Ahmed N. Albatineh

On Similarity Indices and Correction for Chance Agreement

Similarity indices can be used to compare partitions (clusterings) of a data set. Many such indices were introduced in the literature over the years. We are showing that out of 28 indices we were able to track, there are 22 different ones. Even though their values differ for the same clusterings compared, after correcting for agreement attributed to chance only, their values become similar and some of them even become equivalent. Consequently, the problem of choice of the index to be used for comparing different clusterings becomes less important.

Low bone mineral density is associated with balance and hearing impairments

PURPOSE Bone demineralization affects the skeletal system, including the temporal bone, which contains the cochlea and the vestibular labyrinth. However, research on the association of bone mineral density (BMD) with balance and hearing sensitivity is limited with conflicting results. Therefore, we examined the relationship in a population representative sample. METHODS We analyzed 8863 participants to the National Health and Nutrition Examination Survey (1999-2004) aged 40 years and older. Total and head BMD were measured by dual energy x-ray absorptiometry. Balance was evaluated using the Romberg Test of Standing Balance on Firm and Compliant Support Surfaces condition 4, also indicative of vestibular dysfunction. Hearing condition was self-reported. The associations of total and head BMD with balance and hearing were assessed using multiple and multinomial logistic regressions adjusting for covariates. RESULTS On multiple logistic regression, low total BMD was associated with balance impairment (odds ratio [OR], 2.21; 95% confidence interval [CI], 1.43-4.75), especially in older adults (≥65 years old; OR, 3.72; 95% CI, 1.07-12.85). In multinomial regression, low total BMD was associated with report of significant hearing impairment in older adults (OR, 5.30; 95% CI, 1.20-23.26). CONCLUSIONS Low BMD is associated with balance and hearing impairments, especially in older adults.

Factors Explaining Racial/Ethnic Disparities in Rates of Physician Recommendation for Colorectal Cancer Screening

OBJECTIVES Physician recommendation plays a crucial role in receiving endoscopic screening for colorectal cancer (CRC). This study explored factors associated with racial/ethnic differences in rates of screening recommendation. METHODS Data on 5900 adults eligible for endoscopic screening were obtained from the National Health Interview Survey. Odds ratios of receiving an endoscopy recommendation were calculated for selected variables. Planned, sequenced logistic regressions were conducted to examine the extent to which socioeconomic and health care variables account for racial/ethnic disparities in recommendation rates. RESULTS Differential rates were observed for CRC screening and screening recommendations among racial/ethnic groups. Compared with Whites, Hispanics were 34% less likely (P < .01) and Blacks were 26% less likely (P < .05) to receive this recommendation. The main predictors that emerged in sequenced analysis were education for Hispanics and Blacks and income for Blacks. After accounting for the effects of usual source of care, insurance coverage, and education, the disparity reduced and became statistically insignificant. CONCLUSIONS Socioeconomic status and access to health care may explain major racial/ethnic disparities in CRC screening recommendation rates.

Correcting Jaccard and other similarity indices for chance agreement in cluster analysis

Correcting a similarity index for chance agreement requires computing its expectation under fixed marginal totals of a matching counts matrix. For some indices, such as Jaccard, Rogers and Tanimoto, Sokal and Sneath, and Gower and Legendre the expectations cannot be easily found. We show how such similarity indices can be expressed as functions of other indices and expectations found by approximations such that approximate correction is possible. A second approach is based on Taylor series expansion. A simulation study illustrates the effectiveness of the resulting correction of similarity indices using structured and unstructured data generated from bivariate normal distributions.

MCS: A Method for Finding the Number of Clusters

This paper proposes a maximum clustering similarity (MCS) method for determining the number of clusters in a data set by studying the behavior of similarity indices comparing two (of several) clustering methods. The similarity between the two clusterings is calculated at the same number of clusters, using the indices of Rand (R), Fowlkes and Mallows (FM), and Kulczynski (K) each corrected for chance agreement. The number of clusters at which the index attains its maximum is a candidate for the optimal number of clusters. The proposed method is applied to simulated bivariate normal data, and further extended for use in circular data. Its performance is compared to the criteria discussed in Tibshirani, Walther, and Hastie (2001). The proposed method is not based on any distributional or data assumption which makes it widely applicable to any type of data that can be clustered using at least two clustering algorithms.

Immediate rather than delayed memory impairment in older adults with latent toxoplasmosis.

The neurotropic parasite Toxoplasma gondii infects one third of the world population, but its effect on memory remains ambiguous. To examine a potential relationship of the infection with immediate and delayed memory, a population-based study was conducted in 4485 participants of the Third National Health and Nutrition Examination Survey aged 60years and older. Serum anti-Toxoplasma IgG antibodies were measured by enzyme immune assay and verbal memory was assessed using the Mini-Mental State Examination and the East Boston Memory Test. The prevalence of latent toxoplasmosis was 41%; in one way analysis of variance, anti-Toxoplasma IgG antibody levels significantly differed across tertiles for immediate (P=0.006) but not delayed memory scores (P=0.22). In multinomial logistic regression adjusting for covariates, Toxoplasma seropositivity was associated with lower immediate memory performance (OR: 0.65, 95% CI: 0.44, 0.97 for medium tertile and OR: 0.61, 95% CI: 0.37, 0.98 for highest tertile in reference to the lowest tertile), especially in non-Hispanic Whites (OR: 0.56, 95% CI: 0.36, 0.88 for medium tertile and OR: 0.51, 95% CI: 0.30, 0.87 for highest tertile in reference to the lowest tertile). However, no relationship with delayed memory was observed. In conclusion, latent toxoplasmosis is widespread in older adults and may primarily affect immediate rather than delayed memory, particularly in White Americans.

