Correlates of Intimate Partner Violence, Including Psychological Partner Violence, in a Multisite U.S. Cohort of People in HIV Care.

We examined past-year intimate partner violence (IPV), including psychological violence without physical/sexual violence, and health outcomes among people with HIV (PWH) in care in a multi-site U.S. cohort. Between 2016 and 2022, PWH reported 12-month psychological, physical, and sexual IPV in a routine assessment. We used linear and logistic regression models adjusted for age, race/ethnicity, and site to examine relationships with health outcomes. Among 9748 PWH (median age 50 years, 81% cisgender male/16% cisgender female/1% transgender female; 44% non-Hispanic white/36% non-Hispanic Black/15% Hispanic), 9.3% (n = 905) reported any IPV in the past 12 months; half reported psychological IPV without physical/sexual IPV (n = 453). PWH reporting any type of IPV were on average younger than those who did not experience IPV. In adjusted models, any IPV was associated with increased likelihood of unstable housing, HIV viral load detection (HIV viral load ≥ 75 copies/mL), moderate-to-severe depressive symptoms, anxiety with panic symptoms, substance use (methamphetamines, cocaine/crack, illicit opioids, marijuana, heavy episodic/hazardous drinking), and concern about exposure to sexually transmitted infection. PWH reporting any IPV in the past 12 months had 4.2% lower adherence to antiretroviral therapy, 2.4 more HIV-related symptoms, a 1.9 point higher HIV stigma score, and a 9.5% lower quality of life score than those without IPV. We found similar associations among PWH reporting only psychological IPV, without physical/sexual IPV. IPV was common among PWH. Half reporting IPV reported only psychological IPV and had similarly poor outcomes as those reporting physical/sexual IPV, demonstrating the need to assess psychological as well as physical and sexual IPV.

RESUMEN: Examinamos la violencia de la pareja íntima (intimate partner violence, IPV) del año anterior, incluida la violencia psicológica sin violencia física y sexual, así como los resultados sanitarios entre las personas con VIH (people with HIV, PWH) que reciben atención en una cohorte multicéntrica de los Estados Unidos. Entre 2016 y 2022, las PWH informaron situaciones de IPV psicológica, física y sexual durante los 12 meses en una evaluación de rutina. Se utilizaron modelos de regresión lineal y logística ajustados por edad, raza/etnia y centro para examinar las relaciones con los resultados sanitarios. Entre 9748 PWH (mediana de edad de 50 años, 81% de hombres cisgénero/16% de mujeres cisgénero/1% de mujeres transgénero; 44% de blancos no hispanos/36% de negros no hispanos/15% de hispanos), el 9,3% (n = 905) informaron haber sufrido algún tipo de IPV en los últimos 12 meses; la mitad informó situaciones de IPV psicológica sin IPV física y sexual (n = 453). Las PWH que informaron de cualquier tipo de IPV fueron, en promedio, más jóvenes que las que no sufrieron IPV. En los modelos ajustados, cualquier IPV se asoció con una mayor probabilidad de vivienda inestable, detección de carga viral del VIH (carga viral del VIH ≥ 75 copias/ml), síntomas depresivos de moderados a graves, ansiedad con síntomas de pánico, consumo de sustancias (metanfetaminas, cocaína/crack, opioides ilícitos, marihuana, consumo excesivo episódico/peligroso de alcohol) y preocupación por la exposición a infecciones de transmisión sexual. Las PWH que informaron alguna situación de IPV en los últimos 12 meses tuvieron un 4,2% menos de cumplimiento de la terapia antirretrovírica, un 2,4% más de síntomas relacionados con el VIH, una puntuación de estigma del VIH 1,9 puntos más alta y una puntuación de calidad de vida un 9,5% más baja que las que no sufrieron IPV. Se encontraron asociaciones similares entre las PWH que informaron solo IPV psicológica, sin IPV física y sexual. La IPV fue común entre las PWH. La mitad de las personas que informaron IPV solo informaron IPV psicológica y tuvieron resultados igualmente deficientes que los que informaron IPV física y sexual, lo que demuestra la necesidad de evaluar la IPV psicológica, al igual que la IPV física y sexual.

Alcohol and drug use severity are independently associated with antiretroviral adherence in the current treatment era.

