Human Babesiosis in New York State: Review of 139 Hospitalized Cases and Analysis of Prognostic Factors | Infectious Diseases | JAMA Internal Medicine | JAMA Network
[Skip to Navigation]
Figure 1. 
Number of reported (by year of onset) and hospitalized babesiosis cases, New York State, 1982-1993. (Babesiosis became reportable January 1, 1986.)

Number of reported (by year of onset) and hospitalized babesiosis cases, New York State, 1982-1993. (Babesiosis became reportable January 1, 1986.)

Figure 2. 
Number of hospitalized babesiosis cases by month of admission, New York State, 1982-1993.

Number of hospitalized babesiosis cases by month of admission, New York State, 1982-1993.

Table 1. 
Demographic Characteristics of 139 Patients Hospitalized With Babesiosis in New York State, 1982-1993
Demographic Characteristics of 139 Patients Hospitalized With Babesiosis in New York State, 1982-1993
Table 2. 
Admission Signs and Symptoms of 139 Patients Hospitalized With Babesiosis in New York State, 1982-1993
Admission Signs and Symptoms of 139 Patients Hospitalized With Babesiosis in New York State, 1982-1993
Table 3. 
Admission Laboratory Values of 139 Patients Hospitalized With Babesiosis in New York State, 1983-1993
Admission Laboratory Values of 139 Patients Hospitalized With Babesiosis in New York State, 1983-1993
Table 4. 
Complications Observed in 139 Patients Hospitalized With Babesiosis in New York State, 1982-1993
Complications Observed in 139 Patients Hospitalized With Babesiosis in New York State, 1982-1993
Table 5. 
Univariate Analysis of Demographic and Laboratory Clinical Findings in 139 Patients Hospitalized With Babesiosis, by Disease Severity, in New York State, 1982-1993*
Univariate Analysis of Demographic and Laboratory Clinical Findings in 139 Patients Hospitalized With Babesiosis, by Disease Severity, in New York State, 1982-1993*
Table 6. 
Relative Risks (RR) of Prognostic Factors in 139 Patients Hospitalized With Babesiosis, by Multiple Logistic Regression, in New York State, 1982-1993
Relative Risks (RR) of Prognostic Factors in 139 Patients Hospitalized With Babesiosis, by Multiple Logistic Regression, in New York State, 1982-1993
Original Investigation
October 26, 1998

Human Babesiosis in New York State: Review of 139 Hospitalized Cases and Analysis of Prognostic Factors

Author Affiliations

From the Division of Infectious Disease, Wadsworth Center, New York State Department of Health (Drs White and Morse), the Bureau of Communicable Disease Control (Drs Talarico, Birkhead, and Heimberger and Ms Chang), and the Department of Epidemiology, School of Public Health, University at Albany (Drs White, Talarico, Birkhead, and Morse), Albany, NY. Dr Heimberger is now with Abbott Laboratories, Abbott Park, Ill.

Arch Intern Med. 1998;158(19):2149-2154. doi:10.1001/archinte.158.19.2149

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 5×109/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 5×109/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 >5×109/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.

HUMAN BABESIOSIS is a tick-borne malarialike illness caused by Babesia species, which are intraerythrocytic protozoa.1 Although at least 3 species are known to cause infections in humans, Babesia microti is the agent responsible for the majority of human cases of babesiosis in the United States.1-5 Animal reservoirs for B microti in the United States include both the white-footed mouse (Peromyscus leucopus) and the meadow vole (Microtus pennsylvanicus).4,6 The deer tick, or black-legged tick, Ixodes scapularis, is the vector responsible for transmission of babesiosis to humans, and tick bites are the primary mode of transmission in the United States.1,4,7

Infection with B microti causes an acute febrile illness and hemolytic anemia.2,3 The clinical course varies in its severity, but is characterized by fever, myalgias, anorexia, and, frequently, jaundice.8,9 Although babesiosis was once thought to be a self-limited disease,10,11 case reports of severe hemolysis requiring blood transfusions, renal failure, adult respiratory distress syndrome, and death have been associated with human babesiosis.1,5 Previously described risk factors include asplenia and older age, although seroprevalence studies in Connecticut have shown that children have seroprevalence rates that are similar to those of adults.12-15 Other factors, including immunodeficiency syndromes and liver dysfunction,10 may play a role in the severity of the disease.

