Evaluation of the effectiveness of a latrine intervention in the reduction of childhood diarrhoeal health in Nyando District, Kisumu County, Kenya.

Community-led total sanitation (CLTS) is an intervention that strives to end the practice of open defaecation. This study measured the effectiveness of CLTS in Nyando District by examining the association between community open defaecation-free (ODF) status and childhood diarrhoeal illness. A cross-sectional study design was used among households with children ⩽5 years old to ascertain information on acute diarrhoea in the past year (outcome), sanitation and health behaviours. Water testing was conducted to determine Escherichia coli and turbidity levels for 55 water sources. Data were obtained from 210 parents or caregivers from an ODF community and 216 parents or caregivers in a non-ODF community. The non-ODF participants reported a non-significant 16% increased risk of diarrhoea compared with the participants from the ODF community. Children's HIV positivity (adjusted prevalence ratio (aPR) = 2.29; 95% CI 2.07-2.53), unsafe child stool disposal (aPR = 1.92; 95% CI 1.74-2.12) and low household income (aPR = 1.93; 95% CI 1.46-2.56) were associated with diarrhoea, in the non-ODF community. The ODF location had a higher percentage of E. coli in the drinking water compared with the non-ODF location (76.7% vs. 60%). Diarrhoeal disease rates in children ⩽5 years old did not differ by whether a latrine intervention was implemented. Water sampling findings suggest water safety may have decreased the effectiveness of the CLTS' improvement of childhood diarrhoea. Improved water treatment practices, safe stool disposal and education may improve the CLTS intervention in ODF communities and therefore reduced the risk of childhood diarrhoea.

A randomized trial comparing standard outcomes in two treatment models for substance users with tuberculosis.

SETTING: Chicago Department of Public Health (CDPH), TB Control Program. OBJECTIVES: To compare anti-tuberculosis treatment outcomes using two different types of directly observed therapy (DOT) outreach workers. METHODS: Substance users diagnosed with TB from October 1996 to July 2000 were randomized to DOT administered by either 1) CDPH personnel (standard arm) or 2) previous substance-using human immunodeficiency virus/acquired immune-deficiency syndrome outreach workers (enhanced arm). Treatment completion was physician-determined, and adherence was estimated based on risk of missed DOT appointments. RESULTS: Of 94 patients, 46 were randomized to the standard and 48 to the enhanced arm. The standard arm had a significantly higher risk of non-completion of treatment (39% vs. 15%, RR 2.7, 95%CI 1.2-5.8), and a significantly higher risk of missing DOT appointments (RR 2.6, 95%CI 1.4-4.8). For both outcomes, housing instability was a significant predictor in multivariate analyses. CONCLUSIONS: TB treatment completion and adherence among substance users was improved by the enhanced intervention; the familiarity of enhanced-arm DOT workers with the patients' social norms due to their own previous substance use may have made them more effective. Successful DOT in hard-to-reach populations may require strategies that directly address the population's circumstances and utilize DOT workers who are intimately familiar with patients' life situations.

GB virus C infection among young, HIV-negative injection drug users with and without hepatitis C virus infection.

Our study examined the association between GB virus C (GBV-C) and (i) hepatitis C virus (HCV) infection status, (ii) biomedical indicators of liver disease (alanine and aspartate aminotransferases) and (iii) HCV RNA level among young injection drug users (IDUs) recruited using street outreach and respondent-driven methods. Cross-sectional and longitudinal analyses were completed. GBV-C (active or resolved) infection was significantly (P < 0.05) more prevalent among HCV antibody-positive (anti-HCV+) (65.1%) than antibody-negative (anti-HCV-) (32.3%) (OR = 3.9, 95% CI: 2.3-6.9) IDUs. The prevalence of resolved GBV-C infection was highest among those with chronic HCV infection (41.9%), followed by those with resolved HCV infection (34.4%) and significantly lower (P < 0.05) among anti-HCV participants (16.9%). Although not statistically significant (P = 0.13), a similar pattern was observed for active GBV-C infection. No association between GBV-C infection status and biomedical indicators of liver disease or HCV RNA level over time was observed. In conclusion, GBV-C infection prevalence was higher among anti-HCV+ compared to anti-HCV- young IDUs, similar to prior studies among older populations. In particular, chronically HCV-infected young IDUs had an increased rate of GBV-C clearance.

