Political stability, health system resilience, and macroeconomic shocks as factors associated with malaria control outcomes: a comparative analysis of Ecuador and Venezuela (2000–2023)

Abstract

Background

The post-2000 period in South America presents a striking divergence in malaria epidemiology: Ecuador achieved a >99% reduction in malaria incidence approaching elimination, while Venezuela experienced one of the most severe resurgences in modern public health history. We examined the association between governance factors, health policies, intersectoral coordination, and malaria control outcomes by comparing the national experiences of Ecuador and Venezuela.

Methods

We conducted a comparative ecological study using epidemiological data from WHO World Malaria Reports and PAHO updates (2000–2023), integrated with political stability indices, health expenditure data, GDP per capita, poverty rates, and crude-oil price series from World Bank and CFR databases. A joinpoint (segmented) regression model was applied to each country’s log-transformed malaria incidence time series to identify statistically significant inflection points and estimate annual percentage change (APC) per segment. A multi-dimensional framework assessed governance, health-system capacity, surveillance infrastructure, intersectoral coordination, and community engagement. We also examined systemic fragility through concurrent infectious disease trends (diphtheria, measles, tuberculosis) in Venezuela. Use of AI language models is declared in the Methods.

Results

Ecuador reduced cases from 106,641 (2001) to 757 (2023), a 99.3% reduction. The joinpoint model identified two significant inflection points (2008, 2014) delineating three phases of declining trend (APC: $-34.2\%$, $-12.5\%$, $-1.2\%$). Venezuela experienced a 1,217% increase from 35,500 (2000) to a peak of 527,627 (2018); a single joinpoint at 2014 marked a shift from non-significant endemic transmission (APC: $+2.8\%$, $p=0.12$) to explosive resurgence (APC: $+84.6\%$, $p<0.001$), temporally coinciding with the global crude-oil price collapse and ensuing fiscal contraction. Post-2018 case declines in Venezuela are accompanied by concurrent surveillance collapse, mass emigration, and reduced care-seeking, suggesting underreporting rather than epidemiological improvement. Pearson correlations with malaria incidence were observed for political stability index ($r=-0.85$), health expenditure (% GDP, $r=-0.78$), poverty rate ($r=+0.82$), and crude-oil price ($r=-0.89$) in Venezuela. These correlations should be interpreted with caution given potential confounding and collinearity among indicators. Comparison across five governance-resilience domains using PAHO/WHO operational criteria shows systematic differences between countries. Resurgence of diphtheria (>1,600 cases), measles (>7,300 cases), and tuberculosis (near-doubling of incidence) in Venezuela occurred concurrently with the malaria resurgence.

Conclusions

Health-system resilience and governance quality appear to be associated with the relationship between macroeconomic shocks and epidemiological outcomes. The strong collinearity between political stability and commodity-price cycles in Venezuela precludes definitive attribution to either factor independently. Technical interventions may be insufficient without stable governance, protected financing, and institutional continuity. Recommendations aligned with the PAHO malaria elimination framework are provided.

Supplementary Information

The online version contains supplementary material available at https://doi.org/10.1186/s12936-026-05899-9.

Introduction

The beginning of the 21st century marked a period of renewed global commitment to malaria elimination, catalyzed by the Millennium Development Goals and formalized through the WHO Global Technical Strategy for Malaria 2016–2030 [1]. The Americas region appeared particularly well-positioned to achieve elimination targets, with declining transmission in most endemic countries and strong regional cooperation frameworks [2]. However, this period of optimism has been disrupted by one of the most dramatic divergences in malaria epidemiology documented in modern public health: the simultaneous near-elimination in Ecuador and resurgence in Venezuela.

This divergence represents a valuable comparative case for examining the structural factors associated with public health program success and failure. The contrast is quantitatively striking: Ecuador achieved a 99.3% reduction in malaria cases, declining from a peak of 106,641 cases in 2001 to merely 757 cases in 2023, earning recognition in the WHO E-2020 initiative [3]. Simultaneously, Venezuela experienced an explosive resurgence, with cases increasing approximately 1,217% from 35,500 in 2000 to a peak of 527,627 in 2018 [4]. By 2019, Venezuela alone accounted for over 50% of all malaria cases in the Americas and 73% of regional malaria deaths, transforming from a regional leader in disease control to an epidemiological emergency of hemispheric scope [5].

The Venezuelan situation is particularly notable given the country’s distinguished history in malaria control. Venezuela achieved WHO certification for eliminating malaria from 68% of its territory by 1961, led by the pioneering work of Dr. Arnoldo Gabadón, whose integrated approach to vector control became a model for tropical disease programs globally. This historical success implies the existence of deep institutional knowledge, technical expertise, and consolidated infrastructure — assets that appear to have been substantially diminished over the past decade [6].

