The relationship between economic growth and life expectancy has been widely examined in development and health economics, with early cross country research documenting a strong and persistent income–health gradient (26–28). These foundational studies show that higher income levels are generally associated with longer lives, although the strength of this link varies across stages of development. More recent work applies nonlinear and panel methods to capture heterogeneity and structural differences across countries. Evidence from Vietnam indicates that education and healthcare reforms play a central role in longevity gains, with gender gaps shifting over development phases (29). Using broader welfare indicators, Barbier and Mensah (30) find that while environmental health improvements matter, GDP per capita still delivers larger average returns for life expectancy, especially in low and middle income economies. Yeboah et al. (1) show that growth supports longevity but may simultaneously worsen ecological conditions, with delayed benefits from renewable energy. Sector focused results also support a positive income effect, as Akter et al. (31) report that income, aquaculture, and food production raise longevity in exporting nations. Macroeconomic instability works in the opposite direction: Hataminia and Mohammadzadeh Asl (32) link inflation and unemployment to lower life expectancy. Large panel evidence confirms that GDP per capita improves longevity while pollution reduces it, and that health spending effects vary by income group (33). Crisis based analysis further shows that recessions increase mortality and inequality (34). Overall, growth supports life expectancy, but distributional, environmental, and macroeconomic risks shape outcomes.

Health spending is well-known as a major factor in determining life expectancy, although the extent of influence differs across nations and situations. Foundational research in health economics established that health investment functions as a form of human capital that influences longevity outcomes (35, 36). According to Singh et al. (37), Central Europe and the Baltic states report that health expenditures are generally converged with life expectancy; however, there is still a difference in government versus non-government expenditure. Adebayo et al. (38) employ quantile-on-quantile regression on the United States and find that urbanization and CO₂ emissions have negative influences on life expectancy, whereas income and health expenditure have positive effects. In BRICS economies (2000–2019), Kaur et al. (39) determine that government spending on health leads to increases in GDP and GNI per capita, but the effect on life expectancy is not that significant. Moreover, Khan et al. (40) demonstrate in Pakistan (2000–2020) that the health expenditure and sustainable development goals are integrated with life expectancy and there is a short-run causal relationship. The study by Bétila (41), which uses the System-GMM in 48 African countries (2000–2018), illustrates that ICT use improves the effectiveness of health expenditure that leads to the increase of life expectancy and the decrease in mortality. Lastly, Boundioa and Thiombiano (42) apply the threshold analysis (1996–2018) to show that the quality of governance must have critical levels of improvement in life expectancy by spending on public health. Overall, the literature provides consistent evidence that higher health expenditure contributes to improvements in life expectancy, reflecting the critical role of healthcare investment in enhancing population well-being.

A large body of environmental health research has established that exposure to air pollution is closely linked to higher mortality and reduced longevity across countries and regions (43, 44). These early studies laid the foundation for later cross country analyses showing that pollution related health risks remain significant even after accounting for income and demographic factors. More recent work confirms that air pollution interacts with economic and development conditions in shaping life expectancy outcomes. Evidence from Saudi Arabia shows that CO₂ emissions and PM2.5 concentrations are associated with declines in both life expectancy and healthy life expectancy, although policy reforms under Vision 2030 have partly moderated these effects (15). Broad panel results for developing economies also report a statistically significant negative effect of CO₂ emissions on longevity (45). Country studies offer mixed short and long run patterns. In Malaysia, carbon emissions reduce life expectancy in the short run, while health expenditure improves it (46). In contrast, results for Sub Saharan Africa suggest that GDP growth, industrial activity, and emissions can move positively with life expectancy, challenging the Environmental Kuznets Curve hypothesis (47). Distribution sensitive estimates for GCC countries show that emission effects remain negative across quantiles (48). Mediation evidence from Indonesia links emissions to lower human development outcomes despite gains in income and schooling (49). Additional African and SAARC panel studies confirm that emissions generally reduce life expectancy, whereas renewable energy and urban development can offset part of the damage (50, 51).