Bloodstream infections and setting of onset in persons with spinal cord injury and disorder.

STUDY DESIGN: A retrospective cohort study. OBJECTIVE: Health-care-associated (HCA) bloodstream infection (BSI) has been shown to be a distinct epidemiologic category in the general adult population, but few studies have examined specific patient populations. The objective of this study was to assess characteristics associated with BSI that occurred in the hospital (hospital-acquired, HA BSI), from health-care contact outside the hospital (HCA BSI) or in the community (community-acquired, CA BSI) in veterans with spinal cord injury and disorder (SCI&D). SETTING: Two United States Department of Veterans Affairs hospitals. METHODS: All patients with SCI&D with a positive blood culture admitted to study hospitals over a 7-year period (1 October 1997 to 30 September 2004). Demographics, medical characteristics and causative organisms were collected. RESULTS: Four hundred and thirteen episodes of BSI occurred in 226 patients, with a rate of 7.2 BSI episodes per 100 admissions: 267 (64.7%) were HA BSI, 110 (26.6%) were HCA BSI and 36 (8.7%) were CA BSI. Antibiotic resistance was more common in those with HA BSI (65.5%) compared with that in those with HCA (49.1%, P=0.001) and CA BSI (22.2%, P<0.0001). Methicillin resistance in Staphylococcus aureus was highly prevalent; HA BSI (84.5%), HCA BSI (60.6%) and CA BSI (33.3%). CONCLUSION: HCA BSI comprises one-quarter of all BSIs in hospitalized patients with SCI&D. Although those with HCA and CA BSI share similarities, several differences in medical characteristics and causal microorganism are noted. Treatment and management strategies for HCA and CA infections need to vary.

The effect of renal dysfunction on antimicrobial use measurements.

The defined daily dose, a popular measurement of antimicrobial use, may underestimate the use of antimicrobials that are dose-adjusted in patients with renal insufficiency. To evaluate the effect of renal dysfunction on these measures, we performed a retrospective cohort study that involved patients receiving ceftriaxone, levofloxacin, or vancomycin, with use of defined daily doses and 2 methods based on therapy duration--stop-start days (i.e., entire therapy duration) and transaction days (i.e., unique therapeutic days). The vancomycin use rate for patients with renal insufficiency was 36% lower than that of patients with normal renal function for defined daily doses, and it was 23% lower for transaction days; for levofloxacin, there was a 27% rate reduction for the defined daily dose. No significant reduction was noted when the stop-start day method was used. Compared with the defined daily dose method, measures of therapy duration are less affected by renal function and may improve comparisons between populations.

Effects of human T-lymphotropic virus type II on human immunodeficiency virus type 1 phenotypic evolution.

Phenotypic change and broader coreceptor usage by HIV-1 have been associated with disease progression. HIV-1 coreceptor usage by primary isolates obtained from HIV-1-infected and HIV-1/HTLV-II-coinfected individuals was determined. HIV-1 was isolated from 15 of 20 HIV-1-infected and 17 of 24 HIV-1/HTLV-II-coinfected individuals. None of the isolates from either the HIV-1-infected or the coinfected group infected CCR5delta32 PBMCs, suggesting that they all were R5-tropic. Further, both spontaneous and PHA-stimulated production of MIP-1beta and RANTES were similar in HIV-1-infected and coinfected individuals. These data indicate that coinfection with HTLV-II has no effect on HIV-1 coreceptor usage or ex vivo beta-chemokine production.

The natural history of hepatitis D virus infection in Illinois state facilities for the developmentally disabled.

