Assessment of the effect on statistical power of regression model misspecification by using techniques of mathematical statistics and simulation study

Objectives. This paper seeks to assess the effect on statistical power of regression model misspecification in a variety of situations. ^ Methods and results. The effect of misspecification in regression can be approximated by evaluating the correlation between the correct specification and the misspecification of the outcome variable (Harris 2010).In this paper, three misspecified models (linear, categorical and fractional polynomial) were considered. In the first section, the mathematical method of calculating the correlation between correct and misspecified models with simple mathematical forms was derived and demonstrated. In the second section, data from the National Health and Nutrition Examination Survey (NHANES 2007-2008) were used to examine such correlations. Our study shows that comparing to linear or categorical models, the fractional polynomial models, with the higher correlations, provided a better approximation of the true relationship, which was illustrated by LOESS regression. In the third section, we present the results of simulation studies that demonstrate overall misspecification in regression can produce marked decreases in power with small sample sizes. However, the categorical model had greatest power, ranging from 0.877 to 0.936 depending on sample size and outcome variable used. The power of fractional polynomial model was close to that of linear model, which ranged from 0.69 to 0.83, and appeared to be affected by the increased degrees of freedom of this model.^ Conclusion. Correlations between alternative model specifications can be used to provide a good approximation of the effect on statistical power of misspecification when the sample size is large. When model specifications have known simple mathematical forms, such correlations can be calculated mathematically. Actual public health data from NHANES 2007-2008 were used as examples to demonstrate the situations with unknown or complex correct model specification. Simulation of power for misspecified models confirmed the results based on correlation methods but also illustrated the effect of model degrees of freedom on power.^

Social and behavioral determinants of consistent condom use among female commercial sex workers (FCSWs) in Ghana

This study aims at investigating the social and behavioral predictors of consistent condom use among female commercial sex workers (FCSWs) in Ghana. Street commercial sex workers were interviewed in Accra, Kumasi and Techiman. Whereas respondents had attained certain accurate knowledge about HIV transmission routes, misconceptions were still commonly reported. The level of condom education was very low (14%), however consistent condom use (all the time) with clients was relatively high (49.6%), 38.89% reported using condom sometimes and 11.56% reported never using condoms. ^ 277 of the respondent ants did not use condoms all the time. 163 of them reported not using condoms due to refusal by their clients, the remaining 64 respondents did not even request their clients to use condom due to cultural perception of power, lack of authority and the fear of loosing clients. ^ Significant predictive factors associated with consistency of condom use among FCSWs in a multivariate analysis were; age, level of education, religion, and number of customers. Some of the major obstacles to condom use by the FCSWs were refusal by clients, availability of free condoms, trying to communicate trust to their clients, and the lack of empowerment to negotiate safer sex with clients. Some of the respondents may have developed a false sense of safety by subjectively assessing whether their clients were well and do not look sick, but they were unaware that HIV carriers may show no obvious symptoms of illness at all. ^ In summary, this study points to an urgent need for reestablishing effective prevention intervention and some insights of what is required of such program in Ghana. ^

The effect of specific carotenoids on lung cancer risk

A population-based case-comparison study of histologically confirmed lung cancer among white male and female residents of six Texas coastal counties was conducted from 1979-1982. Dietary information as well as information concerning smoking, alcohol consumption, occupational and residential exposures, and family history of cancer was obtained from 149 living cases and 359 comparison subjects.^ These data support the findings of previous studies that reported a protective association of total carotene intake with lung cancer (OR = 4.07, CI = 3.36-4.78), and no association for total Vitamin A or retinol. Of six specific carotenoids examined, these data reveal a statistically significant protective effect for alpha carotene (OR = 3.58, CI = 2.85-4.31), and an elevated, although non-significant effect for beta carotene (OR = 1.46, CI = 0.61-2.31). No consistent significant effect was found for cryptoxanthin, other xanthins, lutein or lycopene.^ Similar results were found for both males and females, and for both squamous cell and adenocarcinoma subtypes, although loss of power resulting from stratification may have rendered the celltype specific results more imprecise. These results should be considered with caution until confirmed by other studies, however, they suggest the importance of evaluating specific carotenoids in future diet-lung cancer investigations. ^

The illness experiences of African Americans residing in an inner city community, Houston, Texas

This dissertation documents health and illness in the context of daily life circumstances and structural conditions faced by African American families living in Clover Heights (pseudonym), an inner city public housing project in the Third Ward, Houston, Texas. Drawing from Kleinman's (1980) model of culturally defined health care systems and using the holistic-content approach to narrative analysis (Lieblich, Tuval- Mashiach, & Zilber, 1998) the purpose of this research was to explore the ways in which social and health policy, economic mobility, the inner city environment, and cultural beliefs intertwined with African American families' health related ideas, behaviors, and practices. I recruited six families using a convenience sampling method (Schensul, Schensul, & LeCompte, 1999) and followed them for fourteen months (2010–2011). Family was defined as a household unit, or those living in the same residence, short or long-term. Single, African American women ranging in age from 29–80 years headed all families. All but one family included children or grandchildren 18 years of age and younger, or children or other relative 18 years of age and older. I also recruited six residents with who I became acquainted over the course of the project. I collected data using traditional ethnographic methods including participant-observation, archive review, field notes, mapping, free-listing, in-depth interviews, and life history interviews. ^ Doing ethnography afforded the families who participated in this project the freedom to construct their own experiences of health and illness. My role centered on listening to, learning from, and interpreting participants' narratives, exploring similarities and differences within and across families' experiences. As the research progressed, a pattern concerning diagnosis and pharmacotherapy for children's behavioral and emotional problems, particularly attention-deficit hyperactivity disorder (ADHD) and pediatric bipolar disorder (PBD), emerged from my formal interactions with participants and my informal interactions with residents. The findings presented in this dissertation document this pattern, focusing on how mothers and families interpreted, organized, and ascribed meaning to their experiences of ADHD and PBD. ^ In the first manuscript presented here, I documented three mothers' narrative constructions of a child's diagnosis with and pharmacotherapy for ADHD or PBD. Using Gergen's (1997) relational perspective I argued that mothers' knowledge and experiences of ADHD and PBD were not individually constructed, but were linguistically and discursively constituted through various social interactions and relationships, including family, spirituality and faith, community norms, and expert systems of knowledge. Mothers' narratives revealed the complexity of children's behavioral and emotional problems, the daily trials of living through these problems, how they coped with adversity and developed survival strategies, and how they interacted with various institutional authorities involved in evaluating, diagnosing, and encouraging pharmaceutical intervention for children's behavior. The findings highlight the ways in which mothers' social interactions and relationships introduced a scientific language and discourse for explaining children's behavior as mental illness, the discordances between expert systems of knowledge and mothers' understandings, and how discordances reflected mothers' ‘microsources of power’ for producing their own stories and experiences. ^ In the second manuscript presented here, I documented the ways in which structural factors, including gender, race/ethnicity, and socioeconomic status, coupled with a unique cultural and social standpoint (Collins, 1990/2009) influenced the strategies this group of African American mothers employed to understand and respond to ADHD or PBD. The most salient themes related to mother-child relationships coalesced around mothers' beliefs about the etiology of ADHD and PBD, ‘conceptualizing responsibility,’ and ‘protection-survival.’ The findings suggest that even though mothers' strategies varied, they were in pursuit of a common goal. Mothers' challenged the status quo, addressing children's behavioral and emotional problems in the ways that made the most sense to them, specifically protecting their children from further marginalization in society more so than believing these were the best options for their children.^