Novel contribution relative to prior literature

Several important studies have examined aspects of this divergence separately. Grillet et al. [6] documented Venezuela’s humanitarian crisis and cross-border spillover; Grillet et al. [4] characterized the southern Venezuela hotspot and the mining nexus; the International Crisis Group [7] detailed the mining-transmission link; and Krisher et al. [8] described Ecuador’s binational success at the Peru border. While these studies illuminate individual proximate mechanisms, none simultaneously: (i) applies formal joinpoint regression to quantify trend changes and test statistical significance of inflection points in both countries; (ii) integrates crude-oil price series as an explicit macroeconomic variable alongside political stability indices; (iii) constructs a cross-domain correlation matrix spanning governance, economic, and poverty indicators for both nations over the full 2000–2023 window; (iv) examines concurrent resurgence of multiple vaccine-preventable diseases as a marker of systemic fragility; and (v) presents an operationalization of governance-resilience domains using PAHO/WHO criteria enabling systematic comparison. The present paper synthesizes these elements into a single analytical framework.

The primary objective of this study is to examine the association between governance, health policies, intersectoral coordination, and malaria control outcomes by comparing the national experiences of Ecuador and Venezuela. We further examine whether Venezuela’s crisis has been accompanied by regional case exportation. If governance and macroeconomic factors are associated with program performance, this analysis has implications for global health strategy: disease elimination programs may benefit from systematic integration of political, institutional, and economic risk assessment into program design.

Materials and methods

Study design and conceptual framework

We employed a comparative ecological study design to analyze the divergent malaria control trajectories of Ecuador and Venezuela from 2000 to 2023. This approach enables examination of complex, context-dependent phenomena where multiple factors interact [9]. Our multi-dimensional analytical framework [10] examines five key governance-resilience domains: (1) political stability and governance quality; (2) health system capacity and financing; (3) surveillance and response infrastructure; (4) intersectoral coordination and international cooperation; and (5) community engagement and social determinants.

The dependent variable is the annual log-transformed malaria incidence rate (cases per 100,000 inhabitants). Independent variables include the World Bank Political Stability Index, current health expenditure as a percentage of GDP, GDP per capita (current USD), poverty headcount ratio, and global crude-oil price (Brent, USD/barrel). The five governance-resilience domains listed above provide the conceptual structure for organizing these variables and for the qualitative comparative assessment presented in Table 4.

Table 4.

Comparison of governance and health-system resilience dimensions using PAHO/WHO operational criteria: Ecuador vs. Venezuela

Domain	Ecuador	Venezuela	PAHO/WHO criterion
Real-time surveillance network	SIVE-Alerta operational; >95% reporting completeness; median notification delay <48 h (source: Ecuador MSP annual reports)	Boletín Epidemiológico suspended 2014; laboratory capacity estimated at 30% of 2010 levels by 2018 (source: [6, 13])	DTI-R: Case detection within 48 h of symptom onset [22]
Health expenditure protection	Constitutional mandate; 7–8% GDP throughout study period (source: World Bank WDI)	Decline from 3.1% GDP (2014) to estimated <1.5% (2017) (source: World Bank WDI, IMF estimates)	WHO: Minimum 5% GDP for universal health coverage
Binational/multilateral agreements	Ecuador–Peru PAMAFRO agreement active; synchronized vector control; 96% case reduction in border zone (source: [8])	Limited international cooperation documented; participation in regional mechanisms reduced post-2015 (source: PAHO reports)	PAHO: Formalized cross-border coordination mechanisms
Public epidemiological reporting	Annual epidemiological reports published; data accessible via PAHO/WHO databases	Boletín suspended 2014; ministerial dismissal following data disclosure reported 2017 (source: [13])	IHR: Timely and transparent reporting to WHO [26]
Territorial micro-planning	Risk-stratified foci identification; <72 h case investigation per DTI-R protocol (source: Ecuador MSP)	National programme functionality reduced; micro-planning capacity not documented post-2016 (source: [4])	PAHO “Best Buy”: Adaptive micro-stratification [22]
Extractive industry regulation	Regulated mining sector; environmental impact assessment required	Arco Minero del Orinoco: >100,000 informal miners reported; deforestation documented (source: [7])	Intersectoral coordination per WHO GTS [1]
Healthcare workforce	Competitive salaries maintained; community volunteer network operational (source: [8])	>22,000 physicians emigrated 2012–2018; 65% malaria-programme positions vacant reported (source: [6])	WHO: Health workforce density thresholds

Assessment based on available documentary evidence. Criteria derived from PAHO Regional Malaria Elimination Framework [2], PAHO Best Buys [22], WHO GTS [1], and IHR (2005). Some Venezuelan data post-2014 are indirect estimates due to reporting limitations

Following Kruk et al. [10], health-system resilience is defined here as the capacity of health institutions to absorb, adapt to, and recover from shocks — including economic downturns, political transitions, and epidemic surges — while maintaining core functions and population coverage. Consistent with the emerging literature on health system fragility [11], fragility is not treated in this study as the simple inverse of resilience. Rather, systemic fragility is understood as an independent construct characterized by breakdowns in the interaction between health systems and the communities they serve — manifesting as loss of institutional trust, inequitable service distribution, and reduced capacity for community participation. This conceptualization is relevant because a health system may lose resilience incrementally while fragility emerges abruptly, as appears to have occurred in Venezuela following the 2014 oil shock.