The relationship between urbanization and health has long been discussed in demographic and development research, with early studies noting that urban growth can generate both infrastructure advantages and environmental or social risks (52). Urbanization influences life expectancy through multiple channels, including access to healthcare, education, sanitation, and employment, while also increasing exposure to congestion and pollution. Recent empirical work highlights these mixed effects. Using Chinese data from 1990 to 2022, Zhang et al. (53) show that education, urbanization, and green growth improve life expectancy when environmental pressures are controlled, whereas CO₂ emissions reduce longevity. Cross country African evidence based on multilevel models indicates that unemployment and HIV prevalence widen life expectancy gaps, but urbanization helps moderate these disparities by improving service access (54). Micro level evidence also suggests that the health effects of aging depend partly on urban context, income, and lifestyle conditions, confirming a mediating role for urban environments (55). However, results are not uniformly positive. Quantile based analysis in highly polluted countries finds that poor air quality and unmanaged urban expansion are associated with lower life expectancy, while governance quality and inclusive development improve outcomes (56). Additional distribution sensitive evidence from Ghana shows that the effects of urbanization vary across health and income levels, producing unequal longevity outcomes (57). Taken together, the literature indicates that urbanization affects life expectancy in uneven ways, depending on environmental management and institutional quality.

Earlier globalization and health research indicates that trade integration can influence population health through several channels, including income growth, technology transfer, and environmental exposure (58). Empirical evidence from different regions generally finds a positive long run association between trade openness and life expectancy, although short run and indirect effects often vary. For Pakistan, cointegration and causality results show that trade openness and foreign direct investment support longer life expectancy over time (59). Similar findings appear in Nigeria, where trade credit linked to export and import activity contributes to longevity gains, highlighting the role of trade financing (60). Panel causality evidence from Latin America also suggests that trade openness significantly improves life expectancy, especially in less advanced economies with higher tax burdens (61). Country studies provide more nuanced results. For China, regime switching models indicate a dual effect in which trade raises income and health outcomes but may also increase emissions that offset part of the benefit (23). Malaysian evidence likewise shows that exports and imports are positively associated with lifespan through their connection with economic growth (62). Broader African panel estimates using System GMM link welfare and longevity improvements with stronger trade and port performance (63). Additional ARDL results for China confirm that trade openness, public spending, and human capital support life expectancy in the long run, with weaker short run effects (64). Overall, most studies report a positive but context dependent relationship between trade openness and longevity.

The accumulated evidence indicates that the existing literature suggests that life expectancy is shaped by a network of interdependent economic, social, environmental, and structural determinants rather than by any single factor in isolation. Economic growth and trade openness affect longevity primarily through their influence on income levels, structural transformation, and fiscal capacity to finance public health systems. Health expenditure operates more directly by improving access to medical services, preventive care, and treatment effectiveness. Environmental pollution contributes to higher disease burden and mortality risk through sustained exposure effects, while urbanization alters health outcomes by reshaping infrastructure availability, service delivery, and living conditions, with results varying by planning quality and institutional effectiveness. These channels interact and often reinforce one another through both direct and indirect transmission mechanisms. Viewed jointly, this evidence aligns with an extended health production function framework in which longevity outcomes depend on multiple coordinated inputs. This integrated perspective motivates the multivariable empirical specification adopted in the following section. Figure 1 presents the conceptual framework of the study.

Although previous studies present useful information about what determines life expectancy, such as economic growth, health spending, environmental pollution, urbanization, and openness to trade (15, 23, 29, 59), there are still several significant gaps. The majority of research is focused on individual nations or states, especially in the Western world, Africa, or Asia, and not on the BRICS economies, which have peculiarities in the form of high rates of economic growth, serious environmental issues, and heterogeneous demographics. Also, current research is mainly based on specific mean econometric models, which cannot quantify different heterogeneous effects of such determinants on various population groups or life expectancy lines. Economic, health, environmental, urban, and trade interactions are usually considered individually, overlooking the complex interactions between them that determine the endpoint of longevity. Conditional distributions of life expectancy are also seldom analyzed but are especially important in the emerging economies, where income inequality, uneven urbanization, and environmental strains are widespread across space and social ranks. Lastly, although a part of the studies examines causality of variables, there are few studies that use powerful panel models that consider cross-sectional dependence, heterogeneity, and dynamic interactions all in the case of BRICS countries. As a result, a methodological study that combines various determinants and takes distributional heterogeneity into account is highly lacking, which creates a significant gap of knowledge about subtle factors in life expectancy in these rapidly changing economies.

