Variance components testing in the longitudinal mixed effects model.

D O Stram; J W Lee

Home/Archive/PubMed

PubMed · 7786999

Variance components testing in the longitudinal mixed effects model.

D O Stram·J W Lee

View original source ↗Export metadata ↓

Abstract

This article discusses the asymptotic behavior of likelihood ratio tests for nonzero variance components in the longitudinal mixed effects linear model described by Laird and Ware (1982, Biometrics 38, 963-974). Our discussion of the large-sample behavior of likelihood ratio tests for nonzero variance components is based on the results for nonstandard testing situations by Self and Liang (1987, Journal of the American Statistical Association 82, 605-610).

Explore related subjects

Age Factors Biometry Child Female Growth Humans Longitudinal Studies Male Models, Statistical Multivariate Analysis Pituitary Gland Probability Sex Characteristics Sex Factors Skull Comparative Study Journal Article Research Support, U.S. Gov't, P.H.S.

Keep this discovery

Explore connections, maps & timelines

BibTeX RIS

D O Stram, J W Lee. 1994. Variance components testing in the longitudinal mixed effects model.. https://pubmed.ncbi.nlm.nih.gov/7786999/

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

KEEP EXPLORING

Related citations

PubMed1998

Oral snuff, smoking habits and alcohol consumption in relation to oral cancer in a Swedish case-control study.

The use of oral snuff is a widespread habit in Sweden. We investigated whether the use of Swedish moist snuff leads to an increasing risk of oral cancer. Other risk factors such as smoking tobacco and alcoholic beverages were also investigated. Our study comprised 410 patients with oral cancer, from the period 1980-1989, and 410 matched controls. All subjects received a mailed questionnaire. The response rates were 96% and 91% for cases and controls, respectively. In the study, a total of 20% of all subjects, cases and controls, were active or ex-snuff users. The univariate analysis did not show any increased risk [odds ratio (OR) 0.7, 95% confidence interval (CI) 0.4-1.1] for active snuff users. We found an increased risk (OR 1.8, CI 1.1-2.7) for oral cancer among active smokers. Alcohol consumption showed the strongest risk for oral cancer. Among consumers of beer, an increased risk of 1.9 (CI 0.9-3.9) was found. Corresponding ORs for wine and liquor were 1.3 (CI 0.9-1.8) and 1.6 (CI 1.1-2.3), respectively. A dose-response effect was observed. Although not statistically significant, a multivariate analysis similarly suggested that the most important risk factors were beer and liquor consumption, followed by smoking.

Age Factors↗

PubMed1998

Oral contraceptives and breast cancer risk in Taiwan, a country of low incidence of breast cancer and low use of oral contraceptives.

One hundred and seventy four (81% of all) pathologically confirmed new incident cases of female breast cancer identified from a medical center in Taipei from February, 1993 to June, 1994 were selected as the case group. Four hundred and fifty three inpatient controls who were without obstetric-gynecological, breast, or malignant diseases were individually matched for each case by age and date of admission. Information was obtained through direct interview and review of medical records. Conditional logistic regression was used to estimate the effects of each risk factor. After adjusting for education level, body mass index, age at menarche and first full-term pregnancy, parity, menopausal status and age at menopause, lifetime lactation, use of lactation inhibition hormones, and family history of breast cancer, breast cancer risk significantly elevated in use of OC before 25 years old and before 1971. In stratified analysis, significantly higher risk were found in OC use before 25 years old and in duration of use less than one year among post-menopausal subjects. Our results support the notion that OC use in early life for younger women and in early calendar years increase breast cancer risk.

Age Factors↗

PubMed1998

Serum distemper virus and parvovirus antibody titers among dogs brought to a veterinary hospital for revaccination.

OBJECTIVE: To determine serum canine distemper virus (CDV) and canine parvovirus (CPV) antibody titers in healthy dogs brought to a veterinary hospital for revaccination. DESIGN: Case series. ANIMALS: 122 dogs. PROCEDURE: Serum antibody titers were measured by means of hemagglutination inhibition (CPV titers) or serum neutralization (CDV titers) at the time dogs were brought to the hospital for revaccination. All dogs had been vaccinated between 271 and 1,665 days previously. Dogs were grouped by age, breed (purebred vs mixed breed), sex, and weight to determine whether these factors were associated with antibody titers. Serum CPV titers > or = 1:80 and serum CDV titers > or = 1:96 were considered protective. RESULTS: Breed, sex, and weight were not significantly associated with serum CPV and CDV titers. Age was significantly associated with CPV titer, with younger dogs having higher titers, but was not associated with CDV titer. Thirty-three of 122 (27%; 95% confidence interval, 19.0 to 34.9%) dogs had a less-than-protective CPV titer. Twenty-five of 117 (21%; 95% confidence interval, 13.6 to 28.4%) dogs had a less-than-protective CDV titer. CLINICAL IMPLICATIONS: Results suggest that, on the basis of serum antibody titers, the current practice of annual revaccination of dogs against CPV and CDV infection should be maintained. Measurement of antibody titers to determine whether revaccination is truly needed would seem justifiable in those dogs that have previously had an adverse reaction to vaccination.

Age Factors↗