]] residential model.
Child underweight and prevalence of undernourishment. For children under five years of age undernourishment is expressed as underweight (measured as weight-for-age), while for higher ages the prevalence of undernourishment is used. The direct effect of undernutrition is protein deficiency, which for children mortality rates are scaled to their underweight status; for higher age groups mortality rates are scaled to the levels of undernourishment. Indirectly, undernourishment enhances the incidence of diarrhea and pneumonia, and the case fatality of malaria, diarrhea and pneumonia. Thees in direct effects are only modeled for children under five.
Child underweight as a result of chronic undernourishment is modeled as a function of improvements in average food intake, the ratio of female to male life expectancy at birth, female enrolment in secondary education and access to clean drinking water (Smith and Haddad, 2000). Based on a normal distribution, the total number of underweight children is divided into a mild, a moderate and a severe underweight group (De Onis and Blossner, 2003).
The prevalence of undernourishment is calculated from per capita food availability and the minimum energy requirements from FAO (FAO, 2003). The calculations uses a lognormal distribution function determined by the mean food consumption and a coefficient of variation. The coefficient of variation decreases over time as a function per capita GDP. Finally, the minimum dietary energy requirement is derived by aggregating the region specific sex-age energy requirements weighted by the proportion of each sex and age group in the total population, including a so-called pregnancy allowance.
Incidence rates of pneumonia, chronic obstructive pulmonary disease (COPD) and lung cancer are increased by indoor air pollution caused by cooking and heating with traditional biomass and coal.
Simultaneously, incidence rates and case fatality rates are increased by child underweight levels.
Incidence rates of diarrhea depend on the different connection levels to drinking water and sanitation facilities, child underweight levels and by temperature increase. The case fatality rates are increased by underweight levels and decreased by the level of oral rehydration therapy.
(iii) Mortality associated with urban air pollution. Mortality rates of lung cancer, cardiopulmonary diseases and acute respiratory infections due to urban air pollution (i.e. PM10 and PM2.5 concentration levels) are derived applying the GBD methodology (Mathers and Loncar, 2006). Based on the emissions of NOx, SO2 and black carbon (chapter 5), PM10 concentration levels are determined using the Global Urban Air quality Model (GUAM). GUAM originates from the GMAPS model (Pandey et al., 2006), that determines PM10 concentration levels by economic activity, population, urbanization and meteorological factors. PM2.5 concentrations are obtained using a region-specific PM10/PM2.5 ratio. Based on these concentration levels and the exposed population, mortality attributable to  the aforementioned causes of death is derived using relative risks. These relative risks are obtained from epidemiological literature (Dockery et al., 1993; Pope et al., 1995).
==GISMO poverty model==
In general, people are considered poor if their consumption or income level falls below some level necessary to meet basic needs, the ‘poverty line’. Poverty headcount, people below a poverty line, is determined by applying a log-normal distribution using per capita income and a GINI coefficient to describe its distribution over a population . The poverty module addresses people living below the international poverty lines of $1.25 and $2 per day at 2005 PPP as defined by the World Bank (Ravallion et al., 2008).
==GISMO education model==
The education model addresses future developments in school enrolment and educational attainment, including literacy rates, for three levels of education: primary, secondary and tertiary. The model tracks the shares of the highest completed education and the average years of schooling per cohort. The enrolment ratios per education level are determined, using cross-sectional relationships with per capita GDP (PPP). The ages at which children attain a certain educational level are assumed to be equal for all regions. Literacy rates are proxied by the share of the population, aged 15+, having completed at least their primary education. Furthermore, to take account of autonomous increases in literacy levels, yearly literacy levels of the population between 15 and 65 increase by 0.3% per year.

Revision as of 11:30, 2 August 2013