Missing value imputation in a data matrix using the regularised singular value decomposition.

Some statistical analysis techniques may require complete data matrices, but a frequent problem in the construction of databases is the incomplete collection of information for different reasons. One option to tackle the problem is to estimate and impute the missing data. This paper describes a form of imputation that mixes regression with lower rank approximations. To improve the quality of the imputations, a generalisation is proposed that replaces the singular value decomposition (SVD) of the matrix with a regularised SVD in which the regularisation parameter is estimated by cross-validation. To evaluate the performance of the proposal, ten sets of real data from multienvironment trials were used. Missing values were created in each set at four percentages of missing not at random, and three criteria were then considered to investigate the effectiveness of the proposal. The results show that the regularised method proves very competitive when compared to the original method, beating it in several of the considered scenarios. As it is a very general system, its application can be extended to all multivariate data matrices. •The imputation method is modified through the inclusion of a stable and efficient computational algorithm that replaces the classical SVD least squares criterion by a penalised criterion. This penalty produces smoothed eigenvectors and eigenvalues that avoid overfitting problems, improving the performance of the method when the penalty is necessary. The size of the penalty can be determined by minimising one of the following criteria: the prediction errors, the Procrustes similarity statistic or the critical angles between subspaces of principal components.

Missing value imputation using least squares techniques in contaminated matrices.

This paper describes strategies to reduce the possible effect of outliers on the quality of imputations produced by a method that uses a mixture of two least squares techniques: regression and lower rank approximation of a matrix. To avoid the influence of discrepant data and maintain the computational speed of the original scheme, pre-processing options were explored before applying the imputation method. The first proposal is to previously use a robust singular value decomposition, the second is to detect outliers and then treat the potential outliers as missing. To evaluate the proposed methods, a cross-validation study was carried out on ten complete matrices of real data from multi-environment trials. The imputations were compared with the original data using three statistics: a measure of goodness of fit, the squared cosine between matrices and the prediction error. The results show that the original method should be replaced by one of the options presented here because outliers can cause low quality imputations or convergence problems.•The imputation algorithm based on Gabriel's cross-validation method uses two least squares techniques that can be affected by the presence of outliers. The inclusion of a robust singular value decomposition allows both to robustify the procedure and to detect outliers and consider them later as missing. These forms of pre-processing ensure that the algorithm performs well on any dataset that has a matrix form with suspected contamination.

Síndrome de vasoconstricción cerebral reversible de evolución mortal / Fulminant reversible cerebral vasoconstriction syndrome

No disponible

Suicide ideation, plans, and attempts among general practice patients with chronic health conditions in Puerto Rico.

BACKGROUND: Little is known about suicidal ideation among general practice patients in Puerto Rico. In this study we examined the rates, severity, and correlates of suicidal ideation, plans, and attempts among general practice patients with chronic illnesses. This is important in targeting appropriate interventions and management approaches to minimize and prevent suicide. METHODS: We screened patients with chronic physical conditions at general practices. Suicidal ideation was assessed with the suicidality module of the Mini International Neuropsychiatric Interview. Major depression was assessed with the Patient Health Questionnaire depression module. The relationship between sociodemographic factors, depression and suicidal ideation was examined with multiple logistic regression analysis. Among the subgroup that acknowledged suicidal ideation, we used multinomial logistic regression analysis to estimate simultaneously the multivariate associations of depression and sociodemographic factors with suicidality risk levels. RESULTS: Of the 2068 patients screened, 15.4% acknowledged recent suicidal ideation. Among this group, 8.6% reported passive ideation, 3.7% active ideation without a plan, and 3.1% active ideation with a plan or attempt. According to multivariate logistic regression, suicidal ideation was higher among patients with moderately severe depression and severe depression than for those with milder symptoms. Patients aged 64 years or younger were over one and a half times more likely to acknowledge suicidal ideation than those aged 65 years and older. Compared with patients having a college degree, those with lower education had a twofold higher risk of suicidal ideation. Multinomial logistic regression analysis indicated that severe depression was associated with a higher likelihood of having a suicide plan or attempt. CONCLUSION: The findings of this study suggest that public health strategies focusing on the systematic identification of patients with increased depression severity and the implementation of evidence-based depression treatments are relevant for minimizing and preventing suicidal behavior among general practice patients with chronic health conditions.

Phenotypic correction of a mouse model for primary hyperoxaluria with adeno-associated virus gene transfer.