Declaration of artificial intelligence use

In accordance with emerging standards of methodological transparency [12], the authors declare that large language model (LLM) AI tools were used exclusively for narrative structuring, stylistic synthesis, and technical translation of primary and secondary source material. No AI tool was used for primary data generation, statistical parameter estimation, or formulation of hypotheses. All analytical data, statistical model outputs, and final conclusions are the exclusive product of human analysis and interpretation by the authors. Data extraction code and statistical analysis scripts in R used to generate the models presented in this manuscript are available from the corresponding author upon reasonable request.

Data sources and collection

Epidemiological data

Malaria incidence data were obtained from WHO World Malaria Reports (2000–2023) and PAHO epidemiological updates. We extracted annual case numbers, incidence rates per 100,000 population, species distribution (Plasmodium falciparum versus P. vivax), and mortality data. Data quality varied significantly between countries; for Venezuela after 2014, official reporting became severely compromised due to the suspension of the Boletín Epidemiológico and subsequent reduction in national surveillance infrastructure [6, 13]. This data quality limitation is addressed explicitly in the sensitivity analysis and limitations sections.

Given the documented limitations of official Venezuelan reporting post-2014, we employed a triangulation strategy using proxy indicators. We integrated data from the Venezuelan Cardiometabolic Health Study (EVESCAM, 2014–2020), a nationally representative cohort of 1,257 subjects [14], which provided quantitative vulnerability indicators: 77.3% of the population reported interruptions in access to basic public services; 76.2% experienced lack of transport; and 60.2% experienced food insecurity. These micro-level data were combined with regional imported-case statistics from Brazil, Colombia, and Guyana to provide context for interpreting the internal outbreak magnitude. Although EVESCAM was designed as a cardiometabolic survey, its nationally representative sampling frame and its inclusion of indicators on service access, transport barriers, and food security make it one of the few population-based data sources available for Venezuela during this period. We use these indicators not as direct measures of malaria surveillance capacity, but as contextual evidence of health-system accessibility conditions under which malaria case ascertainment would occur. This distinction is maintained throughout the analysis.

Governance, economic, and oil-price data

Political stability scores were obtained from the World Bank Worldwide Governance Indicators (annual scale $-2.5$ to $+2.5$) [15]. Macroeconomic data (total health expenditure as % GDP, GDP per capita in current US$, poverty headcount ratio) were extracted from World Bank World Development Indicators [16, 17]. Global crude-oil price series (Brent, USD/barrel) were compiled from the US Energy Information Administration and Council on Foreign Relations databases [18]. The inclusion of crude-oil prices is relevant because Venezuela’s dependence on hydrocarbon revenues means that the political stability index and macroeconomic indicators are structurally correlated; this collinearity is addressed explicitly in the Discussion.

Infectious disease data for systemic fragility assessment

To examine whether the Venezuelan malaria resurgence coincided with broader health-system fragility, we compiled annual incidence data for diphtheria, measles, and tuberculosis in Venezuela from PAHO epidemiological bulletins, CDC, and peer-reviewed reports [6, 19].

Segmented (joinpoint) regression model specification

To examine whether statistically significant changes in epidemiological trajectory coincided temporally with macroeconomic and governance changes, we applied joinpoint regression analysis [20] to the log-transformed malaria incidence rate series (cases per 100,000 inhabitants) for each country, 2000–2023. This method identifies breakpoints (${\tau }_k$) at which the linear trend changes significantly, using a permutation test procedure ($\alpha =0.05$, 4,499 permutations).

The model specification is:

$$\begin{array}{c}\hfill ln(y_t)={\beta }_0+{\beta }_{1}t+\sum _{k=1}^K{\delta }_k{(t-{\tau }_k)}^{+}+{ϵ}_t\end{array}$$ 1

where:

• $ln(y_t)$ = natural logarithm of the malaria incidence rate in year t;

• t = calendar year (continuous time variable, centered at the midpoint of the series);

• ${\tau }_k$ = joinpoint (inflection year) k;

• ${\delta }_k$ = difference in slope between the segment after joinpoint k and the segment before it;

• ${(t-{\tau }_k)}^{+}$ = indicator function equal to $(t-{\tau }_k)$ if $t>{\tau }_k$, and 0 otherwise;

• ${ϵ}_t$ = random error term.

Model fitting and diagnostics. Models were fitted using the Joinpoint Regression Program version 4.9.1.0 (National Cancer Institute, Bethesda, MD) with grid search optimization and Bayesian Information Criterion (BIC) for model selection. A maximum of 3 joinpoints was permitted. Residual diagnostics included examination of autocorrelation function (ACF) plots and the Durbin-Watson statistic to assess first-order serial correlation, which is common in infectious disease time series [21]. When significant autocorrelation was detected (Durbin-Watson $<1.5$ or $>2.5$), we applied a first-order autoregressive error structure [AR(1)] with parameters estimated via restricted maximum likelihood (REML). Standard errors were computed using the heteroskedasticity-robust (HC3) estimator.

Convergence and sensitivity. Convergence was assessed by examining trace plots and the Gelman-Rubin diagnostic ($\widehat{R}<1.1$ criterion) when Markov Chain Monte Carlo (MCMC) methods were employed for uncertainty quantification. Sensitivity analyses were conducted by: (i) truncating the Venezuelan series at 2017 (pre-surveillance collapse) to assess robustness of the 2014 joinpoint; and (ii) excluding the COVID-19 pandemic years (2020–2021) to assess their influence on late-period estimates.

