The outcome following spinal instrumentation in haemodialyzed patients with pyogenic spondylodiscitis.

AIMS: The aim of this study was to evaluate the outcome of spinal instrumentation in haemodialyzed patients with native pyogenic spondylodiscitis. Spinal instrumentation in these patients can be dangerous due to rates of complications and mortality, and biofilm formation on the instrumentation. PATIENTS AND METHODS: A total of 134 haemodialyzed patients aged more than 50 years who underwent surgical treatment for pyogenic spondylodiscitis were included in the study. Their mean age was 66.4 years (50 to 83); 66 were male (49.3%) and 68 were female (50.7%). They were divided into two groups according to whether spinal instrumentation was used or not. Propensity score matching was used to attenuate the potential selection bias. The outcome of treatment was compared between these two groups. RESULTS: A total of 89 patients (66.4%) underwent non-instrumented surgery and 45 (33.5%) underwent instrumented surgery. There were no significant differences in the rates of postoperative complications, except for an increased rate of wound problems in the instrumented group, which was found in the unmatched cohorts (p = 0.034). There were no significant differences in the rate of recurrent infections (p = 0.328 for the unmatched cohort; p = 0.269 for the matched cohort) and mortality rate, including in-hospital (p = 0.713 for the unmatched cohort; p = 0.738 for the matched cohort) and one-year rates (p = 0.363 for the unmatched cohort; p = 0.787 for the matched cohort), between the groups. However, the interval between the initial diagnosis and the first recurrence was significantly longer in the instrumented group (p = 0.008 for the unmatched cohort; p = 0.032 for the matched cohort). CONCLUSION: Instrumented surgery for haemodialyzed patients with pyogenic spondylodiscitis showed similar outcomes, including recurrence and mortality, to non-instrumented surgery, despite the instrumented group having more severe neurological deficit, a larger number of involved levels, and increased kyphotic angle.

Benzene-di-N-substituted carbamates as conformationally constrained substrate analogs of cholesterol esterase.

Benzene-1,2-, 1,3-, and 1,4-di-N-substituted carbamates (1-15) are synthesized as the constrained analogs of gauche, eclipsed, and anti conformations, respectively, for the glycerol backbones of triacylglycerol. Carbamates 1-15 are characterized as the pseudo substrate inhibitors of cholesterol esterase. Long chain carbamates are more potent inhibitors than short chain ones. Comparison of different geometries for benzene-di-substituted carbamates, such as benzene-1,2-di-N-octylcarbamate (3) (ortho-3), benzene-1,3-di-N-octylcarbamate (8) (meta-8), and benzene-1,4-di-N-octylcarbamates (13) (para-13), indicates that inhibitory potencies are as followed: meta-8 > para-13 > ortho-3. Therefore, we suggest that the preferable conformation for the C(sn-1)-O/C(sn-2)-O glycerol backbone in the enzyme-triacylgycerol complex is the eclipsed conformation. Meanwhile, kinetic data indicate that among ortho, meta, and para carbamates, meta carbamates most resemble the substrate cholesterol ester.

Increased sagittal vertical axis is associated with less effective control of acute pain following vertebroplasty.

OBJECTIVES: Although vertebroplasty is very effective for relieving acute pain from an osteoporotic vertebral compression fracture, not all patients who undergo vertebroplasty receive the same degree of benefit from the procedure. In order to identify the ideal candidate for vertebroplasty, pre-operative prognostic demographic or clinico-radiological factors need to be identified. The objective of this study was to identify the pre-operative prognostic factors related to the effect of vertebroplasty on acute pain control using a cohort of surgically and non-surgically managed patients. PATIENTS AND METHODS: Patients with single-level acute osteoporotic vertebral compression fracture at thoracolumbar junction (T10 to L2) were followed. If the patients were not satisfied with acute pain reduction after a three-week conservative treatment, vertebroplasty was recommended. Pain assessment was carried out at the time of diagnosis, as well as three, four, six, and 12 weeks after the diagnosis. The effect of vertebroplasty, compared with conservative treatment, on back pain (visual analogue score, VAS) was analysed with the use of analysis-of-covariance models that adjusted for pre-operative VAS scores. RESULTS: A total of 342 patients finished the 12-week follow-up, and 120 patients underwent vertebroplasty (35.1%). The effect of vertebroplasty over conservative treatment was significant regardless of age, body mass index, medical comorbidity, previous fracture, pain duration, bone mineral density, degree of vertebral body compression, and canal encroachment. However, the effect of vertebroplasty was not significant at all time points in patients with increased sagittal vertical axis. CONCLUSIONS: For single-level acute osteoporotic vertebral compression fractures, the effect of vertebroplasty was less favourable in patients with increased sagittal vertical axis (> 5 cm) possible due to aggravation of kyphotic stress from walking imbalance.Cite this article: Y-C. Kim, D. H. Bok, H-G. Chang, S. W. Kim, M. S. Park, J. K. Oh, J. Kim, T-H. Kim. Increased sagittal vertical axis is associated with less effective control of acute pain following vertebroplasty. Bone Joint Res 2016;5:544-551. DOI: 10.1302/2046-3758.511.BJR-2016-0135.R1.