Substance use in people with HIV (PWH) negatively impacts antiretroviral therapy (ART) adherence. However, less is known about this in the current treatment era and the impact of specific substances or severity of substance use. We examined the associations of alcohol, marijuana, and illicit drug use (methamphetamine/crystal, cocaine/crack, illicit opioids/heroin) and their severity of use with adherence using multivariable linear regression in adult PWH in care between 2016 and 2020 at 8 sites across the US. PWH completed assessments of alcohol use severity (AUDIT-C), drug use severity (modified ASSIST), and ART adherence (visual analogue scale). Among 9400 PWH, 16% reported current hazardous alcohol use, 31% current marijuana use, and 15% current use of ≥1 illicit drugs. In multivariable analysis, current methamphetamine/crystal use, particularly common among men who had sex with men, was associated with 10.1% lower mean ART adherence (p < 0.001) and 2.6% lower adherence per 5-point higher severity of use (ASSIST score) (p < 0.001). Current and more severe use of alcohol, marijuana, and other illicit drugs were also associated with lower adherence in a dose-dependent manner. In the current HIV treatment era, individualized substance use treatment, especially for methamphetamine/crystal, and ART adherence should be prioritized.

Impact and correlates of sub-optimal social support among patients in HIV care.

Social support (SS) predicts health outcomes among patients living with HIV. We administered a brief, validated measure of SS, the Multifactoral Assessment of Perceived Social Support, within a patient-reported outcomes assessment of health domains in HIV care at 4 U.S. clinics in English and Spanish (n = 708). In univariate analysis, low SS was associated with poorer engagement in care, antiretroviral adherence, and health-related quality of life; current methamphetamine/crystal use, depression, anxiety, and HIV stigma (all p < 0.001); any use of either methamphetamines/crystal, illicit opioids, or cocaine/crack (p = 0.001), current marijuana use (p = 0.012), nicotine use (p = 0.005), and concern for sexually transmitted infection exposure (p = 0.001). High SS was associated with undetectable viral load (p = 0.031). Multivariate analyses found low SS independently associated with depression (risk ratio (RR) 3.72, 95% CI 2.93-4.72), lower adherence (RR 0.76, 95% CI 0.64-0.89), poor engagement in care (RR 2.05, 95% CI 1.44-2.96), and having more symptoms (RR 2.29, 95% CI 1.92-2.75). Medium SS was independently associated with depression (RR 2.59, 95% CI 2.00-3.36), poor engagement in care (RR 1.62, 95% CI 1.15-2.29) and having more symptoms (RR 1.75, 95% CI 1.44-2.13). SS assessment may help identify patients at risk for these outcomes.

Anemia risk factors among people living with HIV across the United States in the current treatment era: a clinical cohort study.

BACKGROUND: Anemia is common among people living with HIV infection (PLWH) and is associated with adverse health outcomes. Information on risk factors for anemia incidence in the current antiretroviral therapy (ART) era is lacking. METHODS: Within a prospective clinical cohort of adult PLWH receiving care at eight sites across the United States between 1/2010-3/2018, Cox proportional hazards regression analyses were conducted among a) PLWH free of anemia at baseline and b) PLWH free of severe anemia at baseline to determine associations between time-updated patient characteristics and development of anemia (hemoglobin < 10 g/dL), or severe anemia (hemoglobin < 7.5 g/dL). Linear mixed effects models were used to examine relationships between patient characteristics and hemoglobin levels during follow-up. Hemoglobin levels were ascertained using laboratory data from routine clinical care. Potential risk factors included: age, sex, race/ethnicity, body mass index, smoking status, hazardous alcohol use, illicit drug use, hepatitis C virus (HCV) coinfection, estimated glomerular filtration rate (eGFR), CD4 cell count, viral load, ART use and time in care at CNICS site. RESULTS: This retrospective cohort study included 15,126 PLWH. During a median follow-up of 6.6 (interquartile range [IQR] 4.3-7.6) years, 1086 participants developed anemia and 465 participants developed severe anemia. Factors that were associated with incident anemia included: older age, female sex, black race, HCV coinfection, lower CD4 cell counts, VL ≥400 copies/ml and lower eGFR. CONCLUSION: Because anemia is a treatable condition associated with increased morbidity and mortality among PLWH, hemoglobin levels should be monitored routinely, especially among PLWH who have one or more risk factors for anemia.

Risk of End-Stage Renal Disease in HIV-Positive Potential Live Kidney Donors.

