Imputation of Incident Events in Longitudinal Cohort Studies

Howard, George; McClure, Leslie A.; Moy, Claudia S.; Safford, Monika M.; Cushman, Mary; Judd, Suzanne E.; Kissela, Brett M.; Kleindorfer, Dawn; Howard, Virginia J.; Rhodes, David; Muntner, Paul; Tiwari, Hemant

doi:10.1093/aje/kwr155

Cited by 22 publications

(22 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Because of missing stroke data due to ongoing adjudication of stroke events, as well as differential retrieval of medical records, we utilized multiple imputation techniques to classify in process stroke events using a logistic function predicting the likelihood that an attempted record retrieval would result in an adjudicated stroke. 12 Ten imputed datasets were used, and 22% of reported events were eligible to be imputed.…”

Section: Methodsmentioning

confidence: 99%

Assessing the Performance of the Framingham Stroke Risk Score in the Reasons for Geographic and Racial Differences in Stroke Cohort

McClure

Kleindorfer

Kissela

et al. 2014

Stroke

View full text Add to dashboard Cite

Background and Purpose The most well-known stroke risk score is the Framingham Stroke Risk Score (FSRS), which was developed during the higher stroke risk period of the 1990’s, and has not been validated for blacks. We assessed the performance of the FSRS among participants in the Reasons for Geographic And Racial Differences in Stroke (REGARDS) study to determine whether it is useful in both blacks and whites. Methods Expected annualized stroke rates from the FSRS were compared to observed stroke rates overall and within strata defined by FSRS risk factors (age, sex, systolic blood pressure, use of antihypertensive medications, diabetes, smoking, atrial fibrillation, left ventricular hypertrophy and prevalent coronary heart disease). Results Among 27,748 participants stroke-free at baseline, 715 stroke events occurred over 5.6 years of follow-up. FSRS-estimated incidence rates of stroke were 1.6 times higher than observed for black men, 1.9 times higher for white men, 1.7 times higher for black women and 1.7 times higher for white women. This overestimation was consistent among most subgroups of FSRS factors, although the magnitude of overestimation varied by the risk factor assessed. Conclusions While higher FSRS was associated with higher stroke risk, the FSRS overestimated observed stroke rates in this study, particularly in certain subgroups. This may be due to temporal declines in stroke rates, secular trends in prevention treatments, or differences in populations studied. More accurate estimates of event rates are critical for planning research, including clinical trials, and targeting health-care efforts.

show abstract

Section: Methodsmentioning

confidence: 99%

Assessing the Performance of the Framingham Stroke Risk Score in the Reasons for Geographic and Racial Differences in Stroke Cohort

McClure

Kleindorfer

Kissela

et al. 2014

Stroke

View full text Add to dashboard Cite

show abstract

“…Because of missing stroke data due to ongoing adjudication and differential retrieval of medical records, we applied multiple imputation techniques to classify potential stroke events using a logistic function predicting the likelihood that an attempted record retrieval would result in an adjudicated stroke. 24 The analysis cohort comprised 24,544 participants, excluding those with no follow-up (2%), self-reported stroke at baseline (6%), missing residential questionnaire (8%), and place of birth missing or out of the United States (2%). The end of follow-up for analysis was April 1, 2011, resulting in 141,736 person-years of follow-up.…”

Section: 18mentioning

confidence: 99%

Effect of duration and age at exposure to the Stroke Belt on incident stroke in adulthood