Primary hyperoxaluria type I (PH1) is an inborn error of metabolism caused by deficiency of the hepatic enzyme alanine-glyoxylate aminotransferase (AGXT or AGT) which leads to overproduction of oxalate by the liver and subsequent urolithiasis and renal failure. The current therapy largely depends on liver transplantation, which is associated with significant morbidity and mortality. To explore an alternative treatment, we used somatic gene transfer in a mouse genetic model for PH1 (Agxt1KO). Recombinant adeno-associated virus (AAV) vectors containing the human AGXT complementary DNA (cDNA) were pseudotyped with capsids from either serotype 8 or 5, and delivered to the livers of Agxt1KO mice via the tail vein. Both AAV8-AGXT and AAV5-AGXT vectors were able to reduce oxaluria to normal levels. In addition, treated mice showed blunted increase of oxaluria after challenge with ethylene glycol (EG), a glyoxylate precursor. In mice, AGT enzyme activity in whole liver extracts were restored to normal without hepatic toxicity nor immunogenicity for the 50 day follow-up. In summary, this study demonstrates the correction of primary hyperoxaluria in mice treated with either AAV5 or AAV8 vectors.

Economic cost of dengue public prevention activities in Puerto Rico.

Dengue fever has become a major global public health problem in Puerto Rico. Approximately 5000 suspected cases were reported annually between 2002 and 2007. Vector control is currently the only approach to control the disease and includes prevention education, fumigation, inspections and clean-up campaigns. The annual cost to the public sector of dengue prevention, which includes surveillance and vector control activities, was estimated as part of a study of the economic burden of dengue in Puerto Rico. A telephonic survey was implemented to identify municipalities with vector control programmes and public agencies with dengue surveillance systems. Onsite interviews were conducted using a structured questionnaire. The economic cost of dengue was summarized by line item, function and year from 2002 through 2007. The Puerto Rico Department of Health (PRDH) and 12 municipalities out of 78 conducted vector control activities in different magnitudes during the study years. The cumulative cost of dengue vector control in the public sector was US$ 46.22 million for the years 2002–2007. PRDH spent an average of US$ 1.29 million ($0.33 per capita) per year, while the municipalities spent an average of US$ 6.41 million (US$ 1.64 per capita) per year. Clean-up campaigns had the highest share of average expenditure, followed by fumigation, surveillance and inspection. Puerto Rico’s per capita expenditure on dengue prevention activities is similar to that of other countries in the region. On average, Puerto Rico’s per capita spending on dengue illness is US$ 5.48 compared with US$ 1.97 spent on vector control.

The effect of Medicaid managed care on prenatal care: the case of Puerto Rico.

OBJECTIVES: From 1994 to the year 2000 the government of Puerto Rico implemented a health care reform which included the mandatory enrollment of the entire Medicaid eligible population under Medicaid managed care (MMC) plans. This study assessed the effect of MMC on the use, initiation, utilization, and adequacy of prenatal care services over the reform period. METHODS: Using the vital records of all infants born alive in Puerto Rico from the year 1995-2000, a series of bivariate and multivariate analyses were conducted to assess the effect of insurance status (traditional Medicaid, MMC, private insurance and uninsured) on prenatal care utilization patterns. In order to assess the potential influence of selection bias in generating the health insurance assignments, propensity scores (PS) were estimated and entered into the multivariate regressions. RESULTS: MMC had a generally positive effect on the frequency and adequacy of prenatal care when compared with the experience of women covered by traditional Medicaid. However, the PS analyses suggested that self-selection may have generated part of the observed beneficial effects. Also, MMC reduced but did not eliminate the gap in the amount and adequacy of prenatal care received by pregnant women covered by Medicaid when compared to their counterparts covered by private insurance. CONCLUSIONS: The Puerto Rico Health Reform to implement MMC for pregnant women was associated with a general improvement in prenatal care utilization. However, continued progress will be necessary for women covered by Medicaid to reach prenatal care utilization levels experienced by privately insured women.

Hepatitis C in Puerto Rico: a time for public health action.

Studies investigating the seroprevalence of HCV infection have been carried out in diverse populations, showing an estimated worldwide prevalence of 3%. A seroprevalence survey conducted among randomly selected non-institutionalized adults aged 21-64 years in San Juan, Puerto Rico in 2001-2002 revealed that 6.3% were positive for HCV antibodies. These data suggest that Puerto Ricans are burdened with a significantly greater prevalence of HCV infection compared to the general United States population aged 20-69 years (0.9%-4.3%). This article illustrates data from different sources that taken together establish the need to start addressing HCV infection in Puerto Rico with prompt and decisive public health actions. Some of these include (1) establish hepatitis C prevention as a priority for state and municipal public health authorities, (2) raise awareness and educate target populations about HCV transmission and prevention, (3) increase clinician awareness of the HCV reporting system and the epidemiology and management of hepatitis C, (4) increase availability of diagnosis and treatment facilities, (5) increase access to effective drug treatment services, and (6) develop appropriate control measures to help reduce continued transmission in correctional settings.

