Objective: This study compared the performance of the immunochromatographic strip (ICS) to the Venereal Disease Research Laboratory (VDRL) test and Treponema pallidum haemagglutination assay (TPHA) at a primary health care setting. Methods: The study group was comprised of 150 females randomly drawn from a population of pregnant women attending their first antenatal visit or follow-up visits at West Maternity Hospital in Eldoret Kenya, but without a previous syphilis test during that pregnancy. On-site VDRL, ICS and TPHA tests were performed and immediate treatment provided where appropriate. The performance of the three tests was compared. Results: The sero-prevalence of syphilis as determined by the VDRL test was 3%. There was no significant difference between the ICS and the VDRL test (P > 0.05). The sensitivity and specificity of the ICS test were 80% and 98.6% respectively, while the negative predictive value (NPV) and positive predictive value (PPV) were both 100%. On the other hand, the sensitivity and specificity of the VDRL test were 66.7% and 99.3%, while the NPV and PPV were 80% and 98.6% respectively. The Treponema pallidum haemagglutination assay was used as a reference test and had sensitivity, specificity, NPV and PPV of 100%. Conclusion: The diagnostic accuracy of the ICS compared favorably with theVDRL gold standard. The use of the ICS in Kenya can improve the diagnosis of syphilis in health facilities both with and without laboratories and allow community health care workers to make a rapid diagnosis of the disease, and consequently make immediate therapeutic decisions.