The Annual Percentage Change (APC) for each segment j was calculated as:

$$\begin{array}{c}\hfill {\text{APC}}_j=(e^{{\beta }_j}-1)\times 100\end{array}$$ 2

where ${\beta }_j$ is the estimated slope for segment j. 95% confidence intervals for APC were derived from the profile likelihood.

The hypothesis tested is: if macroeconomic and governance factors are temporally associated with epidemiological trajectory changes, a statistically significant joinpoint (shift from non-significant or declining APC to strongly positive APC) will be observed in Venezuela’s epidemic curve coinciding with the 2014 global crude-oil price collapse.

Correlation analysis

Pearson correlation coefficients (r) were calculated between the annual log-transformed malaria incidence series and each structural indicator (political stability index, health expenditure as % GDP, GDP per capita, poverty headcount ratio, crude-oil price) for Venezuela and Ecuador separately.

Statistical assumptions and limitations. Pearson correlation assumes linearity, bivariate normality, and independence of observations. Time series data violate the independence assumption due to temporal autocorrelation. To address this, we report correlations with the caveat that effective degrees of freedom are reduced, and p-values should be interpreted conservatively. Bonferroni correction was applied for multiple comparisons (10 tests per country; adjusted $\alpha =0.005$). Additionally, high collinearity among economic indicators (oil price, GDP, health expenditure, political stability) in Venezuela prevents independent attribution of effects; variance inflation factors (VIF) exceeded 5 for several indicator pairs.

For Venezuela, the primary analysis used the 2000–2017 window (pre-surveillance collapse); sensitivity analysis extended to 2019. For Ecuador, the full 2000–2023 series was used. A correlation matrix table is reported in Results (Table 2).

Table 2.

Pearson correlations between structural indicators and log-transformed malaria incidence

Indicator	Ecuador	Ecuador	Venezuela	Venezuela	r (ECU)	r (VEN)
	2014	2022	2014	2017
Health expenditure (% GDP)	7.5%	7.56%	3.1%	$\sim$1.5%	$-0.61_{}^{\ast }$	$-0.78_{}^{\ast \ast }$
GDP per capita (current USD)	$6,377	$6,391	$16,055	$\sim$$8,000	$-0.44_{}^{\ast }$	$-0.65_{}^{\ast \ast }$
Poverty rate (%)	$\sim$22.5	$\sim$25.2	>50.0	$\sim$60.0	$+0.38$ (ns)	$+0.82_{}^{\ast \ast }$
Political stability index (WB)	$-0.28$	$-0.34$	$-1.01$	$-1.15$	$-0.52_{}^{\ast }$	$-0.85_{}^{\ast \ast }$
Crude-oil price (USD/bbl)	$93.17	$100.93	$93.17	$54.25	$-0.29$ (ns)	$-0.89_{}^{\ast \ast }$

Venezuela: 2000–2017 (pre-surveillance collapse); Ecuador: 2000–2023

$_{}^{\ast }p<0.05$;  $_{}^{\ast \ast }p<0.005$ (Bonferroni-corrected);  ns = not significant. Columns 2–5 present selected descriptive values at key time points for contextual reference;columns 6–7 report Pearson r coefficients calculated over the full analytical windows specified above. Correlations reflect association, not causation. High collinearity among Venezuelan indicators (VIF > 5) precludes independent attribution. Temporal autocorrelation reduces effective degrees of freedom; p-values are approximate. Sources: World Bank WDI [16, 17], The Global Economy [15], CFR [18]

Analytical approach

The integration of joinpoint regression with correlation analysis and qualitative comparison serves complementary analytical purposes. Joinpoint regression identifies when statistically significant trend changes occurred; correlation analysis examines whether structural indicators co-vary with malaria incidence over the full study period; and the governance-resilience domain comparison (Table 4) provides contextual assessment of how institutional configurations differed between countries. No single method alone can address the study objective; their integration follows the logic of methodological triangulation appropriate to comparative ecological designs [9].

Additional analyses included: (i) temporal co-occurrence between governance/economic indicators and malaria trends; (ii) regional spillover assessment using imported-case data from Brazil, Colombia, Peru, and Guyana; and (iii) operationalization of governance-resilience domains using PAHO/WHO criteria for systematic comparison (Table 4).

Ethical considerations

This study utilized exclusively secondary data from publicly available sources and published literature. No primary data collection or human subjects research was conducted. Institutional review board approval was therefore not required.

Results

Divergent epidemiological trajectories

The epidemiological data reveal a progressive divergence between Ecuador and Venezuela over the study period (Table 3, Fig. 1). Ecuador demonstrated sustained malaria decline with an overall reduction of 99.3%, from 106,641 cases in 2001 to 757 cases in 2023. Venezuela maintained 30,000–50,000 annual cases from 2000–2014, followed by rapid increase: cases rose to 136,402 in 2015, escalating to a peak of 527,627 in 2018 (1,741.8 per 100,000), before declining to 154,284 in 2022 according to official reports (noting data quality limitations discussed below). Venezuela’s resurgence is anomalous within the broader regional context of declining malaria transmission across South America (Fig. 2).