Table 1 presents the variables used to examine the health–economy–environment nexus in the BRICS economies. The selection of indicators was guided by the principles of data completeness, comparability, and reliability across all five countries over 2000–2024, ensuring both cross-sectional coherence and temporal consistency within a balanced panel framework. Life expectancy at birth (LLE) serves as the dependent variable and is drawn from the World Bank’s World Development Indicators, which provide a standardized and internationally comparable measure. LLE is commonly used as an overall indicator of population health because it reflects general mortality conditions and summarizes the average number of years a newborn is expected to live, given current age-specific death rates. Although it does not capture every dimension of health status, it offers a broad and consistent benchmark that incorporates the combined influence of factors such as basic living conditions, healthcare access, and environmental exposure. Its definitional stability and long-term availability also make it well suited for comparative and trend analysis across emerging and developing economies (65, 66).

Among the explanatory variables, GDP per capita (LGDP)—measured in constant 2015 U. S. dollars—was operationalized as a standardized indicator of economic performance and average living standards. Its use allows meaningful comparison across countries and over time by expressing real output per person in inflation-adjusted terms, thereby minimizing price and currency distortions. Government health expenditure per capita (LHE)—reported in current U. S. dollars—captures annual public spending on healthcare goods and services actually consumed within the year, excluding capital investments such as buildings, equipment, or information technology. This operationalization isolates recurrent expenditure that directly reflects the fiscal capacity of governments to sustain health service delivery and coverage (67). Carbon dioxide emissions (LCO₂)—expressed in million metric tons—were selected as the sole environmental indicator because CO₂ data provide complete temporal and cross-country coverage within the World Development Indicators and ensure consistent measurement across all BRICS members. This selection enables comparable long-series estimation of environmental pressure associated with industrial activity and energy consumption, while alternative pollution indicators, such as PM₂.₅ concentrations or water contamination, lack continuity and definitional uniformity for the full study period (68).

Urban population (LUP)—the percentage of people living in areas classified as urban by national statistical authorities—was used to capture the demographic and structural transformation associated with industrialization and expansion of services. This indicator was chosen because it is harmonized by the United Nations Population Division within the World Development Indicators, ensuring consistent definitions of “urban” across countries and through time. Such harmonization minimizes potential measurement bias arising from national classification differences and enhances the comparability of results within a multi-country framework. Trade openness (LTOP)—defined as the sum of exports and imports expressed as a percentage of gross domestic product—was operationalized as a standard ratio widely used in cross-country economic research (69). This specification scales trade volume relative to economic size, allowing meaningful comparison among economies with different output levels and structural compositions. It also provides a transparent and scale-adjusted indicator of external economic integration that aligns with the study’s macro-panel design. Together, these operational definitions and transformations underpin a consistent and balanced dataset, allowing for elasticity-based interpretation within the logarithmic specification and making it possible to examine how economic development, fiscal capacity, urbanization, environmental pressure, and trade integration jointly influence life expectancy across the BRICS economies.

All series are transformed into natural logarithmic form to reduce heteroskedasticity and allow elasticity based interpretation of coefficients. The data are obtained from the World Development Indicators database, which provides internationally comparable cross country statistics on health and development indicators. The final dataset forms a balanced panel covering five BRICS countries over the period 2000 to 2024, with 25 annual observations per country and 125 total observations. Variable definitions were checked for consistency across countries and years, and after verification no missing observations remained, so no interpolation or imputation procedures were required. This balanced structure supports stable multi model and distribution based estimation.

