scholarly journals Population attributable fractions for continuously distributed exposures

2020 ◽  
Vol 9 (1) ◽  
Author(s):  
John Ferguson ◽  
Fabrizio Maturo ◽  
Salim Yusuf ◽  
Martin O’Donnell
Keyword(s):  
Risk Factors ◽  
Risk Factor ◽  
Continuous Exposure ◽  
Minimum Risk ◽  
Population Attributable Fractions ◽  
Graphical Comparison ◽  
Attributable Fractions ◽  
Potential Outcomes Framework ◽  
Prior Risk ◽  
Tobacco Usage

AbstractWhen estimating population attributable fractions (PAF), it is common to partition a naturally continuous exposure into a categorical risk factor. While prior risk factor categorization can help estimation and interpretation, it can result in underestimation of the disease burden attributable to the exposure as well as biased comparisons across different exposures and risk factors. Here, we propose sensible PAF estimands for continuous exposures under a potential outcomes framework. In contrast to previous approaches, we incorporate estimation of the minimum risk exposure value (MREV) into our procedures. While for exposures such as tobacco usage, a sensible value of the MREV is known, often it is unknown and needs to be estimated. Second, in the setting that the MREV value is an extreme-value of the exposure lying in the distributional tail, we argue that the natural estimator of PAF may be both statistically biased and highly volatile; instead, we consider a family of modified PAFs which include the natural estimate of PAF as a limit. A graphical comparison of this set of modified PAF for differing risk factors may be a better way to rank risk factors as intervention targets, compared to the standard PAF calculation. Finally, we analyse the bias that may ensue from prior risk factor categorization, examining whether categorization is ever a good idea, and suggest interpretations of categorized-estimands within a causal inference setting.

scholarly journals Population attributable fraction for continuously distributed exposures

2020 ◽  
Author(s):  
John Ferguson ◽  
Fabrizio Maturo ◽  
Salim Yusuf ◽  
Martin O’Donnell
Keyword(s):  
Risk Factors ◽  
Risk Factor ◽  
Population Attributable Fraction ◽  
Continuous Exposure ◽  
Minimum Risk ◽  
Population Attributable Fractions ◽  
Graphical Comparison ◽  
Attributable Fractions ◽  
Potential Outcomes Framework ◽  
Prior Risk

AbstractWhen estimating population attributable fractions (PAF), it is common to partition a naturally continuous exposure into a categorical risk factor. While prior risk factor categorization can help estimation and interpretation, it can result in underestimation of the disease burden attributable to the exposure as well as biased comparisons across different exposures and risk factors. Here, we propose sensible PAF estimands for continuous exposures under a potential outcomes framework. In contrast to previous approaches, we incorporate estimation of the minimum risk exposure value (MREV) into our procedures. While for exposures such as tobacco usage, a sensible value of the MREV is known, often it is unknown and needs to be estimated. Second, in the setting that the MREV value is an extreme-value of the exposure lying in the distributional tail, we argue that the natural estimator of PAF may be both statistically biased and highly volatile; instead, we consider a family of modified PAFs which include the natural estimate of PAF as a limit. A graphical comparison of this set of modified PAF for differing risk factors may be a better way to rank risk factors as intervention targets, compared to the standard PAF calculation. Finally, we analyse the bias that may ensue from prior risk factor categorization, examining whether categorization is ever a good idea, and suggest interpretations of categorized-estimands within a causal inference setting.

scholarly journals Graphical comparisons of relative disease burden across multiple risk factors

2019 ◽  
Vol 19 (1) ◽  
Author(s):  
John Ferguson ◽  
Neil O’Leary ◽  
Fabrizio Maturo ◽  
Salim Yusuf ◽  
Martin O’Donnell
Keyword(s):  
Risk Factors ◽  
Risk Factor ◽  
Relative Risk ◽  
Disease Burden ◽  
Health Interventions ◽  
Continuous Exposure ◽  
Odds Ratios ◽  
Risk Factor Prevalence ◽  
Attributable Fractions ◽  
Log Odds