New federal regulations allow HIV-positive individuals to be live kidney donors; however, potential candidacy for donation is poorly understood given the increased risk of end-stage renal disease (ESRD) associated with HIV infection. To better understand this risk, we compared the incidence of ESRD among 41 968 HIV-positive participants of North America AIDS Cohort Collaboration on Research and Design followed for a median of 5 years with the incidence of ESRD among comparable HIV-negative participants of National Health and Nutrition Examination III followed for a median of 14 years. We used risk associations from multivariable Cox proportional hazards regression to derive cumulative incidence estimates for selected HIV-positive scenarios (no history of diabetes, hypertension, AIDS, or hepatitis C virus coinfection) and compared these estimates with those from similarly selected HIV-negative scenarios. For 40-year-old HIV-positive individuals with health characteristics that were similar to those of age-matched kidney donors, viral load <400 copies/mL, and CD4+ count ≥500 cells/µL, the 9-year cumulative incidence of ESRD was higher than that of their HIV-negative peers, yet still low: 2.5 versus 1.1 per 10 000 among white women, 3.0 versus 1.3 per 10 000 among white men, 13.2 versus 3.6 per 10 000 among black women, and 15.8 versus 4.4 per 10 000 among black men. HIV-positive individuals with no comorbidities and well-controlled disease may be considered low-risk kidney donor candidates.

A Comparison of Adherence Timeframes Using Missed Dose Items and Their Associations with Viral Load in Routine Clinical Care: Is Longer Better?

Questions remain regarding optimal timeframes for asking about adherence in clinical care. We compared 4-, 7-, 14-, 30-, and 60-day timeframe missed dose items with viral load levels among 1099 patients on antiretroviral therapy in routine care. We conducted logistic and linear regression analyses examining associations between different timeframes and viral load using Bayesian model averaging (BMA). We conducted sensitivity analyses with subgroups at increased risk for suboptimal adherence (e.g. patients with depression, substance use). The 14-day timeframe had the largest mean difference in adherence levels among those with detectable and undetectable viral loads. BMA estimates suggested the 14-day timeframe was strongest overall and for most subgroups although findings differed somewhat for hazardous alcohol users and those with current depression. Adherence measured by all missed dose timeframes correlated with viral load. Adherence calculated from intermediate timeframes (e.g. 14-day) appeared best able to capture adherence behavior as measured by viral load.

Validity assessment of the PROMIS fatigue domain among people living with HIV.

PURPOSE: To evaluate psychometric characteristics and cross-sectional and longitudinal validity of the 7-item PROMIS® Fatigue Short Form and additional fatigue items among people living with HIV (PLWH) in a nationally distributed network of clinics collecting patient reported data at the time of routine clinical care. METHODS: Cross-sectional and longitudinal fatigue data were collected from September 2012 through April 2013 across clinics participating in the Centers for AIDS Research Network of Integrated Clinical Systems (CNICS). We analyzed data regarding psychometric characteristics including simulated computerized adaptive testing and differential item functioning, and regarding associations with clinical characteristics. RESULTS: We analyzed data from 1597 PLWH. Fatigue was common in this cohort. Scores from the PROMIS® Fatigue Short Form and from the item bank had acceptable psychometric characteristics and strong evidence for validity, but neither performed better than shorter instruments already integrated in CNICS. CONCLUSIONS: The PROMIS® Fatigue Item Bank is a valid approach to measuring fatigue in clinical care settings among PLWH, but in our analyses did not perform better than instruments associated with less respondent burden.

Lessons learned from the design and implementation of myocardial infarction adjudication tailored for HIV clinical cohorts.

We developed, implemented, and evaluated a myocardial infarction (MI) adjudication protocol for cohort research of human immunodeficiency virus. Potential events were identified through the centralized Centers for AIDS Research Network of Integrated Clinical Systems data repository using MI diagnoses and/or cardiac enzyme laboratory results (1995-2012). Sites assembled de-identified packets, including physician notes and results from electrocardiograms, procedures, and laboratory tests. Information pertaining to the specific antiretroviral medications used was redacted for blinded review. Two experts reviewed each packet, and a third review was conducted if discrepancies occurred. Reviewers categorized probable/definite MIs as primary or secondary and identified secondary causes of MIs. The positive predictive value and sensitivity for each identification/ascertainment method were calculated. Of the 1,119 potential events that were adjudicated, 294 (26%) were definite/probable MIs. Almost as many secondary (48%) as primary (52%) MIs occurred, often as the result of sepsis or cocaine use. Of the patients with adjudicated definite/probable MIs, 78% had elevated troponin concentrations (positive predictive value = 57%, 95% confidence interval: 52, 62); however, only 44% had clinical diagnoses of MI (positive predictive value = 45%, 95% confidence interval: 39, 51). We found that central adjudication is crucial and that clinical diagnoses alone are insufficient for ascertainment of MI. Over half of the events ultimately determined to be MIs were not identified by clinical diagnoses. Adjudication protocols used in traditional cardiovascular disease cohorts facilitate cross-cohort comparisons but do not address issues such as identifying secondary MIs that may be common in persons with human immunodeficiency virus.

