Introduction
In 2002 the Nursing Council of Kenya, in its vision, proposed that by 2006, enrolled nurses ought to have been upgraded (through in-service training) to diploma-level status. Although this proposal was not fully implemented, nurses in the field thought that this was a requirement for them to remain in employment, hence the rush, which has affected the performance of counseling and testing services for HIV/AIDS.

Materials and methods
Workload data on counseling and testing for HIV that was collected in three consecutive months (June, July, and August 2006) was compared with a similar data set that was collected in three subsequent months (September, October, and November 2006). The data were collected from 14 pilot health facilities for the Project on Integrating Counseling and Testing for HIV Within Family Planning Services (two hospitals, eight health centers, and four dispensaries). Clients attending family planning (FP) services are referred to voluntary counseling and testing (VCT) sites or clinics for those interested or willing to be counseled or tested.

Results and discussion
Comparison of data between June and August 2006 and September–November 2006 showed a sudden decline of utilization data for counseling and testing services from the month of September (nurses undergoing in-service training reported for training on 6 September 2006). Data for the three months (June–August 2006) that preceded the opening of KMTC in September show that a total of 9,089 both new and old FP clients were attended to in the 14 pilot health facilities. Out of 9,089 FP clients, 2,809 of them (accounting for 30.9 percent) were tested. On the other hand, 9,840 FP clients were attended to in the 14 pilot health facilities in the three-month period (September–November 2006) after the opening of KMTC out of which a total of 620 clients (accounting for 6.3 percent of the total FP clients) were tested. During the same period, the number of clients who were referred from FP clinics for counseling and testing were 4,719 for the June–August 2006 period and 3,040 for the September–November 2006 period. The proportion of the clients who were tested out of those referred were 59.5 percent (2,809/4,719*100) for the June–August 2006 period when compared to 20.4 percent (620/3,040*100) for the September–November 2006 period.

Conclusion and policy implication
The difference in the two sample periods (June–August, where 30.9 percent of FP clients were tested, and September–November 2006, where only 6.3 percent of clients were tested) is highly significant (p=0.000). Also the difference between the proportion of the clients who were tested out of those referred (59.5 percent) for the June–August 2006 period is quite significant when compared to the 20.4 percent of clients who were tested (p=0.000) out of those who were referred during the September–November 2006 period. There is need for policy-level intervention to arrest the situation.