Howard¹,

McClure²,

Glymour³

et al. 2013

Neurology

Self Cite

View full text Add to dashboard Cite

Objective: To assess whether there are differences in the strength of association with incident stroke for specific periods of life in the Stroke Belt (SB).Methods: The risk of stroke was studied in 24,544 black and white stroke-free participants, aged 451, in the Reasons for Geographic and Racial Differences in Stroke study, a national populationbased cohort enrolled [2003][2004][2005][2006][2007]. Incident stroke was defined as first occurrence of stroke over an average 5.8 years of follow-up. Residential histories (city/state) were obtained by questionnaire. SB exposure was quantified by combinations of SB birthplace and current residence and proportion of years in SB during discrete age categories (0-12, 13-18, 19-30, 31-45, last 20 years) and entire life. Proportional hazards models were used to establish association of incident stroke with indices of exposure to SB, adjusted for demographic, socioeconomic (SES), and stroke risk factors.Results: In the demographic and SES models, risk of stroke was significantly associated with proportion of life in the SB and with all other exposure periods except birth, ages 31-45, and current residence. The strongest association was for the proportion of the entire life in SB. After adjustment for risk factors, the risk of stroke remained significantly associated only with proportion of residence in SB in adolescence (hazard ratio 1.17, 95% confidence interval 1.00-1.37).Conclusions: Childhood emerged as the most important period of vulnerability to SB residence as a predictor of future stroke. Improvement in childhood health circumstances should be considered as part of long-term health improvement strategies in the SB. The Stroke Belt region (SB) in the southeastern United States has higher stroke mortality than the rest of the United States, 1,2 and national data suggest that similar regional differences exist in stroke incidence. [3][4][5][6] There are several hypothesized reasons for this but data for testing the underlying factors are limited. 2 Research to date has shown that only a portion of the higher stroke mortality can be explained by traditional risk factors.7-10 SB has been defined using state of residence at time of death. Population migration may contribute to differences in regional mortality rates but few US studies have considered the impact of population redistribution on geographic disparities in stroke incidence and mortality.11-14 Some, 12-14 but not all, 11 studies have shown that birthplace is associated with cardiovascular mortality, with individuals born in the Southeast having higher rates than those born in other regions of the United States. With respect to stroke incidence, one national study showed increased incidence of self-reported stroke associated with living in the SB during childhood. 6 It is not known whether there is a "sensitive period" for residence in the SB, i.e., a developmental period during which SB exposure has the greatest consequences for later stroke risk, or whether stroke risk accumulates with each additiona...

show abstract

“…Multiple imputation for suspected stroke events that could not be adjudicated because of inability to retrieve records and for records currently in the adjudication process (approximately 10% of suspected stroke events for each) was used to reduce potential bias; details of this approach are described elsewhere. 12 Standard protocol approvals, registrations, and patient consents. The study was approved by the institutional review boards of all participating universities and all participants provided written informed consent.…”

mentioning

confidence: 99%

Differences in the role of black race and stroke risk factors for first vs recurrent stroke

et al. 2016

View full text Add to dashboard Cite

Objectives: To assess whether black race and other cerebrovascular risk factors have a differential effect on first vs recurrent stroke events.Methods: Estimate the differences in the magnitude of the association of demographic (age, back race, sex) or stroke risk factors (hypertension, diabetes, cigarette smoking, atrial fibrillation, left ventricular hypertrophy, or heart disease) for first vs recurrent stroke from a longitudinal cohort study of 29,682 black or white participants aged 45 years and older.Results: Over an average 6.8 years follow-up, 301 of 2,993 participants with a previous stroke at baseline had a recurrent stroke, while 818 of 26,689 participants who were stroke-free at baseline had a first stroke. Among those stroke-free at baseline, there was an age-by-race interaction (p 5 0.0002), with a first stroke risk 2.70 (95% confidence interval: 1.86-3.91) times greater for black than white participants at age 45, but no racial disparity at age 85 (hazard ratio 5 0.91; 95% confidence interval: 0.70-1.18). In contrast, there was no evidence of a higher risk of recurrent stroke at any age for black participants (p . 0.05). The association of traditional stroke risk factors was generally similar for first and recurrent stroke. Conclusion:The association of age and black race differs substantially on first vs recurrent stroke risk, with risk factors playing a similar role. Risk factors for first stroke have been well studied, 1 providing a framework for primary stroke prevention. Multivariable risk functions developed in the Framingham 2 and Cardiovascular Health Study 3 cohorts have identified age, sex, hypertension, diabetes, cigarette smoking, left ventricular hypertrophy, atrial fibrillation, and heart disease as predictors of incident stroke in a stroke-free population. A limitation of those studies was their inability to address the role of race (black race specifically) as a risk factor for stroke; however, there are substantial differences in stroke incidence between black and white persons, with incident stroke in black persons approximately 3 times higher between the ages of 45 and 65 years, but with disparities attenuating by age 85. [4][5][6][7][8] While substantial data document the risk factors for first stroke, there are fewer data establishing the risk factors for recurrent stroke (that is, risk factors for a subsequent stroke in a population with a previous stroke). The

show abstract

Imputation of Incident Events in Longitudinal Cohort Studies

Cited by 22 publications

References 11 publications

Assessing the Performance of the Framingham Stroke Risk Score in the Reasons for Geographic and Racial Differences in Stroke Cohort

Assessing the Performance of the Framingham Stroke Risk Score in the Reasons for Geographic and Racial Differences in Stroke Cohort

Effect of duration and age at exposure to the Stroke Belt on incident stroke in adulthood

Differences in the role of black race and stroke risk factors for first vs recurrent stroke

Contact Info

Product

Resources

About