Graphical methods for comparing confounding in two or more designs

In many circumstances it is necessary to design an experiment with partial confounding of parameter estimates. In this situation, one would like to have tools to assess and compare candidate designs graphically. In this paper we discuss two graphical representations of design properties, and illustrate their application with two examples.     

Management of Allergic Respiratory Disease by the Nurse Practitioner

The results of several studies have established the validity of the SCOFF questionnaire (a 5-question screening tool for eating disorders), but researchers need to explore further replicability using the US version in the graduate school population. In this study, the authors asked 335 graduate students attending the Northwestern student health clinic on the Chicago campus to complete a written survey anonymously. A total of 305 (91%) patients completed the survey. The sensitivity and specificity for the SCOFF was 53.3% and 93.2%, respectively. This produced a PPV (the proportion who tested positive on the screen and actually had an eating disorder) of 66.7% and an NPV (the proportion of those who tested negative on the screen and actually did not have an eating disorder) of 88.7%. More than 80% of respondents were dissatisfied with their shape and weight, with over one third having a moderate to severe body image disturbance. The SCOFF is an easy instrument to administer that health care providers can use to screen for eating disorders in the primary care setting.     

Appropriate use of bright light promotes a durable adaptation to night-shifts and accelerates readjustment during recovery after a period of night-shifts

Abstract

The present study examines the stability of bright light circadian readjustment during two consecutive dim light night-work periods and circadian synchronization during the recovery after a night-shift period. A sample of 10 subjects was divided into 2 groups (control group: 6 subjects; experimental group: 4 subjects). All subjects worked during 5 days, between 23:00 and 07:00 h and then went to sleep. Subjects received 2500-3000 lux between 02:00 and 05:00 h during 5 days for the control group and 3 days for the experimental group. During recovery after the night-shift periods, three cycles of bright light were administered at two different times: 12:00-15:00 h for two of six subjects from the control group and 10:00-13:00 h for all subjects (4) of the experimental group. By the fifth cycle of night-work the maximum of urinary aMT6s excretion that occurs at 05:00 h in the baseline condition was shifted to 12:00 h for the control and experimental groups (delay in hours: 7±1.6 (control); 7±1 (experimental)). This result suggests that three cycles of bright light are sufficient to induce a significant phase delay and that this delay remained stable when night-work proceeded under dim light. The phase delay of the circadian aMT6s excretion by exposure to bright light was accompanied by an improvement of the quality of day sleep and level of cognitive and psychomotor performances for control and experimental groups. No significant difference was observed in the two groups for daytime sleep and nocturnal performance. The two bright light periods used during the three days of recovery induced a complete synchronization in five of six subjects. One subject showed a partial synchronization probably because he remained at the laboratory under dim light during the day and had few family and social contacts.     

Some small-sample results on a bounded influence rank regression method

Naranjo and HeUmansperger (1994) recently derved a bounded influence rank regression method and suggested how hypotheses about the regression coefficients might be tested. This brief note reports some simulation results on how their procedure performs when there is one predictor. Even when the error term is highly skewed, good control over the Type I error probability is obtained Power can be high relative to least squares regression when the error term has a heavy tailed distribution .and the predictor has a symmetric distribution However, if the predictor has a skewed distribution, power can be relatively low even when the distribution of the error term is heavy tailed. Despite this, it is argued that their method provides an important and useful alternative to ordinary least squares as well as other robust regression methods.     

A Qualitative Analysis of College Women's Leaving Processes in Abusive Relationships

Abstract

Objective: This study assessed the process of leaving an abusive dating relationship utilizing a qualitative design. Methods: Participants included 123 college women in abusive dating relationships who participated at the beginning and end of a 10-week academic quarter. Results: Qualitative content analyses were used to analyze the transcribed responses to an open-ended question about women's leaving processes over the interim period. A variety of categories and themes emerged for women in different stages of the leaving process, consistent with the Transtheoretical Model of Change and Investment Model. Data also underscored women's lack of acknowledgment, minimization, and normalization of abuse. Conclusions: These data demonstrate the importance of dating violence intervention and prevention programming on college campuses and offer information that may be useful to college health providers who assist women in abusive dating relationships.     

Mental Health and Clinical Correlates in Lesbian,Gay, Bisexual,and Queer Young Adults

Abstract

Objective: This study examined the prevalence of mental health disorders and their clinical correlates in a university sample of lesbian, gay, bisexual, and queer (LGBQ) students. Participants: College students at a large public university. Methods: An anonymous, voluntary survey was distributed via random e-mail generation to university students during April and May of 2011. LGBQ students were compared with their heterosexual counterparts on psychological and physical status as well as academic performance. Results: LGBQ students reported worse depressive symptoms, higher levels of perceived stress, considered themselves less attractive, and were more likely to be overweight. LGBQ students were significantly more likely to report histories of affective, substance use, and certain anxiety disorders as well as compulsive sexual behavior and compulsive buying. Conclusions: The higher rates of many psychiatric conditions among LGBQ students underscore the need for universities to provide LGBQ students a nonjudgmental environment to discuss sexual orientation and health issues.     

The relation between early abuse and adult sexuality

One thousand and thirty‐two (566 non‐Asian; 466 Asian) undergraduates were assessed in a confidential laboratory setting regarding levels of physical abuse, emotional abuse, sexual abuse, neglect, and a wide range of sexuality variables. The purpose was to examine relationships between specific forms of early abuse and later adult sexuality, and to assess potential gender and ethnic differences in these associations. Independent of other forms of abuse, sexual abuse in females was significantly positively related to sexual drive and experience, range of sexual fantasies, liberal sexual attitudes, frequency of intercourse and masturbation, and likelihood of engaging in unrestricted sexual behaviors and fantasies, and was negatively related to virginity status. Among males, emotional abuse was significantly associated with poor body image and sexual dissatisfaction, independent of the other forms of abuse. There were no significant differences in associations between early abuse and sexuality between persons of Southeast Asian and European ancestry.     

On square ordinal contingency tables: a comparison of social class and income mobility for the same individuals

Summary.  Square contingency tables with matching ordinal rows and columns arise in particular as empirical transition matrices and the paper considers these in the context of social class and income mobility tables. Such tables relate the socio-economic position of parents to the socio-economic position of their child in adulthood. The level of association between parental and child socio-economic position is taken as a measure of mobility. Several approaches to analysis are described and illustrated by UK data in which interest focuses on comparisons of social class and income mobility tables that are derived from the same individuals. Account is taken of the use of the same individuals in the two tables. Additionally comparisons over time are considered.     

Risk-Reducing Legal Documents: Controlling Personal Health and Financial Resources

Every adult can eliminate the risk of losing control of personal health decisions and financial resources by having a last will and testament, health proxy, power of attorney, and living trust. Yet evidence suggests that less than a majority of adult Americans have these documents. We surveyed 900 adult Americans (25 plus years old) to determine what factors predict the possession of these risk-reducing documents. In total, 46% had a living will and testament, 32% a durable power of attorney, 30% a health proxy, and 18% a living trust. Events that prompt people with wealth to secure these documents, such as increasing age, getting married, having children, and experiencing a life-threatening event were the strongest predictors of having a will. These were also among the strongest predictors of having a health proxy, along with a personal history of being proactive, and a family history that fostered trust of attorneys. Affluence, age, and education were the strongest predictors of having a living trust, and the strongest correlate of having a durable power of attorney was having faced a life-threatening event, positive experiences with attorneys, and higher educational achievement.