Spatiotemporal and demographic effects on avian malaria prevalence in blue tits

Based on 15 years of breeding data from southern Sweden, this study reveals that avian malaria prevalence in blue tits is significantly shaped by host age, sex, and local habitat, with infection rates increasing over time and accumulating in older birds, thereby underscoring the critical value of long-term research for understanding host-parasite dynamics.

The Gist

Imagine a tiny, invisible neighborhood of parasites living inside blue birds called Blue Tits. These birds are like the residents of a small town, and the parasites are like uninvited guests who move in and sometimes cause trouble.

This paper is a 26-year-long detective story about these guests. The researchers in Sweden watched over 2,500 Blue Tits to answer a simple question: Who gets the most uninvited guests, and why?

Here is the story of their findings, broken down into simple concepts:

1. The Three Types of "Guests"

The parasites aren't all the same. Think of them as three different types of troublemakers, each arriving via a different delivery service:

• Haemoproteus: Delivered by biting midges (tiny, mosquito-like flies).
• Plasmodium: Delivered by mosquitoes.
• Leucocytozoon: Delivered by black flies.

Sometimes, a bird gets all three at once (a "triple infection"), which is like having three different gangs moving into your house at the same time.

2. The Age Factor: "The Long Game"

The researchers found that older birds are much more likely to have these parasites than young birds.

• The Analogy: Imagine a library. A young bird has only been to the library for a few weeks, so it has only borrowed a few books. An older bird has been visiting for years; it has a massive collection of books on its shelves.
• The Finding: The birds aren't necessarily getting sick more often as they age; they are just accumulating these infections over time, like collecting stamps. The older they get, the more "stamps" (parasites) they have in their collection. Interestingly, the gap between young and old birds is getting wider every year, meaning the older birds are collecting stamps faster than before.

3. The Gender Factor: "The Hormone Effect"

When it came to the black fly parasite (Leucocytozoon), male birds were much more likely to be infected than females.

• The Analogy: Think of male birds as the "loud party hosts" of the town. During breeding season, males have high levels of testosterone to defend their territory and attract mates. This hormone boost is like turning on a neon sign that says, "Come party here!"
• The Finding: The black flies seem to love the "party atmosphere" of the males. However, for the other two parasites, there was no difference between boys and girls. Why? Because those other parasites often infect the birds when they are just babies (nestlings), before they have developed their "party hormones."

4. The Location Factor: "The Neighborhood Effect"

The study looked at three different forests (Åsen, Öved, and Revinge). Even though they were close neighbors (less than 5km apart), the infection rates were very different.

• The Analogy: Imagine three houses on the same street. One house is right next to a swamp (Revinge), one is on a dry hill (Öved), and one is in between.
• The Finding: The house next to the swamp (Revinge) had the most mosquitoes and midges, so the birds there had the most parasites. The house on the dry hill had fewer pests. This shows that where you build your nest matters a lot, even if you are just a few miles away from your neighbor.

5. The Big Trend: "The Rising Tide"

The most surprising discovery was that everyone is getting more infected over time.

• The Analogy: Imagine the tide in the ocean is slowly rising. Even if you are on a high cliff (an older bird) or a low beach (a younger bird), the water is getting higher for everyone.
• The Finding: Over the last 26 years, the number of infected birds has gone up significantly across all three forests. The researchers suspect this is due to climate change. Warmer weather is likely helping the mosquitoes and flies survive longer and breed more, making the "uninvited guests" more common every year.

The Bottom Line

This study teaches us that to understand how diseases spread in nature, you can't just look at a snapshot in time. You need to watch the movie over decades.

• Old birds have more parasites because they've been around longer.
• Male birds get more of one specific type because of their hormones.
• Wet neighborhoods have more parasites than dry ones.
• The whole town is getting more infected as the climate warms up.

It's a reminder that nature is a complex web, and small changes in the weather or the age of a bird can ripple out to change the entire ecosystem.

Technical Summary

1. Problem Statement

While the ubiquity of parasites is well-established, the specific mechanisms by which host demographic factors (age, sex) and spatiotemporal variables shape parasite prevalence patterns require further investigation. Specifically, there is a knowledge gap regarding:

• How host age and sex influence the odds of infection with different avian malaria genera (Haemoproteus, Plasmodium, Leucocytozoon).
• The dynamics of coinfections (specifically triple-genus infections) and whether hosts accumulate these over time.
• Whether observed temporal increases in malaria prevalence (previously documented at a single site) are generalizable across neighboring habitats at a small spatial scale (<5 km).
• How these demographic patterns interact with long-term temporal trends.