Table 2 indicates definite differences in the mean values and dispersion of the indicators. The mean of life expectancy is high (4.236), and the standard deviation is low (0.092), with the result meaning that there are similar health outcomes between BRICS. Both carbon emissions and GDP per capita indicate high standard deviation (1.144) and mean (7.241), respectively, as well as economic and environmental disparities. The mean (5.577) of health expenditure is large with a broad range of the government expenditures. Conversely, the means of urbanization and openness to trade are moderate (4.054 and 3.751) with low standard deviations (0.359 and 0.288) and have more stable trends. Such differences explain why quantile regression is more suitable because average-based procedures would fail to capture such variations.

Table 3 presents the connection of life expectancy and all the independent variables. The lifespan is positively correlated with GDP (0.304), health expenditure (0.189), carbon emissions (0.538), and negatively correlated with urbanization (−0.163). The data demonstrate that economic expansion, enhanced healthcare finance, and urbanization positively influence longevity, whereas trade exposure may introduce adverse health or environmental consequences. The positive association between life expectancy and emissions reflects the outcomes of the initial stages of industrialization in emerging economies.

The relatively high correlations among GDP, health expenditure, and urbanization indicate some degree of interdependence among these variables and raise the possibility of multicollinearity affecting coefficient estimates. We accounted for this issue by applying robust estimation methods—specifically, Panel-Corrected Standard Errors (PCSE), Driscoll–Kraay Standard Errors (DKSE), and Feasible Generalized Least Squares (F-GLS)—in order to produce consistent standard errors and reliable inference in the presence of correlated regressors, heteroskedasticity, and cross-sectional dependence. Furthermore, the quantile-regression approach focuses on conditional distributions of life expectancy rather than mean effects, which helps lessen the sensitivity of estimates to potential multicollinearity among strongly related predictors. These combined steps help ensure that the relationships identified are stable and reflective of underlying structural patterns rather than distortions from correlated explanatory variables.

A significant economic model in the field of health economics is the Grossman (35) Model, which explains the interaction between health and economic behavior. It considers health as stock capital, which individuals invest to improve well-being and productivity. Although health is a natural process that worsens with age, it can be retained or enhanced by using investments like medical care, healthy lifestyles, and education, which are explained by the model. Building on this view, Smith and Dunt (70) proposed the health production function that incorporates both the health expenditures and non-health inputs as determinants of health outcomes. In the recent past, Shaari et al. (71) added health investments to their model, emphasizing that they enhance life expectancy. Equation (1) reflects the Grossman health function. Health output = f ( health inputs ) (1)

Extending the model to include economic, environmental, and social variables, the health output (life expectancy at birth) can be modeled using government spending on health and government variables: L E it = f ( GDP , HE , CO 2 , UP , TOP ) (2)

In Equation (2) LE = life expectancy which is a dependent variable and GDP = gross domestic product, CO2 = carbon dioxide emission, UP = urbanization and TOP = trade openness are the independent variables.

Equation (3) represents the logarithmic format. LL E it = β 0 + β 1 LGD P it + β 2 LH E it + β 3 LC O 2 it + β 4 LU P it + β 5 LTO P it + ε it (3)

We use the logarithmic form of representation in this work to stabilize the variation of data and linearize nonlinear relationships. In addition, it enables us to examine coefficients as elasticity effects or percentages.