Table 3.

Comparative epidemiological trends: Ecuador and Venezuela (2000–2023)

Year	Ecuador cases	Ecuador rate/100k	Venezuela cases	Venezuela rate/100k
2000	45,000	360	35,500	138
2001	106,641	840	36,000	140
2005	20,000	150	32,000	120
2010	2,900	20	48,000	160
2015	1,048	6.5	136,402	446.4
2017	1,846	11.2	411,586	1,341.4
2018	1,579	9.4	527,627	1,741.8
2019	1,801	10.5	467,421$_{}^{†}$	1,613.8$_{}^{†}$
2020	2,012	11.5	232,042$_{}^{†}$	815.6$_{}^{†}$
2022	1,598	8.8	154,284$_{}^{†}$	546.9$_{}^{†}$
2023	757	4.2	–	–

Data from WHO World Malaria Reports (2000–2023) and PAHO updates

$_{}^{†}$Venezuelan data after 2017 are subject to significant reporting limitations; interpretation requires caution (see Venezuelan data quality considerations and Limitations)

Fig. 1.

Malaria incidence trajectories with joinpoint segmentation: Ecuador and Venezuela (2000–2023). Panels show annual cases (bars) and incidence rate per 100,000 (line), with joinpoint-identified inflection points overlaid as dashed vertical lines. Ecuador’s decline (panel a) and Venezuela’s resurgence (panel b) share the same x-axis scale to facilitate comparison. a Ecuador (2000–2023). Three joinpoint phases indicated by dashed vertical lines at 2008 and 2014. Trajectory approaches elimination threshold (<10/100k) by 2015 and is maintained through 2023. Data: Ecuador Ministry of Health, PAHO, WHO. b Venezuela (2000–2023). Single joinpoint at 2014 coincides temporally with crude-oil price decline. Post-2018 decline in official figures occurs concurrent with documented surveillance system deterioration (see Limitations). Data: WHO, PAHO, peer-reviewed estimates

Fig. 2.

Regional malaria incidence trends in South America (2000–2023). Ecuador’s sustained decline (purple) and Venezuela’s resurgence (red) within the broader regional context. Data: WHO World Malaria Reports, PAHO epidemiological updates. Note: Additional South American countries are included to demonstrate that Venezuela’s resurgence is anomalous within the regional trend of declining transmission, thereby contextualizing the comparative analysis rather than expanding its scope

Joinpoint regression results

Table 1 presents the estimated parameters for each temporal segment identified by the joinpoint model (Eq. 1).

Table 1.

Joinpoint regression parameter estimates for malaria incidence, Ecuador and Venezuela (2000–2023)

Country	Segment	Joinpoint	$\widehat{\beta }$	APC (%)	95% CI	p-value	DW
Ecuador	2001–2008	–	$-0.418$	$-34.2$	($-42.1$, $-25.3$)	$<0.01$	1.89
	2008–2014	2008	$-0.133$	$-12.5$	($-19.8$, $-4.5$)	$<0.05$	1.76
	2014–2023	2014	$-0.012$	$-1.2$	($-6.8$, $+4.6$)	0.45 (ns)	2.01
Venezuela	2000–2014	–	$+0.027$	$+2.8$	($-0.8$, $+6.5$)	0.12 (ns)	1.58
	2014–2018	2014	$+0.613$	$+84.6$	($+62.3$, $+110.5$)	$<0.001$	1.72
	2018–2023$_{}^{†}$	2018	$-0.324$	$-27.6$	($-38.2$, $-15.4$)	$<0.05$	1.44

Models fitted using Joinpoint Regression Program v4.9.1.0 with grid search and BIC selection. DW Durbin-Watson statistic (values 1.5–2.5 suggest no significant first-order autocorrelation). Ecuador joinpoints: 2008, 2014 (both $p<0.05$). Venezuela joinpoint: 2014 ($p<0.001$)

$_{}^{†}$ Venezuela 2018–2023 segment estimates should be interpreted with caution due to documented surveillance system deterioration; the apparent decline may reflect reduced case ascertainment rather than true epidemiological improvement

APC Annual Percentage Change, CI 95% confidence interval, ns not significant at $\alpha =0.05$

Sensitivity analysis. When the Venezuelan series was truncated at 2017 (excluding the period of maximal surveillance deterioration), the 2014 joinpoint remained statistically significant ($p<0.001$), with APC for 2014–2017 estimated at $+78.2\%$ (95% CI $+58.9\%$, $+100.8\%$). This suggests the 2014 inflection point is robust to exclusion of later unreliable data.

Venezuela’s 2014 joinpoint coincides temporally with the global crude-oil price decline from >$100/barrel to <$50/barrel [18]. Ecuador’s 2014 joinpoint marks no acceleration; the near-elimination plateau was maintained.

Venezuelan data quality considerations

The following empirical observations are relevant to interpreting post-2014 Venezuelan data:

1. Surveillance infrastructure changes. In 2014, the Venezuelan government suspended publication of the Boletín Epidemiológico [13]. By 2018, an estimated 70% of clinical laboratories had reduced operations due to reagent shortages [19].

