The long-term impact and effectiveness of rotavirus vaccination in Malawi: an interrupted time-series and case-control analysis

This study demonstrates that the monovalent rotavirus vaccine (Rotarix) in Malawi provided moderate, age-dependent protection against hospitalisation for diarrhoea over seven years, with significant impact limited to infants and no measurable influence from the concurrent switch from trivalent to bivalent oral poliovirus vaccine.

The Gist

The Big Picture: A Shield That Fades

Imagine Malawi's children are a group of travelers walking through a storm of Rotavirus (a nasty germ that causes severe diarrhea). Before 2012, this storm was so strong it sent thousands of kids to the hospital every year.

In October 2012, Malawi introduced a vaccine (Rotarix®) as a shield to protect these travelers. This study is like a long-term weather report, looking back over seven years to see:

1. Did the shield work?
2. Did the shield wear out over time?
3. Did a change in another medicine (the Polio vaccine) mess with the Rotavirus shield?

The Main Findings

1. The Shield Works Best for the Little Ones

Think of the vaccine as a fresh, sturdy umbrella.

• For babies (under 1 year): The umbrella is brand new and strong. The study found that for babies, the vaccine reduced hospital visits by 37%. It's like having a superhero shield that keeps the rain off almost completely.
• For older toddlers (1 to 5 years): The umbrella starts to get a few holes in it. The protection drops significantly. By the time kids are over a year old, the vaccine isn't doing much to stop them from getting sick. The study showed that while the total number of sick kids went down, the type of kid getting sick changed.

2. The "Age Shift" Phenomenon

Here is a tricky part that needs a metaphor. Imagine a crowded room where everyone is trying to dodge a ball being thrown at them.

• Before the vaccine: The ball hit everyone, but mostly the small, slow kids (babies) because they couldn't dodge.
• After the vaccine: The babies put on their shields and dodge the ball easily. But the ball keeps flying! Now, the ball is hitting the older kids who didn't have shields yet.
• The Result: The study found that while babies got sick less often, the average age of the kids in the hospital went up. The disease didn't disappear; it just "shifted" to older children whose vaccine protection had faded away.

3. The Polio Vaccine Switch (The "Sidekick" Test)

In 2016, Malawi changed the type of Polio vaccine they gave to babies. They switched from a "Triple-Action" version (tOPV) to a "Double-Action" version (bOPV).

• The Fear: Scientists worried that the "Triple-Action" Polio vaccine might be fighting with the Rotavirus vaccine in the baby's tummy, making the Rotavirus shield weaker. They hoped the new "Double-Action" version would stop this fight.
• The Reality: The study checked this carefully and found no difference. Whether the baby got the old Polio vaccine or the new one, the Rotavirus shield worked exactly the same. The "sidekick" didn't change the outcome.

Why Does the Shield Fade?

The authors suggest a few reasons why the vaccine is less effective in Malawi compared to rich countries:

• The "Gut Garden": In low-income settings, babies' guts are full of many different bacteria (like a very crowded garden). This might make it hard for the live vaccine virus to take root and grow strong.
• The Environment: Even with a vaccine, if the water isn't clean or the toilets aren't safe (WASH issues), the virus is so strong that it can overwhelm the shield.
• Time: Like a sunhat that gets bleached by the sun, the protection from the vaccine simply wears off as the child gets older.

What Should We Do Next?

The study concludes that while the vaccine is a great success for babies, we can't just rely on it forever. To keep the "travelers" safe as they grow up, we might need:

1. A "Booster" Shot: Giving an extra dose of the vaccine later (maybe when they are 9 months or a year old) to patch the holes in the umbrella.
2. Better "Infrastructure": Improving clean water and sanitation so the storm itself isn't quite so violent.
3. New Strategies: Maybe giving the vaccine at birth (neonatal schedule) instead of waiting until 6 weeks, to get the shield on before the storm hits.

The Bottom Line

The Rotavirus vaccine in Malawi is a hero for babies, saving many from the hospital. However, its power fades as children get older, and changing the Polio vaccine didn't fix that. To protect children for the long haul, we need to combine better vaccines with better clean water and sanitation.

Technical Summary

1. Problem Statement

Rotavirus remains a leading cause of severe childhood diarrhoea and hospitalization globally, with a disproportionate burden in Low- and Middle-Income Countries (LMICs). While the monovalent rotavirus vaccine (Rotarix®) was introduced in Malawi in October 2012, there is limited evidence regarding its sustained long-term effectiveness and population-level impact more than a decade post-introduction in high-transmission settings.

Furthermore, a critical knowledge gap exists regarding the interaction between rotavirus vaccines and concurrent oral polio vaccines (OPV). Trivalent OPV (tOPV) contains a type 2 poliovirus component that has been hypothesized to interfere with the immunogenicity of rotavirus vaccines when administered concurrently. Malawi switched from tOPV to bivalent OPV (bOPV) in April 2016. It was unclear whether this switch improved Rotarix® performance or if the interference persisted.

2. Methodology

The study employed a dual-methodological approach using data from the Queen Elizabeth Central Hospital (QECH) in Blantyre, Malawi, covering the period from 1997 to 2019.

