Methods

Top Methods Results Discussion Author & Article Info References

Patients

We identified six HIV-seronegative patients with self-reported HIV infection who were followed in the Boston City Hospital Clinical AIDS Program during a 10-month period in 1992 and 1993; one patient had been previously identified in 1990. Boston City Hospital does not routinely document HIV infection before starting medical care and does not release a hard copy of the HIV antibody test result for the medical record.

Patient serum samples were tested for antibodies to HIV-1 by the Abbott HIV-1 enzyme immunoassay (Abbott Park, Illinois); p24 antigen was tested by the Coulter antigen-capture assay (Hialeah, Florida). In patients 4 and 5, we used culture and polymerase chain reaction to test for HIV-1 [18].

Case-Control Study

We compared 7 patients with self-reported, undocumented HIV infection with 70 controls who were randomly selected by computer from approximately 500 clinic patients receiving care for HIV infection over the same time period. We reviewed the medical records to collect data on demographic characteristics, risk behavior, CD4 counts, and psychosocial characteristics, such as history of chronic depression and previous suicide attempt. "Chronic depression" was defined as the use of antidepressant drugs or clear documentation of depression in the medical record by a provider from the mental health team. This study was approved by the Human Studies Committee, Department of Health and Hospitals of the City of Boston.

Survey of Hospital Practices on Documenting HIV Infection

Physicians working with HIV-infected patients at 10 other hospitals in the Boston area completed a survey that contained questions on local hospital practices for generating a hard copy of the HIV test result for the medical record, documenting HIV infection before instituting care, and the number of recognized cases of factitious HIV disease. Results were confirmed by telephone interview.

Comparisons between the case patients and controls were done using the mean ±SD and the Student t-test; we used the Fisher exact test to compare discrete variables. We calculated odds ratios and 95% CIs by standard methods [19].

Results

Top Methods Results Discussion Author & Article Info References

Patients with Factitious HIV Infection

The seven documented HIV-seronegative patients received care in the Clinical AIDS Program for a mean of 9.2 months (Table 1). Three of the seven patients were identified as HIV seronegative by routine HIV screening test for entry into an AIDS clinical trial (patients 2, 3, and 6); four were identified because of persistently high CD4 counts or lack of previous documentation of the HIV test result (patients 1, 2, 5, and 7); and one was identified because of suspicion of factitious HIV disease despite documentation of a previous positive HIV test result (patient 4). Several patients reported that they believed their blood was routinely checked for evidence of HIV infection when they entered the HIV clinic.

All of the patients in this study gave a history of a positive HIV antibody test result from another site. Three patients were tested at an anonymous test site, 2 while incarcerated, 1 at a drug treatment program, and 1 at a neighborhood health center. Only the HIV test result in patient 4 could be documented, and although this result was positive, no personal identifiers were present on the hard copy; subsequent test results for HIV antibodies, p24 antigen, and HIV culture were negative.

Medical and Psychosocial Problems

Six of the seven patients with factitious HIV infection gave a history of multiple, nonspecific medical symptoms that were consistent with HIV infection. Four patients had a history of viral hepatitis, 3 reported oral thrush, 2 had the HIV wasting syndrome, and 1 (patient 4) gave a history of Pneumocystis carinii pneumonia.

Three of the 7 patients seronegative for HIV had been incarcerated, 5 were unemployed, 1 was living in a half-way house, and 2 were intermittently homeless. Four of the patients were actively seeking disability benefits, 2 had been given expedited referrals for drug treatment, and 6 had increased access to prescription drugs because of their alleged HIV infection.

Four of the 7 patients with factitious HIV disease had chronic depression, and 3 had attempted suicide. Two of the patients reported grief over the death of their HIV-infected sexual partner; 2 patients acknowledged physical or emotional abuse and expressed homicidal ideation toward their HIV-infected sexual partner. Patient 4 had a history of factitious illness, gave an elaborate history of treatment and prophylaxis for Pneumocystis carinii pneumonia, but had a CD4 count of 1310 cells/mm³, suggesting a diagnosis of the AIDS Munchausen syndrome.

