PubMed Health⌕ Search
Home/Archive/PubMed

PubMed · 9883556

A generalized estimating equations approach for testing ordered group effects with repeated measurements.

T Park·D W Shin·C G Park
View original source ↗Export metadata ↓

Abstract

In repeated measures studies, we are often interested in comparing group effects in which groups are associated with a certain order relation. We propose testing procedures for ordered group effects using the generalized estimating equations (GEE) approach of Liang and Zeger (1986, Biometrika 73, 13-22). The order-constrained GEE estimators of group effects are approximated by the isotonic regression of the unconstrained GEE estimators. Based on these constrained estimators, we construct test statistics for detecting ordered group effects. The limiting distributions of the test statistics are mixtures of chi-square distributions. A Monte Carlo experiment shows improved performances of the proposed tests over the usual chi-square tests in detecting ordered group effects. The proposed test procedures are illustrated by familial polyposis supplementation trial data.

Explore related subjects

Adenomatous Polyposis ColiBiometryChi-Square DistributionDietary FiberHumansMonte Carlo MethodRandomized Controlled Trials as TopicVitaminsJournal ArticleResearch Support, Non-U.S. Gov't

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

T Park, D W Shin, C G Park. 1998. A generalized estimating equations approach for testing ordered group effects with repeated measurements.. https://pubmed.ncbi.nlm.nih.gov/9883556/

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

KEEP EXPLORING

Related citations

PubMed2007

Familial adenomatous polyposis patients have high levels of arachidonic acid and docosahexaenoic acid and low levels of linoleic acid and alpha-linolenic acid in serum phospholipids.

Familial adenomatous polyposis (FAP) provides a model of APC inactivation as an early genetic event for the approximately 85% of colorectal cancers that develop from polyps. Abnormal fatty acid composition of tissues and serum phospholipids has been linked to cancer risk. Our aim was to describe the composition of fatty acids in serum phospholipids in 38 colectomized FAP patients as compared to 160 healthy subjects. Mean fatty acid intakes were similar between the groups. Colectomy was done on average 16 years prior to inclusion, and 18% were diagnosed with colorectal cancer at colectomy. The levels (weight %) of linoleic and alpha-linolenic acid were higher among the reference subjects (difference: 3.96, 95% confidence interval (CI) = 2.87, 5.04, and difference: 0.06, 95% CI = 0.04, 0.08, respectively), and the levels of arachidonic and docosahexaenoic acid were lower (difference: -3.70, 95% CI = -4.35, -3.06, and difference: -5.26, 95% CI = -6.25, -4.28, respectively) as compared to the FAP patients (all p < or = 0.0001). The abnormal fatty acid composition was not related to time since colectomy, intestinal reconstruction or history of colorectal cancer for any of the fatty acids assessed. Compositional differences in the fatty acid profile of serum phospholipids have not been described before in FAP patients. Further studies are needed to confirm these findings and assess clinical significances of a possible distorted fatty acid metabolism, including a potentially different dietary need of essential fatty acids. The relevance of these findings for APC induced cancers remains unclear.

Adenomatous Polyposis Coli↗
PubMed2007

Balance between endoscopic and genetic information in the choice of ileorectal anastomosis for familial adenomatous polyposis.

BACKGROUND AND OBJECTIVES: The number of rectal polyps and the site of mutations in the APC (Adenomatous polyposis coli) gene have been used to guide the surgical management in patients with familial adenomatous polyposis (FAP). The aim of this study is to assess the utility of the APC mutation screening compared to the degree of the rectal polyposis in surgical decision making. METHODS: The post-surgical courses of 25 patients submitted to subtotal colectomy with ileorectal anastomosis (IRA) were reviewed. Preservation of the rectum was prospectively decided on the basis of well-defined endoscopic criteria. The number of rectal polyps was assessed preoperatively and every 6-12 months. APC gene was screened for mutations by heteroduplex analysis, single strand conformation polymorphism, in vitro synthesized protein (IVSP), and DNA sequencing. Patients negative for APC mutations were tested for MYH mutations. RESULTS: On the basis of preoperative polyp rectal count we categorized patients as follows: Group I, 5 or fewer adenomas; Group II, 6-9 adenomas; Group III, 10 or more adenomas. After a follow-up ranging from 12 to 225 months we have observed a significant difference of recurrent rectal adenomas between Groups I-II versus III. No difference was detected among patients of Group I and II. The mean number of adenomas/year/patient was 0.67, 1.62, and 9.29 for Group I, II, and III, respectively. Carpeting polyposis of the rectal stump developed in three patients with APC mutation at codon 1309 and two of them needed later proctectomy. Diffuse rectal polyposis was observed in one patient with mutation at exon 9 who had 10 small polyps at time of surgery. Mutation at the 5'-end of APC (codons 144-232), mutation of MYH and unknown APC or MYH mutation were correlated with a low number of polyps both at presentation and follow-up. No IRA patients developed rectal cancer. CONCLUSIONS: In our experience fewer than 10 rectal polyps at presentation can predict a favorable outcome after IRA. Identification of specific germ-line APC or MYH mutation can address the choice of surgical treatment.

Adenomatous Polyposis Coli↗