Human cases of babesiosis have been reported in the United States since 1966,16 and although the disease is not reportable to the Centers for Disease Control and Prevention, it has been a reportable illness in New York State (NYS) since 1986.17,18 While there have been numerous individual case reports describing outcomes, geographic distribution, and the clinical and demographic trends associated with babesiosis,6,7,10,15,17,19-22 we report on the largest series of hospitalized patients with babesiosis. In this study, we examined the hospital records of babesiosis cases in NYS from 1982-1993. The objective of this study was to review the epidemiological and clinical characteristics of hospitalized patients with babesiosis and to document prognostic indicators that may be associated with severe disease.

Notifiable disease surveillance for human babesiosis

Human cases of B microti infection became legally reportable from physicians, laboratories, and hospitals in NYS in January 1986. Cases are reported to county health departments and the NYS Department of Health (NYSDOH). Additional cases were voluntarily reported from 1983-1986, when the NYSDOH offered reference laboratory services to confirm cases. The case definition requires a clinically compatible illness (may include fever, chills, fatigue, myalgia, and jaundice secondary to hemolytic anemia lasting several days to a few months), with the documentation of B microti parasites on a peripheral thin blood smear or immunofluorescent antibody titers of 1:128 or higher. Confidential case report forms are completed by physicians and include basic demographic, clinical, and laboratory data. Only those patients reported as being hospitalized in NYS between 1982 and 1993 were analyzed for the purposes of this study.

Statewide planning and research cooperative system

The Statewide Planning and Research Cooperative System (SPARCS) provides information on the discharge diagnosis of individuals hospitalized for specific communicable diseases based on International Classification of Diseases, Ninth Revision (ICD-9) disease codes. The SPARCS database of patients hospitalized between 1982 and 1993 was reviewed for all babesiosis cases, including ICD-9 codes of 088.8 (other arthropod-borne disease, unspecified) and 136.8 (other specified infections and parasitic disease). Babesiosis cases in the notifiable disease and SPARCS databases were then matched by name, address, and date of birth. Medical records for matched cases were requested from hospitals and reviewed for babesiosis case confirmation as defined above.

Descriptive analysis

All hospital charts of patients meeting the case definition were reviewed. Demographic data and clinical history, including symptoms, medical and medication history, and information about tobacco and alcohol use, were recorded. Physical examination findings on admission, laboratory data, treatment modalities, and complications were also collected. Laboratory data included complete blood cell counts, serum glucose and electrolyte levels, arterial blood gas data, liver function study results, serum and urinary bilirubin levels, serum urea nitrogen and creatinine levels, and chest radiographs; the lowest values of hematocrit, white blood cell count, and platelet levels; and the highest reticulocyte count and serum glucose, creatinine, bilirubin, aspartate aminotransferase, lactate dehydrogenase, and parasitemia levels. Clinical cutpoints for normal laboratory values used in the analysis were determined using standard clinical laboratory references.23-29

Prognostic factors for severe outcome

Hospitalized patients were classified as having mild or severe disease as the outcome variable. Severe disease was defined as a hospitalization ending in death, lasting for more than 2 weeks, or resulting in an intensive care unit stay of more than 2 days. In addition, the NYSDOH death certificate registry was reviewed to identify deaths after hospitalization during the 1982-1995 period by matching on the patient's name, date of birth, and street address. Cases with a reported date of onset were analyzed to determine whether the duration of time between onset and treatment was associated with a severe outcome.

Statistical analysis

Statistical analyses were conducted with SAS software (SAS Inc, Cary, NC) for χ2 tests and BMDP software (California-Princeton Fulfillment Service, Ewing, NJ) for logistic regression analyses. χ2 Tests were performed to evaluate the relationship between potential risk factors and disease outcome. Univariate relative risks (RRs) and 95% confidence intervals (CIs) were computed. All variables with P values of .05 or less, age at admission, and sex were included in a stepwise multiple logistic regression model to estimate the adjusted RRs of prognostic variables associated with severe outcome.

