We review three common methods to estimate predicted probabilities following confounder-adjusted logistic regression: marginal standardization (predicted probabilities summed to a weighted average ...reflecting the confounder distribution in the target population); prediction at the modes (conditional predicted probabilities calculated by setting each confounder to its modal value); and prediction at the means (predicted probabilities calculated by setting each confounder to its mean value). That each method corresponds to a different target population is underappreciated in practice. Specifically, prediction at the means is often incorrectly interpreted as estimating average probabilities for the overall study population, and furthermore yields nonsensical estimates in the presence of dichotomous confounders. Default commands in popular statistical software packages often lead to inadvertent misapplication of prediction at the means.
Using an applied example, we demonstrate discrepancies in predicted probabilities across these methods, discuss implications for interpretation and provide syntax for SAS and Stata.
Marginal standardization allows inference to the total population from which data are drawn. Prediction at the modes or means allows inference only to the relevant stratum of observations. With dichotomous confounders, prediction at the means corresponds to a stratum that does not include any real-life observations.
Marginal standardization is the appropriate method when making inference to the overall population. Other methods should be used with caution, and prediction at the means should not be used with binary confounders. Stata, but not SAS, incorporates simple methods for marginal standardization.
Objective We used validated sensitive and specific questions associated with clinically confirmed diagnoses of unexplained vulvar pain (vulvodynia) to compare the cumulative incidence of vulvar pain ...and prevalence of care-seeking behavior in Boston metropolitan area (BMA) and in Minneapolis/Saint Paul metropolitan area (MSP) from 2001 through 2005 using census-based data, and 2010 through 2012, using outpatient community-clinic data, respectively. Study Design We received self-administered questionnaires from 5440 women in BMA and 13,681 in MSP, 18-40 years of age, describing their history of vulvar burning or pain on contact that persisted >3 months that limited/prevented intercourse. Results By age 40 years, 7-8% in BMA and MSP reported vulvar pain consistent with vulvodynia. Women of Hispanic origin compared to whites were 1.4 times more likely to develop vulvar pain symptoms (95% confidence interval, 1.1–1.8). Many women in MSP (48%) and BMA (30%) never sought treatment, and >50% who sought care with known health care access received no diagnosis. Conclusion Using identical screening methods, we report high prevalence of vulvar pain in 2 geographic regions, and that access to health care does not increase the likelihood of seeking care for chronic vulvar pain.
Good practices for quantitative bias analysis LASH, Timothy L; FOX, Matthew P; MACLEHOSE, Richard F ...
International journal of epidemiology,
12/2014, Letnik:
43, Številka:
6
Journal Article
Recenzirano
Odprti dostop
Quantitative bias analysis serves several objectives in epidemiological research. First, it provides a quantitative estimate of the direction, magnitude and uncertainty arising from systematic ...errors. Second, the acts of identifying sources of systematic error, writing down models to quantify them, assigning values to the bias parameters and interpreting the results combat the human tendency towards overconfidence in research results, syntheses and critiques and the inferences that rest upon them. Finally, by suggesting aspects that dominate uncertainty in a particular research result or topic area, bias analysis can guide efficient allocation of sparse research resources. The fundamental methods of bias analyses have been known for decades, and there have been calls for more widespread use for nearly as long. There was a time when some believed that bias analyses were rarely undertaken because the methods were not widely known and because automated computing tools were not readily available to implement the methods. These shortcomings have been largely resolved. We must, therefore, contemplate other barriers to implementation. One possibility is that practitioners avoid the analyses because they lack confidence in the practice of bias analysis. The purpose of this paper is therefore to describe what we view as good practices for applying quantitative bias analysis to epidemiological data, directed towards those familiar with the methods. We focus on answering questions often posed to those of us who advocate incorporation of bias analysis methods into teaching and research. These include the following. When is bias analysis practical and productive? How does one select the biases that ought to be addressed? How does one select a method to model biases? How does one assign values to the parameters of a bias model? How does one present and interpret a bias analysis?. We hope that our guide to good practices for conducting and presenting bias analyses will encourage more widespread use of bias analysis to estimate the potential magnitude and direction of biases, as well as the uncertainty in estimates potentially influenced by the biases.