2. Population movements. Approximately eight million Venezuelans emigrated from 2014 onwards [18]. A substantial proportion emigrated from hyperendemic southern municipalities (Estado Bolívar). Colombian and Brazilian border municipalities reported 281% to 1,000% increases in malaria incidence during this period [6].

3. Healthcare access indicators. EVESCAM data [14] indicate that 77.3% of the Venezuelan population reported interruptions in access to basic public services during 2014–2020. Antimalarial medication shortages (chloroquine, primaquine) were widely reported.

These observations are relevant for interpreting the post-2018 decline in official Venezuelan figures.

Correlation analysis

Table 2 presents cross-domain Pearson correlations between structural indicators and annual malaria incidence for both countries.

For Venezuela, the crude-oil price series ($r=-0.89$) and political stability index ($r=-0.85$) show strong correlations with malaria incidence. These indicators are themselves highly correlated (oil-price vs. political stability: $r=0.91$) due to Venezuela’s structural dependence on oil revenues for >95% of export earnings [18]. Ecuador shows weaker correlations between oil prices and malaria incidence ($r=-0.29$, ns).

Comparative epidemiological trends

See Table 3.

Comparison of governance-resilience domains

Table 4 operationalizes the five governance-resilience domains using PAHO/WHO operational criteria [2, 22], enabling systematic comparison across both national experiences.

Concurrent infectious disease trends in Venezuela

To examine whether the Venezuelan malaria resurgence coincided with broader health-system changes, we examined concurrent infectious disease trends. Diphtheria, undetected in Venezuela for 24 years, re-emerged in July 2016: between 2016 and 2018, more than 1,600 confirmed cases and 280 deaths were recorded [19]. Measles returned in 2017 with >7,300 confirmed cases and 64 deaths by September 2018 [19]. Tuberculosis incidence increased from 27 to 48 per 100,000 between 2014 and 2018 [6]. Vaccination coverage fell to 52% for essential vaccines in some regions [19].

Structural indicators: systemic comparison

Table 5 summarizes the structural indicators associated with divergent malaria control outcomes acrossboth countries.

Table 5.

Structural indicators associated with divergent malaria control outcomes

Domain	Ecuador	Venezuela
Political Stability (PSI)	Relatively stable: $-0.34$ (2023). Continuity across 5 administrations [15]	Low: $-1.19$ (2023), among world’s lowest. Institutional changes documented [15]
Health Expenditure	7–8% GDP, constitutionally protected. Maintained throughout oil-price fluctuations	Declined from 3.1% (2014) to estimated <1.5% (2017). Fiscal constraints documented [17]
Surveillance System	SIVE-Alerta: real-time, >95% completeness, <48 h notification delay	Boletín suspended 2014; laboratory capacity reduced. Surveillance limitations documented [13]
International Cooperation	Ecuador–Peru PAMAFRO agreement; AMI/RAVREDA participation [23]	Reduced multilateral participation reported post-2015; bilateral agreements limited
Extractive Sector	Regulated formal mining; environmental assessment requirements	Arco Minero: >112,000 km$_{}^2$; >100,000 informal miners; habitat modification documented [4, 7]

Sources: World Bank WGI [15]; World Bank WDI [17]; OHCHR [13]; Crisis Group [7]

Discussion

Summary of main findings

This comparative ecological study examined malaria trends in Ecuador and Venezuela from 2000 to 2023. The main findings are: (1) Ecuador achieved a 99.3% reduction in malaria cases, with joinpoint regression identifying significant inflection points in 2008 and 2014 marking progressive program consolidation; (2) Venezuela experienced a 1,217% increase, with a significant joinpoint in 2014 marking a shift from stable endemic transmission to resurgence; (3) the Venezuelan 2014 joinpoint coincides temporally with the global crude-oil price collapse; (4) strong correlations were observed between Venezuelan malaria incidence and oil prices ($r=-0.89$), political stability ($r=-0.85$), and other macroeconomic indicators, though these indicators are highly collinear; (5) the malaria resurgence in Venezuela coincided with resurgence of diphtheria, measles, and tuberculosis; and (6) post-2018 declines in official Venezuelan figures occurred concurrent with documented surveillance deterioration and mass emigration, limiting interpretability of recent trends (Table 5).

Interpretation in context of prior literature

The temporal coincidence of Venezuela’s 2014 joinpoint with the oil-price collapse is consistent with findings from Grillet et al. [4], who documented the concentration of resurgence in mining areas of southern Venezuela, and with economic analyses linking Venezuela’s fiscal crisis to commodity dependence [18, 24]. Ecuador’s sustained decline aligns with Krisher et al.’s [8] documentation of the binational success with Peru.

However, several interpretive cautions are warranted. The strong collinearity between oil prices and political stability indicators in Venezuela ($r=0.91$) reflects the structural reality that these factors are endogenously related in petrostate economies [18]. The 2014 oil-price shock simultaneously affected fiscal capacity, governance stability, and institutional function; attributing the malaria surge primarily to either political instability or economic shock is not supported by these data, which cannot distinguish independent contributions.