Human babesiosis in New York State: Review of 139 hospitalized cases and analysis of prognostic factors.

BACKGROUND: Babesiosis infections are infrequent, occur in limited geographic locations, and range from asymptomatic infection to severe illness and death. METHODS: Descriptive clinical and epidemiological information on human babesiosis cases was collated from state communicable disease reports and medical records of patients hospitalized from 1982 to 1993. Univariate and multivariate analyses were performed to determine prognostic factors associated with severe disease outcome (hospitalization ending in death, duration of hospitalization > 14 days, or intensive care unit stay > 2 days). RESULTS: Between 1982 and 1993, 139 patients were hospitalized with babesiosis in New York State. Nine patients (6.5%) died, 35 (25.2%) were admitted to the intensive care unit, and 35 (25.2%) required hospitalization for more than 14 days. Mean age at first hospitalization was 62.5 years. Sixty-two percent were male, and 91% resided in Suffolk County, Long Island. The most common symptoms were fatigue/malaise/weakness (91%), fever (91%), shaking chills (77%), and diaphoresis (69%). Past medical records showed that 52% of patients had a history of chronic disease; 12% had a history of Lyme disease; 12% had undergone a splenectomy; and 2% had undergone a blood transfusion. There was a 12- to 14-day delay between onset of symptoms and initiation of appropriate antibiotic treatment. Univariate analyses showed alkaline phosphatase levels greater than 125 U/L, white blood cell counts greater than 5 x 10(9)/L, history of cardiac abnormality, history of splenectomy, presence of heart murmur, and parasitemia values of 0.04 or higher to be significantly associated with disease severity. Multiple logistic regression analyses indicated that male sex, alkaline phosphatase values greater than 125 U/L, and white blood cell counts greater than 5 x 10(9)/L remained strong predictors of severe outcome. CONCLUSIONS: Human babesiosis is a rare but debilitating and potentially fatal illness, especially in the elderly. Prompt disease diagnosis and treatment are essential but are often delayed, as seen in our series. This delay reinforces the need for enhanced public and physician education targeted toward residents and visitors to the few high-risk geographic areas where disease and Ixodes scapularis ticks are endemic. Patients presenting with certain prognostic indicators (male sex, alkaline phosphatase values > 125 U/L, and white blood cell counts >5 x 10(9)/L) require comprehensive and aggressive medical care to prevent further deterioration. Since babesiosis is only 1 of 3 currently recognized diseases transmitted by I scapularis ticks, primary prevention recommendations will also reduce human exposure to Lyme disease and human granulocytic ehrlichiosis.

High prevalence of syphilis detected through a jail screening program. A potential public health measure to address the syphilis epidemic.

BACKGROUND: In June 1990, a syphilis initiative was undertaken to help control New York's most extensive syphilis epidemic since the 1940s. METHODS: To evaluate the usefulness of syphilis screening in local jails, were reviewed demographic and syphilis screening data from a county jail in an area with a high prevalence of syphilis that has routinely tested incoming inmates. RESULTS: Of 12,685 inmates, 9797 (77%) were screened for syphilis, and 321 (3.3%) had a positive test result; 258 (80%) of the positive results were confirmed. Data were available for 244 of the inmates with a confirmed positive result: 162 (67%) had newly diagnosed syphilis (overall rate, 1.6%), 112 of whom had early syphilis; 50 (20%) had been previously treated for syphilis; and 32 (13%) were unavailable for follow-up. Of 162 inmates with newly diagnosed syphilis, 122 (75%) were treated in jail, and 40 were treated after their release from jail. The median time from screening to treatment was 17 days. The median jail stay was 45 days for inmates who were evaluated for treatment vs 5 days for those who were unavailable for follow-up. CONCLUSIONS: Screening inmates for syphilis was a productive public health measure, as inmates accounted for 20% of the county's syphilis morbidity. Given the high prevalence of syphilis among inmates and the inability to reach them for treatment after release, strategies are needed to rapidly screen and treat inmates before their release from jail.