Background: Despite a vast air pollution epidemiology literature to date and the recognition that lower-socioeconomic status (SES) populations are often disproportionately exposed to pollution, there ...is little research identifying optimal means of adjusting for confounding by SES in air pollution epidemiology, nor is there a strong understanding of biases that may result from improper adjustment. Objective: We aim to provide a conceptualization of SES and a review of approaches to its measurement in the U.S. context and discuss pathways by which SES may influence health and confound effects of air pollution. We explore bias related to measurement and operationalization and identify statistical approaches to reduce bias and confounding. Discussion: Drawing on the social epidemiology, health geography, and economic literatures, we describe how SES, a multifaceted construct operating through myriad pathways, may be conceptualized and operationalized in air pollution epidemiology studies. SES varies across individuals within the contexts of place, time, and culture. Although no single variable or index can fully capture SES, many studies rely on only a single measure. We recommend examining multiple facets of SES appropriate to the study design. Furthermore, investigators should carefully consider the multiple mechanisms by which SES might be operating to identify those SES indicators that may be most appropriate for a given context or study design and assess the impact of improper adjustment on air pollution effect estimates. Last, exploring model contraction and expansion methods may enrich adjustment, whereas statistical approaches, such as quantitative bias analysis, may be used to evaluate residual confounding.
Celotno besedilo
Dostopno za:
CEKLJ, DOBA, IZUM, KILJ, NUK, OILJ, PILJ, PNG, SAZU, SIK, UILJ, UKNU, UL, UM, UPUK, VSZLJ
IMPORTANCE: Estimates of lifetime risk may help raise awareness of the extent to which educational inequalities are associated with risk of cardiovascular disease (CVD). OBJECTIVE: To estimate ...lifetime risks of CVD according to categories of educational attainment. DESIGN, SETTING, AND PARTICIPANTS: Participants were followed from 1987 through December 31, 2013. All CVD events (coronary heart disease, heart failure, and stroke) were confirmed by physician review and International Classification of Diseases codes. A total of 13 948 whites and African Americans who were 45 to 64 years old and free of CVD at baseline were included from 4 US communities (Washington County, Maryland; Forsyth County, North Carolina; Jackson, Mississippi; and suburbs of Minneapolis, Minnesota). The data analysis was performed from June 7 to August 31, 2016. EXPOSURES: Educational attainment. MAIN OUTCOMES AND MEASURES: We used a life table approach to estimate lifetime risks of CVD from age 45 through 85 years according to educational attainment. We adjusted for competing risks of death from underlying causes other than CVD. RESULTS: The sample of 13 948 participants was 56% female and 27% African American. During 269 210 person-years of follow-up, we documented 4512 CVD events and 2401 non-CVD deaths. Educational attainment displayed an inverse dose-response relation with cumulative risk of CVD, which became evident in middle age, with the most striking gap between those not completing vs completing high school. In men, lifetime risks of CVD were 59.0% (95% CI, 54.0%-64.1%) for grade school, 52.5% (95% CI, 47.7%-56.8%) for high school education without graduation, 50.9% (95% CI, 47.3%-53.9%) for high school graduation, 47.2% (95% CI, 41.5%-52.5%) for vocational school, 46.4% (95% CI, 42.8%-49.6%) for college with or without graduation, and 42.2% (95% CI, 36.6%-47.0%) for graduate/professional school; in women, 50.8% (95% CI, 45.7%-55.8%), 49.3% (95% CI, 45.1%-53.1%), 36.3% (95% CI, 33.4%-39.1%), 32.2% (95% CI, 26.0%-37.3%), 32.8% (95% CI, 29.1%-35.9%), and 28.0% (95% CI, 21.9%-33.3%), respectively. Educational attainment was inversely associated with CVD even within categories of family income, income change, occupation, or parental educational level. CONCLUSIONS AND RELEVANCE: More than 1 in 2 individuals with less than high school education had a lifetime CVD event. Educational attainment was inversely associated with the lifetime risk of CVD, regardless of other important socioeconomic characteristics. Our findings emphasize the need for further efforts to reduce CVD inequalities related to educational disparities.
