Truncated logistic regression.

T J O'Neill; S C Barry

Home/Archive/PubMed

PubMed · 7662842

Truncated logistic regression.

T J O'Neill·S C Barry

View original source ↗Export metadata ↓

Abstract

Truncated binary data occurs when a group of individuals, who each have a binary response, are observed only if one or more of the individuals has a positive response. In this paper the group will be taken to be a motor vehicle accident and the binary response taken to be survival or death. We compare two regression techniques that can be used for truncated binary data. The first procedure, conditional logistic regression (Breslow and Day, 1980, Statistical Methods in Cancer Research. 1: The Analysis of Case-Control Studies. No. 32. Lyon: IARC) conditions on the actual number of deaths, and has been previously used for this type of data. The second procedure, truncated logistic regression, conditions on there being at least one death. It is computationally simpler than conditional logistic for groups of size greater than two and can be considerably more efficient. A major difference between the two methods is that only truncated logistic regression requires a knowledge of group level covariates and allows estimation of group level effects.

Explore related subjects

Accidents, Traffic Australia Humans Mathematics Models, Statistical Regression Analysis Safety Journal Article Research Support, Non-U.S. Gov't

Keep this discovery

Explore connections, maps & timelines

BibTeX RIS

T J O'Neill, S C Barry. 1995. Truncated logistic regression.. https://pubmed.ncbi.nlm.nih.gov/7662842/

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related citations

PubMed2006

Review of 509 mandibular fractures treated at the University Hospital, Alexandria, Egypt.

Mandibular fracture is one of the most common facial skeletal injuries. Although its main causes are road crashes and violence, the relation between these causes varies from one country to another. We made a retrospective study of the medical records and radiographs of 509 patients treated for mandibular fracture at the University of Alexandria Hospital between 1991 and 2000. The data that we collected included age, sex, aetiology, date of injury, anatomical site of the fracture, associated maxillofacial trauma, and treatment. The prevalence of mandibular fractures was higher in male subjects in all age groups, and the male:female ratio was 3.6:1. Most fractures were sustained by men in the age group 21-30 years and girls between 0 and 10 years, and the monthly incidence was constant. Road crashes were the main cause, followed by falls and assaults. Fractures of the angle were the most common (22%) followed by parasymphyseal fractures (21%) and the lowest was in the coronoid region (1%).

Accidents, Traffic↗

PubMed2005

Driving under the influence of chlormethiazole.

This article describes a case of driving under the influence of the sedative-hypnotic-anticonvulsant drug chlormethiazole. The suspect, who was a physician, was driving dangerously on a busy highway and caused a traffic collision. When apprehended by the police, the man had bloodshot and glazed eyes and pupil size was enlarged. He could not answer the questions properly and his gait was unsteady. A roadside breath-alcohol screening test was positive but an evidential breath-alcohol test conducted about one hour later was below the legal limit for driving of 0.10 mg/L (10 microg/100 mL or 0.021 g/210 L). Because of the special circumstances of the traffic crash and the man's appearance and behaviour, the police suspected that drugs other than alcohol were involved and obtained a venous blood sample for toxicological analysis. The blood contained 0.23 mg/g alcohol, which is above the legal limit for driving in Sweden 0.20 mg/g (20 mg/100 mL or 0.020 g/100 mL), and codeine was also present at a therapeutic concentration of 0.02 mg/L. The conflict between the clinical signs of impairment and the toxicology report prompted a reanalysis of the blood sample with major focus on sedative-hypnotic drugs. Analysis by capillary GC-NPD identified chlormethiazole at a concentration of 5mg/L, the highest so far encountered in traffic cases in Sweden. In 13 other impaired driving cases over 10 years the mean (median) and range of concentrations of chlormethiazole were 1.6 mg/L (1.6 mg/L) and 0.3-3.3 mg/L. This case report underscores the need to consider clinical observations and the person's behaviour in relation to the toxicology report when interpreting and testifying in drug-impaired driving cases.

Accidents, Traffic↗

PubMed2005

Complete bronchial rupture in a child: report of a case.

Tracheobronchial rupture due to blunt chest trauma is a rare and serious injury in children. The diagnosis is usually difficult and may be overlooked because of the variability of symptoms and findings. Fiberendoscopy is useful in children with stable tracheal or bronchial ruptures. However, in the emergency situation, fiberendoscopy may not be appropriate, and thoracotomy and primary anastomosis may be the best option.

Accidents, Traffic↗