There is rich evidence that disease status affects contact behavior in humans and animals, but this phenomenon has not previously been included in epidemiological models.  We explore the relationship between partnership dynamics and transmission of a sexually-transmitted disease (STD), and consider situations where disease status influences pairing behavior.  Starting from a model of partnership formation, we use a simple timescale approximation to obtain analytic expressions for the incidence rate, basic reproductive ratio (R₀), and endemic prevalence (i_¥) of an epidemic for four cases of increasing disease-induced behavioral change.  Such behavioral change can therefore be incorporated into STD models with minimal effort.

In addition to these new methods, two main insights arise from our analysis:

1.      Effects of disease-induced changes of behavior are illustrated by simulating STD epidemics.  Behavioral changes are found to have greater impact on epidemics of transient, highly-transmissible infections (e.g. gonorrhea, chlamydia) than on chronic, less-transmissible infections (e.g. HIV, HSV-2).

2.      When disease does not alter behavior, the pair-formation model reduces to the classical “frequency-dependent” transmission model when our timescale approximation holds.  Our derivation thus clarifies conditions under which frequency-dependence accurately represents the fundamental pair-based nature of STD transmission.  Exploring different scenarios, we conclude that caution should be exercised when using frequency-dependent transmission to model transient infections such as most bacterial STDs.