Assessing secular trends in blood pressure: A multiple-imputation approach

Daniel F. Heitjan, J. Richard Landis

Research output: Contribution to journalArticle

19 Citations (Scopus)

Abstract

The National Center for Health Statistics makes available data from three national health evaluation surveys that it has conducted since 1960: NHES I (1960–1962), NHANES I (1971–1975), and NHANES II (1976–1980). There has been considerable interest in using these data to assess secular trends in cardiovascular risk factors such as blood pressure (BP). Unfortunately, underlying trends in BP are confounded with trends in physician treatment of hypertension over the same period; in the early 1960s it was rare to treat hypertension, whereas by the late 1970s it had become quite common. Our approach to estimating the underlying trends is to take untreated BP to be the variable of interest and to consider it missing in those subjects who are under treatment. We then use a multiple-imputation scheme to construct estimates of trend parameters that adjust for the incompleteness of the original data. Because our imputations depend on certain model features that the data cannot address, we form estimates under different models and compare the results. Our analyses suggest that trend estimates are sensitive to the assumed model, and naive estimates that do not adjust for treatment trends appear to be overly optimistic.

Original languageEnglish (US)
Pages (from-to)750-759
Number of pages10
JournalJournal of the American Statistical Association
Volume89
Issue number427
DOIs
StatePublished - Jan 1 1994

Fingerprint

Multiple Imputation
Blood Pressure
Hypertension
Estimate
Health
Feature Model
Incompleteness
Imputation
Risk Factors
Trends
Multiple imputation
Blood pressure
Secular trend
Statistics
Evaluation
Model

Keywords

  • Bayesian bootstrap
  • Hot deck
  • Incomplete data
  • Missing data
  • Observational study
  • Predictive-mean matching

ASJC Scopus subject areas

  • Statistics and Probability
  • Statistics, Probability and Uncertainty

Cite this

Assessing secular trends in blood pressure : A multiple-imputation approach. / Heitjan, Daniel F.; Landis, J. Richard.

In: Journal of the American Statistical Association, Vol. 89, No. 427, 01.01.1994, p. 750-759.

Research output: Contribution to journalArticle

@article{9b9dcb121efd44eeb7d19be38865cdb9,
title = "Assessing secular trends in blood pressure: A multiple-imputation approach",
abstract = "The National Center for Health Statistics makes available data from three national health evaluation surveys that it has conducted since 1960: NHES I (1960–1962), NHANES I (1971–1975), and NHANES II (1976–1980). There has been considerable interest in using these data to assess secular trends in cardiovascular risk factors such as blood pressure (BP). Unfortunately, underlying trends in BP are confounded with trends in physician treatment of hypertension over the same period; in the early 1960s it was rare to treat hypertension, whereas by the late 1970s it had become quite common. Our approach to estimating the underlying trends is to take untreated BP to be the variable of interest and to consider it missing in those subjects who are under treatment. We then use a multiple-imputation scheme to construct estimates of trend parameters that adjust for the incompleteness of the original data. Because our imputations depend on certain model features that the data cannot address, we form estimates under different models and compare the results. Our analyses suggest that trend estimates are sensitive to the assumed model, and naive estimates that do not adjust for treatment trends appear to be overly optimistic.",
keywords = "Bayesian bootstrap, Hot deck, Incomplete data, Missing data, Observational study, Predictive-mean matching",
author = "Heitjan, {Daniel F.} and Landis, {J. Richard}",
year = "1994",
month = "1",
day = "1",
doi = "10.1080/01621459.1994.10476808",
language = "English (US)",
volume = "89",
pages = "750--759",
journal = "Journal of the American Statistical Association",
issn = "0162-1459",
publisher = "Taylor and Francis Ltd.",
number = "427",

}

TY - JOUR

T1 - Assessing secular trends in blood pressure

T2 - A multiple-imputation approach

AU - Heitjan, Daniel F.

AU - Landis, J. Richard

PY - 1994/1/1

Y1 - 1994/1/1

N2 - The National Center for Health Statistics makes available data from three national health evaluation surveys that it has conducted since 1960: NHES I (1960–1962), NHANES I (1971–1975), and NHANES II (1976–1980). There has been considerable interest in using these data to assess secular trends in cardiovascular risk factors such as blood pressure (BP). Unfortunately, underlying trends in BP are confounded with trends in physician treatment of hypertension over the same period; in the early 1960s it was rare to treat hypertension, whereas by the late 1970s it had become quite common. Our approach to estimating the underlying trends is to take untreated BP to be the variable of interest and to consider it missing in those subjects who are under treatment. We then use a multiple-imputation scheme to construct estimates of trend parameters that adjust for the incompleteness of the original data. Because our imputations depend on certain model features that the data cannot address, we form estimates under different models and compare the results. Our analyses suggest that trend estimates are sensitive to the assumed model, and naive estimates that do not adjust for treatment trends appear to be overly optimistic.

AB - The National Center for Health Statistics makes available data from three national health evaluation surveys that it has conducted since 1960: NHES I (1960–1962), NHANES I (1971–1975), and NHANES II (1976–1980). There has been considerable interest in using these data to assess secular trends in cardiovascular risk factors such as blood pressure (BP). Unfortunately, underlying trends in BP are confounded with trends in physician treatment of hypertension over the same period; in the early 1960s it was rare to treat hypertension, whereas by the late 1970s it had become quite common. Our approach to estimating the underlying trends is to take untreated BP to be the variable of interest and to consider it missing in those subjects who are under treatment. We then use a multiple-imputation scheme to construct estimates of trend parameters that adjust for the incompleteness of the original data. Because our imputations depend on certain model features that the data cannot address, we form estimates under different models and compare the results. Our analyses suggest that trend estimates are sensitive to the assumed model, and naive estimates that do not adjust for treatment trends appear to be overly optimistic.

KW - Bayesian bootstrap

KW - Hot deck

KW - Incomplete data

KW - Missing data

KW - Observational study

KW - Predictive-mean matching

UR - http://www.scopus.com/inward/record.url?scp=21844510757&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=21844510757&partnerID=8YFLogxK

U2 - 10.1080/01621459.1994.10476808

DO - 10.1080/01621459.1994.10476808

M3 - Article

AN - SCOPUS:21844510757

VL - 89

SP - 750

EP - 759

JO - Journal of the American Statistical Association

JF - Journal of the American Statistical Association

SN - 0162-1459

IS - 427

ER -