Regression to the mean of repeated ambulatory blood pressure monitoring in five studies

Moore, Myles N; Atkins, Emily; Salam, Abdul; Callisaya, Michele L.; Hare, James L.; Marwick, Thomas H.; Nelson, Mark; Wright, Leah; Sharman, James E.; Rodgers, Anthony

doi:10.1097/hjh.0000000000001977

Cited by 38 publications

(29 citation statements)

References 30 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In brief, RTM can be characterized by the high likelihood of values close to the average (mean) of the dependent value for extreme levels of an independent variable [34][35][36]. RTM frequently leads to inaccurate conclusions, wrong inferences, and spurious associations.…”

Section: Introductionmentioning

confidence: 99%

“…RTM frequently leads to inaccurate conclusions, wrong inferences, and spurious associations. RTM is observed both in cross-sectional and longitudinal data; however, it is more commonly discussed for repeated measures in the presence or absence of an intervention or a treatment [34][35][36]. Given that RTM has the potential to lead to wrong conclusions, it has been examined extensively in the literature [37][38][39].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Diminished Returns of Parental Education in Terms of Youth School Performance: Ruling out Regression toward the Mean

et al. 2020

View full text Add to dashboard Cite

Background: Minorities’ Diminished Returns (MDRs) refer to systemically weaker effects of socioeconomic status (SES) indicators on various developmental, behavioral, and health outcomes of ethnic minorities compared to non-Hispanic (non-Latino) Whites. Similar MDRs also exist for the effects of parental education on the school performance of ethnic minority youth. Aim: To assess whether regression toward the mean (RTM) has any role in explaining the diminished effects of parental education on the school performance of Black and Hispanic youth relative to non-Hispanic White youth. Materials and methods: Data for this cross-sectional study came from the Monitoring the Future survey (MTF, 2017), a nationally representative survey of American youth in 12th grade. The sample included 10,262 youth who were 12th graders (typically 17–18 years old). The independent variable was parental education with five categories: Some high school, High school graduate, Some college, College graduate, and Graduate school. The outcome was self-reported school performance measured as grade point average (GPA). Ethnicity was the effect modifier. Analysis of variance (ANOVA) and the Tukey Post Hoc test was used to analyze the data. Data visualization (line graphs) was used to visualize the shape of youth GPA as a function of parental education levels across ethnic groups. Results: While a perfect stepwise increase was seen in youth school performance as a result of parental education improvement, this pattern differed considerably across ethnic groups. Such a perfect stepwise increase in youth school performance as a result of the incremental increase in parental education was missing for Black and Hispanic youth. The shape of the association between parental education and youth school performance ruled out regression toward the mean (RTM) as an explanation for the observed diminished effects of parental education on the school performance of Black and Hispanic youth. Conclusion: Diminished returns of parental education on the school performance of Black and Hispanic youth cannot be explained by regression toward the mean. Other factors and contextual processes, such as segregation, discrimination, racism, and poor quality of schools in urban areas, should be investigated in future research.

show abstract

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Diminished Returns of Parental Education in Terms of Youth School Performance: Ruling out Regression toward the Mean

et al. 2020

View full text Add to dashboard Cite

show abstract

“…Baseline BP cannot generally be considered the best estimate of future untreated BP level, from which treatment efficacy can be judged, now no matter how it is measured. Further, in a separate article ( [31]), we demonstrate regression to the mean even with ambulatory BP monitoring. Therefore, as past trials showed unequivocal benefits of treating high BP, whether based on single or repeated measures, treatment should in general be initiated promptly among patients with high BP.…”

Section: Discussionmentioning

confidence: 60%

“…The results also have relevance for clinical trialscontrol groups will continue to be needed as some degree of regression to the mean is inevitable. Regression to the mean is less but still substantial following repeated measures, better office measures and ambulatory measures [31].…”

Section: Discussionmentioning

confidence: 98%

Effects of blood pressure lowering on cardiovascular events, in the context of regression to the mean