The study resorts to a rigorous panel econometric methodology to examine the various impacts of economic growth, healthcare spending, air pollution, urbanization, and open trade on life expectancy in BRICS countries during 2000–2024. It commences by analyzing the cross-sectional dependence (CSD) and homogeneity of slope to consider interdependencies and structural differences across the BRICS countries. The CSD test as suggested by Pesaran (72) is a test that is used to determine whether the shocks in one country have an influence on the other in the panel, and this is mostly applicable to highly intertwined economies. The slope homogeneity test (73) is used in testing the homogeneity of the slope coefficients across nations to assist in deciding whether heterogeneous estimators are appropriate. Having confirmed the cross-sectional correlations and heterogeneity, panel unit root tests are then used to test the stationarity of the variables. Three tests are used in complement: Levin, Lin, and Chu (LLC) (74), Cross-sectionally Augmented IPS (CIPS) (75), and Augmented Dickey-Fuller (ADF), which together confirm that all variables are integrated of order one, I(1). Subsequently, the panel cointegration test by Pedroni is performed in order to reveal the long-term equilibrium relationship between life expectancy, GDP, health expenditure, emissions, urbanization, and trade openness (76, 77). The findings affirm that these variables are co-moved over the long run among BRICS countries.

The study uses quantile regression to investigate asymmetric and heterogeneous effects of economic, health, and environmental factors on life expectancy (78). In contrast to traditional mean-based estimators, quantile regression permits the study of the way the impacts of explanatory factors differ across the conditional distribution of life expectancy. The method is especially applicable to the BRICS situation, where nations are at varying levels of growth and where they have significant differences in terms of income, urbanization, and health infrastructure. The approach identifies the subtle effects of GDP, healthcare spending, emissions, urbanization, and trade openness on nations with low, medium, and high rates of life expectancy by examining the effects at various quantiles (e.g., lower, median, and upper tails).

Mathematically, the conditional quantile functionality is written as for Equation 4: Q LLE it ( τ ∣ X it ) = α i ( τ ) + X it ′ β ( τ ) , 0 < τ < 1 (4) where Q 2 LLE it ( τ ∣ X it ) denotes the τ-th conditional quantile of life Expectancy, α i ( τ ) captures country fixed effects, and β(τ) represents the vector of slope parameters that vary across quantiles.

The estimation is obtained by solving the minimization problem as shown in Equation 5: β ̂ ( τ ) = arg min β ∑ i = 1 N ∑ t = 1 T ρ τ ( LLE it − α i ( τ ) − X it ′ β ( τ ) ) (5)

Where ρ τ ( μ ) = μ ( τ − I { μ < 0 } ) is the quantile loss function.

Lastly, models are robust by use of various estimation methods. The possible heteroskedasticity, autocorrelation, and cross-sectional correlation problems are addressed by Panel-Corrected Standard Errors (PCSEs) (79), Driscoll-Kraay Standard Errors (DKSEs) (80), and Feasible Generalized Least Squares (F-GLS) (81), which prove the results to be stable. In an effort to investigate the causation direction further, the Dumitrescu-Hurlin (D-H) panel causality test is used (82). This approach determines both unidirectional and bidirectional causal relationships between variables and offers insights on the dynamic relationships between life expectancy, economic growth, health expenditure, urbanization, emissions, and trade openness in BRICS countries. This intensive panel econometric system as a whole makes the results robust and policy-relevant.

Several specification considerations were taken into account when designing the empirical strategy. First, potential endogeneity cannot be fully ruled out, as reverse relationships may exist between health outcomes and macroeconomic variables. Due to data and instrument constraints in a multi country macro panel setting, the estimates are interpreted as conditional associations rather than strict causal effects. Second, correlation diagnostics indicate moderate to high associations among some regressors, particularly GDP, health expenditure, and urbanization. To reduce inference bias from cross sectional dependence, heteroskedasticity, and residual correlation, multiple robust estimators including PCSE, DKSE, and F-GLS are applied. Third, CO₂ emissions may partly capture industrialization and development intensity rather than pure environmental damage, and coefficient interpretation is therefore made with this structural channel in mind. Quantile regression is selected because the primary objective is to examine distributional heterogeneity rather than average effects.

It should be noted that the quantile regression framework is designed to identify heterogeneous conditional relationships across the distribution of life expectancy and is not, by itself, a causal identification strategy. While robustness estimators and panel causality tests are employed to examine directional patterns, the reported coefficients should be interpreted as distribution specific associations. Accordingly, policy implications are drawn with appropriate caution and focus on structural relationships rather than definitive causal magnitudes.