Reply to Ng and Tan.

TO THE EDITOR—We thank Drs Ng and Tan for their interesting data from Singapore and for their feedback on our review [1, 2]. The high rates of accuracy and acceptability reported from Singapore are encouraging. We agree that pricing and willingness to pay are essential considerations that are likely to vary substantially from one setting to the next. We also agree that it will be unfortunate if high pricing reduces the end-user uptake of this very promising new option for homebased human immunodeficiency virus (HIV) testing. Although limited willingness-to-pay data are available [3], data on actual consumer uptake in different settings are lacking. In addition, the task shifting associated with HIV self-testing is not insignificant, and we have yet to produce data on the cost or cost-effectiveness of self-testing programs to the public health system. Ultimately, we see the development of >1 market for self-testing. At one endwill bemiddle-class consumerswilling and able to afford the higher cost of a high-specification product. At the other end are consumers living on >2 US dollars a day, for whom even the smallest costs can be a major deterrent. Even in a single country, such as Singapore, consumers coming from disadvantaged groups may need to be targeted with more proactive approaches to maximize the utility of self-testing, for instance, subsidized or free product. This model of parallel markets is similar to that currently used for the marketing and distribution of condoms. Posttest support may also need to be delivered quite differently based on setting, with the Internet, smartphone applications, telephone hotlines, and faceto-face services among the possible support options. We anticipate a growing market, with different test kits being developed to serve self-testing and different approaches to supporting self-testing programs developed and evaluated. It is also likely that the cost per kit will reduce as volume increases and different products are developed. We agree that although cost must be factored into the calculations, cost implications should not deter us from implementing self-testing programs, but rather must be a consideration as we move forward with the scale-up of HIV self-testing.