Knowledge and attitudes of healthcare workers about influenza: why are they not getting vaccinated?

In a survey of employees at a chronic care psychiatric facility following an influenza outbreak, previous influenza vaccination (relative risk [RR], 69.7; 95% confidence interval [CI95], 25.2 to 192.4), age > or = 50 years (RR, 2.4; CI95, 1.3 to 4.5), and knowledge that vaccine does not cause influenza (RR, 2.3; CI95, 1.3 to 3.7) were the factors most predictive of current influenza vaccination. Medical personnel were less likely than nonmedical employees to be vaccinated (RR, 0.5; CI95, 0.3 to 0.9). Educational efforts to address misconceptions about vaccination plus more vigorous administrative measures to vaccinate employees should be considered.

Biomechanical testing of the lumbosacral spine.

OBJECTIVE: To assess various forms of anterior and posterior sacral fixation and to study the influence of anterior lumbosacral fixation and posterior pedicle fixation at L5 in conjunction with lumbosacral fixation. SUMMARY OF BACKGROUND DATA: Moments at the lumbosacral junction are high in the long constructs requiring lumbosacral fixation. The purpose of this study was to assess bending moments in flexion-extension and lateral bending and rotational forces at the lumbosacral junction involving a variety of long constructs to the lumbosacral junction. The incidence of pseudarthrosis in such constructs in the adult spine literature ranges from 7% to 40%. METHODS: An alignment jig was designed to display three-dimensional motion in the three orthogonal planes. Nine constructs of five specimens each were tested. These consisted of fixation at T12-L5-S1 (construct 1), T12-L5-S1 with anterior L5-S1 fixation and grafting (construct 2), T12-L5-S1, S2 with and without L5-S1 fixation and grafting anterior (constructs 3 and 4, respectfully), T12-S1, S2 with and without L5-S1 anterior grafting and fixation (constructs 5 and 6, respectfully), T12 Jackson intrasacral fixation with or without L5-S1 grafting anteriorly at the anterior fixation (constructs 7 and 8, respectfully), and T12-L5-S1, S2 fixation with anterior grafting only (construct 9). RESULTS: The use of anterior fixation statistically increased stiffness in extension. There was a trend toward increasing stiffness in constructs with anterior fixation (two anterior anterior-oblique L5-S1 screws) and in other loading modes as well. Failure to use L5 screw fixation significantly decreased torsional rigidity in long constructs without anterior fixation. CONCLUSIONS: In long constructs, particularly in scoliosis surgery requiring lumbosacral fixation, the addition of anterior fixation at L5-S1 is recommended. The addition of L5 fixation in addition to sacral fixation significantly decreases rotational stresses and is recommended as well.

A retrospective cohort study of cancer incidence among New York State Farm Bureau members.

Cancer incidence from 1973 through 1983 in 18,811 New York Farm Bureau members was examined using a retrospective cohort study design. The observed number of cancers for all age groups was 72% of the expected, and the major deficits in incidence occurred for lung (52% of expected), gastrointestinal (67% of expected), and bladder (78% of expected). Similar deficits have been reported by other researchers. Unlike other studies, we did not find a significant excess of cancer of any site. Given the healthy worker effect and the small numbers of incident tumors at some sites, the Standardized Cancer Incidence Ratios that were over 100 in value (i.e., lip, melanoma of the skin, prostate, multiple myeloma) merit further investigation. This study differs from previous research in population, setting, and method. Nonetheless, the general pattern of results is consistent with the findings of other investigations.

A retrospective cohort study of mortality among New York State Farm Bureau members.