Case-Control Study

The seven patients with factitious HIV infection had a lower mean age than the 70 controls (32.0 ±4.8 years compared with 37.9 ±6.9 years; P = 0.07) and were more likely to be white (71.4% compared with 37.1%; odds ratio, 4.7; P = 0.08) and a homosexual or bisexual man (42.8% compared with 11.4%; odds ratio, 5.8; P = 0.06). No differences were noted for the number of injection drug users (71.4% compared with 70.0%) in each group. Patients seronegative for HIV were more likely to have an HIV-infected sexual partner (85.7% compared with 28.6%; odds ratio, 15.0; P = 0.005) and higher absolute CD4 counts (782 ±290 cells/mm³ compared with 263 ±235 cells/mm³; difference, 519; P < 0.001) with a median of 740 cells/mm³ and 230 cells/mm³, respectively. Seronegative patients were also more likely to have a history of chronic depression (57.1% compared with 31.4%; odds ratio, 2.9; P = 0.2) or previous suicide attempt (42.8% compared with 7.1%; odds ratio, 9.8; P = 0.02).

Survey of Hospital Practices on Documenting HIV Disease

All 10 of the other hospitals surveyed gave hard copies of the HIV antibody test result to the ordering physician, but only 7 placed a hard copy of the result in the patient's medical record (Table 2). Self-reported HIV infection was accepted unless the CD4 count was high or factitious HIV disease was suspected; only hospital H had a policy of routinely documenting HIV infection in all patients before initiating care. Cases of factitious HIV infection had been documented in 8 of the 10 other hospitals surveyed.

Discussion

Top Methods Results Discussion Author & Article Info References

We described seven patients with factitious HIV infection who were followed in our HIV clinic for 2 to 29 months. Because we could not document the initial HIV antibody test result for six of the patients, the exact cause of their self-reported HIV disease could not be determined. Possible causes include a false-positive HIV antibody test result, confusion over the initial HIV test result, factitious disease, depression, or survivor's guilt related to the illness or death of a loved one with AIDS [7-17]. Of note, several patients reportedly believed that their HIV infection was documented on admission to the HIV clinic.

Although the frequency of false-positive HIV test results has been reduced by improvements in the enzyme immunoassay and Western blot confirmation [13], false-positive HIV test results are still possible. The widespread use of anonymous HIV testing preserves confidentiality but makes accurate documentation of HIV infection difficult [1, 5, 6]. However, as shown in our study, even when confidential HIV testing is used, documentation may be difficult because a hard copy of the HIV test result is not present in the medical record [2, 3, 20].

The seven patients with factitious HIV disease usually had a CD4 count greater than 500 cells/mm³, and six had medical symptoms consistent with HIV infection. In addition, many had problems with addiction, depression, anxiety, guilt, and serious psychiatric disease. Furthermore, HIV-seronegative persons living, caring, or grieving for loved ones with AIDS may have delusions of HIV infection or may assume that they too must have HIV infection [7, 8].

Because patients with AIDS often have preferred access to drug treatment, prescription drugs, social security disability insurance, housing, and comprehensive medical care, the rate of malingering may increase and reach extremes, such as the recently described AIDS Munchausen syndrome [9, 10].

Our data suggest several possible causes for factitious HIV infection, and the survey results indicate that this problem may be more widespread than currently appreciated. Because of the limitation of the current HIV anti-body tests, use of anonymous HIV testing, and recent re-ports of a lymphocyte depletion syndrome that may mimic AIDS [21], we urge health care providers to clearly document the presence of HIV infection before initiating care.

Presented in part at the 33rd Interscience Congress on Antimicrobial Agents and Chemotherapy, New Orleans, Louisiana, October 1993.