Family meals have been found to be associated with a number of health benefits for children; however, associations with obesity have been less consistent, which raises questions about the specific ...characteristics of family meals that may be protective against childhood obesity. The current study examined associations between interpersonal and food-related family dynamics at family meals and childhood obesity status.
The current mixed-methods, cross-sectional study included 120 children (47% girls; mean age: 9 years) and parents (92% women; mean age: 35 years) from low-income and minority communities. Families participated in an 8-day direct observational study in which family meals were video-recorded in their homes. Family meal characteristics (eg, length of the meal, types of foods served) were described and associations between dyadic (eg, parent-child, child-sibling) and family-level interpersonal and food-related dynamics (eg, communication, affect management, parental food control) during family meals and child weight status were examined.
Significant associations were found between positive family- and parent-level interpersonal dynamics (ie, warmth, group enjoyment, parental positive reinforcement) at family meals and reduced risk of childhood overweight. In addition, significant associations were found between positive family- and parent-level food-related dynamics (ie, food warmth, food communication, parental food positive reinforcement) and reduced risk of childhood obesity.
Results extend previous findings on family meals by providing a better understanding of interpersonal and food-related family dynamics at family meals by childhood weight status. Findings suggest the importance of working with families to improve the dyadic and family-level interpersonal and food-related dynamics at family meals.
Abstract Purpose The purpose of the study was to describe parent and adolescent involvement in food preparation for the family and to examine whether adolescents' food preparation involvement was ...related to their dietary quality (e.g., fruit and vegetable intake, sugar-sweetened beverage consumption, and various common nutrients) and eating patterns (e.g., frequency of breakfast, family meals, fast food intake). Methods Data from two linked population-based studies, Eating and Activity in Teens 2010 and Families and Eating and Activity among Teens were used in cross-sectional analyses. Mothers (n = 1,875), stepmothers (n = 18), fathers (n = 977), stepfathers (n = 105), and adolescents (n = 2,108) from socioeconomically and racially/ethnically diverse households participated in the study. Adolescents completed food frequency questionnaires and surveys in school. Parents individually completed surveys by mail or phone. Linear regression was used to estimate differences in adolescent dietary quality and eating patterns between those who do and do not engage in meal preparation. Results Parent and adolescent report of “usually preparing food for the family” was related to several sociodemographic characteristics, including race/ethnicity (minority populations), parent education (college or higher), parent employment status (part time or stay-at-home caregiver), household size (≤3 children), and adolescent gender (female). Adolescent involvement in food preparation for the family was significantly associated with several markers of better dietary quality and better eating patterns. In contrast, parent involvement in food preparation for the family was unrelated to adolescent dietary intake. Conclusions Results suggest that involving adolescents in food preparation for the family is related to better adolescent dietary quality and eating patterns. Public health interventions and health care providers may want to encourage adolescents to help with food preparation for the family. Additionally, adolescents may benefit from interventions/programs that teach cooking skills in order to increase the likelihood of participating in food preparation for the family.
Objectives
To estimate the cost of dementia and the extra cost of caring for someone with dementia over the cost of caring for someone without dementia.
Design
We developed an evidence‐based ...mathematical model to simulate disease progression for newly diagnosed individuals with dementia. Data‐driven trajectories of cognition, function, and behavioral and psychological symptoms were used to model disease progression and predict costs. Using modeling, we evaluated lifetime and annual costs of individuals with dementia, compared costs of those with and without clinical features of dementia, and evaluated the effect of reducing functional decline or behavioral and psychological symptoms by 10% for 12 months (implemented when Mini‐Mental State Examination score ≤21).
Setting
Mathematical model.
Participants
Representative simulated U.S. incident dementia cases.