• Data Source: Surveillance data from children under five years hospitalized with acute gastroenteritis (AGE). Stool samples were tested for Rotavirus A antigen using Enzyme Immunoassay (EIA).
• Study Periods:
  • Pre-vaccine: July 1997 – December 2012.
  • Post-vaccine: January 2013 – December 2019 (after coverage exceeded 80%).
  • OPV Switch Analysis: Comparison of periods before and after the tOPV-to-bOPV transition (April 2016).

Analytical Designs:

1. Interrupted Time-Series (ITS) Analysis:

   • Objective: To estimate the population-level impact of vaccine introduction on rotavirus hospitalization rates.
   • Model: Generalized linear regression with negative binomial distribution to account for overdispersion.
   • Controls: Adjusted for seasonality (harmonics), secular trends, and test-negative diarrhoeal cases (to control for changes in healthcare-seeking behavior and surveillance intensity).
   • Counterfactual: Predicted hospitalization trends had the vaccine not been introduced.

2. Test-Negative Case-Control Study:

   • Objective: To estimate individual-level vaccine effectiveness (VE).
   • Cases: Children with AGE testing positive for rotavirus.
   • Controls: Children with AGE testing negative for rotavirus.
   • Exposure: Completion of the two-dose Rotarix® schedule (verified via health passports).
   • Interaction Analysis: A subset analysis (n=1,622) specifically assessed the interaction between Rotarix® and OPV formulation (tOPV vs. bOPV) using logistic regression, adjusting for age, sex, birth month/year, and nutritional status (MUAC).

3. Key Contributions

• Long-term Evaluation: Provides one of the longest post-introduction evaluations (7+ years) of Rotarix® in a high-burden, low-income setting.
• Dual-Method Approach: Combines population-level impact (ITS) with individual-level effectiveness (Test-Negative) using consistent age strata, offering a robust validation of findings.
• OPV Interference Assessment: Specifically investigates the impact of the national switch from tOPV to bOPV on rotavirus vaccine performance, a factor rarely evaluated in post-licensure studies.
• Age-Stratified Dynamics: Highlights the shifting epidemiology of rotavirus disease, specifically the decline in infant protection and the potential shift of burden to older children.

4. Key Results

A. Population-Level Impact (Interrupted Time-Series)

• Overall Impact: A 23% reduction (95% CI: 10–34%) in rotavirus hospitalization rates among children <5 years in the post-vaccine period compared to the pre-vaccine period.
• Age-Specific Impact:
  • Infants (<1 year): Significant reduction of 37% (95% CI: 25–47%).
  • Older Children (≥1 year): No protective effect; instead, a non-significant 39% increase in hospitalizations was observed (95% CI: -70 to 24%), suggesting a shift in disease burden to older age groups.
• Temporal Trend: Vaccine impact remained relatively consistent over the 7-year period, with the strongest effects observed in the second year post-introduction.

B. Individual-Level Effectiveness (Test-Negative Case-Control)

• Overall VE: 52% (95% CI: 18–71%) for two doses in children <5 years.
• Age-Specific VE:
  • Infants (<1 year): 67% (95% CI: 36–82%).
  • Older Children (>1 year): 29% (95% CI: -136 to 74%), which was not statistically significant.
• Conclusion: Protection is robust in infancy but wanes significantly or becomes negligible in children older than one year.

C. OPV Formulation Switch (tOPV vs. bOPV)

• Interaction: No measurable interaction was found between the OPV formulation and Rotarix® effectiveness.
• Statistical Finding: The adjusted Odds Ratio (aOR) for bOPV vs. tOPV was 1.07 (95% CI: 0.85–1.34), indicating that switching to bOPV did not significantly alter Rotarix® performance.
• Nutritional Status: Children with Severe Acute Malnutrition (SAM) had significantly lower odds of RVGE hospitalization compared to well-nourished peers (aOR 0.60), though this finding requires cautious interpretation.

5. Significance and Implications

• Sustained Infant Protection: Rotarix® provides moderate to strong protection for Malawian infants, validating the current immunization schedule for the first year of life.
• Waning Immunity & Epidemiological Shift: The study confirms that protection declines rapidly after infancy. The observed increase in hospitalizations among children >1 year suggests a "delayed primary infection" phenomenon, where unvaccinated or waning-immunity children are infected later in life.
• OPV Interference: The lack of improvement in VE following the switch from tOPV to bOPV suggests that OPV formulation alone is not the primary driver of reduced rotavirus vaccine performance in LMICs. Other factors, such as gut microbiota diversity or environmental enteric dysfunction, likely play larger roles.
• Policy Recommendations:
  • Booster Doses: The data supports the consideration of a booster dose (e.g., at 14 weeks or 9 months) to sustain protection beyond infancy.
  • Neonatal Vaccination: Alternative schedules, such as neonatal vaccination (e.g., RV3-BB), may offer better durability and cost-effectiveness in high-burden settings.
  • Holistic Approach: Vaccine strategies must be paired with strengthened Water, Sanitation, and Hygiene (WASH) interventions to reduce the overall disease burden and mitigate the impact of waning immunity.

Limitations: The study relied on hospital data from an urban center (Blantyre), which may have better WASH conditions than rural areas, potentially inflating impact estimates. Additionally, the wide confidence intervals for older age groups reflect limited statistical power due to lower case numbers in that stratum.