Molecular surveillance of Falciparum malaria in Rwanda: Shifts in parasite prevalence and risk factors between the 2014/15 and 2019/20 Rwanda Demographics and Health Surveys

By applying ultra-sensitive molecular surveillance to dried blood spots from the 2019/20 Rwanda Demographics and Health Survey, researchers found that despite rising artemisinin resistance markers, national *Plasmodium falciparum* prevalence significantly decreased by 53% compared to 2014/15, with infection risk strongly linked to male sex, lower socioeconomic status, and lower elevation.

The Gist

Imagine Rwanda as a vast, green garden. For years, the gardeners (health officials) have been fighting a persistent weed called Malaria. They've done a great job: they've pulled up many weeds, sprayed the garden, and put up fences (mosquito nets) to keep the pests out.

However, there's a problem. The tools the gardeners usually use to check for weeds are like naked eyes. They can only see the big, obvious weeds that are already causing trouble. But what if there are tiny, invisible seedlings hiding underground that the naked eye can't see? These hidden seedlings could grow back and ruin the garden later.

This paper is about a team of scientists who decided to use a super-microscope (a molecular test called qPCR) to look at the soil of the Rwandan garden. They didn't just look at the big weeds; they looked for the tiny, invisible seeds.

Here is the story of what they found, broken down simply:

1. The "Super-Microscope" vs. The "Naked Eye"

In 2019 and 2020, the government did a big health survey (like a census). They usually tested people for malaria using standard tools (rapid tests and microscopes). These tools said, "Great news! Only 1% of people have malaria."

But the scientists in this paper took leftover blood samples from that survey and ran them through their super-microscope.

• The Result: The super-microscope found that 7.7% of people actually had malaria.
• The Analogy: It's like checking a swimming pool for dirt. The naked eye sees clear water. The super-microscope reveals that the water is actually full of tiny, invisible specks of dust. Most of these people didn't even feel sick; they were carrying "invisible" infections.

2. The Good News: The Garden is Getting Cleaner

The scientists compared their new "super-microscope" results from 2019–2020 with similar results from 2014–2015.

• Then (2014-15): The hidden infection rate was 16.3%.
• Now (2019-20): The hidden infection rate dropped to 7.7%.
• The Takeaway: Even though the "invisible" weeds are still there, the gardeners' hard work paid off. They cut the malaria population in half! This proves that the fences (bed nets) and the sprays (treatments) are working.

3. Who is Still Carrying the Seeds?

The scientists looked at who was most likely to have these hidden infections. They found a few patterns:

• Men vs. Women: Men were more likely to have the infection than women.
• The "Rich" vs. The "Poor": People with less money and less education were more likely to have malaria. Think of it like living in a house with a leaky roof; it's harder to keep the rain (mosquitoes) out.
• The Location: People living in lower, warmer valleys were at higher risk than those living on high, cool mountains. Mosquitoes love the warm, low places.
• Age: Young adults (15–24) were more likely to be infected than older adults.

4. The "Invisible" Threat

The paper highlights a tricky situation. Rwanda is fighting a new enemy: drug-resistant malaria. It's like the weeds are learning to survive the weed-killer. Even though the total number of malaria cases is going down, the scientists are worried that the remaining "invisible" weeds might be the ones that are toughest to kill.

5. Why This Matters

The main lesson of this paper is: Don't stop looking just because you can't see the problem.

If the government only relied on their "naked eye" tests, they would think the malaria problem is almost solved (1% vs 7.7%). But because they used the "super-microscope," they know there is still a lot of work to do.

The Bottom Line:
Rwanda has done an amazing job reducing malaria, but the battle isn't over. By using high-tech tools to find the "invisible" infections, health officials can now target their efforts exactly where they are needed—like shining a flashlight in the dark corners of the garden to pull out the last few stubborn weeds before they grow back.

Technical Summary

1. Problem Statement

Rwanda is a malaria-endemic country facing a complex epidemiological landscape. While national malaria case numbers have declined significantly since 2018, the country is simultaneously grappling with emerging threats, including climate change impacts, insecticide resistance, and the emergence of Plasmodium falciparum partial resistance to artemisinin (mediated by PfKelch13 mutations).

A critical gap exists in current surveillance: standard Demographic and Health Surveys (DHS) in Rwanda rely on Rapid Diagnostic Tests (RDT) and microscopy, which are limited to women (15–49 years) and children (<5 years) and often fail to detect low-density, asymptomatic infections. Previous research indicated that molecular methods could reveal a much higher burden of infection than standard diagnostics. This study aims to update national prevalence estimates using ultra-sensitive molecular surveillance on a representative adult population and compare these findings with a parallel study from 2014–15 to assess temporal trends and risk factors.

2. Methodology

Study Design and Sampling:

• Data Source: Residual dried blood spots (DBS) originally collected for HIV testing during the 2019–20 Rwanda Demographic and Health Survey (RDHS).
• Population: The study analyzed 7,127 adult DBS samples (representing a weighted national population of ~7,122 individuals). This included both males and females, addressing the demographic gap in standard malaria testing.
• Sampling Strategy: A down-sampling strategy was employed to ensure national representativeness while oversampling high-prevalence clusters.
  • 50% of residual DBS from 500 GPS-located clusters were randomly selected.
  • Additional samples were added from 16 clusters identified as "high prevalence" (≥15% by standard RDT/microscopy) to ensure sufficient power for resistance marker sequencing in future studies.
• Comparison: The study design, laboratory protocols, and statistical methods were identical to a parallel study of the 2014–15 RDHS, allowing for direct temporal comparison.