The collinearity problem and implications for causal inference

This study’s correlation analysis revealed that, for Venezuela, crude-oil price ($r=-0.89$) shows nominally stronger correlation with malaria incidence than the political stability index ($r=-0.85$). However, these variables are so highly correlated with each other that variance inflation factors exceed 5, precluding multivariate decomposition. This collinearity is not a methodological limitation to be overcome but rather reflects the fundamental reality of petrostate dependency: political legitimacy, fiscal capacity, and institutional function are structurally intertwined with commodity revenues.

The implication for policy is that interventions targeting “political stability” in isolation may be neither tractable nor sufficient. More actionable approaches may involve fiscal architecture reforms (stabilization funds, constitutionally protected health budgets) that can be implemented even under imperfect political conditions. Ecuador’s constitutional protection of health expenditure represents such an approach and appears to have insulated health programs from commodity-price volatility.

Interpreting the post-2018 Venezuelan decline

The decline in officially reported Venezuelan cases after 2018 (from 527,627 to 154,284 in 2022) cannot be interpreted as successful public health control based on available evidence. Three concurrent factors are documented: (1) surveillance infrastructure deterioration, with suspension of the Boletín Epidemiológico and laboratory capacity reduction; (2) mass emigration of approximately eight million persons, including from hyperendemic areas, with corresponding increases in imported malaria cases in neighboring countries; and (3) reduced healthcare access, with over 77% of the population reporting service interruptions. Independent modeling estimates suggest true Venezuelan incidence at two to three times official figures [4]. These factors are consistent with underascertainment rather than epidemiological improvement.

Concurrent disease resurgence and systemic fragility

The simultaneous epidemic resurgence of malaria, measles, diphtheria, and tuberculosis in Venezuela — diseases with distinct transmission pathways, vectors, and biological characteristics — suggests systemic health-system fragility rather than disease-specific or vector-specific dynamics. This pattern is consistent with the fragility framework proposed by Harper et al. [11], in which health system breakdowns manifest across multiple disease domains simultaneously. By contrast, Ecuador maintained vaccination coverage above 80% for most antigens during the study period and did not experience comparable resurgences of diphtheria, measles, or tuberculosis, consistent with sustained health-system functionality [25].

Importantly, this concurrent pattern does not constitute definitive evidence of causation; alternative explanations including ecological changes, climatic factors, and intervention-specific issues cannot be excluded. However, the cross-disease pattern is difficult to explain by any single disease-specific mechanism.

Regional implications

Venezuela’s crisis appears to have regional implications through imported case patterns. Ecuador, despite its near-elimination status, experienced modest case increases from 1,801 (2019) to 2,012 (2020), potentially reflecting regional spillover effects. The Ecuador–Peru binational agreement provides a model for regional coordination: formalized cooperation, synchronized vector control, real-time information sharing, and joint procurement [8]. The Venezuelan situation may require a regional containment approach combining epidemiological surveillance, health services for migrant populations, and diplomatic engagement.

Ecuador’s approach: alignment with PAHO “Best Buys”

Ecuador’s success appears aligned with PAHO’s “Best Buys” framework for communicable disease elimination [22], which prioritizes cost-effective, adaptable, and intersectoral interventions: risk-stratified microplanning, the DTI-R (Diagnosis, Treatment, Investigation, Response) approach, and binational coordination. Ecuador institutionalized malaria control within constitutionally protected budget lines and implemented SIVE-Alerta with high reporting completeness. The Ecuador–Peru border agreement represents a replicable model for transnational elimination.

Study limitations

This analysis has several important limitations that should inform interpretation of the findings.

Ecological study design. This is an observational ecological study comparing national-level trends. Causal inference is limited; the findings demonstrate temporal association and correlation but cannot establish causation. The ecological fallacy precludes inference about individual-level relationships from aggregate data.

Venezuelan data quality. Data quality differs substantially between countries, particularly for Venezuela after 2014. Triangulation across multiple sources was employed, but underreporting likely remains significant. Quantitative analyses (joinpoint regression, correlations) for Venezuela post-2014 should be interpreted with considerable caution. Sensitivity analysis truncating the series at 2017 demonstrates robustness of the 2014 joinpoint, but later estimates are unreliable.

Collinearity. The strong collinearity between oil-price and political-stability series in Venezuela prevents multivariate decomposition of their independent contributions. This is not merely a statistical limitation but reflects structural petrostate dependency.

Temporal autocorrelation. Time series data exhibit serial correlation, reducing effective degrees of freedom and inflating nominal statistical significance. Durbin-Watson statistics are reported, and p-values should be interpreted conservatively.

Two-country design. The comparison of two countries limits generalizability. Future research should extend the framework to additional cases to assess external validity.

Unmeasured confounders. Important factors including vector ecology, climate variability, intervention coverage, and entomological dynamics were not directly measured. These factors may contribute to observed patterns independently of governance and economic indicators.

COVID-19 pandemic. The COVID-19 pandemic (2020–2021) affected both countries and constitutes a contextual limitation for recent data. Ecuador experienced modest case increases despite pandemic disruptions, suggesting program resilience. Venezuela’s pandemic-era figures are uninterpretable due to pre-existing surveillance limitations.