Unannounced telephone-based pill counts: a valid and feasible method for monitoring adherence.

Phone-based unannounced pill counts to measure medication adherence are much more practical and less expensive than home-based unannounced pill counts, but their validity has not been widely assessed. We examined the validity of phone versus home-based pill counts using a simplified protocol streamlined for studies embedded in clinical care settings. A total of 100 paired counts were used to compare concordance between unannounced phone and home-based pill counts using interclass correlations. Discrepancy analyses using χ(2) tests compared demographic and clinical characteristics across patients who were concordant between phone and home-based pill counts and patients who were not concordant. Concordance was high for phone-based and home-based unannounced total pill counts, as well as individual medication counts and calculated adherence. This study demonstrates that a simplified phone-based pill count protocol can be implemented among patients from a routine clinical care setting and is a feasible means of monitoring medication adherence.

Evaluation of the single-item self-rating adherence scale for use in routine clinical care of people living with HIV.

The self-rating scale item (SRSI) is a single-item self-report adherence measure that uses adjectives in a 5-point Likert scale, from "very poor" to "excellent," to describe medication adherence over the past 4 weeks. This study investigated the SRSI in 2,399 HIV-infected patients in routine care at two outpatient primary HIV clinics. Correlations between the SRSI and four commonly used adherence items ranged from 0.37 to 0.64. Correlations of adherence barriers, such as depression and substance use, were comparable across all adherence items. General estimating equations suggested the SRSI is as good as or better than other adherence items (p's <0.001 vs. <0.001-0.99) at predicting adherence-related clinical outcomes, such as HIV viral load and CD4(+) cell count. These results and the SRSI's low patient burden suggest its routine use could be helpful for assessing adherence in clinical care and should be more widespread, particularly where more complex instruments may be impractical.

Racial and ethnic disparities in COVID-19 disease incidence independent of comorbidities, among people with HIV in the US.

OBJECTIVES: To define the incidence of clinically-detected COVID-19 in people with HIV (PWH) in the US and evaluate how racial and ethnic disparities, comorbidities, and HIV-related factors contribute to risk of COVID-19. DESIGN: Observational study within the CFAR Network of Integrated Clinical Systems cohort in 7 cities during 2020. METHODS: We calculated cumulative incidence rates of COVID-19 diagnosis among PWH in routine care by key characteristics including race/ethnicity, current and lowest CD4 count, and geographic area. We evaluated risk factors for COVID-19 among PWH using relative risk regression models adjusted with disease risk scores. RESULTS: Among 16,056 PWH in care, of whom 44.5% were Black, 12.5% were Hispanic, with a median age of 52 years (IQR 40-59), 18% had a current CD4 count < 350, including 7% < 200; 95.5% were on antiretroviral therapy, and 85.6% were virologically suppressed. Overall in 2020, 649 PWH were diagnosed with COVID-19 for a rate of 4.94 cases per 100 person-years. The cumulative incidence of COVID-19 was 2.4-fold and 1.7-fold higher in Hispanic and Black PWH respectively, than non-Hispanic White PWH. In adjusted analyses, factors associated with COVID-19 included female sex, Hispanic or Black identity, lowest historical CD4 count <350 (proxy for CD4 nadir), current low CD4/CD8 ratio, diabetes, and obesity. CONCLUSIONS: Our results suggest that the presence of structural racial inequities above and beyond medical comorbidities increased the risk of COVID-19 among PWHPWH with immune exhaustion as evidenced by lowest historical CD4 or current low CD4:CD8 ratio had greater risk of COVID-19.

Measuring depression levels in HIV-infected patients as part of routine clinical care using the nine-item Patient Health Questionnaire (PHQ-9).

