Dataset Information

Derivative estimation for longitudinal data analysis: Examining features of blood pressure measured repeatedly during pregnancy.

ABSTRACT: Estimating velocity and acceleration trajectories allows novel inferences in the field of longitudinal data analysis, such as estimating change regions rather than change points, and testing group effects on nonlinear change in an outcome (ie, a nonlinear interaction). In this article, we develop derivative estimation for 2 standard approaches-polynomial mixed models and spline mixed models. We compare their performance with an established method-principal component analysis through conditional expectation through a simulation study. We then apply the methods to repeated blood pressure (BP) measurements in a UK cohort of pregnant women, where the goals of analysis are to (i) identify and estimate regions of BP change for each individual and (ii) investigate the association between parity and BP change at the population level. The penalized spline mixed model had the lowest bias in our simulation study, and we identified evidence for BP change regions in over 75% of pregnant women. Using mean velocity difference revealed differences in BP change between women in their first pregnancy compared with those who had at least 1 previous pregnancy. We recommend the use of penalized spline mixed models for derivative estimation in longitudinal data analysis.

SUBMITTER: Simpkin AJ

PROVIDER: S-EPMC6099422 | biostudies-literature | 2018 May

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Derivative estimation for longitudinal data analysis: Examining features of blood pressure measured repeatedly during pregnancy.

Simpkin Andrew J AJ Durban Maria M Lawlor Debbie A DA MacDonald-Wallis Corrie C May Margaret T MT Metcalfe Chris C Tilling Kate K

Statistics in medicine 20180520

Estimating velocity and acceleration trajectories allows novel inferences in the field of longitudinal data analysis, such as estimating change regions rather than change points, and testing group effects on nonlinear change in an outcome (ie, a nonlinear interaction). In this article, we develop derivative estimation for 2 standard approaches-polynomial mixed models and spline mixed models. We compare their performance with an established method-principal component analysis through conditional ...[more]

PMID: 29781174

Similar Datasets

Project description:BackgroundRaised blood pressure (BP) affects approximately 10% of pregnancies worldwide, and a high proportion of affected women develop pre-eclampsia. This study aimed to evaluate the feasibility of self-monitoring of BP in pregnancy in women at higher risk of pre-eclampsia.MethodsThis prospective cohort study of self-monitoring BP in pregnancy was carried out in two hospital trusts in Birmingham and Oxford and thirteen primary care practices in Oxfordshire. Eligible women were those defined by the UK National Institute for Health and Care Excellence (NICE) guidelines as at higher risk of pre-eclampsia. A total of 201 participants were recruited between 12 and 16 weeks of pregnancy and were asked to take two BP readings twice daily three times a week through their pregnancy. Primary outcomes were recruitment, retention and persistence of self-monitoring. Study recruitment and retention were analysed with descriptive statistics. Survival analysis was used to evaluate the persistence of self-monitoring and the performance of self-monitoring in the early detection of gestational hypertension, compared to clinic BP monitoring. Secondary outcomes were the mean clinic and self-monitored BP readings and the performance of self-monitoring in the detection of gestational hypertension and pre-eclampsia compared to clinic BP.ResultsOf 201 women recruited, 161 (80%) remained in the study at 36 weeks or to the end of their pregnancy, 162 (81%) provided any home readings suitable for analysis, 148 (74%) continued to self-monitor at 20 weeks and 107 (66%) at 36 weeks. Self-monitored readings were similar in value to contemporaneous matched clinic readings for both systolic and diastolic BP. Of the 23 who developed gestational hypertension or pre-eclampsia and self-monitored, 9 (39%) had a raised home BP prior to a raised clinic BP.ConclusionsSelf-monitoring of BP in pregnancy is feasible and has potential to be useful in the early detection of gestational hypertensive disorders but maintaining self-monitoring throughout pregnancy requires support and probably enhanced training.

Project description:ImportanceDiscrepancies in blood pressure (BP) estimates lead to incomparable BP assessment.ObjectiveTo determine intraindividual discrepancies in BP estimates and classifications based on different BP estimation protocols.Design, setting, and participantsThis cross-sectional study was a secondary analysis of data from the May Measurement Month Taiwan in 2017 and 2018, which were cross-sectional survey campaigns at pharmacies nationwide to raise awareness of high BP. Participants were volunteers aged 20 years or older. Analysis was conducted from February 2 to August 7, 2020.ExposurePharmacist-measured sitting BP using oscillometric sphygmomanometers.Main outcomes and measuresA total of 7 BP estimation protocols were assessed according to the latest American College of Cardiology (ACC), Chinese Hypertension League (CHL), European Society of Cardiology (ESC), International Society of Hypertension, Japanese Society of Hypertension, and National Institute of Health and Care Excellence (NICE) hypertension guidelines, and the proposed Averaging the Lowest Two systolic readings protocol. According to BP classification schemes of ESC and ACC guidelines, intraindividual discrepancies were identified if classification inconsistencies among 7 BP estimates were present.ResultsOf 81 041 participants, 62 647 adults with 3 BP readings were included. The median (interquartile range) age was 59.0 (46.0-69.0) years, and 31 922 (51.5%) were women. The intraindividual maximum mean (SD) differences in systolic/diastolic BP estimates among the seven protocols were 4.8 (4.3)/3.3 (3.1) mm Hg. The highest prevalence of BP of 140/90 mm Hg or higher was by CHL (16 405 participants [26.2%]) and the lowest was by Averaging the Lowest Two (13 996 participants [22.3%]; P < .001); while the highest prevalence of 130/80 mm Hg or higher was by NICE (37 232 participants [59.4%]) and the lowest prevalence was by Averaging the Lowest Two (32 788 participants [52.4%]; P < .001). Compared with the other 6 estimates, Averaging the Lowest Two reclassified 7.3% to 15.8% of participants designated as 140/90 mm Hg or higher to less than 140/90 mm Hg, and 4.9% to 14.1% of those as 130/80 mm Hg or higher to less than 130/80 mm Hg. Intraindividual discrepancies in classifications occurred in 19 815 participants (31.6%) with the ESC classification and 16 401 participants (26.2%) with the ACC BP classification. Classification agreements were the lowest between NICE (κ coefficient, 0.667 [95% CI, 0.662-0.671]) and ESC protocols (κ coefficient, 0.705 [95% CI, 0.701-0.709]).Conclusions and relevanceThis cross-sectional study of adults in Taiwan found that different BP estimation protocols led to considerable intraindividual discrepancies in BP estimates and classifications. These findings suggest that the Averaging the Lowest Two protocol is less likely to overestimate BP and could serve as a prudent recommendation for BP estimation.

Dataset Information

Derivative estimation for longitudinal data analysis: Examining features of blood pressure measured repeatedly during pregnancy.

Publications

Derivative estimation for longitudinal data analysis: Examining features of blood pressure measured repeatedly during pregnancy.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets