Using Median Regression to Obtain Adjusted Estimates of Central Tendency for Skewed Laboratory and Epidemiologic Data

McGreevy, Katharine; Lipsitz, Stuart R.; Linder, Jeffrey A.; Rimm, Eric B.; Hoel, David G.

doi:10.1373/clinchem.2008.106260

Cited by 61 publications

(51 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The hypothesis is well justified in this case, since the measurements are performed by direct observation of the particles and the presence of pollutants can be excluded. Adjustments like those proposed by McGreevy et al (2008) are thus unnecessary. Mean debris diameter (D d ) and its standard deviation can thus be used to describe location and spread of data and are listed in Table 3. SEM observation of the sectioned and polished debris proves that most of the particles are hollow and show a foamy inner morphology.…”

Section: Resultsmentioning

confidence: 99%

Bridges of debris in the EDD process: Going beyond the thermo-electrical model

Gatto

Bassoli

Denti

et al. 2013

Journal of Materials Processing Technology

View full text Add to dashboard Cite

Section: Resultsmentioning

confidence: 99%

Bridges of debris in the EDD process: Going beyond the thermo-electrical model

Gatto

Bassoli

Denti

et al. 2013

Journal of Materials Processing Technology

View full text Add to dashboard Cite

“…Quantile regression was used to obtain adjusted medians (interquartile ranges) and the Wald test was used to test differences in the adjusted medians between CKD patients and controls [19]. Age, gender, race, high school education, current cigarette smoking, weekly alcohol consumption, physical activity (≥twice per week), BMI, low-density lipoprotein (LDL) cholesterol, serum glucose, systolic blood pressure, history of hypertension, diabetes, CVD, and medication uses were adjusted in these analyses.…”

Section: Methodsmentioning

confidence: 99%

Elevated Plasma Levels of Endostatin Are Associated with Chronic Kidney Disease

Chen

Hamm²,

Kleinpeter³

et al. 2012

Am J Nephrol

View full text Add to dashboard Cite

Background/Aims: Angiogenesis may play an important role in the renal repair process after injury. We investigated the association between plasma endostatin, an endothelial-specific antiangiogenic factor, and chronic kidney disease (CKD). Methods: We compared plasma endostatin levels in 201 CKD patients and 201 controls. CKD was defined as estimated glomerular filtration rate (eGFR) <60 ml/min/1.73 m² or presence of albuminuria (≥30 mg/24 h). Results: After adjustment for established CKD risk factors, the median (interquartile range) of plasma endostatin was 276.7 ng/dl (199.3–357.5) in patients with CKD and 119.4 ng/dl (103.7–134.6) in controls without CKD (p < 0.0001 for group difference). log-transformed plasma endostatin was significantly and inversely correlated with eGFR (r = –0.83, p < 0.0001) and positively correlated with log-transformed urine albumin (r = 0.66, p < 0.0001) in the study participants. In addition, one standard deviation increase in log-transformed plasma endostatin (0.55 ng/dl) was associated with a decline in eGFR of –26.2 ml/min and an increase in urine albumin of 3.26 mg/ 24 h after adjusting for multiple covariables. Furthermore, the multivariable-adjusted odds ratio for CKD comparing the highest tertile (≥131.4 ng/dl) to the two lower tertiles of plasma endostatin was 21.6 (95% CI: 10.2–45.5; p < 0.0001). Conclusion: These data indicate that elevated plasma endostatin is strongly and independently associated with CKD. Prospective cohort studies and clinical trials are warranted to further examine the causal relationship between endostatin and risk of CKD and to develop novel interventions targeting circulating endostatin aimed at reducing CKD risk.

show abstract

“…Multivariable quantile regression modeling was used to estimate the relationship between sociodemographic and healthrelated factors and the median of the LSMA-C scores. 16 Variables with P-values less than 0.25 in the univariable analyses were entered simultaneously into the multivariable model and were subsequently removed using a backward stepwise method.…”

Section: Statistical Analysesmentioning

confidence: 99%