Notifiable disease surveillance for babesiosis and sparcs data

From 1982 to 1993, 186 cases of babesiosis were reported to the NYSDOH. Among the reported cases, 144 patients (77.4%) were hospitalized (Figure 1). Medical records were obtainable for 139 (96.5%) of the hospitalized patients.


Demographic characteristics of the hospitalized patients showed the mean±SD age on admission to be 62.5±15.0 years (median age, 66 years; range, 23-89 years) (Table 1). Patients were predominantly male (61.6%) and white (94.2%), and 94.9% resided in Suffolk (90.6%), Nassau, or Westchester counties. The majority of patients were admitted during June through August, with very few admissions during winter (Figure 2). The mean±SD hospital stay was 11.7±11.0 days (median, 9 days; range, 2-100 days). Sixteen (30.8%) of 52 patients with severe outcome were hospitalized during the first 6 years of the study interval, as opposed to 36 (69.2%) during the last 6 years (Figure 1).

Clinical features
Admission Status

The most common signs and symptoms on admission were nonspecific symptoms of fatigue/malaise/weakness (91.2%), fever (90.6%), shaking chills (76.6%), diaphoresis (69.2%), and nausea/anorexia (57.3%) (Table 2). Clinical findings on physical examination included temperature higher than 38°C (55.8%), heart murmur (20.1%), hepatomegaly (14.4%), and splenomegaly (10.8%). Only 6 patients (4.3%) had jaundice listed. A review of medical records indicated that 51.8% had a history of chronic disease, 11.9% had a history of Lyme disease, and 11.7% had undergone splenectomies. No patient had evidence of human immunodeficiency virus infection. More than one third (37.5%) of the patients reported having a tick bite within 30 days prior to their hospitalization.

The most common laboratory abnormalities on admission are shown in Table 3. The mean hematocrit was 0.34 and the mean hemoglobin was 113 g/L. Fifty-nine percent of the patients had hematocrit levels lower than 0.35, 35% had total white blood cell counts lower than 5×109/L, and 60.5% had platelet counts lower than 100×109/L. Liver function studies on admission showed elevations of mean levels of lactate dehydrogenase (572 U/L), aspartate aminotransferase (80 U/L), alkaline phosphatase (113 U/L), and total bilirubin (296 µmol/L [17.3 mg/dL]).

Hospital Course

During hospitalization, the mean hematocrit of the patients decreased to 0.27, with 90.3% decreasing to less than 0.35. The mean change from hematocrit on admission to the lowest value during hospitalization was 6.1 percentage points. Total mean bilirubin values increased from 296 to 361 µmol/L [17.3-21.1 mg/dL]. Thirty-five patients (25.2%) were admitted to intensive care units during hospitalization, and 8 (5.8% of total cases) required intubation. The majority of treated patients received clindamycin hydrochloride (n=110, 79.1%) and quinine sulfate (n=106, 76.3%). For persons reporting onset dates, the average time from onset until hospital admission was 10.6 days; until the start of treatment with clindamycin, it was 12.1 days; and until treatment with quinine, it was 14 days. Of those patients with available treatment dates (n=110, 79.1%), 7 (5%) started recommended treatment before hospital admission, 57 (41%) on the day of admission, and 46 (33%) only later in their hospital course (average, 3.2 days after admission).


Fifty-four hospitalized patients (38.8%) had some type of complication (Table 4). Congestive heart failure (n=15, 10.9%) and acute respiratory distress syndrome (n=11, 8.0%) were the most common complications. Nine patients died during hospitalization, for a fatality rate of 6.5%. The death certificate analysis identified 29 patients who died from 1982-1995. The mean duration from date of hospital admission to date of death was 3.6 years. Nine patients (31%) had cancer listed as the underlying cause of death, 6 patients had coronary heart disease, and only 1 patient had babesiosis.

Prognostic Factors for Severe Outcome

Univariate analysis of laboratory results and clinical evaluations showed the following factors (Table 5) to be associated with disease severity: parasitemia values of 0.04 or greater (RR, 2.48; 95% CI, 1.36-4.54), white blood cell counts higher than 5×109/L (RR, 2.11; 95% CI, 1.17-3.83), alkaline phosphatase values greater than 125 U/L (RR 2.10; 95% CI, 1.28-3.44), prior cardiac abnormalities (RR, 1.67; 95% CI, 1.10-2.53), history of splenectomy (RR, 1.80; 95% CI, 1.15-2.83), and presence of heart murmur (RR, 1.61; 95% CI, 1.04-2.48).