Abstract Background Population attributable fractions (PAF) measure the proportion of disease prevalence that would be avoided in a hypothetical population, similar to the population of interest, but where a particular risk factor is eliminated. They are extensively used in epidemiology to quantify and compare disease burden due to various risk factors, and directly influence public policy regarding possible health interventions. In contrast to individual specific metrics such as relative risks and odds ratios, attributable fractions depend jointly on both risk factor prevalence and relative risk. The relative contributions of these two components is important, and usually needs to be presented in summary tables that are presented together with the attributable fraction calculation. However, representing PAF in an accessible graphical format, that captures both prevalence and relative risk, may assist interpretation. Methods Taylor-series approximations to PAF in terms of risk factor prevalence and log-odds ratio are derived that facilitate simultaneous representation of PAF, risk factor prevalence and risk-factor/disease log-odds ratios on a single co-ordinate axis. Methods are developed for binary, multi-category and continuous exposure variables. Results The methods are demonstrated using INTERSTROKE, a large international case control dataset focused on risk factors for stroke. Conclusions The described methods could be used as a complement to tables summarizing prevalence, odds ratios and PAF, and may convey the same information in a more intuitive and visually appealing manner. The suggested nomogram can also be used to visually estimate the effects of health interventions which only partially reduce risk factor prevalence. Finally, in the binary risk factor case, the approximations can also be used to quickly convert logistic regression coefficients for a risk factor into approximate PAFs.

Preventable Risk Factors for Dementia: Population Attributable Fractions in a Brazilian Population-Based Study

2021 ◽  
Author(s):  
Wyllians Vendramini Borelli ◽  
Vanessa Bielefeldt Leotti ◽  
Matheus Zschornack Strelow ◽  
Márcia Lorena Fagundes Chaves ◽  
Raphael Machado Castilhos
Keyword(s):  
Risk Factors ◽  
Population Based ◽  
Brazilian Population ◽  
Population Based Study ◽  
Population Attributable Fractions ◽  
Attributable Fractions

scholarly journals Regional variations in the economic burden attributable to excess weight, physical inactivity and tobacco smoking across British Columbia

2016 ◽  
Vol 36 (4) ◽  
pp. 76-86 ◽  
Author(s):  
H. Krueger ◽  
J.M. Koot ◽  
D.P. Rasali ◽  
S. E. Gustin ◽  
M. Pennock
Keyword(s):  
Risk Factors ◽  
British Columbia ◽  
Tobacco Smoking ◽  
Economic Burden ◽  
Physical Inactivity ◽  
Excess Weight ◽  
Prevalence Rates ◽  
Population Attributable Fractions ◽  
Attributable Fractions ◽  
Social Environmental

Introduction Prevalence rates of excess weight, tobacco smoking and physical inactivity vary substantially by geographical region within British Columbia (B.C.). The purpose of this study is to determine the potential reduction in economic burden in B.C. if all regions in the province achieved prevalence rates of these three risk factors equivalent to those of the region with the lowest rates. Methods We used a previously developed approach based on population-attributable fractions to estimate the economic burden associated with the various risk factors. Sexspecific relative risk and age/sex-specific prevalence data was used in the modelling. Results The annual economic burden attributable to the three risk factors in B.C. was about $5.6 billion in 2013, with a higher proportion of this total attributable to excess weight ($2.6 billion) than to tobacco smoking ($2.0 billion). While B.C. has lower prevalence rates of the risk factors than any other Canadian province, there is significant variation within the province. If each region in the province were to achieve the best prevalence rates for the three risk factors, then $1.4 billion (24% of the $5.6 billion) in economic burden could be avoided annually. Conclusion There are notable disparities in the prevalence of each risk factor across health regions within B.C., which were mirrored in each region’s attributable economic burden. A variety of social, environmental and economic factors likely drive some of this geographical variation and these underlying factors should be considered when developing prevention programs.

