Dataset Information

Combining self-reported and objectively measured survey data to improve hypertension prevalence estimates: Portuguese experience.

ABSTRACT:

Background

Accurate data on hypertension is essential to inform decision-making. Hypertension prevalence may be underestimated by population-based surveys due to misclassification of health status by participants. Therefore, adjustment for misclassification bias is required when relying on self-reports. This study aims to quantify misclassification bias in self-reported hypertension prevalence and prevalence ratios in the Portuguese component of the European Health Interview Survey (INS2014), and illustrate application of multiple imputation (MIME) for bias correction using measured high blood pressure data from the first Portuguese health examination survey (INSEF).

Methods

We assumed that objectively measured hypertension status was missing for INS2014 participants (n = 13,937) and imputed it using INSEF (n = 4910) as auxiliary data. Self-reported, objectively measured and MIME-corrected hypertension prevalence and prevalence ratios (PR) by sex, age group and education were estimated. Bias in self-reported and MIME-corrected estimates were computed using objectively measured INSEF data as a gold-standard.

Results

Self-reported INS2014 data underestimated hypertension prevalence in all population subgroups, with misclassification bias ranging from 5.2 to 18.6 percentage points (pp). After MIME-correction, prevalence estimates increased and became closer to objectively measured ones, with bias reduction to 0 pp - 5.7 pp. Compared to objectively measured INSEF, self-reported INS2014 data considerably underestimated prevalence ratio by sex (PR = 0.8, 95CI = [0.7, 0.9] vs. PR = 1.2, 95CI = [1.1, 1.4]). MIME successfully corrected direction of association with sex in bivariate (PR = 1.1, 95CI = [1.0, 1.3]) and multivariate analyses (PR = 1.2, 95CI = [1.0, 1.3]). Misclassification bias in hypertension prevalence ratios by education and age group were less pronounced and did not require correction in multivariate analyses.

Conclusions

Our results highlight the importance of misclassification bias analysis in self-reported hypertension. Multiple imputation is a feasible approach to adjust for misclassification bias in prevalence estimates and exposure-outcomes associations in survey data.

SUBMITTER: Kislaya I

PROVIDER: S-EPMC8028082 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:BACKGROUND:Sedentary behavior (SED) is a potential risk factor for poor pregnancy outcomes. We evaluated the validity of several common and one new method to assess SED across three trimesters of pregnancy. METHODS:This cohort study of pregnant women measured objective and self-reported SED each trimester via thigh-worn activPAL3 micro (criterion), waist-worn Actigraph GT3X, and self-report from the Pregnancy Physical Activity Questionnaire (PPAQ) and the de novo Sedentary Behavior Two Domain Questionnaire (SB2D). SED (hours per day) and percent time in SED (SED%) from activPAL were compared to GT3X, SB2D, and PPAQ using Pearson's r, ICC, Bland-Altman analysis, and comparison of criterion SED and SED% across tertiles of alternative methods. RESULTS:Fifty-eight women (mean age 31.5 ± 4.8 years; pre-pregnancy BMI 25.1 ± 5.6 kg/m2; 76% white) provided three trimesters of valid activPAL data. Compared to activPAL, GT3X had agreement ranging from r = 0.54-0.66 and ICC = 0.52-0.65. Bland-Altman plots revealed small mean differences and unpatterned errors, but wide limits of agreement (greater than ±2 h and ± 15%). The SB2D and PPAQ had r < 0.5 and ICC < 0.3 vs. activPAL SED, with lower agreement during the 2nd and 3rd trimesters, and performed poorly in Bland-Altman analyses. SED% from the modified SB2D performed best of the self-reported instruments with modest mean differences, r ranging from 0.55 to 0.60, and ICCs from 0.31-0.33; though, limits of agreement were greater than ±35%. Significant trends in activPAL SED were observed across increasing tertiles of SB2D SED in the 1st and 3rd trimesters (both p ≤ 0.001), but not the 2nd trimester (p = 0.425); and for PPAQ SED in the 1st and 2nd trimesters (both p < 0.05), but not the 3rd trimester (p = 0.158). AcitvPAL SED and SED% increased significantly across tertiles of GT3X SED and SED% as well as SB2D SED% (all p-for-trend ≤ 0.001). CONCLUSIONS:Compared to activPAL, waist-worn GT3X produced moderate agreement, though similar mean estimates of SED across pregnancy. Self-report questionnaires had large absolute error and wide limits of agreement for SED hr./day; SB2D measurement of SED% was the best self-report method. These data suggest activPAL be used to measure SED when possible, followed by GT3X, and - when necessary - SB2D assessing SED% in pregnancy. TRIAL REGISTRATION:www.clinicaltrials.gov NCT03084302 on 3/20/2017.

Project description:ObjectivesAlthough previous studies have reported lower mortality and morbidity in people with higher daily step counts, the association between frailty and objectively measured step counts has not been evaluated well. We investigated the association between step counts and prevalence of frailty in community-dwelling older adults.DesignA cross-sectional study.SettingThe Kyoto-Kameoka study in Japan.ParticipantsWe used data of 3,616 Japanese older adults, aged 65 years or older, with valid daily step count data, obtained by an accelerometer-based pedometer.MeasurementsThe step count during 4 or more days was objectively obtained by a validated triaxial accelerometer. Participants were classified by quartiles (Qs) based on their step counts. Frailty was defined using the Fried phenotype (FP) model and the Kihon Checklist (KCL). We evaluated the association between prevalence of frailty and step counts using multivariate logistic regression and the restricted cubic spline model.ResultsMean step counts across low-to-high Qs of distribution were 1,759, 2,988, 4,377, and 7,200 steps/day, respectively. The prevalence of frailty, as defined by the FP model and KCL, was 11.3% and 26.8%, respectively. After adjusting for confounders, there was a negative association between the odds ratio (OR) and prevalence of frailty, as defined by the FP model among people with higher step counts (Q1: reference; Q2: OR = 0.73; 95% confidence interval (CI) = 0.56-0.96; Q3: OR = 0.56; 95% CI = 0.42-0.76; and Q4: OR = 0.41; 95% CI = 0.30-0.57; P for trend <.001). The mean step count of the population was 4,081. The OR of frailty for a 1,000-steps/day increment was 0.74 (95% CI = 0.58-0.91) and 0.85 (95% CI = 0.72-0.97) below 4,000 steps and above 4,000 steps, respectively. In the spline model, this relationship was similar between the FP model and KCL.ConclusionThese findings suggest that slightly increasing the current step count, as by 1,000 steps/day (about 10 minutes of activity), may potentially prevent frailty. J Am Geriatr Soc 68:2310-2318, 2020.