Myotube depolarization generates reactive oxygen species through NAD(P)H oxidase; ROS-elicited Ca2+ stimulates ERK, CREB, early genes.

Controlled generation of reactive oxygen species (ROS) may contribute to physiological intracellular signaling events. We determined ROS generation in primary cultures of rat skeletal muscle after field stimulation (400 1-ms pulses at a frequency of 45 Hz) or after depolarization with 65 mM K+ for 1 min. Both protocols induced a long lasting increase in dichlorofluorescein fluorescence used as ROS indicator. Addition of diphenyleneiodonium (DPI), an inhibitor of NAD(P)H oxidase, PEG-catalase, a ROS scavenger, or nifedipine, an inhibitor of the skeletal muscle voltage sensor, significantly reduced this increase. Myotubes contained both the p47phox and gp91phox phagocytic NAD(P)H oxidase subunits, as revealed by immunodetection. To study the effects of ROS, myotubes were exposed to hydrogen peroxide (H2O2) at concentrations (100-200 microM) that did not alter cell viability; H2O2 induced a transient intracellular Ca2+ rise, measured as fluo-3 fluorescence. Minutes after Ca2+ signal initiation, an increase in ERK1/2 and CREB phosphorylation and of mRNA for the early genes c-fos and c-jun was detected. Inhibition of ryanodine receptor (RyR) decreased all effects induced by H2O2 and NAD(P)H oxidase inhibitors DPI and apocynin decreased ryanodine-sensitive calcium signals. Activity-dependent ROS generation is likely to be involved in regulation of calcium-controlled intracellular signaling pathways in muscle cells.

Drug treatment, health, and social service utilization by substance abusing women from a community-based sample.

BACKGROUND: Substance abuse is an escalating problem among poor urban Latina women; little is known about their access to drug treatment and to needed social and health services. OBJECTIVE: Our objectives were to (1) examine the need and use of substance abuse treatment, health services, government entitlement programs, and social service programs among cocaine and heroin using Puerto Rican women and (2) identify whether service use predicts their prospective entry into drug treatment. RESEARCH DESIGN: This was a 3-wave longitudinal study of community substance abusing women evaluated on substance abuse and dependence using diagnostic measures, and hair and urine toxicological screens. Information was collected on self-reported need and receipt of substance abuse treatment, social services, general health services, and government entitlement programs. SUBJECTS: A community sample of cocaine-, crack-, and/or heroin-using women from copping areas in low-income urban centers of Puerto Rico were interviewed in 1997-1998 with 2 follow-up periods. RESULTS: Drug treatment, health, and social service utilization were low relative to need for services throughout all data waves. Social service utilization predicted prospective entry into drug treatment but not contacts with general health services or government entitlement programs. CONCLUSION: Drug-abusing women in low-income urban areas in Puerto Rico have substantial unmet substance abuse treatment, health, and social service needs. Mandated treatment by social service agencies may explain their clients' higher likelihood of entering drug treatment. Building linkages between service sectors to augment entry into drug treatment is essential for meeting the complex needs of this underserved population.

Understanding hard-core drug use among urban Puerto Rican women in high-risk neighborhoods.

The objectives of the current study were to determine the incidence and prevalence of hard-core drug (HCD) use and to explore the predictors of HCD use in a three-wave longitudinal study of women ages 18-35 living in urban areas of high drug traffic in Puerto Rico. Prevalence of HCD use was determined by positive self-report or positive toxicological tests for crack, cocaine, or heroine at baseline and follow-up periods. Incidence density was calculated using Poisson methods. Predictors of HCD use were identified using the generalized estimating equation (GEE) approach. Prevalence rates fluctuated between 16.3 at baseline and 12.6 and 14.6 at subsequent waves. Incidence of crack/cocaine or heroine ranged from 7.4 to 6.3 per 100 person years. Low education, unemployment, current alcohol use, and severe partner violence predict incident HCD use. Therefore, prevention programs for HCD use must consider strategies to promote social mobility and empowerment and to mitigate violence against women.