A retrospective cohort study was conducted to examine mortality among 18,811 male farm owners and operators in New York State from 1973-1984. Farm Bureau membership lists were used to identify the study population, and vital status was determined through record linkage with death certificate and motor vehicle files. The comparison group consisted of the 1980 United States Census population of men who resided in the same towns as did the farmers. The results indicated that the study cohort experienced fewer than the expected numbers of deaths overall and for each major cause category except accidents. Specific causes with significant mortality deficits included cancer of the lung (standardized mortality ratio [SMR] = 47.0); diabetes mellitus (SMR = 57.5); ischemic heart disease (SMR = 65.3); bronchitis, emphysema, and asthma (SMR = 26.7); and cirrhosis of the liver (SMR = 29.7). The only specific cause with a significantly elevated mortality was accidents other than motor vehicle (SMR = 146.5). The investigation differs from previous research in method, setting, and population, but the pattern of findings is generally consistent with that of other studies.

Estimating the mortality rate of hepatitis C using multiple data sources.

The New York State hospital discharge database and the multiple cause-of-death file were used to estimate the mortality rate of hepatitis C in New York State excluding New York City in 1997. The mortality rate with hepatitis C was severely underestimated when each data source was used alone. Applying the capture recapture method using the hospital discharge database and the multiple cause-of-death file appears to be an efficient method to estimate the mortality rate with hepatitis C.

Characteristics that predict risk of breast cancer before and after the menopause.

In a case-control study of 1868 breast cancer patients and 3391 control patients we searched for characteristics that predicted risk of breast cancer diagnosed before and after menopause. Common to increased risk of this disease in both periods of womanhood were: early menarche and late menopause; delayed marriage and first childbirth; more nulliparity or reduced gravidity and parity; reduced frequency of abortions; shorter overall child-bearing interval; more advanced education, higher socioeconomic status, and more contraceptive usage; and familial tendencies toward the disease. Breast cancer patients diagnosed before menopause were leaner than controls at age 20 and at time of diagnosis, but breast cancer risk in the postmenopausal period was related to increased weight-for-height at diagnosis and greater weight-for-height at diagnosis and greater weight gain since age 20. Postmenopausal breast cancer patients had a longer interval between first and second childbirths. Frequency and duration of the gravid state, inversely related to breast cancer risk, were largely dependent on contraceptive practices rather than unexplained infertility per se. Whether the breast cancer reaches diagnosis before or after menopause, the bulk of evidence examined here supports the view that it has a common cause and is subject to modifying influences over the long period of cancer latency.

PIP: Case-control studies are reported of 1868 breast cancer patients and 3391 unaffected women from the San Francisco Bay Area that identify clues to the etiology of breast cancer and distinguish characteristics of patients with premenopausal, paramenopausal, or postmenopausal dates of 1st diagnosis. Patients with breast cancer diagnosed during 1970-1977 were interviewed at home, as were twice as many control patients from the same hospitals matched to the cancer patients by age (within 5 years), race (white or black), and time of hospitalization (within 7 months). Rates of participation through interview were about 70%, somewhat higher for breast patients than for control patients. Common to increased risk of breast cancer in periods of womanhood were: early menarche and late menopause; delayed marriage and 1st childbirth; more nulliparity or reduced gravidity and parity; reduced frequency of abortions; shorter overall childbearing interval; more advanced education, higher socioeconomic status, and more contraceptive usage; and familial tendencies toward the disease. Breast cancer patients diagnosed before menopause were leaner than controls at age 20 and at time of diagnosis, but breast cancer risk in the postmenopausal period was related to increased weight-for-height at diagnosis and greater weight gain since age 20. Postmenopausal breast cancer patients had a longer interval between 1st and 2nd childbirths. Frequency and duration of the gravid state, inversely related to breast cancer risk, were largely dependent on contraceptive practices rather than unexplained infertility per se. Whether the breast cancer reaches diagnosis before or following menopause, the bulk of the evidence examined supports the view that it has a common cause and is subject to modifying influences over the long period of cancer latency.

Epidemiologic characteristics of breast cancer in three menopausal stages: preliminary findings.

A preliminary analysis of data from 1,477 patients with breast cancer and 2,695 control women examined potential characteristics of high risk for breast cancer to determine whether prediction differed depending on whether the women were premenopausal, paramenopausal (intermediate), or postmenopausal. Since information was not yet available on real or presumed menopausal status, women were characterized by age: women less than 45 years as premenopausal; women from 45 through 54 years as paramenopausal; and women 55 years and older as postmenopausal. Height, weight at age 20, number of stillbirths, number of abortions, and history of cancer of any site in the subject's mother did not affect the risk of breast cancer. Generally, highest level of education, number of pregnancies, and number of live births seemed to predict risk of breast cancer. Weightbody mass index, age at menarche, and age at first pregnancy may predict breast cancer risk differently depending on the menopausal stage in which onset occurs.