2. Methodology

Study System and Data Collection:

• Host: Blue tits (Cyanistes caeruleus).
• Location: Three field sites near the Stensoffa Research Station in Southern Sweden (Åsen, Öved, and Revinge), separated by at least 5 km.
• Timeframe: A 26-year period spanning 15 breeding seasons (1996–2000, 2007–2011, 2017–2021).
• Sample Size: 2,498 sampling events from 607 nest boxes.
• Demographics: Birds were categorized by age (first-year breeders vs. older birds) and sex. Recapture status was also recorded.

Parasite Screening:

• Molecular Detection: DNA extractions were followed by multiplex PCR to screen for the three malaria genera (Haemoproteus, Plasmodium, Leucocytozoon). Samples were screened in duplicate to account for low-intensity infections.
• Genotyping: Sanger sequencing of the cytb gene was performed on a subset (n=772) to identify genetic lineages.
• Outcome Variables: Infection status for each genus individually and a binary variable for triple-genus coinfection.

Statistical Analysis:

• Modeling: Generalized Linear Mixed Models (GLMMs) with a binomial distribution and logit link function were fitted using the lme4 package in R.
• Predictors: Host age, sex, mean-centered sampling year, and their interactions. Site was included as a fixed effect (with Revinge as the reference). Nest box was included as a random effect to control for shared environmental conditions.
• Model Selection: A multi-model comparison approach using the MuMIn package. Models with $\Delta$AIC $\le$ 2 were retained, and model averaging was applied to estimate predictor effects.
• Diagnostics: Residuals were checked using the DHARMa package for uniformity, dispersion, and zero inflation.

3. Key Contributions

• Long-term Temporal Scope: The study utilizes a rare 26-year dataset to analyze temporal trends in avian malaria, moving beyond short-term snapshots.
• Multi-Parasite Focus: It simultaneously analyzes three distinct malaria genera with different vectors (biting midges, mosquitoes, and black flies) and their complex coinfection dynamics.
• Interaction Effects: It specifically investigates how demographic factors (age/sex) interact with time, revealing that age-related infection patterns are not static but evolve over decades.
• Spatial Generalization: It tests the generalizability of previously observed temporal increases across multiple sites with varying micro-habitats.

4. Key Results

Demographic Effects:

• Age: Older birds had significantly higher odds of infection with Haemoproteus and Plasmodium and were more likely to harbor triple infections compared to first-year breeders. This suggests an accumulation of infections over time.
  • Interaction: The gap in Haemoproteus prevalence between older and younger birds widened slightly over time (significant Age $\times$ Year interaction).
  • Exception: No significant age effect was found for Leucocytozoon, possibly due to hosts clearing these infections or a narrow detection window during the breeding season.
• Sex: Males had significantly higher odds of Leucocytozoon infection than females, potentially linked to elevated testosterone levels during the breeding season. No significant sex bias was found for Haemoproteus or Plasmodium across all sites (though a site-specific effect was noted for Plasmodium at Revinge).

Spatiotemporal Effects:

• Temporal Trend: Across all three sites, the prevalence of all three parasite genera and triple infections increased significantly over the 26-year study period.
• Spatial Variation:
  • Revinge (lowest elevation, near water) had the highest prevalence of Haemoproteus and Plasmodium.
  • Åsen and Öved had significantly lower prevalence for these two genera, likely due to differences in habitat (e.g., elevation, vegetation, water availability) affecting vector abundance.
  • Leucocytozoon showed no significant site variation, likely because its vector (black flies) breeds in flowing water present at all sites.

Coinfection Dynamics:

• Triple infections were more common in older birds, supporting the hypothesis that hosts accumulate multiple parasite lineages over their lifespan.

5. Significance and Implications

• Climate Change Indicator: The consistent increase in prevalence across all sites, despite no major human-mediated landscape changes, strongly suggests that climate warming is driving increased transmission rates, corroborating findings from previous single-site studies.
• Importance of Long-term Data: The study highlights that short-term studies would fail to detect the evolving nature of age-related infection patterns (e.g., the widening gap between young and old birds for Haemoproteus).
• Vector Ecology: The results underscore the critical role of local habitat features (standing water vs. flowing water) in determining which parasite genera dominate a specific site.
• Host-Parasite Coevolution: The accumulation of infections in older birds suggests that while birds may develop resistance to infection intensity (as seen in previous blue tit studies), they do not necessarily clear infections, leading to a higher prevalence of chronic, low-level infections in the older population.

In conclusion, this paper demonstrates that avian malaria prevalence is a dynamic system shaped by the interplay of host demographics, local vector habitats, and long-term climatic trends, necessitating multi-decadal, multi-site monitoring to fully understand host-parasite epidemiology.