Stepwise logistic regression analysis for each significant variable in the univariate analysis, age at admission, and sex indicated that male sex, alkaline phosphatase values higher than 125 U/L, and white blood cell counts higher than 5×109/L remained strong predictors of severe outcome (Table 6).

The duration of time between onset and treatment in those cases that were not associated with a severe outcome (mean±SD, 10.7±8.0 days; range, 0-32 days) was shorter, but not significantly different (P=.26) from the duration of time between onset and treatment in those cases in which there was a severe outcome (13.3±13.9 days; range,1-72 days).


This study represents the largest case series of patients hospitalized with clinical babesiosis ever reported. Babesiosis is 1 of at least 3 diseases (including Lyme disease and ehrlichiosis) transmitted by the deer tick, or black-legged tick, I scapularis. While most exposures of healthy individuals to deer ticks infected with B microti may result in asymptomatic or mild infections, elderly or immunocompromised individuals may face a more serious, debilitating infection.17 The need for enhanced public and physician education that is targeted toward high-risk patients and visitors to areas where I scapularis ticks and disease are endemic is underscored by the observation that diagnosis and treatment were delayed by an average of 12 to 14 days between onset and appropriate antibiotic treatment.

The current study identified demographic, clinical, and laboratory criteria that were associated with increased relative risks for severe disease outcome, using death, hospitalization for more than 14 days, or a stay in the intensive care unit for longer than 2 days to define severe outcome. Approximately 73% of these hospitalized patients were older than 55 years, and almost 62% were male. Results from univariate analyses indicated that patients with parasitemia values in excess of 0.04, with white blood cell counts in excess of 5×109/L, and alkaline phosphatase values higher than 125 U/L had a 2-fold greater RR of a severe outcome. Individually, hospitalized patients with a history of splenectomy, a history of cardiac abnomalities, and the presence of heart murmur were associated with RRs of 1.8, 1.67, and 1.61, respectively, compared with hospitalized patients with babesiosis without these conditions.

Data analyzed by backward stepwise multiple logistic regression indicated that male patients were more than 4 times more likely to have a severe outcome. Risk factors associated with the male sex were shown to be greater than the RR associated with parasitemia values higher than 0.04 (RR, 2.32), but not as great as the RR associated with elevated alkaline phosphatase values (RR, 7.72). These indicators for disease severity in the patient with babesiosis may be useful to the medical provider in assessing levels of care necessary to manage the infection. Decisions to provide more aggressive therapy may be justifiable in the presence of these prognostic indicators. We did not find age to be a risk factor for severe outcome, perhaps because our study included only hospitalized patients, who may have been already selected for severity of disease. Also, while there was a 2.5-day difference, we did not find a significant association between the duration of time from disease onset to treatment and the development of severe outcome. This suggests that underlying physical conditions of the patient, rather than prompt institution of appropriate therapy, may contribute more to the likelihood of the development of a severe outcome in patients with babesiosis.

Several recent studies have used a statistical review of clinical information to identify patients likely to have a poor or severe outcome. Martin et al30 investigated hemolytic uremic syndrome in Minnesota and were able to examine potential predictors of disease severity. They found that children with elevated polymorphonuclear-leukocyte counts on hospital admission were at higher risk for severe disease and poor outcome through both univariate and multivariate analyses. Through the use of the bivariable and stepwise logistic regression analyses, Hamel et al31 developed a simple prognostic scoring system to identify comatose patients at high risk for poor outcomes using data that were available during hospitalization. Such risk stratification was shown to offer physicians, patients, and patients' families valuable information that could contribute to decisions and allocations of health care resources affecting patient care. Marsh et al32 reviewed malaria in hospitalized African children and applied World Health Organization criteria for severe and complicated malaria. These criteria included 10 clinical or laboratory-based data available during the patient's hospitalization. Univariate and multivariate logistic regression analyses were used to identify the criteria associated with the greatest prognostic value for severe, life-threatening malaria. Farr et al33 examined patients with pneumonia and looked for associations between death and 42 hospital admission variables. Univariate associations were identified for several variables, including, but not limited to, older age, respiratory rates greater than 30/min, decreased diastolic and systolic blood pressure, elevated serum urea nitrogen levels, and leukocyte counts lower than 4×109/L.