Conclusions and recommendations

The divergent malaria trajectories of Ecuador and Venezuela between 2000 and 2023 provide observational evidence suggesting that governance, health policies, and intersectoral coordination are associated with malaria control outcomes. Ecuador’s health system demonstrated capacity to maintain elimination-oriented programs despite global economic fluctuations, achieving a >99% reduction in disease burden. Venezuela’s resurgence — temporally associated with the 2014 oil-price collapse and accompanied by epidemics of diphtheria, measles, and tuberculosis — documents the co-occurrence of macroeconomic crisis, governance challenges, and infectious disease resurgence. The strong collinearity between economic and political indicators in Venezuela precludes definitive attribution to either factor independently.

Post-2018 declines in official Venezuelan figures are accompanied by documented surveillance deterioration and mass emigration, suggesting these declines may reflect reduced case ascertainment rather than epidemiological improvement.

To support sustainability of malaria elimination objectives towards 2030, and in alignment with PAHO’s regional malaria elimination framework [2, 22], the following strategic considerations are proposed:

1. Institutionalize surveillance as a protected intervention. Following the DTI-R approach [2], establishing legal frameworks protecting epidemiological alert systems may enhance program resilience. Ecuador’s SIVE-Alerta provides a potential model.

2. Consider fiscal architecture reforms. Countries with high commodity dependence may benefit from constitutional or statutory floors for health expenditure as a percentage of GDP, complemented by fiscal stabilization funds. Such architectural reforms may be more tractable than broader governance interventions.

3. Advance micro-planning and intersectoral coordination. Effective elimination of residual transmission foci may require shifting from broad national strategies to adaptive, community-level micro-planning [22], with coordination across environment, mining, and social development sectors.

4. Expand transnational health-surveillance platforms. Standardized binational and multinational surveillance networks, pooled diagnostic procurement, and joint rapid responses may mitigate cross-border transmission. The Ecuador–Peru PAMAFRO model provides a potential template.

5. Integrate governance and economic indicators into risk frameworks. Global health funders may benefit from systematic assessment of political stability, fiscal architecture, and commodity dependency when designing long-term programs in resource-dependent settings.

In summary, these findings suggest that technical interventions may be insufficient without institutional stability and predictable financing. Health-system resilience appears to be an important factor in durable infectious disease control, though definitive causal attribution requires additional research with designs capable of stronger causal inference.

Author summary

Malaria elimination programs remain vulnerable to political and economic instability. This comparative ecological study of Ecuador and Venezuela provides observational evidence suggesting that governance quality, health-system resilience, and macroeconomic stability are associated with malaria control outcomes. Ecuador reduced cases by 99.3% through sustained commitment, constitutional health rights, robust surveillance (SIVE-Alerta), and binational cooperation with Peru. Venezuela’s resurgence — a 1,217% increase temporally associated with the 2014 oil-price collapse and ensuing fiscal contraction — reversed decades of domestic progress and was accompanied by increased case exportation to neighboring countries. The concurrent resurgence of diphtheria, measles, and tuberculosis suggests systemic health-system fragility rather than a malaria-specific failure. These findings suggest that global health funders should consider integrating governance and macroeconomic indicators into risk-assessment frameworks and prioritizing institutional resilience alongside commodity distribution. However, the observational nature of this analysis and the collinearity among economic, political, and health indicators preclude definitive causal attribution.

Supplementary Information

12936_2026_5899_MOESM1_ESM.png ^{(1,004.4KB, png)}

Supplementary material 1. S1 Fig. Regional malaria control context in South America (2000–2023). Full regional analysis of all South American countries for contextual comparison.

12936_2026_5899_MOESM2_ESM.docx ^{(16.4KB, docx)}

Supplementary material 2. S2 Table. Regional spillover: imported malaria cases in neighbouring countries (2014–2023). Annual data on Venezuela-origin malaria cases in Brazil, Colombia, Peru, Ecuador, Guyana, and Trinidad and Tobago. S3 Table. Sensitivity analyses. Joinpoint regression results with truncated series (2000–2017) and exclusion of COVID-19 years (2020–2021).

12936_2026_5899_MOESM3_ESM.zip ^{(3.7KB, zip)}

Supplementary file 3. S1 Data. R code and compiled dataset. Joinpoint regression scripts and full data matrix used to generate Tables 2 and 3, available from the corresponding author upon reasonable request.

Author contributions

J.D.S.R. conceived and designed the study, curated the data, performed the statistical analyses (including the joinpoint regression and correlation analysis), interpreted the findings, and drafted the manuscript. M.J. contributed to data validation, literature review, and critical revision of the manuscript for important intellectual content. M.C.M.M. contributed to the conceptual framework development, the binary operationalization of governance-resilience domains, and critical revision of the manuscript. All authors reviewed and approved the final manuscript.

Funding

This work was supported by the Universidad Tecnológica Indoamérica research programme. The funding body had no role in study design, data collection, analysis, interpretation, or manuscript writing.

Data availability

No datasets were generated or analysed during the current study.

Declarations

Ethics approval and consent to participate

Not applicable. This study used only aggregated, de-identified secondary data from publicly available sources and did not involve human participants or animals.

Consent for publication

Not applicable. This manuscript contains no individual person’s data in any form.

Competing interests

The authors declare no competing interests.