Correct Classification Of Cases Research Articles

New insights into SARS-CoV-2 Lumipulse G salivary antigen testing: accuracy, safety and short TAT enhance surveillance.

The rapid, accurate and safe detection of SARS-CoV-2 is the key to improving surveillance and infection containment. The aim of the present study was to ascertain whether, after heat/chemical inactivation, SARS-CoV-2N antigen chemiluminescence (CLEIA) assay insaliva remains a valid alternative to molecular testing. In 2022, 139 COVID-19 inpatients and 467 healthcare workers were enrolled. In 606 self-collected saliva samples (Salivette), SARS-CoV-2 was detected by molecular (TaqPath rRT-PCR) and chemiluminescent Ag assays (Lumipulse G). The effect of sample pre-treatment (extraction solution-ES or heating) on antigen recovery was verified. Salivary SARS-CoV-2 antigen assay was highly accurate (AUC=0.959, 95% CI: 0.943-0.974), with 90% sensitivity and 92% specificity. Of the 254 antigen positive samples, 29 were false positives. We demonstrated that heterophilic antibodies could be a cause of false positive results. A significant antigen concentration decrease was observed after ES treatment (p=0.0026), with misclassification of 43 samples. Heat had a minimal impact, after treatment the correct classification of cases was maintained. CLEIA SARS-CoV-2 salivary antigen provides accurate, timely and high-throughput results that remain accurate also after heat inactivation, thus ensuring a safer work environment. This supports the use of salivary antigen detection by CLEIA in surveillance programs.

Correlation of radiographic measurements of structures of the equine foot with lesions detected on magnetic resonance imaging.

There are few studies on the correlations between radiographic measurements of the foot and abnormalities of specific structures found with magnetic resonance imaging (MRI). To document the relationship between radiographic measurements of the equine foot and the presence of lesions in the foot on MRI. We hypothesised that different radiographic measurements would be associated with specific lesions detected by MRI. Retrospective analysis of radiographs and MRI studies. Seventy-four feet from 52 lame horses were included. Twenty parameters were measured on radiographs, whereas the signal intensity, homogeneity and size of each structure in the foot were evaluated on magnetic resonance images. The data were analysed using simple linear correlation analysis and classification and regression trees (CARTs). Linear correlations were found between the navicular bone compacta thickness and injuries of the deep digital flexor tendon, collateral sesamoidean ligament, navicular spongiosa and navicular bone proximal border. Long-toed horses had a high incidence of lesions involving the spongiosa and proximal border of the navicular bone. Elongation of the navicular bone was associated with proximal and distal border injuries. A reduced palmar angle and increased angle between the middle and distal phalanx were observed in horses with alterations of collateral ligaments of the distal interphalangeal joint and navicular bone spongiosa, respectively. For each structure under investigation, CARTs predicting the presence of MRI pathology based on radiographic measurements had excellent performance, with >80% correct classification of cases, when using one of 3 data sources. This study demonstrated a relationship between radiographic measurements of the foot and the presence of lesions detected on MRI, while CARTs illustrated that different radiographic measurements were associated with different MRI lesions.

Variables related to sexual prejudice among Mexican health science students

BackgroundThe stigmatization and discrimination of non-heterosexual persons is a reality in some institutions of the Health Services, and among health sciences students. ObjectivesTo describe and predict the level of sexual prejudice in health sciences students, taking into account a set of qualitative and numerical variables on socio-demographic data, sexual life, social life, university (private or public) the student's major (medicine or psychology), and clinical aspects. MethodologyA socio-demographic and life-history data questionnaire, an 8-item homophobia scale and a 16-item internalized homonegativity scale were applied to a non-probabilistic sample composed of 231 health sciences students. The predictive models were estimated by analyses of multinomial and ordinal regression. ResultsTwelve percent of participants exhibited an attitude of open rejection towards non-heterosexual persons (including 0.9% who exhibited extreme rejection). Non-heterosexual orientation, having non-heterosexual friends and acceptance of one's own homosexual desires were variables associated with lower levels of open rejection towards non-heterosexual persons. Only the two latter variables were significant predictive variables; they explained 21% of the variance in the ordinal regression model and 27% in the multinomial regression model. The percentage of the correct classification of cases of acceptance was high but the percentage of the correct classification of cases of rejection was low. ConclusionThe level of open rejection towards non-heterosexual persons is low. An exclusively heterosexual identity, affirming not to share aspects of the sexual sphere and not having personal contact with the stigmatized subject are determinants of open rejection. There exist other variables that were not taken into account in this study, as is deduced by the high percentage of unexplained variance.