Human babesiosis, which is a rarely diagnosed, protozoal zoonotic disease, has become a significant community health problem in certain locations along the northeastern coastal areas of the United States.1,5,7,17 Infections with B microti and other strains are also seen in the upper Midwest and northern Pacific coastal areas.6,21,22 The frequency of clinical diagnosis is directly related to the geographic distribution of the pathogen and the medical community's collective awareness of the disease.1,4,18,19 Fortunately, most infections with B microti in healthy individuals cause minimal symptoms,5 but for those patients who may be elderly, have a history of splenectomy, or may be otherwise immunocompromised,1-3,17,20 including those with HIV infection, infection with babesiosis can be particularly debilitating.

This study provides some clinical guidance; however, certain limitations should be mentioned. As an observational study, there was a lack of systematic data collection and recording in hospital charts. Potential comorbidity from other tick-borne pathogens may have affected outcomes and may not have been recognized, diagnosed, or recorded in the chart. We cannot rule out that some missing information in charts resulted in misclassification of outcome measures. However, the impact of potential misclassification is probably small, as severe events would be likely to be recorded. This study was also limited to the 77% of patients with babesiosis who were hospitalized and was not controlled for by stage of disease on admission.

Identification of a patient's significant travel history (to an area where babesiosis is endemic) is crucial for the inclusion of babesiosis in the differential diagnosis. Once babesiosis is diagnosed, the presence of parasitemia values greater than 0.04, alkaline phosphatase values higher than 125 U/L, or white blood cell counts higher than 5×109/L should help alert the health care provider to those patients who may require special attention because of the higher risk for death. Physicians should consider babesiosis in patients who present with a fever of unknown origin and who reside or have traveled in endemic areas if they have undergone a splenectomy or are elderly, immunocompromised, or debilitated.

Accepted for publication March 12, 1998.

Determination of laboratory parasitemia values was provided by Jorge Benach, PhD, NYSDOH, State University of New York at Stony Brook. Sean Meldrum, MS, NYSDOH, assisted on the coordination of initial requests for hospital charts of affected patients.

Reprints: Dennis J. White, PhD, Griffin Laboratory, Wadsworth Center, New York State Department of Health, Empire State Plaza, Albany, NY 12201-0509.

