An interactive explainer

Rebounds in malaria transmission

A rebound is when malaria burden, after a period of effective control, climbs back not just toward but above its level before the intervention. Immunity fades while transmission is suppressed, so once that protection is removed clinical cases can, for a time, run higher than if there had been no intervention at all.1

The three drivers of a rebound

Rebounds are driven by three conditions acting together

Acquired immunity to clinical malaria is developed and maintained by repeated exposure. Preventative interventions, such as bed nets and indoor residual spraying, lower exposure and reduce cases. As the population is protected, population-level immunity falls: acquired immunity wanes, and newly born individuals develop immunity more slowly. Three drivers shape the dynamics of a rebound:

Select a corner for detail.

1 Baseline transmission potential

The intrinsic potential for transmission

Evidence

Acquired immunity to clinical malaria is built and maintained by exposure, so a population at higher baseline transmission holds more of it. That immunity is what a rebound puts at risk: the more a population has acquired, the more it can lose while control suppresses exposure, and the more clinical disease can return once control stops and exposure climbs back. Rebound risk is therefore greatest where pre-control immunity was high, broadly, in higher-transmission settings.1

Mechanism

In transmission models, the equilibrium level of acquired clinical immunity is a saturating function of exposure: it rises steeply at low transmission, then plateaus.3 A higher baseline therefore means more immunity has been built up, so there is more to lose when control suppresses exposure.

When control is later removed, exposure returns close to that high baseline, now acting on a population whose immunity has fallen. Both terms point the same way, so the rebound after withdrawal is largest where baseline transmission was highest.

2 Strength & duration of control

How effective the preventative interventions are, and how long they run for

Evidence

This driver concerns preventative interventions: bed nets and indoor residual spraying, which lower people's exposure to infectious bites through both direct personal protection and a community effect on the mosquito population. Because they reduce exposure, they also slow the acquisition and maintenance of immunity. Interventions that treat disease without reducing exposure, such as case management, do not erode immunity in the same way. The stronger and longer the reduction in exposure, the further population immunity falls.

Mechanism

Lower exposure reduces population immunity in two ways:

Two properties of the programme set how far immunity falls. A larger reduction in exposure lowers the level immunity settles toward, so it has further to fall. A longer programme matters because immunity changes over years: a short programme lowers it only slightly, while a sustained one lets immunity approach its suppressed level and adds successive birth cohorts with little acquired immunity. A strong, sustained programme therefore opens the largest gap between the lowered immunity and its original level.1

3 Speed of withdrawal & loss of protection

How quickly residual protection fades once a programme stops

Evidence

When a programme is stopped, its protection does not disappear at once: some residual protection remains and then declines. Whether a rebound follows depends on whether that residual protection is lost before immunity has recovered. If it fades slowly, immunity can rebuild beneath it and total protection stays near baseline; if it is lost quickly while immunity is still low, total protection falls below its starting level.

Mechanism

The operational decision to stop can be abrupt, but how fast the protection already in place is then lost depends on the intervention:

The faster this residual protection is lost, the less time immunity has to recover, and the larger the rebound. A slower loss of protection lets immunity rebuild and softens or prevents it.

Illustrative — not for decision making

Explore a rebound

A toy model of total protection over time

Total protection (top line) is acquired immunity (green) plus intervention protection (orange); the dashed line is the pre-intervention baseline. Set the three drivers to see whether total protection dips below baseline after control is withdrawn, producing a rebound.

Baseline transmission45
Control strength70%
Control duration6 yr
Speed of protection loss-
- -

Total protection over time

Protection from immunity Protection from intervention Rebound (dip below baseline) Total protection

While control is in place, the intervention more than offsets the fading immunity, so total protection stays above baseline. After control stops, its protection is lost faster than immunity recovers, so total protection falls below baseline. Because transmission has by then returned to its original level, this shortfall in protection is realised as clinical cases rising above their pre-intervention level: the rebound.

What to notice

Immunity declines during control. While exposure is suppressed, acquired immunity (green) declines through less boosting and slower acquisition in newborns (driver 2). The intervention (orange) more than offsets this at first, so total protection rises above baseline even as the immunity beneath it falls.

Rebound depends on competing rates. After control ends, intervention protection (orange) decays while immunity (green) recovers. If protection is lost faster than immunity is regained, total protection falls below baseline, a rebound. Slower loss of protection, or faster immunity recovery, reduces it.

Largest at high baseline transmission. Sweep the baseline slider: the dip deepens as transmission rises, for two reasons that compound. More immunity is acquired before control, so more is lost during it, and a higher force of infection returns when control stops, acting on the less-immune population.1

Non-linearities also shape the cases. This chart tracks total protection; the clinical cases it implies do not follow one-for-one. During the rebound period, the non-linear relationships between transmission (EIR), immunity and clinical disease also shape how cases respond. Clinical disease, for instance, is highest at moderate transmission rather than the very highest.1 These links are traced in the non-linearities explainer.

Resurgence after control is relaxed is well documented. Malaria burden has risen in several settings when vector control was scaled back or stopped, and fallen again when it resumed.2,7 A true rebound, with burden climbing above its pre-intervention level from lost immunity, is predicted by transmission models and would develop over years.1

Methods. A deliberately simple, illustrative toy model showing the shape of a rebound, not a specific setting; not to be used for decision making. Protection is shown in relative units on an arbitrary scale.

References.

  1. Ghani et al, 2009. Loss of population levels of immunity to malaria as a result of exposure-reducing interventions: consequences for interpretation of disease trends. PLoS ONE 4:e4383. doi.org/10.1371/journal.pone.0004383
  2. Cohen et al, 2012. Malaria resurgence: a systematic review and assessment of its causes. Malaria Journal 11:122. doi.org/10.1186/1475-2875-11-122
  3. Griffin et al, 2010. Reducing Plasmodium falciparum malaria transmission in Africa: a model-based evaluation of intervention strategies. PLoS Medicine 7:e1000324. doi.org/10.1371/journal.pmed.1000324
  4. Griffin et al, 2014. Estimates of the changing age-burden of Plasmodium falciparum malaria disease in sub-Saharan Africa. Nature Communications 5:3136. doi.org/10.1038/ncomms4136
  5. Sherrard-Smith et al, 2018. Systematic review of indoor residual spray efficacy and effectiveness against Plasmodium falciparum in Africa. Nature Communications 9:4982. doi.org/10.1038/s41467-018-07357-w
  6. Charles et al / mrc-ide. malariasimulation: an individual-based model of Plasmodium falciparum transmission (Griffin model). R package. github.com/mrc-ide/malariasimulation
  7. Namuganga et al, 2021. The impact of stopping and starting indoor residual spraying on malaria burden in Uganda. Nature Communications 12:2635. doi.org/10.1038/s41467-021-22896-5