Multicenter ALK Testing in Non–Small-Cell Lung Cancer: Results of a Round Robin Test

The anaplastic lymphoma kinase (ALK) inhibitor crizotinib has recently received approval for the treatment of patients with locally advanced or metastatic ALK-positive non-small-cell lung cancer (NSCLC). As the therapeutic prescription postulates the detection of ALK rearrangements, reliable diagnostic approaches are of utmost importance. With this study, we present the data of the first German ALK-round robin test based on genomic DNA in situ hybridization (ISH). The application of immunohistochemistry (IHC) for ALK protein detection was optional and not required for certification. Two tissue microarrays, each consisting of the same 10 NSCLC but in different arrangement of the cases, were generated (five unequivocally ALK-positive and five unequivocally ALK-negative cases). ISH-based results and optional IHC data had to be submitted within 10 working days. A successful participation (certification) was reached if at least 19 of the 20 possible points were scored (2 points for a correct case classification). Fifty-three of 59 participants (89.8%) provided their data for ISH-ALK detection within the submission period. Thirty-two of 53 participants (60.3%) received at least 19 points required for certification. Remarkably, the range of cells with aberrant ALK signal configuration was broad in ALK-negative (0-13%) and in ALK-positive cases (15-95%). Thirty-five participants supported the round robin test with optional ALK IHC results, which displayed a great heterogeneity in the ALK ISH-positive cases. In essence, our ALK ISH round robin test clearly demonstrates that there is accumulating need for improvement of ALK testing. Although ISH may be regarded as a well-established procedure, its broad application in a diagnostic setting is challenging and requires standardized methods and harmonized interpretation to achieve sound results for therapeutic decisions. The same is true for ALK IHC which, however if standardized, might improve the diagnostic approach.

Validity of computerized diagnoses, procedures, and drugs for inflammatory bowel disease in a northern California managed care organization

Resources for studying inflammatory bowel disease (IBD) are needed in evaluations of drug safety including traditional drugs and new biologics agents. We developed an IBD registry, with ascertainment from computerized visit information. We sought to characterize the positive predictive value (PPV) of IBD case-finding using computerized data compared with chart review. We identified 2906 persons aged 89 years or younger with one or more IBD diagnoses in computerized visit data during the period of 1996-2002. The diagnosis of IBD was confirmed through chart review. Adopting chart review as the gold standard, the validity of computerized encounter data to determine IBD was estimated. Among the 2906 study subjects with one or more ICD-9 diagnosis codes of 555 or 556 in computerized data, 81% were confirmed as having IBD by chart review. Defining cases as those who underwent two or more visits without regard to diagnostic procedures or drug utilization maximized the correct classification of cases (PPV, 95%). The quality of IBD diagnoses in computerized data is adequate to meet the aims of a wide range of research studies.

Artificial Neural Network Model for the Prediction of Obsessive-Compulsive Disorder Treatment Response

Several patients with obsessive-compulsive disorder (OCD) who are refractory to adequate treatment with first-line treatments are considered treatment-resistant. Further surveys were to be implemented to explore the outcome predictors of the antiobsessional response. Such study was aimed at building a model suitable to predict the final outcome of a mixed OCD pharmacologic and nonpharmacologic treatment approaches. We studied 130 subjects with OCD who underwent pharmacologic (with selective serotonin reuptake inhibitors alone or with selective serotonin reuptake inhibitors and risperidone at low dosage) and/or behavioral therapy (using exposure and response prevention techniques). The following variables were used as predictors: symptoms dimension, as resulting from the Yale-Brown Obsessive-Compulsive Scale items factor analysis; neuropsychologic performances; and epidemiologic variables. The treatment response arising from 3 to 6 months of therapy was used as dependent variable. A conventional logistic regression was used to define a previsional model of treatment response and multilayer perceptrons and to supervise an artificial neural network technique. The 46.9% of the sample resulted to be refractory to treatment. Results obtained with the logistic regression model showed that the only predictors of treatment outcome are hoarding symptoms, repeating rituals, and counting compulsions. Furthermore, using all the variables considered in the models, multilayer perceptrons showed highly better predictive performance as compared with the logistic regression models (93.3% vs 61.5%, respectively, of correct classification of cases). Complex interactions between different clinical and neuropsychologic variables are involved in defining OCD treatment response profile, and nonlinear and interactive modeling strategies, that is, supervised artificial neural networks, seem to be more suitable to investigate this complexity than linear techniques.

