pregnancies) and to encourage other secondary prevention efforts. In 1990, state health officials also rejected the CDC's 1989 recommendation for risk-based assessments—which attempted to concentrate efforts on encouraging testing among individuals in specific self-reported high-risk categories—in favor of a broader, universal approach that sought to gain the consent to test among all women at state-regulated clinics. In September 1990, the health department sent out a "Dear Colleague" letter to all physicians in the state, urging them to counsel any patient who had sex with more than one person in the last ten years, or who had ever used illicit drugs, to be tested for HIV. Furthermore, the state launched the OB Initiative in 1990, a postpartum program at 24 hospitals in high-seroprevalence areas to counsel and test women who indicated they had not been tested during their pregnancies.
The results of these voluntary programs, though, proved disappointing. Testing rates in 1991 and 1992 ranged from 14% to 66% (Healton et al., 1996) at the women's clinics, and except for Harlem Hospital's program, which persuaded over 90% of postpartum women to test, the OB Initiative was equally ineffective. Although the merits of case finding versus prevention were debated, it was increasingly evident to public health officials in the early 1990s that such voluntary case finding strategies needed strengthening.
At the same time that these programs directed at individual behavior change were being initiated, a number of efforts were undertaken by the state health department to more accurately focus community-wide prevention and treatment efforts. One innovation developed by the AIDS Institute and the state's epidemiology unit was a Community Needs Index, which took into account newborn seroprevalence rates and hospital discharge data in constructing a profile of high-, medium-, and low-risk neighborhoods. The index was then used in program development, the expansion of specially designated AIDS centers at hospitals and community health centers, and the distribution of state funds to community-based organizations in high-risk neighborhoods.
From an epidemiological perspective, universal newborn screening was still regarded as effective and relevant. Beginning in 1990, though, as treatment options for HIV-infected infants became more widely accepted, the tension between the epidemiological and the clinical utility of newborn screening (in very broad terms, the polarization of surveillance and prevention versus case finding and treatment) grew within the state health department. These issues had percolated within the larger health care community since 1988 (Krasinski et al., 1988), but now they were gaining a wider audience. What began as an internal debate within the health care community in the late 1980s evolved into a very public debate by 1993.
By 1990, Commissioner Axelrod was having second thoughts about the state's blinded seroprevalence study. The Fifth International AIDS Conference in