OBJECTIVE: We sought to define the natural history of hepatitis D virus infection in an institutionalized, developmentally disabled population and to identify other prognostic factors. METHODS: A retrospective cohort study was conducted on 231 hepatitis B virus carriers, 65 of whom were also infected with hepatitis D virus, at thirteen Illinois state facilities for the developmentally disabled. Demographic, clinical and laboratory data from 1986 to 1998 were obtained by chart review. Cox regression analysis was used to compare those with and without hepatitis D virus infection in terms of overall mortality, mortality from hepatic disease, and risk of developing chronic hepatitis and cirrhosis and to identify other potential prognostic factors. RESULTS: Residents with hepatitis D virus infection were more likely to die of liver disease than uninfected residents (11% vs 0.6%, respectively; relative hazard, 15.2; 95% confidence interval, 1.8-126.6), although there was no significant difference in overall mortality. Twenty-one percent of residents with hepatitis D virus infection were diagnosed to have cirrhosis or chronic hepatitis compared with 9% of those uninfected (relative hazard 2.5, 95% confidence interval 1.2-5.2). Among the other variables tested, none was predictive of risk of dying of liver disease, and only seropositivity for hepatitis B e antigen was predictive of risk of developing cirrhosis or chronic hepatitis. CONCLUSIONS: In an institutionalized, developmentally disabled population of hepatitis B virus carriers, hepatitis D virus infection is associated with a greater risk of liver-associated mortality and of developing chronic liver disease than that associated with hepatitis B virus carriage alone.

Body composition in HIV-infected women.

Although loss of lean body mass is a common complication of human immunodeficiency virus (HIV) infection that can occur across the disease trajectory, few studies have characterized the body composition of HIV-infected women. We used bioelectrical impedance analysis to characterize the body composition of HIV-infected (n = 56) and uninfected (n = 12) women who were matched on percentage of ideal body weight. The HIV-infected women did not differ from the uninfected women by height-adjusted fat mass or body cell mass. Intergroup comparisons among the HIV-infected women showed that underweight women had significantly less fat mass than did normal-weight women but did not significantly differ with respect to body cell mass. Among all HIV-infected women, CD4(+) lymphocyte count was positively correlated with fat mass (r = 0.32, P = 0.01) but not with body cell mass. No significant correlations were found between any body-composition parameter and plasma viral load. Our findings suggest that, unlike men, HIV-infected underweight women show a preferential loss of fat mass and a relative preservation of body cell mass. This altered pattern of weight loss may relate to higher premorbid fat stores in women and/or hormonal differences.

The contribution of assay variation and biological variation to the total variability of plasma HIV-1 RNA measurements. The Women Infant Transmission Study Clinics. Virology Quality Assurance Program.

OBJECTIVES: To assess the specific contributions of assay variation and biological variation to the total variation of plasma HIV-1 RNA measured by the Roche Monitor assay and the extent to which batch assays reduced both assay variability and total variability compared with real-time determinations. DESIGN: A retrospective analysis of data obtained from three trials conducted by the Adult and Pediatric AIDS Clinical Trials Groups (ATCG), the Women and Infants Transmission Study (WITS) and the NIAID-sponsored Virology Quality Assurance Program. METHODS: Within-subject variation was assessed from stored, serially collected plasma samples from 663 subjects enrolled in the ACTG and WITS studies. Interassay and intra-assay variation were estimated from two of the clinical trials and 22 laboratories that participated in a quality assurance program and were used to estimate the effect of real-time testing on total variation. RESULTS: The total variation (standard deviation) from a random effects model was 0.26 log10 RNA copies/ml. The estimated interassay variation was 0.08 log10 and intra-assay variation was 0.12 log10 RNA copies/ml. Biological variation accounted for 56-80% of total variation. The effect of real-time testing compared with batch testing was minimal. CONCLUSION: Our estimates of total within-subject HIV-1 RNA variation support the current recommendation to obtain at least two specimens, preferably obtained less than 2 weeks apart, for viral RNA measurement before starting therapy. The major contribution of biological variation to the total variation supports the use of real-time HIV-1 RNA assays, provided that consistent specimen collection procedures are followed and acceptable assay proficiency is maintained.