Villar  BFWhite  DJBenach  JL Human babesiosis. Sun  Ted Progress in Clinical Parasitology Vol 2 Philadelphia, Pa Field & Wood Inc Medical1991;129- 143Google Scholar
Gombert  MEGoldstein  EJCBenach  JL  et al.  Human babesiosis: clinical and therapeutic considerations.  JAMA. 1982;2483005- 3007Google ScholarCrossref
Benach  JLHabicht  GS Clinical characteristics of human babesiosis.  J Infect Dis. 1981;144481Google ScholarCrossref
Dammin  GJSpielman  ABenach  JLPiesman  J The rising incidence of clinical Babesia microti infection.  Hum Pathol. 1981;12398- 400Google ScholarCrossref
Cahill  K Babesiosis: unappreciated even in endemic areas.  J Community Health. 1995;20315- 320Google ScholarCrossref
Steketee  RWEckman  MRBurgess  EC  et al.  Babesiosis in Wisconsin: a new focus of disease transmission.  JAMA. 1985;2532675- 2678Google ScholarCrossref
Krause  PJTelford  SRRyan  R  et al.  Geographical and temporal distribution of babesial infection in Connecticut.  J Clin Microbiol. 1991;291- 4Google Scholar
Marcus  LCMattia  AR A 63-year-old man with fever, sweats and shaking chills.  N Engl J Med. 1993;329194- 199Google ScholarCrossref
Grunwaldt  E Babesiosis on Shelter Island.  N Y State J Med. 1977;1320- 1321Google Scholar
Ruebush  TKJuranek  DDChisholm  ESSnow  PCHealy  GRSulzer  AJ Human babesiosis on Nantucket Island: evidence for self-limited and subclinical infections.  N Engl J Med. 1977;297825- 827Google ScholarCrossref
Mathewson  HOAnderson  AEHazard  GW Self-limited babesiosis in a splenectomized child.  Pediatr Infect Dis. 1984;3148- 149Google ScholarCrossref
Bredt  ABWeinstein  WMCohen  S Treatment of babesiosis in asplenic patients.  JAMA. 1981;2451938- 1939Google ScholarCrossref
Rowin  KSTanowitz  HBRubinstein  AKunkle  MWittner  M Babesiosis in asplenic hosts.  Trans R Soc Trop Med Hyg. 1984;78442- 444Google ScholarCrossref
Teutch  SMEtkind  PBurwell  EL  et al.  Babesiosis in post-splenectomy hosts.  Am J Trop Med Hyg. 1980;29738- 741Google Scholar
Rosner  FZarrabi  MHBenach  JLHabicht  GS Babesiosis in splenectomized adults: review of 22 reported cases.  Am J Med. 1984;76696- 701Google ScholarCrossref
Scholtens  RGBraff  EHHealy  GRGleason  N A case of babesiosis in man in the United States.  Am J Trop Med Hyg. 1968;17810- 813Google Scholar
Meldrum  SCBirkhead  GSWhite  DJBenach  JLMorse  DL Human babesiosis in New York State: an epidemiological description of 136 cases.  Clin Infect Dis. 1992;151019- 1023Google ScholarCrossref
Filstein  MRBenach  JLWhite  DJ  et al.  Serosurvey for human babesiosis in New York.  J Infect Dis. 1980;141518- 521Google ScholarCrossref
Popovsky  MALindberg  LESyrek  ALPage  PL Prevalence of Babesia antibody in a selected blood donor population.  Transfusion. 1988;2859- 61Google ScholarCrossref
Ruebush  TKJuranek  DDSpielman  APiesman  JHealy  GR Epidemiology of human babesiosis on Nantucket Island.  Am J Trop Med Hyg. 1981;30937- 940Google Scholar
Quick  REHerwaldt  BLThomford  JW  et al.  Babesiosis in Washington State: a new species of Babesia Ann Intern Med. 1993;119284- 290Google ScholarCrossref
Persing  DHHerwaldt  BLGlaser  C  et al.  Infection with a Babesia-like organism in northern California.  N Engl J Med. 1995;332298- 303Google ScholarCrossref
Bakerman  S ABCs of Interpretive Laboratory Data. 2nd ed. Greenville, NC Interpretive Laboratory Data Inc1984;
Beutler  ELichtman  MAColler  BSKips  TG Williams Hematology. 5th ed. New York, NY McGraw-Hill Book Co1995;
Fischbach  F A Manual of Laboratory and Diagnostic Tests. 4th ed. Philadelphia, Pa JB Lippincott1992;
Jacobs  DSKasten  BLDermott  WRWolfson  WL Laboratory Test Handbook. 2nd ed. Stow, Ohio LexiComp1990;
Rubenstein  EFederman  D Scientific American Medicine.  New York, NY Scientific American Inc1993;
Wallach  J Interpretation of Diagnostic Tests: A Synopsis of Laboratory Medicine. 5th ed. Boston, Mass Little Brown & Co Inc.
Wygarden  JDSmith  LH Cecil's Textbook of Medicine. 17th ed. Philadelphia, Pa WB Saunders Co1985;
Martin  DLMacDonald  KLWhite  KESoler  JTOsterholm  MT The epidemiology and clinical aspects of the hemolytic uremic syndrome in Minnesota.  N Engl J Med. 1990;3231161- 1167Google ScholarCrossref
Hamel  MBGoldman  LTeno  J  et al.  Identification of comatose patients at high risk for death or severe disability.  JAMA. 1995;2731842- 1848Google ScholarCrossref
Marsh  KForster  DWaruiru  C  et al.  Indicators of life-threatening malaria in African children.  N Engl J Med. 1995;3321399- 1404Google ScholarCrossref
Farr  BMSloman  AJFisch  MJ Predicting death in patients hospitalized for community-acquired pneumonia.  Ann Intern Med. 1991;115428- 436Google ScholarCrossref