California has the second highest number of persons living with HIV/AIDS in the United States (U.S.) after New York [1]. By the end of 2008, there were 100,366 adults and adolescents reported to be living with HIV or AIDS in California. San Francisco and Los Angeles counties account for more than half of persons living with HIV/AIDS in California with 14,440 (14%) in San Francisco and 36,705 (37%) in Los Angeles County. As the state where AIDS was first described [2] and with continued high rates of disease, changes in the HIV epidemic in California over the last three decades have often preceded changes in the rest of the country. Therefore, understanding the demographic, transmission, and clinical trends among persons living with HIV/AIDS in California is important to the U.S. epidemic as a whole. However, the HIV epidemic in California varies by region. HIV-infected persons in Los Angeles County are more likely to be Latino than in San Francisco or other California counties. In San Francisco, men account for 92% of persons living with HIV/AIDS compared to 88% in Los Angeles County and 87% in the rest of California. There are proportionally more people exposed to HIV through heterosexual contact and non-men who have sex with men (MSM) injection drug use in Los Angeles County and the other California counties compared to San Francisco, where approximately 86% of people living with HIV are either MSM or MSM injection drug users [3-5].

The Centers for Disease Control and Prevention implemented the Medical Monitoring Project (MMP) in 2004 in response to an Institute of Medicine recommendation that representative data on persons in HIV care in the U.S. was necessary to adequately track and compare clinical data, monitor and target prevention among HIV-infected individuals, evaluate service needs, and compare data across geographic regions [6,7]. MMP was designed to be a national, multi-site, comprehensive, population-based supplemental HIV/AIDS surveillance project.

We compared MMP data from California by region with respect to demographic characteristics of patients in care for HIV/AIDS and described their medical care experiences, HIV treatment use, clinical care and outcomes, and need for ancillary and support services. MMP’s unmet need, clinical care and outcomes, and sociodemographic data can be used by community planning councils, policymakers, HIV care providers, and others to target care and prevention services and to inform the HIV prevention community planning process at the local, state and national levels.

We analyzed MMP interview data from 2007-2008 collected in California. The San Francisco Department of Public Health and the Los Angeles County Department of Public Health conducted MMP in their respective counties and the California State Office of AIDS conducted MMP in the remaining counties. The MMP sampling methodology has been previously described [8]. Briefly, facilities were selected for inclusion from each of the three California project areas using stratified probability proportional to size sampling to ensure that both smaller and larger volume providers would be included. Only facilities that provided HIV medical care to patients over the age of 18 were included in the sampling frame. HIV care facilities that provided only inpatient care, facilities that had closed or were outside the study area, correctional and work-release facilities, tribal facilities and facilities on military installations were excluded from the sampling frame. HIV-infected patients 18 years of age or older from these facilities were then randomly selected for participation.

Between 2007 and 2008, 899 people receiving medical care for HIV/AIDS were interviewed: 329 (37%) in San Francisco, 333 (37%) in Los Angeles and 237 (26%) in other California counties Patient interview response rate was 43% in San Francisco, 45% in Los Angeles, and 25% in other California counties. Patients sampled from 31, 22 and 42 outpatient medical facilities completed interviews in San Francisco, Los Angeles and in other California counties respectively. Overall, the project response rate was low (38% in San Francisco, 40% in Los Angeles, and 17% in other California counties).

Using the MMP standardized sampling methodology and data collection instrument, we were able to characterize and assess care received, clinical outcomes and the need for supportive services among people receiving care for HIV/AIDS in California and make regional comparisons. Overall, the majority of participating patients reported receiving standard of care CD4 and viral load monitoring, high rates of treatment use, undetectable HIV viral loads and CD4 counts indicative of good immune function and treatment efficacy.

These results are encouraging and demonstrate improvements in HIV care since the HIV Cost and Utilization Study (HCSUS), the first examination of HIV care in a nationally representative sample in 1997-1998. Shapiro et al. concluded that access to care, especially among vulnerable populations such as Blacks, Latinos, women, the uninsured and Medicaid-insured patients, was suboptimal [9]. For example, 41% of patients in HCSUS with CD4 cell counts below 500 cells/mm³ reported that they had not received antiretroviral therapy at the initial interview. At subsequent interviews, 15% of this population had not received ART. In contrast, 90% of all MMP patients, regardless of CD4 cell counts, reported current ART use which may indicate recent changes in access to and acceptance of ART among patients, revised recommendations regarding optimal time to initiate ART use or medical provider practices [10].

However, the need for supportive services among the California MMP sample was similar to HIV patients in care in other studies including HCSUS and the national MMP patients [7,11-13]. Sixty-seven percent of patients in HCSUS reported need for at least one supportive service and 40% of those who needed a service reported not receiving it (contingent unmet need) while 27% of the total sample reported an unmet need (prevalent unmet need). Similarly, among MMP patients, 70% reported need for any supportive service with 43% reporting contingent unmet need and 29% 

reporting prevalent unmet need. MMP patients in California indicated need for the same top four supportive services as MMP patients nationally. As a state with a significant burden of HIV disease, need for supportive services in California will likely continue and may be expected to grow as survival with HIV improves and the population living with HIV ages. Therefore, it will be important to monitor the ability of HIV care agencies and providers in California to provide these services as the economic challenges in the state and at the Federal level continue. MMP data can assist Ryan White-funded HIV/AIDS agencies to target supportive services to the areas of greatest need in California.