Hepatitis C virus infection in Chicago women with or at risk for HIV infection: evidence for sexual transmission.

BACKGROUND AND OBJECTIVES: The importance of sexual transmission of hepatitis C virus (HCV) infection is unclear. We attempted to define its role in women with or at risk for HIV infection. GOAL OF THIS STUDY: To ascertain if high-risk sexual behavior was independently associated with HCV infection. STUDY DESIGN: Risk factors were assessed cross-sectionally in Chicago women newly enrolled in the Women's Interagency HIV Study. Women who had (n = 243) or were at risk for HIV infection (n = 53) were tested for HCV antibodies (Ab). RESULTS: Of 296 women, 123 (42%) were HCV Ab positive; prevalence was 90% in women who injected drugs (IDU) compared with 12% in noninjectors (odds ratio [OR], 64.0, 95% confidence interval [CI], 29.9 to 137.0). A multivariate model showed associations with IDU (OR, 110.3, 95% CI, 33.3 to 365.8), prior gonorrhea (OR, 3.6, 95% CI, 1.4 to 8.9), and sex with a male IDU (OR, 2.7, 95% CI, 1.1 to 7.0). CONCLUSION: Injection drug use is the strongest predictor of HCV infection, but sexual risk factors are also independently associated.

Ciprofloxacin resistance in methicillin-resistant Staphylococcus aureus: associated factors and resistance to other antibiotics.

At the University of Illinois Hospital, antibiotic susceptibility testing was retrospectively performed on 254 stored clinical methicillin-resistant Staphylococcus aureus (MRSA) isolates cultured from 1985 through 1990 to characterize resistance to ciprofloxacin and other antibiotics. In case-control analyses, inpatients with and without ciprofloxacin-resistant strains were compared. Ciprofloxacin-resistance increased from 7% before 1988 to 83% in 1990. A sudden increase in resistance to trimethoprim-sulfamethoxazole also occurred in 1988, and by 1990, 65% of strains were resistant to both antibiotics. In 95 patients with recent MRSA isolation (70 acquired nosocomially, 25 acquired in the community), ciprofloxacin resistance was more common in the nosocomial group (80% v 60%, P < 0.05). In that group, no host or in-hospital factors were associated with ciprofloxacin resistance. Among community cases, a greater proportion with ciprofloxacin-resistant MRSA had diabetes mellitus (60% v 0%, P = 0. 002). Thus, with use, ciprofloxacin resistance emerged rapidly in MRSA and developed particularly among strains resistant to trimethoprim-sulfamethoxazole. Combined resistance to these antibiotics, uncommon in previous reports, severely limits oral therapy as an option for MRSA carriage or infection.

Retrospective evaluation of therapies for Staphylococcus aureus endocarditis.

We retrospectively evaluated antiinfective therapy for methicillin-sensitive (MSSA) and methicillin-resistant Staphylococcus aureus (MRSA) endocarditis in 54 patients who had 57 treatment courses for the disease. Three treatments were assessed: 27 nafcillin-treated courses of MSSA endocarditis, 18 vancomycin-treated courses of MSSA endocarditis, and 11 vancomycin-treated courses of MRSA endocarditis. At baseline, patients with MSSA treated with vancomycin had more chronic conditions (p<0.01), a lower frequency of intravenous drug use (p<0.01), a lower hematocrit concentration (p<0.05), and a higher serum creatinine concentration (p<0.05) than the nafcillin group. Vancomycin-treated patients had a higher complication rate during therapy (p<0.05) and a longer duration in an intensive care unit (p<0.01) than the nafcillin group. The trend was for a higher complete response rate in the nafcillin group (74% vs 50%, p=0.12), but no difference in mortality (22% vs 28%, p=0.73). Patients with MRSA infection treated with vancomycin had higher mortality than those with MSSA who received that drug (55% vs 28%, p=0.24). Patients with vancomycin-treated MSSA endocarditis may have a poorer outcome than those who receive nafcillin, but this may be influenced by different or more severe clinical features.