Salam

Atkins

Sundström

et al. 2019

Journal of Hypertension

Self Cite

View full text Add to dashboard Cite

on behalf of the Blood Pressure Lowering Treatment Trialists' Collaboration Ã Objective: To assess the clinical relevance of regression to the mean for clinical trials and clinical practice. Methods: MEDLINE was searched until February 2018 for randomized trials of BP lowering with over 1000 patientyears follow-up per group. We estimated baseline mean BP, follow-up mean (usual) BP amongst patients grouped by 10 mmHg strata of baseline BP, and assessed effects of BP lowering on coronary heart disease (CHD) and stroke according to these BP levels. Results: Eighty-six trials (349 488 participants), with mean follow-up of 3.7 years, were included. Most mean BP change was because of regression to the mean rather than treatment. At high baseline BP levels, even after rigorous hypertension diagnosis, downwards regression to the mean caused much of the fall in BP. At low baseline BP levels, upwards regression to the mean increased BP levels, even in treatment groups. Overall, a BP reduction of 6/3 mmHg lowered CHD by 14% (95% CI 11-17%) and stroke by 18% (15-22%), and these treatment effects occurred at follow-up BP levels much closer to the mean than baseline BP levels. In particular, more evidence was available in the SBP 130-139 mmHg range than any other range. Benefits were apparent in numerous high-risk patient groups with baseline mean SBP less than 140 mmHg. Conclusion: Clinical practice should focus less on pretreatment BP levels, which rarely predict future untreated BP levels or rule out capacity to benefit from BP lowering in high cardiovascular risk patients. Instead, focus should be on prompt, empirical treatment to maintain lower BP for those with high BP and/or high risk.

show abstract

“…Power deficit may be one explanation for the insignificant SBP results; and regression to the mean in repeated BP measurements may affect SBP more than DBP due to higher variability. [45][46][47] Potential mechanistic underpinnings include that DBP may be more affected by hormonal pathways mediated by metabolites; and that atherosclerosis and stiff arteries and arterioles explain increasingly more of SBP with increasing age, drowning out other pathways. The specific potential roles of 3 lipid categories are described below.…”

Section: The Evidence For the Role Of Lipidomics In Hypertensionmentioning

confidence: 99%

Global Plasma Metabolomics to Identify Potential Biomarkers of Blood Pressure Progression

Yang

Salihović

Fall

et al. 2020

ATVB

View full text Add to dashboard Cite

Objective: The pathophysiology of hypertension remains incompletely understood. We investigated associations of circulating metabolites with longitudinal blood pressure (BP) changes in the Prospective Investigation of the Vasculature in Uppsala Seniors cohort and validated the findings in the Uppsala Longitudinal Study of Adult Men cohort. Approach and Results: Circulating metabolite levels were assessed with liquid- and gas-chromatography coupled to mass spectrometry among persons without BP-lowering medication at baseline. We studied associations of baseline levels of metabolites with changes in BP levels and the clinical BP stage between baseline and a follow-up examination 5 years later. In the discovery cohort, we investigated 504 individuals that contributed with 757 observations of paired BP measurements. The mean baseline systolic and diastolic BPs were 144 (19.7)/76 (9.7) mm Hg, and change in systolic and diastolic BPs were 3.7 (15.8)/−0.5 (8.6) mm Hg over 5 years. The metabolites associated with diastolic BP change were ceramide, triacylglycerol, total glycerolipids, oleic acid, and cholesterylester. No associations with longitudinal changes in systolic BP or BP stage were observed. Metabolites with similar structures to the 5 top findings in the discovery cohort were investigated in the validation cohort. Diacylglycerol (36:2) and monoacylglycerol (18:0), 2 glycerolipids, were associated with diastolic BP change in the validation cohort. Conclusions: Circulating baseline levels of ceramide, triacylglycerol, total glycerolipids, and oleic acid were positively associated with longitudinal diastolic BP change, whereas cholesterylester levels were inversely associated with longitudinal diastolic BP change. Two glycerolipids were validated in an independent cohort. These metabolites may point towards pathophysiological pathways of hypertension.

show abstract

Regression to the mean of repeated ambulatory blood pressure monitoring in five studies

Cited by 38 publications

References 30 publications

Diminished Returns of Parental Education in Terms of Youth School Performance: Ruling out Regression toward the Mean

Diminished Returns of Parental Education in Terms of Youth School Performance: Ruling out Regression toward the Mean

Effects of blood pressure lowering on cardiovascular events, in the context of regression to the mean

Global Plasma Metabolomics to Identify Potential Biomarkers of Blood Pressure Progression

Contact Info

Product

Resources

About