Influence of Computer-Aided Detection False-Positives on Reader Performance and Diagnostic Confidence for CT Colonography

The objective of our study was to investigate whether an increasing number of computer-aided detection (CAD) false-positives decreases reader sensitivity, specificity, and confidence for nonexpert readers of CT colonography (CTC). Fifty CTC data sets (29 men; mean age, 65 years), 25 of which contained 35 polyps > or = 5 mm, were selected in which CAD had 100% polyp sensitivity at two sphericity settings (0 and 75) but differed in the number of false-positives. The data sets were read by five readers twice: once at each sphericity setting. Sensitivity, specificity, report time, and confidence before and after second-read CAD were compared using the paired exact and Student's t test, respectively. Receiver operating characteristic (ROC) curves were generated using reader confidence (1-100) in correct case classification (normal or abnormal). CAD generated a mean of 42 (range, 3-118) and 15 (range, 1-36) false-positives at a sphericity of 0 and 75, respectively. CAD at both settings increased per-patient sensitivity from 82% to 87% (p = 0.03) and per-polyp sensitivity by 8% and 10% for a sphericity of 0 and 75, respectively (p < 0.001). Specificity decreased from 84% to 79% (sphericity 0 and 75, p = 0.03 and 0.07). There was no difference in sensitivity, specificity, or reader confidence between sphericity settings (p = 1.0, 1.0, 0.11, respectively). The area under the ROC curve was 0.78 (95% CI, 0.70-0.86) and 0.77 (0.68-0.85) for a sphericity of 0 and 75, respectively. CAD added a median of 4.4 minutes (interquartile range [IQR], 2.7-6.5 minutes) and 2.2 minutes (IQR, 1.2-4.0 minutes) for a sphericity of 0 and 75, respectively (p < 0.001). CONCLUSION. CAD has the potential to increase the sensitivity of readers inexperienced with CTC, although specificity may be reduced. An increased number of CAD-generated false-positives does not negate any beneficial effect but does reduce efficiency.

Heavy Metals and Mineral Elements Not Included on the Nutritional Labels in Table Olives

The average contents, in mg/kg edible portion (e.p.), of elements not considered for nutritional labeling in Spanish table olives were as follows: aluminum, 71.1; boron, 4.41; barium, 2.77; cadmium, 0.04; cobalt, 0.12; chromium, 0.19; lithium, 6.56; nickel, 0.15; lead, 0.15; sulfur, 321; tin, 18.4; strontium, 9.71; and zirconium, 0.04. Sulfur was the most abundant element in table olives, followed by aluminum and tin (related to green olives). There were significant differences between elaboration styles, except for aluminum, tin, and sulfur. Ripe olives had significantly higher concentrations (mg/kg e.p.) of boron (5.32), barium (3.91), cadmium (0.065), cobalt (0.190), chromium (0.256), lithium (10.01), nickel (0.220), and strontium (10.21), but the levels of tin (25.55) and zirconium (0.039) were higher in green olives. The content of contaminants (cadmium, nickel, and tin) was always below the maximum limits legally established. The discriminant analysis led to an overall 86% correct classification of cases (80% after cross-validation).

Effect of visualization display colour on polyp conspicuity at virtual colonoscopy

To investigate the effect of different colour three-dimensional (3D) displays on polyp detection at virtual colonoscopy (VC). Five VC trained observers were shown "brief flashes" (lasting 0.2s) of 125 3D endoluminal image snap-shots, repeated for each of six display colours (750 images total). One hundred images contained a single polyp (diameter range 5-42 mm) and 25 contained no polyp ("normal"). Images were reviewed in random order over five reading sessions, readers recording either normality or presence and location of a polyp. Multilevel logistic regression was used to examine any influence of colour on polyp detection stratified according to polyp size (medium 5-9 mm/large >or=10mm). The kappa statistic was used to assess effect of colour on observer agreement. Individual reader polyp detection rates ranged between 75-94%. Compared to the default pink "soft tissue" display, the odds of polyp detection were 0.65 (CI 0.41,1.01) for green, 0.82 (0.53,1.30) for blue, 1 (0.63,1.59) for red, 1.12 (0.7,1.79) for monochrome, and 1.15 for yellow (0.72,1.84). Overall, there was no significant difference between the displays (p=0.11). Including normal cases, there was no overall difference in correct case classification between the six colours (p=0.44). The odds of detecting large versus medium polyps was significantly greater for 3/5 observers; odds ratio (OR) 2.84-10.1, although unaffected by display colour (p=0.3). The background colour display generally has a minimal effect on polyp detection at VC, although green should be avoided.