Increased vertical transmission of human immunodeficiency virus from hepatitis C virus-coinfected mothers. Women and Infants Transmission Study.

To determine if hepatitis C virus (HCV) infection affects vertical transmission of human immunodeficiency virus (HIV), 487 HIV-infected pregnant women in the prospective, multicenter, Women and Infants Transmission Study had HCV antibody (anti-HCV by second-generation ELISA) and HCV RNA (by quantitative polymerase chain reaction) measured in peripartum maternal plasma; 161 (33%) were anti-HCV-positive. HIV vertical transmission occurred from 42 HCV-infected mothers (26.1%) versus 53 HCV-uninfected mothers (16.3%; odds radio [OR], 1.82; P = .01). In a logistic regression model that included maternal drug use, a potential confounder, HCV infection was marginally associated with perinatal HIV transmission (OR, 1.64; P = .05), whereas drug use was not. Women who transmitted HIV had higher levels of HCV RNA (median, 721,254 copies/mL) than those who did not (337,561 copies/mL; P = .01). Maternal HCV infection is associated with increased HIV vertical transmission. Further studies are needed to ascertain if HCV directly affects perinatal HIV transmission or is a marker for another factor, such as maternal drug use.

An international collaborative study of the effects of coinfection with human T-lymphotropic virus type II on human immunodeficiency virus type 1 disease progression in injection drug users.

To determine whether human T-lymphotropic virus (HTLV) type II coinfection affects progression of human immunodeficiency virus type 1 (HIV) infection, longitudinal data on 370 HIV-infected injection drug users (IDUs) with known HIV seroconversion dates from four cohort studies were pooled. HTLV infection was determined by EIA and confirmed and typed by Western blot. Proportional hazards models were used to determine whether HTLV-II infection was associated with AIDS or AIDS-related mortality. Regression analyses were used to compare declines in CD4 cell percents in singly and dually infected persons. Of 370 IDUs, 61 (16%) were HTLV-II-coinfected. During follow-up, 43 (12%) developed and 24 (6%) died of AIDS. HTLV-II coinfection was not associated with progression to AIDS (relative hazard [RH], .82; 95% confidence interval [CI], 0.34-1.94]) or AIDS mortality (RH, 1.69; 95% CI, 0.62-4.60). Rates of decline in CD4 cell percent were similar in singly and dually infected IDUs. These results suggest that HTLV-II does not affect the progression of HIV infection.

Investigation of proviral load in individuals infected with human T-lymphotropic virus type II.

Human T-lymphotrophic virus type II (HTLV-II) has not yet been associated with any disease. Little is known about the proviral loads of HTLV-II in vivo and its relationship, if any, to lack of pathogenicity. We determined the HTLV-II proviral copy number in peripheral blood lymphocyte (PBL) samples from 49 HTLV-II-infected individuals, of whom 25 were coinfected with human immunodeficiency virus type 1 (HIV-1). The HTLV-II copy numbers were determined by polymerase chain reaction (PCR) amplification of end-point dilutions of PBL lysates, followed by hybridization to a 32P-labeled HTLV-II-specific probe. The proviral copy number for the 49 samples ranged from < 0.02 to 200 per 1000 PBLs; 6% had < 0.02, 16% had 0.02, 20% had 0.2, 18% had 2, 31% had 20, and 8% had 200 copies per 1000 PBLs. The distributions of HTLV-II copy numbers in the coinfected and singly infected subgroups were not significantly different (Wilcoxon rank sum, p = 0.24). In the coinfected subgroup, there was no significant correlation between the HTLV-II proviral load and the counts of CD4-positive lymphocytes or CD8-positive lymphocytes (Spearman Coefficient = 0.26, p = 0.20; = 0.091, p = 0.67, respectively). Our data demonstrate the presence of a wide range of viral loads in HTLV-II-infected individuals. The high viral loads (> or = 20 copies/1000 lymphocytes) detected in 39% of our samples suggest that the low pathogenicity of HTLV-II is not related to the presence of low viral loads in the infected subjects. Our data from the HIV-1 coinfected individuals show no apparent effect of HIV-1 on HTLV-II proviral loads.