Little is known about the psychometric properties of depression instruments among persons infected with HIV. We analyzed data from a large sample of patients in usual care in two US cities (n=1467) using the nine-item Patient Health Questionnaire (PHQ-9) from the PRIME-MD. The PHQ-9 had curvilinear scaling properties and varying levels of measurement precision along the continuum of depression measured by the instrument. In our cohort, the scale showed a prominent floor effect and a distribution of scores across depression severity levels. Three items had differential item functioning (DIF) with respect to race (African-American vs. white); two had DIF with respect to sex; and one had DIF with respect to age. There was minimal individual-level DIF impact. Twenty percent of the difference in mean depression levels between African-Americans and whites was due to DIF. While standard scores for the PHQ-9 may be appropriate for use with individual HIV-infected patients in cross-sectional settings, these results suggest that investigations of depression across groups and within patients across time may require a more sophisticated analytic framework.

Lipoatrophy and lipohypertrophy are independently associated with hypertension.

OBJECTIVE: Lipoatrophy and lipohypertrophy are associated with metabolic abnormalities, but little is known about their impact on hypertension. We conducted this study to determine the associations of lipoatrophy and lipohypertrophy with hypertension. METHODS: A cross-sectional study of HIV-infected patients who completed a self-report body morphology assessment was performed. We defined hypertension as a clinical diagnosis, or a mean systolic blood pressure (BP) > 140 mmHg or diastolic BP > 90 mmHg in the previous 6 months. We used logistic regression to examine the association between hypertension and body morphology. RESULTS: Among 347 patients, there were 2278 BP readings in 6 months. In adjusted analyses, patients with moderate lipoatrophy [odds ratio (OR) 4.3; P = 0.03] or moderate lipohypertrophy (OR 4.3; P = 0.006) had four times the odds, and patients with mild lipohypertrophy (OR 2.3; P = 0.03) had twice the odds of having hypertension compared with patients without changes. We hypothesized that the impact of lipohypertrophy on hypertension was mediated, in part, through body mass index (BMI). When BMI was included in the analysis, increased BMI was significantly associated with hypertension (OR = 1.1; P < 0.001 per kg/m(2)), and the association between lipohypertrophy and hypertension was no longer present. However, the association between moderate lipoatrophy and hypertension was strengthened (OR = 5.5; P = 0.01). CONCLUSIONS: Lipoatrophy and lipohypertrophy are independently associated with hypertension and there is a dose-response effect with more severe lipoatrophy and lipohypertrophy. The association between lipohypertrophy (but not lipoatrophy) and hypertension appears to be mediated by BMI. Our results suggest that patient-based body morphology assessments are related to hypertension and may have potential implications for cardiovascular disease.

Lipoatrophy among HIV-infected patients is associated with higher levels of depression than lipohypertrophy.

OBJECTIVES: We sought to determine the association between body morphology abnormalities and depression, examining lipoatrophy and lipohypertrophy separately. METHODS: An observational cross-sectional study of 250 patients from the University of Washington HIV Cohort was carried out. Patients completed an assessment including measures of depression and body morphology. We used linear regression analysis to examine the association between lipoatrophy or lipohypertrophy and depression. Analysis of variance was used to examine the relationship between mean depression scores and lipoatrophy and lipohypertrophy in 10 body regions. RESULTS: Of 250 patients, 76 had lipoatrophy and 128 had lipohypertrophy. Mean depression scores were highest among patients with moderate-to-severe lipoatrophy (16.4), intermediate among those with moderate-to-severe lipohypertrophy (11.7), mild lipohypertrophy (9.9) and mild lipoatrophy (8.5), and lowest among those without body morphology abnormalities (7.7) (P=0.002). After adjustment, mean depression scores for subjects reporting moderate-to-severe lipoatrophy were 9.2 points higher (P<0.001), scores for subjects with moderate-to-severe lipohypertrophy were 4.8 points higher (P=0.02), and scores for subjects with mild lipohypertrophy were 2.8 points higher (P=0.03) than those for patients without body morphology abnormalities. Facial lipoatrophy was the body region associated with the most severe depression scores (15.5 vs. 8.9 for controls; P=0.03). CONCLUSIONS: In addition to long-term cardiovascular implications, body morphology has a more immediate effect on depression severity.

Integrating a web-based, patient-administered assessment into primary care for HIV-infected adults.