Assessing the effectiveness of statistical classification techniques in predicting future employment of participants in the temporary assistance for needy families program

Three classification techniques (Chi‐square Automatic Interaction Detection [CHAID], Classification and Regression Tree [CART], and discriminant analysis) were tested to determine their accuracy in predicting Temporary Assistance for Needy Families program recipients' future employment. Technique evaluation was based on proportion of correctly predicted cases and 6 other criteria. Sustained employment was associated with both recipients' subjective work‐related skills assessments and their objective test scores, and with their educational levels and marital status. Correct case classification by model was 74% for CHAID, 75% for CART, and 68% for discriminant analysis. Results suggest these classification tools can be useful supplements in assessments of clients' employment potentials.

Homocysteine, vitamin B12 and folate status in pediatric acute lymphoblastic leukemia

The cause of majority of acute leukemias is unknown, but likely to involve interaction of environment, hematopoitic development and weak susceptibility loci within an individual's genetic constitution. The present study evaluates the association between plasma levels of homocysteine, folate and vitamin B12 and acute lymphoblastic leukemia. Plasma levels of homocysteine, folate and vitamin B12 were compared between cases of acute lymphoblastic leukemia and age and sex matched normal controls. Homocysteine levels were measured by solid immunoassay, while folate and vitamin B12 levels were determined by radioassay. Folate levels were significantly among cases as compared to control group (8.56 +/- 4.35) vs (14.04 +/- 2.62) ng/ml, P< 0.001). Although individually vitamin B12 and homocysteine were not significant different between cases and controls, the combined effect of all three parameters was significantly different (P< 0.001), with 83.3% of correct classification of cases and controls was obtained by discriminate function analysis. The data provide evidence for the role of folate, vitamin B12 and homocysteine levels in acute lymphoblastic leukemia, suggesting that gene-environment interaction may be an important factor in the development of acute lymphoblastic leukemia.

Visual function and dysfunction in early and late age-related maculopathy

Late age-related maculopathy (ARM) is responsible for the majority of blind registrations in the Western world among persons over 50 years of age. It has devastating effects on quality of life and independence and is becoming a major public health concern. Current treatment options are limited and most aim to slow progression rather than restore vision; therefore, early detection to identify those patients most suitable for these interventions is essential. In this work, we review the literature encompassing the investigation of visual function in ARM in order to highlight those visual function parameters which are affected very early in the disease process. We pay particular attention to measures of acuity, contrast sensitivity (CS), cone function, electrophysiology, visual adaptation, central visual field sensitivity and metamorphopsia. We also consider the impact of bilateral late ARM on visual function as well as the relationship between measures of vision function and self-reported visual functioning. Much interest has centred on the identification of functional changes which may predict progression to neovascular disease; therefore, we outline the longitudinal studies, which to date have reported dark-adaptation time, short-wavelength cone sensitivity, colour-match area effect, dark-adapted foveal sensitivity, foveal flicker sensitivity, slow recovery from glare and slower foveal electroretinogram implicit time as functional risk factors for the development of neovascular disease. Despite progress in this area, we emphasise the need for longitudinal studies designed in light of developments in disease classification and retinal imaging, which would ensure the correct classification of cases and controls, and provide increased understanding of the natural course and progression of the disease and further elucidate the structure-function relationships in this devastating disorder.

Karyometry detects subvisual differences in chromatin organisation state between non-recurrent and recurrent papillary urothelial neoplasms of low malignant potential

Aim: To analyse nuclear chromatin texture in non-recurrent and recurrent papillary urothelial neoplasms of low malignant potential (PUNLMPs). Materials: Ninety three karyometric features were analysed on haematoxylin and eosin stained...

Can we mitigate animal–vehicle accidents using predictive models?