Control of nosocomial infections in an intensive care unit in Guatemala City.

We tested the effectiveness of specific vs. general infection control interventions in a teaching hospital in Guatemala City. After 3 months of prospective surveillance, we implemented targeted interventions (i.e., modification of respiratory tract care and use of a closed urinary catheter drainage system), an educational program focused on respiratory intervention, and general interventions (i.e., aseptic technique). The rate of nosocomial pneumonia, the most common nosocomial infection, decreased from 33% (41 of 123 patients) before intervention to 16% (21 of 130 patients) after intervention (P = .001). Although the frequency of hand washing increased from 5% to 63% (P < .001), the rates of other types of nosocomial infections did not change significantly. The combination of targeted respiratory intervention and an intense, focused educational campaign reduced the rate of nosocomial pneumonia. General improvements in hygiene and hand washing rates, or even implementation of a closed urinary drainage system without focused education, may not be sufficient to reduce infection rates in intensive care units in developing countries.

The spectrum of medical conditions and symptoms before acquired immunodeficiency syndrome in homosexual and bisexual men infected with the human immunodeficiency virus.

The full range and occurrence of medical conditions in persons infected with human immunodeficiency virus (HIV) before they develop illnesses that define acquired immunodeficiency syndrome (AIDS) have not been systematically or completely described. In a retrospective and prospective cohort study, 1,073 homosexual and bisexual men in three US cities were interviewed and examined twice per year from January 1988 to September 1992. Study participants were from San Francisco, California (273 HIV-seropositive and 432 HIV-seronegative men), Denver, Colorado (107 positive and 129 negative men), and Chicago, Illinois (54 positive and 78 negative men). A total of 305 HIV-positive men had specifiable dates of HIV seroconversion (mean of 15.3 months between the last negative and the first positive HIV antibody test). Besides much increased incidences of thrush (incidence relative risk (IRR) = 23.3) and hairy leukoplakia (IRR = 551), the following conditions also occurred significantly more frequently in HIV-positive men than in HIV-negative men: anal herpes (incidence density (ID) = 10.7/100 person-years; IRR = 7.7); sinusitis requiring antibiotics (ID = 6.2/100 person-years; IRR = 2.1); anal warts (ID = 5.8/100 person-years; IRR = 2.7); seborrhea (ID = 3.8/100 person-years; IRR = 6.6); community-acquired pneumonia (ID = 1.4/100 person-years; IRR = 2.7); skin cancers (ID = 1.0/100 person-years; IRR = 2.2); and seizures, often apparently "breaking through" prior anticonvulsant therapy (ID = 0.8/100 person-years; IRR = 5.6). First episodes of hairy leukoplakia, thrush, and skin cancer occurred at low mean CD4 counts (mean counts were less than 350 cells/microliters) and late in HIV infection (mean times were more than 8 years after HIV seroconversion). Many medical problems, some not widely appreciated, occur in HIV-infected men before they develop AIDS-defining illnesses, signifying considerable morbidity from pre-AIDS HIV infection.

Use of therapeutic and prophylactic drugs for AIDS by homosexual and bisexual men in three US cities.