Laboratory Techniques:

• Extraction: DNA was extracted from three 6mm DBS punches using Chelex.
• Detection: Species-specific quantitative real-time PCR (qPCR) targeting the P. falciparum 18S ribosomal RNA gene.
• Sensitivity: Assays were run for 45 cycles to detect ultra-low-density infections.
• Quantification: Parasitemia was estimated based on a standard curve using a diagnostic plasmid (6 copies of the 18S rRNA gene per parasite).
• Thresholds: Primary analysis used a Cycle Threshold (CT) of 45. A secondary conservative analysis used a CT of ≤40 (approx. ≥1 parasite/µL).

Statistical Analysis:

• Weighting: Complex survey weights were applied, combining HIV sampling weights, inverse propensity of selection weights, and transmission intensity weights to correct for selection bias and ensure national representativeness.
• Variables: Analyzed covariates included individual demographics (sex, age, wealth, education), household factors (bed net ownership, water source), and cluster-level ecological factors (elevation, temperature, precipitation, urban/rural status).
• Software: Analysis was performed in R 4.4.1.

3. Key Contributions

• First Molecular Surveillance in 2019–20 RDHS: Provides the first ultra-sensitive molecular prevalence estimates for P. falciparum in Rwanda using the 2019–20 DHS dataset.
• Inclusion of Adult Males: Expands the surveillance scope beyond women and children to include adult males, a demographic often excluded from standard malaria surveys but critical for understanding reservoirs.
• Direct Temporal Comparison: Offers a head-to-head comparison with the 2014–15 molecular study, isolating the impact of national control interventions over a 5-year period.
• Validation of Molecular Surveillance: Demonstrates the significant discrepancy between standard diagnostics (RDT/microscopy) and molecular methods in detecting the true burden of asymptomatic malaria.

4. Key Results

Prevalence Estimates:

• National Prevalence: The weighted national P. falciparum prevalence was 7.7% (95% CI: 6.8%–8.7%) using a CT cutoff of 45.
  • Using the conservative CT ≤40 cutoff, prevalence was 7.3% (95% CI: 6.4%–8.2%).
• Comparison to Standard Diagnostics: Standard RDHS diagnostics (RDT/microscopy) reported a national prevalence of <1%. In the female 15–49 age group, RDT/microscopy detected 0.5–1.2%, whereas qPCR detected 6.1%.
• Temporal Trend: National prevalence decreased by 53% compared to the 2014–15 RDHS (which was 16.3%).
• Geographic Distribution:
  • Prevalence varied by district, ranging from 0.0% in Musanze to 17.4% in Ruhango.
  • The Eastern and Southern provinces had the highest burden.
  • P. falciparum was detected in 49% of all clusters (244/500).

Parasitemia Characteristics:

• Infections were predominantly low-density. The median parasitemia was 7.3 parasites/µL.
• 77% of positive samples had <100 parasites/µL.

Risk Factors (2019–20):

• Significant Predictors of Infection: Male sex, lower household wealth, lower educational attainment, and residence at lower elevations.
• Protective Factors: Female sex, secondary/higher education, bed net ownership, and residence above 1500m elevation.
• Age: Individuals aged 15–24 were significantly more likely to be infected than those >24 years.
• Comparison to 2014–15: While risk factors remained similar (sex, wealth, education, elevation), the magnitude of association for many factors (e.g., bed net ownership, temperature) was reduced in 2019–20, suggesting that control interventions have mitigated the impact of these variables.

5. Significance and Implications

• Evidence of Control Success: The 53% reduction in molecular prevalence between 2014–15 and 2019–20 indicates that Rwanda's intensified malaria control efforts (mass ITN distribution, IRS, free treatment) have been highly effective, even in the face of emerging drug resistance and climate challenges.
• Hidden Reservoir: Despite the decline, a substantial reservoir of submicroscopic infection (7.7% vs. <1% by RDT) persists. This hidden reservoir, particularly in adults and low-density infections, poses a risk for transmission and resurgence.
• Surveillance Policy: The study strongly advocates for the integration of molecular diagnostics (qPCR) into national health surveys and surveillance systems. Relying solely on RDT/microscopy significantly underestimates the malaria burden and may lead to premature relaxation of control measures.
• Future Directions: The findings support the need for continued targeted interventions in high-burden districts (Eastern and Southern provinces) and the inclusion of adult males in surveillance. The dataset is now being used to screen for PfKelch13 and other antimalarial resistance markers.

Conclusion:
The study confirms that while Rwanda has made significant progress in reducing malaria burden, a hidden reservoir of low-density P. falciparum infections remains. Ultra-sensitive molecular surveillance is essential for accurately monitoring progress and guiding elimination strategies in the post-elimination phase.