PubMed Health⌕ Search
Home/Archive/PubMed

PubMed · 10369501

Presenting statistical uncertainty in trends and dose-response relations.

S Greenland·K B Michels·J M Robins·C Poole·W C Willett
View original source ↗Export metadata ↓

Abstract

When one estimates the effects of a polytomous exposure, it is common practice to express all effects relative to a baseline or reference level. Certain authors have challenged this practice and proposed alternatives, which we review here. One alternative, the "floating absolute risk" method, can supply useful statistics and trend graphs, but it does not yield valid confidence intervals for relative risks. All categorical methods have further shortcomings when the exposure is continuous, however. These shortcomings can be addressed by plotting or tabulating confidence limits for points on a flexible curve fitted to the uncategorized data.

Explore related subjects

Birth WeightBreast NeoplasmsCase-Control StudiesConfounding Factors, EpidemiologicData Interpretation, StatisticalFemaleHumansMathematical ComputingOdds RatioRiskJournal Article

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

S Greenland, K B Michels, J M Robins, C Poole, W C Willett. 1999-06-15. Presenting statistical uncertainty in trends and dose-response relations.. https://doi.org/10.1093/oxfordjournals.aje.a009761

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

KEEP EXPLORING

Related citations

PubMed2007

Infant growth, development and tooth emergence patterns: A longitudinal study from birth to 6 years of age.

OBJECTIVE: To test the association of child's growth/development status and socioeconomic variables with dental emergence patterns. DESIGN: A sample of 359 children belonging to a birth cohort of over 5000 children born in 1993 in Pelotas (Brazil) was studied in 1998-1999. Previously trained observers collected birthweight, head circumference at birth, birth height, gestational age, smoking during pregnancy, child's sex and skin colour, height by age at 6 months of age, caries experience in the deciduous teeth, number of emerged teeth at 6 and 12 months of age and emergence stages of the first permanent molars at 6 years of age. Family income, parental educational attainment and anthropometric measurements were collected at children's homes. Poisson regression models were applied in order to select best predictors of pairs of emerged teeth at 6 and 12 months of age and emerged (yes/no) first permanent molars at 6 years of age. RESULTS: Children shorter than 49cm had, on average, less pairs of emerged teeth at 6 months of age. Twelve-month-old children with birth height < or =49cm, presenting height-for-age deficit at 6 months of age and the females also presented less pairs of emerged teeth. Stunted children at 6 months of age were more likely to have non-emerged first upper left and lower right permanent molars at 6 years of age. CONCLUSIONS: Important developmental and nutritional variables at birth and throughout the early life may predict the numbers of emerged teeth children have in their oral cavity.

Birth Weight↗
PubMed2007

Long-term neurodevelopmental outcome of triplets.

OBJECTIVE: To analyze the incidence of neurodevelopmental disabilities in triplets and to find out possible connection between the outcome and perinatal events. DESIGN: Retrospective cohort study of 94 triplets and their outcome at 24-144 months of age correlated with gestational age, birth weight, pregnancy complications, early neonatal period, neonatal cranial ultrasound, period of birth (1985-1995, 1996-2000) and type of antenatal care. RESULTS: Sixty-two triplets are healthy, 15 suffer cerebral palsy (CP) and 17 minimal cerebral dysfunction (MCD). Adverse outcome correlates significantly with prematurity, low birth weight and maternal age. In multivariate analysis, both cerebral palsy and minor disabilities correlate significantly with early neonatal complications, neonatal cranial ultrasound with later CP (p<0.01), and MCD with preterm rupture of membranes (p=0.047). Children conceived spontaneously do worse than those born after assisted reproduction (p=0.004), those born in the time period 1996-2000 do better than those born before (p=0.021). Seventy-seven percent (77%) of newborns delivered in the time period 1996-2000 and after level 1 antenatal care was introduced, compared with 54% being delivered in the time period before 1996 and with less meticulous types of antenatal care, remain healthy (p=0.015). CONCLUSION: Triplets are still at high risk for long-term neurodevelopmental complications. Stringent perinatal care might appear important determinant of their long-term outcome.

Birth Weight↗