Intermediate outcomes of a chronic disease self-management program for Spanish-speaking older adults in South Florida, 2008-2010.

Introduction The prevalence and negative health effects of chronic diseases are disproportionately high among Hispanics, the largest minority group in the United States. Self-management of chronic conditions by older adults is a public health priority. The objective of this study was to examine 6-week differences in self-efficacy, time spent performing physical activity, and perceived social and role activities limitations for participants in a chronic disease self-management program for Spanish-speaking older adults, Tomando Control de su Salud (TCDS). Methods Through the Healthy Aging Regional Collaborative, 8 area agencies delivered 82 workshops in 62 locations throughout South Florida. Spanish-speaking participants who attended workshops from October 1, 2008, through December 31, 2010, were aged 55 years or older, had at least 1 chronic condition, and completed baseline and post-test surveys were included in analysis (N = 682). Workshops consisted of six, 2.5-hour sessions offered once per week for 6 weeks. A self-report survey was administered at baseline and again at the end of program instruction. To assess differences in outcomes, a repeated measures general linear model was used, controlling for agency and baseline general health. Results All outcomes showed improvement at 6 weeks. Outcomes that improved significantly were self-efficacy to manage disease, perceived social and role activities limitations, time spent walking, and time spent performing other aerobic activities. Conclusion Implementation of TCDS significantly improved 4 of 8 health promotion skills and behaviors of Spanish-speaking older adults in South Florida. A community-based implementation of TCDS has the potential to improve health outcomes for a diverse, Spanish-speaking, older adult population.

Confidence interval estimation for the population coefficient of variation using ranked set sampling: a simulation study

In this paper, an evaluation of the performance of several confidence interval estimators of the population coefficient of variation (τ) using ranked set sampling compared to simple random sampling is performed. Two performance measures are used to assess the confidence intervals for τ, namely: width and coverage probabilities. Simulated data were generated from normal, log-normal, skew normal, Gamma, and Weibull distributions with specified population parameters so that the same values of τ are obtained for each distribution, with sample sizes n=15, 20, 25, 50, 100. A real data example representing birth weight of 189 newborns is used for illustration and performance comparison.

Toxoplasma gondii seropositivity and cognitive functions in school-aged children.

Toxoplasma gondii (T. gondii) infects one-third of the world population, but its association with cognitive functions in school-aged children is unclear. We examined the relationship between Toxoplasma seropositivity and neuropsychological tests scores (including math, reading, visuospatial reasoning and verbal memory) in 1755 school-aged children 12-16 years old who participated to the Third National Health and Nutrition Examination Survey, using multiple linear regressions adjusted for covariates. Toxoplasma seroprevalence was 7·7% and seropositivity to the parasite was associated with lower reading skills (regression coefficient [β] = -5·86, 95% confidence interval [CI]: -11·11, -0·61, P = 0·029) and memory capacities (β = -0·86, 95% CI: -1·58, -0·15, P = 0·017). The interaction between T. gondii seropositivity and vitamin E significantly correlated with memory scores. In subgroup analysis, Toxoplasma-associated memory impairment was worse in children with lower serum vitamin E concentrations (β = -1·61, 95% CI: -2·44, -0·77, P < 0·001) than in those with higher values (β = -0·12, 95% CI: -1·23, 0·99, P = 0·83). In conclusion, Toxoplasma seropositivity may be associated with reading and memory impairments in school-aged children. Serum vitamin E seems to modify the relationship between the parasitic infection and memory deficiency.

Intermediate outcomes of chronic disease self-management program offered by members of the Healthy Aging Regional Collaborative in South Florida.

Currently, 80% of adults over the age of 65 have at least one chronic disease. The Chronic Disease Self-management Program (CDSMP) focuses on increasing self-efficacy for managing chronic disease. Few studies have evaluated the effectiveness of CDSMP when offered by multiple agencies, as a collaborative effort, in community-based settings. Seven agencies delivered 108 CDSMP workshops at 81 sites from October 1, 2008, to December 31, 2010. A total of 811 participants were eligible for analysis. Participants completed surveys at baseline and week 6, the end of instruction. Controlling for agency effect and general health at baseline, the general linear model was used to assess the significance of outcomes at 6 weeks. Outcomes showing significant improvement included self-efficacy to manage disease (p = .001), self-efficacy to manage emotions (p = .026), time spent walking (p = .008), and perceived social/role activities limitations (p = .001). Findings showed that CDSMP is an effective program at improving self-efficacy, increasing physical activity, and decreasing limitations.