Health effects of low-level exposure to polychlorinated biphenyls.

A polychlorinated biphenyl (PCB) spill resulting from a transformer explosion in Syracuse, New York, with no subsequent fire, provided an opportunity for the examination of the effects of low-level PCB exposure without the confounding presence of furans and dioxins. The incident provided 52 individuals exposed to PCB among building personnel, police, firemen, and public utility employees. Sixty-eight nonexposed were matched to the exposed group by sex, age, employer, and job description. Data were collected on the exposed relative to their activities at the spill site, their location, possible routes of exposure, duration of exposure, and subsequent health effects. Exposed and nonexposed were interviewed for past medical history and relevant symptoms. Blood chemistries were studied inclusive of SGOT, SGPT, total protein, CBC, cholesterol, and triglyceride levels, as well as a fasting blood PCB level measurement. Six weeks after the spill, exposed and nonexposed were reinterviewed and had their blood work repeated except for the CBC and PCB levels. Exposed and nonexposed laboratory results were unremarkable. Some transient skin irritation believed to be associated with PCBs was noted. There were significant PCBs in blood level trends for occupation, age, duration of exposure, and level of alcohol consumption. Triglyceride level was highly correlated with PCB level. This relationship held when age and alcohol consumption were controlled for.

Intussusception, rotavirus diarrhea, and rotavirus vaccine use among children in New York state.

OBJECTIVE: To describe epidemiologic features of intussusception and rotavirus diarrhea in New York, to examine the baseline incidence and trends over time, and to ascertain whether an excess of cases occurred in the 9 months of vaccination with the newly licensed rotavirus vaccine. METHODS: Hospital discharge data from 1989 through 1998 were reviewed for children (<1 year old) whose primary or secondary diagnosis was coded as intussusception or rotavirus diarrhea. Characteristics of patients admitted for intussusception and rotavirus diarrhea were compared, and trends over time were examined. For a subset of patients, medical records and vaccine histories for intussusception hospitalizations from October 1998 through June 1999 were analyzed. The number of intussusception cases attributable to rotavirus vaccine was calculated based on the penetration of the vaccine (21%) and a range of excess risks of intussusception among vaccinated children as estimated by the National Immunization Program (NIP). RESULTS: From 1989 through 1998, 1450 intussusception-associated hospitalizations were reported in children <1 year old (average annual incidence 5.4/10 000). Among these children, 47% were treated medically and 53% had surgery, with 9% needing surgical resection. The incidence of intussusception declined over time from 6.1 per 10 000 in 1989 to 3.9 per 10 000 in 1998. Intussusception hospitalizations occurred throughout the year, whereas rotavirus-associated hospitalizations peaked from February to April. Of 20 patients with intussusception whose hospitalization charts were reviewed, 5 had received rotavirus vaccine. All 5 were hospitalized after their first dose of vaccine, were admitted before 7 months of age, were white, and had private insurance. A total of 81 cases of intussusception occurred during the 9-month period of rotavirus vaccination, compared with 78 during the same period in the prevaccination year. The number of excess intussusception cases observed (n = 3) was lower than expected using the NIP estimate of excess risk (1.8) among rotavirus vaccinated children (n = 12) but not significantly different from the risks identified in the NIP cohort studies (1 in 12 000). CONCLUSION: Our data suggest that in New York the rate of intussusception has declined, and approximately 1 child in 2600 develops intussusception before 1 year of age. The different seasonality between intussusception and rotavirus-related hospitalizations suggests that if any causal association exists, it must be small. Unlike other studies, analysis of New York hospitalized discharge data failed to show an appreciable increase in the incidence of intussusception after introduction of the rotavirus vaccine.

Coccidioidomycosis in New York State.

Coccidioidomycosis, a systemic fungal disease caused by Coccidioides immitis, is endemic in the southwestern United States and in parts of Mexico and Central and South America. Only sporadic cases have been reported in areas (including New York) where the disease is not endemic. We used hospital discharge records and state mycology laboratory data to investigate the characteristics of C. immitis infections among New York State residents. From 1992 to 1997, 161 persons had hospital discharge diagnoses of coccidioidomycosis (ICD9 Code 114.0 - 114.5, 114.9). From 1989 to 1997, 49 cultures from patients were confirmed as C. immitis; 26 of these patients had traveled to disease-endemic areas. Fourteen of 16 isolates had multilocus genotypes similar to those of Arizona isolates, which corroborates the travel-related acquisition of the disease. Our results indicate that coccidioidomycosis may be more common in New York residents than previously recognized. Increased awareness among health-care providers should improve timely diagnosis of coccidioidomycosis and prevention of associated illnesses and deaths among patients in nondisease-endemic areas.

