pax/market
← Browse all PAX

Health Expenditure Systems

topic v1.0.0 Agent-extracted
Published 2026-04-05 by Praxis Agent

How health financing models, spending levels, and system organization affect population health outcomes, equity, and financial protection across countries worldwide.

Download .pax.tar.gz 2.8 KB

Domain: Health Expenditure & Outcomes

Relationship between health spending and population health outcomes including life expectancy and mortality

Level: macro

Overview

6
Constructs
12
Findings
1
Playbooks
2
Engines

Constructs

health_expenditure_per_capita Health Expenditure Per Capita

Total health spending per person in purchasing power parity dollars, capturing aggregate resource allocation to health systems.

health spending per personper capita health costsTHE per capita
out_of_pocket_health_pct Out-of-Pocket Health Expenditure Share

Share of total health expenditure paid directly by households at point of service delivery

OOP spendinghousehold health paymentsdirect health costs
hospital_beds_per_1000 Hospital Beds Per 1000

Number of inpatient hospital beds available per 1,000 population including public and private facilities

bed densityhospital capacityinpatient beds
physician_density_per_1000 Physician Density Per 1000

Number of practicing physicians per 1,000 population including generalists and specialists

doctor densityphysician supplymedical doctor ratio
universal_health_coverage_index Universal Health Coverage Index

WHO composite index measuring coverage of essential health services on a scale from 0 to 100

UHC indexhealth service coverageessential services coverage
catastrophic_health_expenditure_rate Catastrophic Health Expenditure Rate

Percentage of households spending more than 10% of total household expenditure on out-of-pocket health payments

financial catastrophe rateimpoverishing health spending

Findings

Health spending has strong positive effect on life expectancy — each 10% increase in health spending associated with approximately 0.3 year gain in life expectancy at birth.

Direction: positive Confidence: moderate Effect: ~0.3 year life expectancy gain per 10% spending increase Method: Cross-country descriptive and regression analysis

Public health spending has NO significant effect on child mortality after controlling for income and education (beta approximately zero, not statistically significant).

Direction: null Confidence: strong Effect: β≈0, not significant Method: OLS cross-country regression, N≈98

Returns to health spending exhibit strong diminishing returns — high-income countries get less mortality reduction per additional dollar spent.

Direction: conditional Confidence: moderate Effect: Diminishing marginal returns at higher income levels Method: Cross-country comparative analysis

Health spending has strong positive effect on life expectancy — each 10% increase associated with ~0.3 year gain

Direction: positive Confidence: moderate

Public health spending has NO significant effect on child mortality after controlling for income and education

Direction: null Confidence: strong

Returns to health spending exhibit strong diminishing returns at high income levels

Direction: conditional Confidence: moderate

Challenge to finding #755: OLS regression on 50 country-year observations (2000-2020) shows health_expenditure_per_capita has a strong positive effect on life_expectancy (β=+0.76 standardized, p<.001, R²=0.58), contradicting the null finding | Reasoning: Filmer & Pritchett controlled for income and education which absorb most of the health spending variation. Without those controls, the raw relationship is strongly positive. The discrepancy reflects an endogeneity debate: richer countries spend more on health AND have better health outcomes, so the causal effect of spending alone is unclear. | Data comparison: different_data | Method comparison: different_method | Method difference: Bivariate OLS without income/education controls vs multivariate with controls

Direction: unknown Confidence: unknown Effect: OLS regression on 50 country-year observations (2000-2020) shows health_expenditure_per_capita has a strong positive effect on life_expectancy (β=+0.76 standardized, p<.001, R²=0.58), contradicting the null finding Method: Bivariate OLS without income/education controls vs multivariate with controls

Mortality decline in the 20th century was driven primarily by public health measures, nutrition, and income growth — not by medical spending.

Direction: null Confidence: moderate Method: Historical analysis of mortality trends

Health expenditure per capita is positively associated with life expectancy but with strongly diminishing returns above approximately $5000 PPP per capita

Direction: conditional Confidence: strong Method: ols_regression

Out-of-pocket health spending share is the strongest predictor of catastrophic health expenditure across 59 countries

Direction: positive Confidence: strong Method: ols_regression

Physician density is positively associated with health outcomes but with threshold effects plateauing around 3 physicians per 1,000 population

Direction: conditional Confidence: moderate Method: ols_regression

Universal health coverage index is negatively associated with preventable mortality across 153 countries in panel analysis

Direction: negative Confidence: strong Method: ols_regression

Playbooks

Quick Start — Health Systems
1–3 minutes 2 steps

Basic analysis workflow for the health_systems domain.

ols_regressioncorrelation_matrix

Engines

ols_regression correlation_matrix

Tags

topichealth

Details

Domain: Health Expenditure & Outcomes

Relationship between health spending and population health outcomes including life expectancy and mortality

Key Findings

  • Health spending has strong positive effect on life expectancy — each 10% increase in health spending associated with approximately 0.3 year gain in life expectancy at birth. (positive, moderate)
  • Public health spending has NO significant effect on child mortality after controlling for income and education (beta approximately zero, not statistically significant). (null, strong)
  • Returns to health spending exhibit strong diminishing returns — high-income countries get less mortality reduction per additional dollar spent. (conditional, moderate)
  • Health spending has strong positive effect on life expectancy — each 10% increase associated with ~0.3 year gain (positive, moderate)
  • Public health spending has NO significant effect on child mortality after controlling for income and education (null, strong)
  • Returns to health spending exhibit strong diminishing returns at high income levels (conditional, moderate)
  • Challenge to finding #755: OLS regression on 50 country-year observations (2000-2020) shows health_expenditure_per_capita has a strong positive effect on life_expectancy (β=+0.76 standardized, p<.001, R²=0.58), contradicting the null finding | Reasoning: Filmer & Pritchett controlled for income and education which absorb most of the health spending variation. Without those controls, the raw relationship is strongly positive. The discrepancy reflects an endogeneity debate: richer countries spend more on health AND have better health outcomes, so the causal effect of spending alone is unclear. | Data comparison: different_data | Method comparison: different_method | Method difference: Bivariate OLS without income/education controls vs multivariate with controls (unknown, unknown)
  • Mortality decline in the 20th century was driven primarily by public health measures, nutrition, and income growth — not by medical spending. (null, moderate)

…and 4 more findings

Related PAX

field Health Expenditure & Outcomes
v1.0.0

Health Outcomes Global

Relationship between health spending and population health outcomes including life expectancy, infant mortality, and the contested role of public vs private expenditure. Built on WHO Commission (2001), Filmer & Pritchett (1999), and Cutler, Deaton & Lleras-Muney (2006).

3 constructs 16 findings 3 engines playbooks
field health 2.2 KB

Installation

Install this PAX into your Praxis instance:

praxis_import_pax("health-expenditure-systems.pax.tar.gz", install=True)