We did identify some areas where the characteristics of interviewed MMP patients differed from the characteristics of persons in the HIV/AIDS case registries, indicating that some populations may have been under or overrepresented in MMP. However, some of these differences in population demographics may reflect characteristics of patients in HIV care as compared to all persons reported with HIV/AIDS. The case registry includes all reported HIV/AIDS cases regardless of whether they are engaged in care; therefore, it is possible that the differences seen in the MMP participants reflect true differences in access to care and/or care-seeking behaviors by population. For example, African-Americans are underrepresented in HIV care as indicated by more delayed testing for HIV and less ART use which results in poorer survival rates compared to people of other race/ethnicities in California [3,14]. In addition, the greater proportion of persons reporting a mixed or other race could reflect the opportunity during a face-to-face interview to describe race/ethnicity in more detail than is routinely collected on case report forms where race/ethnicity is recorded by the provider or abstracted from a medical record.

We were also able to identify some significant regional differences in the need and unmet need for supportive services. Identification of regional differences has implications for delivery of HIV care services as supportive services need to be culturally specific to the populations served. For example, in San Francisco, most services need to be tailored to the needs of men who have sex with men whereas Los Angeles care providers may have a greater need to ensure readily available Spanish language services. In addition, the higher proportion of younger clients in Los Angeles relative to San Francisco and the rest of the state could have implications for care service planning. However, notwithstanding the younger care clientele in Los Angeles, 70-80 percent of clients from all California regions were over 40 years of age, indicating that the specific needs of older persons living with HIV/AIDS will need to be addressed.

The original site of HIV diagnosis may also explain some of the regional differences found. The majority of HIV/AIDS cases are diagnosed at private facilities in all regions of California. Traditionally, HIV partner notification services are offered by public health departments and at counseling, testing and referral sites at the time of initial HIV diagnosis. Private medical providers are less likely to offer these services. Among MMP respondents, the total proportion reporting being offered partner services (19%) was low as has been reported in other studies [15,16] perhaps because of where these patients were originally diagnosed or because patients were diagnosed before HIV partner services were offered in their county. Additionally, partner notification services have been standard procedure for STD prevention and control and therefore may be more routinely offered in the smaller California counties where HIV, STD and other communicable disease activities are more fully integrated.

There are several limitations associated with this project. Most importantly, while MMP was designed to produce a representative sample of patients receiving HIV care nationally and within each jurisdiction, recruitment of both facilities and patients has proven difficult in California. As a result of low facility and patient response rates, our data may not be representative or generalizable to all patients in HIV care in California. Facilities were most likely to refuse participation due to time constraints involving sampling and recruiting patients. Additionally, patient response rates were higher in San Francisco and Los Angeles than the rest of the state. Thus estimates of indicators representing the combined California data may over-represent the experience of patients in these jurisdictions relative to the underlying California-wide population of people receiving care. At the patient level, patients were often lost to follow up, did not see their medical provider during the recruitment period, failed to attend scheduled interviews or not well enough to participate. CDC is currently evaluating changes to the MMP sampling methodology that may improve response rates. For example, adjustments to facility sampling to prevent the same facilities from being sampled every year are being considered and, recently, the option for a telephone survey has been added to the MMP protocol to increase access to patients. Careful examination of barriers to participation will be needed to increase response rates to ensure representativeness.

In addition, we have not verified patients’ self reported clinical information and it is likely that some patients are not able to accurately recall clinical results or might confuse medical terminology. Further analyses that compare the interview data with the medical record abstraction data will be able to assess the extent of this limitation. Another possible limitation is that, by nature of the sampling process, patients in regular medical care are most likely to be included and reached for participation. Therefore, the results may not be generalizable to patients who do not access care consistently.

Despite these limitations, the information available from MMP interview data contributes to the overall picture of HIV care among patients living with HIV/AIDS in California, information that is unavailable through routine surveillance activities. While the MMP respondents might not be representative of all persons receiving HIV care in California, these data provide evidence that HIV care in all regions of California is generally good and that those patients who are linked to and retained in care are receiving standard of care. Additionally, identifying differences by region can facilitate the sharing of best practices among local health jurisdictions to improve patient care and satisfaction. For example, MMP data show care providers in Los Angeles were more likely to have discussed safe sex during the last 12 months than other California regions. Additionally, data show greater levels of unmet need for social services among patients in other California counties than in San Francisco or Los Angeles. The next step will be to incorporate the information gathered from the comprehensive MMP medical chart abstractions. This additional information will describe in greater depth the HIV care experiences among patients in California and also allow us to validate the self-reported information gathered from the interview.