OBJECTIVE: To determine the use of AIDS drugs and therapies by populations with relatively good access to health care. DESIGN: Prospective cohort study, with interview and examination twice a year since 1988. SETTING: Two public-health departments (San Francisco Department of Health and Denver Disease Control Service) and a private clinic (Howard Brown Memorial Clinic, Chicago). PARTICIPANTS: HIV-seropositive homosexual and bisexual men in San Francisco (311 men), Denver (120 men) and Chicago (59 men). INTERVENTIONS: HIV counseling and testing at each visit. MAIN OUTCOME MEASURES: Time and duration of use of drugs used for AIDS and Pneumocystis carinii pneumonia (PCP) treatment and prophylaxis. RESULTS: Zidovudine and pentamidine use increased from 1987 through 1989 in all three cities. In San Francisco in 1987, only 17 out of 110 (15%) HIV-seropositive men without AIDS reported taking zidovudine. By 1990, over 90% of AIDS patients and approximately 80% of HIV-seropositive men with low CD4+ cell counts (< 200 x 10(6)/l) had taken zidovudine; most men who by 1990 had never taken zidovudine (82%) or PCP prophylaxis (95%) had not been recommended these therapies because they did not have symptoms and their absolute CD4+ cell counts were > 200 x 10(6)/l. However, overall in the three cities, only 68% of the AIDS patients and 63% of the men with low CD4+ cell counts had taken zidovudine for more than 6 months by 1990. Most men who had stopped taking zidovudine (67%) did so because of toxicity; however, 64% of respondents gave reasons other than drug toxicity as a or the sole reason why they discontinued zidovudine. CONCLUSIONS: AIDS therapeutic and prophylactic drugs were increasingly (and appropriately) recommended to and accepted by these cohorts after 1987, but had limited consistent use because of toxicity, adverse side-effects, and several other less readily appreciated reasons. These data do not indicate that zidovudine use in San Francisco would mainly account for the observed slowing in the rate of increase of AIDS cases in homosexual and bisexual men in this city after 1987.

A comparison of clinical virulence of nosocomially acquired methicillin-resistant and methicillin-sensitive Staphylococcus aureus infections in a university hospital.

OBJECTIVES: To compare the clinical virulence of nosocomially acquired methicillin-resistant Staphylococcus aureus (MRSA) and methicillin-sensitive S aureus (MSSA) infections in 1989. DESIGN: A retrospective comparison of host factors, in-hospital exposures, sites of infections, and outcomes of patients with nosocomial MRSA and MSSA infections. SETTING: University of Illinois Hospital, Chicago, Illinois. PARTICIPANTS: Forty-four adult patients with nosocomial S aureus infections. RESULTS: The 22 MRSA-infected and 22 MSSA-infected persons were similar regarding mean age, gender, underlying diseases, and exposure to surgery. Before developing infection, MRSA-infected persons were more likely to have received antibiotics (73% compared with 27%, odds ratio = 7.1, 95% confidence interval [CI95] = 2.0-25.8 p = .003) and to have stayed in the hospital > 2 weeks (64% compared with 18%, odds ratio = 7.9, CI95 = 2.0-31.6, p = .002). Bacteremia was the most common presentation in the MRSA and MSSA groups (55% and 59%, respectively). Infectious complications and death were infrequent in both groups. CONCLUSIONS: MRSA and MSSA strains infect patients with similar demographic features and underlying diseases, but MRSA infections are significantly more common among patients with previous antibiotic therapy and a prolonged preinfection hospital stay. Clinical presentations and outcomes did not differ significantly between the 2 groups. Thus, similar to studies in the early 1980s, our findings do not suggest greater intrinsic virulence of MRSA.

HIV seropositivity in community-recruited and drug treatment samples of injecting drug users.

OBJECTIVE: To assess the representativeness of drug use treatment samples for measuring HIV seroprevalence among injecting drug users (IDU) in community settings. DESIGN: Seroprevalence was determined in two cross-sectional, convenience samples including an unlinked survey of IDU entrants to all publicly-funded drug-treatment programs and a survey of community-recruited IDU. METHODS: Unconditional logistic regression [odds ratio (OR)] was used to calculate unadjusted and adjusted OR to measure the association between HIV seropositivity and site of recruitment. RESULTS: Between 1988 and 1989, 25% of 870 community-recruited IDU were seropositive, compared with 13% of 671 entrants to drug-treatment programs. This twofold risk of HIV seropositivity among community-recruited IDU remained after adjustment for sample differences in gender, race-ethnicity, and age group (adjusted OR, 2.09; 95% confidence interval, 1.58-2.78). CONCLUSIONS: These results suggest the importance of extending HIV surveillance outside of drug-treatment facilities. Active serologic surveillance may be feasible by coupling recent saliva and fingerstick sampling techniques with existing community outreach education efforts.