Table 1.

Hypothetical Scenarios for Assessment of the Effect of Missing Cases on a Transmission Network Inferred From Mycobacterium tuberculosis Sequencing Data on Extensively Drug-Resistant Tuberculosis, KwaZulu-Natal, South Africa, 2011–2014

Scenario	Changes to Complete Transmission Network Model or Sampling Procedures
1. Cases missing at random	No changes to model terms.
2. Cases missing by transmission	No changes to model terms. Sample from complete network nonrandomly using degree to define sampling weights.
	I. Highly connected cases (“high transmitters”) more likely to be sampled: sampling weighted by degree
	II. Poorly connected cases (“low transmitters”) more likely to be sampled: sampling weighted by inverse degree
3. Cases missing by smear status	No changes to model terms. Increase proportion of smear-positive cases in complete network relative to sampled network.
4. Latent, unmeasured (super-spreading) factor	Add model term corresponding to unmeasured factor strongly related to transmission in a minority of cases. Vary strength and prevalence of factor.
	I. Unmeasured factor that increases transmission by a factor of 10 (prevalence: 10%)
	II. Unmeasured factor that increases transmission by a factor of 20 (prevalence: 10%)
	III. Unmeasured factor that increases transmission by a factor of 40 (prevalence: 10%)