Providers routinely under diagnose at risk behaviors and outcomes, including depression, suicidal ideation, substance abuse, and poor medication adherence. To address this, we developed a web-based, self-administered patient-reported assessment tool and integrated it into routine primary care for HIV-infected adults. Printed results were delivered to providers and social workers immediately prior to patient appointments. The assessment included brief, validated instruments measuring clinically relevant domains including depression, substance use, medication adherence, and HIV transmission risk behaviors. Utilizing the Institute for Healthcare Improvement's Plan-Do-Study-Act (PDSA) approach to quality improvement, we addressed issues with clinic flow, technology, scheduling, and delivery of assessment results with the support of all levels of clinic staff. We found web-based patient-reported assessments to be a feasible tool that can be integrated into a busy multi-provider HIV primary care clinic. These assessments may improve provider recognition of key patient behaviors and outcomes. Critical factors for successful integration of such assessments into clinical care include: strong top-level /ort from clinic management, provider understanding of patient-reported assessments as a valuable clinical tool, tailoring the assessment to meet provider needs, communication among clinic staff to address flow issues, timeliness of delivery, and sound technological resources.

Developing a relational XML schema for sharing HIV clinical data.

Access to multi-site clinical data regarding treatment and outcomes of HIV-infected patients in routine care is required to support clinical research to improve the treatment of HIV. As part of the NIAID-funded CFAR Network of Integrated Clinical Systems (CNICS), we have developed a relational XML Schema to extend the existing observational research repository and to integrate real-time clinical information from electronic medical records (EMRs) at six Centers for AIDS Research (CFAR) into the repository. The schema will aid the expansion of the research repository beyond the initial sites, and the development process may facilitate the use of multi-site repositories to study other chronic diseases.

Physician experience in the care of HIV-infected persons is associated with earlier adoption of new antiretroviral therapy.

OBJECTIVE: Recent advances in antiretroviral therapy have led to effective but increasingly complex strategies for the treatment of HIV infection. In a previous study, we demonstrated that physicians' experience in the care of patients with AIDS improves survival. We conducted this study to determine whether greater physician experience is associated with earlier adoption and appropriate use of new antiretroviral treatment regimens. DESIGN: Retrospective medical record review of a population-based sample of HIV-infected individuals who received antiretroviral treatment between December 1995 and May 1997 by primary care physicians practicing throughout the state of Washington. We classified antiretroviral regimens observed into one of four categories based on national treatment guidelines. RESULTS: The use of new antiretroviral treatment regimens significantly increased during the study period; 22% of patients were treated with a protease inhibitor (PI)-based regimen or an alternative PI- or nonnucleoside reverse transcriptase inhibitor (NNRTI)-based regimen between December 1995 and November 1996, compared with 57% between April and May 1997 (p < .001). After controlling for CD4 count and the calendar period of treatment, patients cared for by physicians with greater HIV experience were significantly more likely to receive PI-based regimens or alternative PI- or NNRTI-based antiretroviral regimens (p = .02). Use of PI-based regimens was also associated with lower CD4 count (p < .001) and treatment after January 1997 (p = .02), but independent of patient demographic characteristics and the geographic location of physicians' practices. CONCLUSIONS: Greater physician experience in the care of persons with HIV infection is associated with earlier adoption of new antiretroviral treatment regardless of whether physicians practice in a rural or urban area.

Physicians' experience with the acquired immunodeficiency syndrome as a factor in patients' survival.

BACKGROUND: Previous studies have found that patients with the acquired immunodeficiency syndrome (AIDS) who are admitted to hospitals that admit many such patients have lower mortality rates than patients in hospitals with less experience with AIDS. We examined the relation between physicians' experience with AIDS and the survival of their patients with AIDS. METHODS: We studied 403 adult male patients enrolled in a staff-model health maintenance organization in whom first AIDS-defining illnesses were diagnosed from 1984 through mid-1994; we determined that these illnesses met the 1987 case definition of the Centers for Disease Control. We defined three levels of experience for the patients' 125 primary care physicians according to their experience with AIDS during residency training and the cumulative number of patients with AIDS they had cared for in their practices. RESULTS: The median survival of the patients of physicians with the least experience in the management of AIDS was 14 months, as compared with 26 months for the patients of physicians with the most experience (P<0.001). Controlling for the severity of illness and the year of diagnosis, we found that the patients cared for by physicians with the most experience had a 31 percent lower risk of death than the patients cared for by physicians with the least experience (P<0.02). Among 244 patients with an AIDS-defining illness diagnosed from 1989 through 1994, after adjustment for the CD4+ cell count and the severity of illness, the risk of death was 44 percent lower for patients of the most experienced physicians than for patients of the least experienced (P<0.02). CONCLUSIONS: The experience of primary care physicians in the management of AIDS is significantly associated with survival among their patients.