scholarly journals What to deploy to prevent hip OA; population attributable fractions of risk factors for incident clinical and radiographic hip OA

2019 ◽  
Vol 27 ◽  
pp. S271-S272
Author(s):  
J. Runhaar ◽  
M. Hall ◽  
D. Schiphof ◽  
F. Saberi Hosnijeh ◽  
J. van Meurs ◽  
...  
Keyword(s):  
Risk Factors ◽  
Population Attributable Fractions ◽  
Attributable Fractions

Risk factors for sporadicVibrio parahaemolyticusgastroenteritis in east China: a matched case-control study

2014 ◽  
Vol 143 (5) ◽  
pp. 1020-1028 ◽  
Author(s):  
W. X. YAN ◽  
Y. DAI ◽  
Y. J. ZHOU ◽  
H. LIU ◽  
S. G. DUAN ◽  
...  
Keyword(s):  
Risk Factors ◽  
Case Control Study ◽  
Multivariable Analysis ◽  
Antibiotic Use ◽  
Population Based ◽  
Case Control ◽  
Population Attributable Fractions ◽  
Attributable Fractions ◽  
Matched Case ◽  
Control Study

SUMMARYTo determine risk factors for sporadicVibrio parahaemolyticusgastroenteritis, we conducted a population-based case-control study in sentinel hospital surveillance areas of Shanghai and Jiangsu province, China. Seventy-one patients with diarrhoea and confirmedV. parahaemolyticusinfections were enrolled, and they were matched with 142 controls for gender, age and residential area. From the multivariable analysis,V. parahaemolyticusinfections were associated with antibiotics taken during the 4 weeks prior to illness [odds ratio (OR) 8·1, 95% confidence interval (CI) 1·2–56·4)], frequent eating out (OR 3·3, 95% CI 1·1–10·1), and shellfish consumption (OR 3·2, 95% CI 1·0–9·9), with population-attributable fractions of 0·09, 0·25, and 0·14, respectively. Protective factors included keeping the aquatic products refrigerated (OR 0·4, 95% CI 0·1–0·9) and pork consumption (OR 0·2, 95% CI 0·1–0·8). Further study of the association ofV. parahaemolyticusgastroenteritis with prior antibiotic use and shellfish consumption is needed.

scholarly journals Lifestyle Risk Factors and the Population Attributable Fractions for Overweight and Obesity in Chinese Students of Zhejiang Province

2021 ◽  
Vol 9 ◽  
Author(s):  
Fang Gu ◽  
Siliang Zhou ◽  
Ke Lou ◽  
Rui Deng ◽  
Xingxiu Li ◽  
...  
Keyword(s):  
Risk Factors ◽  
Chinese Students ◽  
Lifestyle Factors ◽  
Overweight And Obesity ◽  
Zhejiang Province ◽  
Lifestyle Risk Factors ◽  
Population Attributable Fractions ◽  
Attributable Fractions ◽  
Lifestyle Risk ◽  
The Relationship