Assessment of AIDS knowledge in selected New York State sexually transmitted disease clinics.

A 1988 survey regarding the level of knowledge about acquired immunodeficiency syndrome (AIDS) among 505 clients of sexually transmitted disease (STD) clinics demonstrated 80% correct responses. However, 13% thought AIDS could be transmitted via sharing utensils, and 51% were unaware that acquiring another STD could increase one's risk of human immunodeficiency virus (HIV) infection. In addition, only 10% always used condoms. An average 69% of patients obtained their AIDS information from newspapers, magazines, television, or radio; and 30% acquired information from medical or clinic staffs. In response to this survey, a videotape on AIDS was developed to show to patients of STD clinics while they are waiting to see a physician.

Application of the Cox model as a predictor of relative risk of coronary heart disease in the Albany Study.

Patients in long term studies of coronary heart disease may have different levels of risk during the course of study. Smoking habits, blood pressure, and obesity may change drastically during this period. The multiple logistic model, the most commonly used model for the analysis of coronary heart disease studies, does not consider survival time in assessment of the dependent covariates and does not account for the censoring which usually occurs in such studies. We propose a Cox model with time-dependent covariates to model the risk of coronary heart disease in the Albany study. The Cox model we fitted evaluates the patients' risk on the basis of the data at the last visit. With this methodology, we can evaluate whether it is advantageous for individuals to modify their risk of disease by their effecting changes in their covariates, that is to stop smoking, lose weight, change diet and so on. The important covariates that explain the risk of coronary heart disease were the same in our model as in the models used in the earlier reports. The estimated relative risks were slightly higher in most cases and lend more support to the need to encourage patients to achieve a better covariate state.

Typhoid fever at a resort hotel in New York: a large outbreak with an unusual vehicle.

The largest outbreak of typhoid fever in the United States since 1981 occurred in 1989 among guests and staff at a New York hotel. There were 43 culture-confirmed and 24 probable cases among guests, 1 culture-confirmed case and 1 asymptomatic culture-positive case among hotel employees, and 1 culture-confirmed secondary case. Twenty-one persons were hospitalized and 2 had bowel perforation. Breakfast on 13 June was the only meal consumed by all ill persons (relative risk, infinite; P = .004). In a case-control study, case-patients were more likely than controls to have consumed orange juice (odds ratio, 5.6; 95% confidence interval, 1.1-54.7), which had been prepared in a 208-L container with ample opportunity for hand contact. No other food was associated with illness. S. typhi was isolated from the stool of an asymptomatic food worker who handled orange juice but who was not known to be a typhoid carrier. S. typhi is a foodborne pathogen with continuing potential to cause large outbreaks in the United States.

Survival and mortality patterns of an acquired immunodeficiency syndrome (AIDS) cohort in New York State.

The survival experience and causes of death of acquired immunodeficiency syndrome (AIDS) patients were studied using a cohort of 3,699 AIDS patients in New York State, excluding New York City, whose illness was diagnosed before January 1990 at age 13 years or older. The median length of survival for all cases was 11.5 months, and survival increased over time from 5.3 months pre-1984 to 9.3 months in 1984-1986 and to 13.2 months in 1987-1989. In a Cox proportional hazards model, risk of dying was higher for persons aged 35 years or more at diagnosis and for persons with a diagnosis other than Pneumocystis carinii pneumonia or Kaposi's sarcoma whose illness was diagnosed before 1986. In this AIDS cohort, 2,834 (77 percent) persons died before 1991; 87 percent of the death certificates listed human immunodeficiency virus (HIV)/AIDS or an AIDS indicator disease as one of the multiple causes of death. The finding that 13 percent of the death certificates did not mention AIDS/HIV suggests that use of death certificates alone to count HIV-related deaths would result in an undercount. The recent expansion of the federal AIDS case definition is expected to add HIV-infected persons who die from conditions, such as recurrent pneumonia, that were not included in the earlier definition.