Measurements
Value of informal care, out‐of‐pocket expenditures, Medicaid expenditures, and Medicare expenditures.
Results
From time of diagnosis (mean age 83), discounted total lifetime cost of care for a person with dementia was $321,780 (2015 dollars). Families incurred 70% of the total cost burden ($225,140), Medicaid accounted for 14% ($44,090), and Medicare accounted for 16% ($52,540). Costs for a person with dementia over a lifetime were $184,500 greater (86% incurred by families) than for someone without dementia. Total annual cost peaked at $89,000, and net cost peaked at $72,400. Reducing functional decline or behavioral and psychological symptoms by 10% resulted in $3,880 and $680 lower lifetime costs than natural disease progression.
Conclusion
Dementia substantially increases lifetime costs of care. Long‐lasting, effective interventions are needed to support families because they incur the most dementia cost.
IMPORTANCE: Testosterone therapy is increasingly prescribed in patients without a diagnosis of hypogonadism. This therapy may be associated with increased risk of venous thromboembolism (VTE) through ...several mechanisms, including elevated hematocrit levels, which increase blood viscosity. OBJECTIVE: To assess whether short-term testosterone therapy exposure is associated with increased short-term risk of VTE in men with and without evidence of hypogonadism. DESIGN, SETTING, AND PARTICIPANTS: This case-crossover study analyzed data on 39 622 men from the IBM MarketScan Commercial Claims and Encounter Database and the Medicare Supplemental Database from January 1, 2011, to December 31, 2017, with 12 months of follow-up. Men with VTE cases who were free of cancer at baseline and had 12 months of continuous enrollment before the VTE event were identified by International Classification of Diseases codes. Men in the case period were matched with themselves in the control period. Case periods of 6 months, 3 months, and 1 month before the VTE events were defined, with equivalent control periods (6 months, 3 months, and 1 month) in the 6 months before the case period. EXPOSURES: National drug codes were used to identify billed testosterone therapy prescriptions in the case period (0-6 months before the VTE) and the control period (6-12 months before the VTE). MAIN OUTCOMES AND MEASURES: The main outcome in this case-only experiment was first VTE event stratified by the presence or absence of hypogonadism. RESULTS: A total of 39 622 men (mean SD age, 57.4 14.2 years) were enrolled in the study, and 3110 men (7.8%) had evidence of hypogonadism. In age-adjusted models, testosterone therapy use in all case periods was associated with a higher risk of VTE in men with (odds ratio OR, 2.32; 95% CI, 1.97-2.74) and without (OR, 2.02; 95% CI, 1.47-2.77) hypogonadism. Among men without hypogonadism, the point estimate for testosterone therapy and VTE risk in the 3-month case period was higher for men younger than 65 years (OR, 2.99; 95% CI, 1.91-4.68) than for older men (OR, 1.68; 95% CI, 0.90-3.14), although this interaction was not statistically significant (P = .14). CONCLUSIONS AND RELEVANCE: Testosterone therapy was associated with an increase in short-term risk for VTE among men with and without hypogonadism, with some evidence that the association was more pronounced among younger men. These findings suggest that caution should be used when prescribing testosterone therapy.
Recent methodological innovation is giving rise to an increasing number of applied papers in medical and epidemiological journals in which natural direct and indirect effects are estimated. However, ...there is a longstanding debate on whether such effects are relevant targets of inference in population health. In light of the repeated calls for a more pragmatic and consequential epidemiology, we review three issues often raised in this debate: (i) the use of composite cross-world counterfactuals and the need for cross-world independence assumptions; (ii) interventional vs non-interventional identifiability; and (iii) the interpretational ambiguity of natural direct and indirect effect estimates. We use potential outcomes notation and directed acyclic graphs to explain 'cross-world' assumptions, illustrate implications of this assumption via regression models and discuss ensuing issues of interpretation. We argue that the debate on the relevance of natural direct and indirect effects rests on whether one takes as a target of inference the mathematical object per se, or the change in the world that the mathematical object represents. We further note that public health questions may be better served by estimating controlled direct effects.
PDF