Objectives: To assess the relationship between modifiable lifestyle factors and risk of overweight/obesity in Chinese students, and to evaluate the predicting prevalence of overweight if the lifestyle risk factors were removed.Methods: A cross-sectional survey was conducted among 40,141 students in grade three and above (8–24yrs) in 2019 in Zhejiang Province, China. Physical examination was performed, and a self-administered questionnaire was used to collect lifestyle information, including dietary behavior, physical activity, TV watching, sleeping, smoking, drinking, and tooth-brushing habits. Logistic regression models were performed to assess the relationship between overweight/obesity and a series of lifestyle factors. Population attributable fractions (PAFs) were used to calculate the predicting prevalence of overweight/obesity if lifestyle risk factors were removed.Results: The prevalence of overweight/obesity of participants was 25.5% (male 32.3%, female 18.1%). Overweight/obesity were associated with adverse lifestyle factors, such as watch TV ≥1 h/day (OR = 1.14, 95% CI: 1.11–1.22), insufficient sleep (OR = 1.14, 95% CI: 1.11–1.22), and irregular toothbrushing habits (OR = 1.19, 95% CI: 1.01–1.39). Based on the calculated PAFs, the predicted prevalence of overweight/obesity would decline moderately if lifestyle factors were modified, with the magnitudes of decrease vary by sex, age and residence. Generally, a larger reduction was estimated if the sleeping time was increased and TV time was reduced, with the prevalence of overweight/obesity decreased by 1.1% (95% CI: 0.7, 1.5%) and 0.9% (95% CI: 0.6, 1.2%), respectively.Conclusions: Predicted prevalence of overweight/ obesity in Chinese students may decrease if modifiable lifestyle risk factors were removed. The attributable risk for obesity of lifestyle behaviors varied in age, sex and residence groups. The findings of this study may provide insights for planning and optimizing future obesity intervention endeavors.

scholarly journals 1396. Risk Factors for Brain Abscess: A Nationwide Population-based Nested Case–Control Study

2019 ◽  
Vol 6 (Supplement_2) ◽  
pp. S507-S508
Author(s):  
Jacob Bodilsen ◽  
Michael Dalager-Pedersen ◽  
Diederik van de Beek ◽  
Matthijs C Brouwer ◽  
Henrik Nielsen
Keyword(s):  
Risk Factors ◽  
Brain Abscess ◽  
Case Control Study ◽  
Population Based ◽  
Case Control ◽  
Population Attributable Fractions ◽  
Attributable Fractions ◽  
Nested Case Control ◽  
Population Controls ◽  
Control Study

Abstract Background Knowledge of risk factors for brain abscess is limited and relies on single-center cohorts without control groups. Methods We accessed nationwide medical registries to conduct a population-based nested case–control study of risk factors for brain abscess. We applied risk set sampling for selection of population controls (1:10) individually matched by age, sex, and area of residence. Conditional logistic regression was used to compute adjusted odds ratios (aOR) with 95% confidence intervals (CIs). Next, population attributable fractions were calculated. Results We identified 1,384 brain abscess patients in Denmark from 1982 through 2016 and 13,839 matched population controls. The median age was 50 years (interquartile range 33–63) and 37% were female. Cases often had a Charlson comorbidity score>2 (16%) compared with controls (3%). Adjusted ORs were: head trauma 2.15 (1.72–2.70), neurosurgery 19.3 (14.3–26.0), dental infection 4.61 (3.39–6.26) or surgery 2.57 (1.71–3.84), ear-nose-throat infection 3.81 (3.11–4.67) or surgery 2.85 (2.21–3.70), congenital heart disease 15.6 (9.57–25.4), diabetes mellitus 1.74 (1.33–2.29), alcohol abuse 2.22 (1.58–3.11), liver disease 2.37 (1.53–3.68), kidney disease 2.04 (1.30–3.20), and lung abscess or bronchiectasis 8.15 (3.59–18.5). The aORs were 4.12 (3.37–5.04) and 8.77 (5.66–13.6) for solid and hematological cancer, 12.0 (6.13–23.7) for HIV, and 5.71 (4.22–7.75) for immuno-modulating treatments. Risks were twice as high when risk factors were observed within 5 years before brain abscess. Population attributable fractions showed that neurosurgery (12%), solid cancer (11%), ear-nose-throat infections (7%) and immuno-modulating treatments (5%) were substantial contributors to occurrence of brain abscess. Conclusion Important risk factors included neurosurgery, cancer, ear-nose-throat infections and immuno-modulating treatments Disclosures All authors: No reported disclosures.

Sign in / Sign up

Export Citation Format

Share Document