Summary Vehicle collisions with wild animals are a serious problem that justifies the widespread application of mitigation measures such as road fencing and provision of crossing structures. Models that predict the best location for mitigation measures can improve wildlife survival and road safety. A database of 2067 records of animal–vehicle collisions was used to create two data sets at different spatial scales. The first comprised records of road sections of 1 km length with high rates of collision in combination with road sections with a low number of collisions. The second comprised records of collision and no collision incidence at points on the road system at a 0·1‐km scale. Logistic regression was used to investigate the relationship between incidence of collision and measured habitat features in each data set. The models were validated with a subset of the original data not used in developing the models. Road sections with high collision rates were associated with areas having high forest cover, low crop cover, low numbers of buildings and high habitat diversity. The fitted model achieved a significant predictive success during validation (χ2 = 4·82, 1 d.f., P= 0·028), with more than 70% correct classification of cases. Specific collision points typically had no guard‐rails or lateral embankments, were not near underpasses, crossroads or buildings, and featured hedges or woodland near the road. The fitted model also showed a significant predictive power in validation (64% correct classification, χ2 = 9·51, 1 d.f., P= 0·002) and accurately predicted 85·1% of collision points. Synthesis and applications. Predictive models of animal–vehicle collision locations should be used at both a landscape level and a local scale during the process of road design and implementation of mitigation measures. Modelling of collision risk could inform decisions on road alignment and on the exact location of crossing structures for mammals, to improve wildlife survival and road safety. This is the first study integrating both landscape and local scales of analysis for the variables associated with animal–vehicle collisions.

Cytological grading of breast cancer in Giemsa-stained fine needle aspiration smears.

A potential cytological nuclear grading based on a semi-quantitative evaluation of three basic nuclear features, size of cell nuclei, anisonucleosis and the proportion of nucleoli-containing-nuclei, was tested on 74 Giemsa-stained fine needle aspiration of breast smears for its reliability in establishing the malignant potential of breast cancer. The prognostic impact of DNA-ploidy and S-phase fraction was also assessed. A good correlation between the three basic nuclear features, DNA-ploidy, S-phase fraction, cytological nuclear grade and histological grade, was shown. Using the cytological nuclear grade proposed, correct classification of cases between low histological grade (HG I) and high histological grade (HG II + HG III) was achieved in 79.73%. A statistically significant difference in 5-year survival rate was also observed between low malignancy grade and high malignancy grade breast cancer patients, regardless of the grading method used. DNA-ploidy and S-phase fraction were not statistically significant in establishing the malignant potential of breast cancer.

Normative Studies Research Project Test Battery: Detection of Dementia in African American and European American Urban Elderly Patients

The present study evaluated the psychometric properties and clinical utility of the Normative Studies Research Project test battery (NSRP) for use with urban medical patients. A greater proportion of African American patients and persons with fewer years of education were represented in the present sample as compared to previous investigations of similar screening batteries (e.g., Iowa Battery; Eslinger et al., 1985). Seventy-four of the subjects were designated as cognitively intact and 89 were cognitively impaired. Principal components analyses revealed a two-factor solution for the intact persons, whereas a one-factor solution was obtained for the impaired patient group. Logistic regression analysis resulted in 76% correct classification of cases; sensitivity was 77% and specificity was 74%. The NSRP battery is a clinically useful tool for patients in urban medical settings.

The return of thalidomide: are birth defects surveillance systems ready?

In the 1960s, thalidomide caused limb deficiencies in thousands of infants worldwide. The limb deficiencies were frequently of the intercalary type. As a result, numerous countries started birth defect surveillance programs. In 1967, the Centers for Disease Control (CDC) started the Metropolitan Atlanta Congenital Defects Program (MACDP), a population-based surveillance system, to provide early warning against new teratogens. Recent studies have shown that thalidomide may be beneficial for a range of conditions, including cancer and AIDS, and it may once again become widely available. Here, we examine the ability of MACDP to detect an increase in the birth prevalence of limb deficiency as an early warning of fetal exposure to thalidomide. We calculated base rates for all limb deficiencies, for bilateral nonsyndromic intercalary or preaxial deficiencies, and for all nonsyndromic intercalary limb deficiencies among Atlanta infants born from 1968 through 1993. We used relative risk estimates from previous studies and a range of pregnancy exposure rates for thalidomide. We tested the statistical power of MACDP to detect subtle changes in the birth prevalence of these defects using Poisson and cumulative sum (CUSUM) techniques. The base rates for all limb deficiencies, for bilateral intercalary or preaxial deficiencies, and for all intercalary limb deficiencies, were 0.53, 0.035, and 0.022/1,000, and the estimated relative risks were 175, 4,570, and 8,180, respectively. We varied the assumed rate of exposure to thalidomide from 1/10,000 to 5/100. With a 1/1,000 exposure rate, both Poisson and CUSUM techniques will detect a rate change in intercalary limb deficiency in about 6 months of monitoring, and a rate change in bilateral intercalary or preaxial deficiencies in about 12 months of monitoring. When monitoring all limb deficiencies, a pregnancy exposure rate of 3.5% or less would go unnoticed by the Poisson method and would take more than 50 years for the CUSUM method to signal an alarm with a 1/1,000 exposure rate. However, for rates of exposure less than 1/1,000, a progressively longer period of time or larger sample are needed to detect a rate change by both methods. Our findings highlight the importance of enlarging the monitored population and correct case classification in birth defects surveillance.

DSM-III-R personality disorders in panic and obsessive-compulsive disorder: A comparison study

Forty-eight patients with panic disorder/agoraphobia (PAD) and 30 with obsessive-compulsive disorder (OCD) were assessed for DSM-III-R axis II personality disorders (PD) and the presence of the same anxiety disorder in the relatives of probands (homotypic disorders). No specific personality disorder was present significantly more often in either of the two groups. Agoraphobia was not associated with higher rates of axis II disorders in PAD patients. Duration of illness did not influence the presence of a PD in patients of both groups. Secondary cases of the same anxiety disorder were significantly more common among first-degree relatives of PAD patients. A discriminant analysis performed on the most frequent personality traits of both groups provided a correct classification of cases of 97.4%. Our results do not support the hypothesis of PD as secondary to anxiety disorders and confirm previous findings of a lack of specificity between DSM-III-R axis II categories and OCD and PAD.

Attentional bias and agoraphobic avoidance: The role of cognitive style

Research has suggested that clinically anxious individuals tend to exhibit an attentional bias in processing threat-relevant information. Agoraphobics, in particular, have demonstrated consistent patterns of attention deployment variously described as fear stimulus generalization and hysterical cognitive style. The present study was designed to identify attentional and affective mediators of agoraphobic behavior using recent guidelines for information-processing research with anxiety-disordered populations. Ten adult outpatients meeting DSM-III-R/DIS criteria for panic disorder with agoraphobic avoidance and ten patients diagnosed with generalized anxiety disorder (GAD) were identified from 85 consecutive referrals. The GAD patients constituted a psychiatric control group. All subjects participated in a sequence of diagnostic interviews and completed a battery of tests that included the Embedded Figures Test (a measure of perceptual field dependence), WAIS-R “perceptual-organizational” factor subtests, and the SCL-90-R. Results showed that perceptual field dependence and phobic symptomatology were significant predictors of agoraphobia accounting for 86% of the variance. Overall correct classification of cases into diagnostic groups was 95%. Contrary to previous findings, phobic anxiety was found to be a distinct feature of the syndrome, and agoraphobics did not demonstrate significantly greater psychopathology than GAD patients. The results were interpreted within the framework of current information-processing heuristics for anxiety disorders.

Antenatal prediction of women at increased risk for infants with low birth weight: A discriminant analysis of maternal characteristics known at 24 weeks

Prenatal factors known to be associated with delivery of an infant with low birth weight (LBW) were studied in 165 women who were delivered of infants of 2,500 gm or less, as well as 154 women who were delivered of term infants with normal birth weights (greater than 2,500 gm). Data included information related to the clinical course of the pregnancy through 24 weeks' gestation. Of 25 characteristics measured prenatally, 10 variables were identified which indicated differences between the LBW and normal birth weight groups sufficient to make them suitable for a discriminant analysis. Following further analysis and attribute reduction, eight variables were selected as the final set of discriminating attributes. These are: maternal age, height, and weight, the patient's perception of her own parental treatment as a child, presence of uterine anomalies, bleeding during pregnancy, suspected multiple gestation, and number of previous pregnancies ending with a gestational age of 13 through 36 weeks. On the basis of coefficients derived for each of these variables, an equation was developed to derive a discriminant score for each patient. With the use of such equations, 73.5% of the known cases were correctly classified into groups with low and high birth weight. A similar high correct classification of cases was made when suspected multiple gestation was excluded as a variable.