Retrospective study of prevalence of Community Associated Methicillin Resistant Staphylococcus Aureus (CA-MRSA) seen in Texas MedClinics, San Antonio, TX from April 2006-August 2009

Purpose. To evaluate the presence of Community Associated–Methicillin Resistant Staphylococcus Aureus, CA-MRSA, in abscesses and skin and soft tissue infections presenting at 9 urgent care clinics in San Antonio, TX. ^ Methods. During the 40-month retrospective study (April 2006 to August 2009), wound cultures collected in 9 urgent care centers were evaluated for MRSA growth, antibiotics prescribed, follow up wound care, and antibiotic prescribing habits by physicians for all patients presenting with abscesses and skin/soft tissue infections. ^ Results. Across 9 urgent care centers in San Antonio, TX, 36,797 abscesses and cases of skin and soft tissue infections were treated during 40 months. Of the 36,797 cases, 9290 patients had wound cultures sent with 5,630 cultures sent to Texas MedClinic’s primary lab. Of the 5630 cultures sent to their primary lab, this reflected a prevalence of 4727 (84 %) cultures were positive for MRSA. Of the 9290 patients who had a wound culture sent (April 10th, 2006 to August 31st, 2009), a total of 4,307 antibiotics were prescribed. The top five antibiotics prescribed for CA-MRSA were Bactrim (55.5%), Clindamycin (18.4%), Bactroban (5%), Amoxicillin (3.5%), and Doxycycline (3%) representing 85.4% of the antibiotics prescribed. 8809/9290 (94.8%) of patients required no more than 3 follow up visits. Of the 33 physicians working full time during the entire study period, 29/33 (87.8%) of the physicians were family medicine physicians and represented varied prescribing rates of antibiotics between 11-76% with 26/33 (78.8%) of physicians prescribing antibiotics greater than 40% of the time.^ Conclusions. Abscesses and soft tissue infections are a common presenting complaint to urgent care centers. This study reveals that antibiotic-prescribing practices can be improved with physician education since this high prevalence was not known previously. Also, treating abscesses with limited packing has been shown to be a viable option in this particular circumstance and would be open field for additional clinical research. Due to the high prevalence of CA-MRSA skin and soft tissue infections among patients presenting to urgent care centers presumptive treatment for MRSA is indicated. Increasing levels of resistance to penicillin antibiotics is concerning and warrants alternative antibiotic management strategies.^

The effect of exposure to antibiotics on incidence and spontaneous clearance of childhood Helicobacter pylori infection

It is claimed often in the H. pylori literature that spontaneous clearance (infection loss without attempts to treat) is uncommon, though little evidence supports this claim. Emerging evidence suggests that spontaneous clearance may be frequent in young children; however, factors that determine persistence of untreated H. pylori infection in childhood are not well understood. The author hypothesized that antibiotics taken for common infections cause spontaneous clearance of H. pylori infection in children. The Pasitos Cohort Study (19982005) investigated predictors of acquisition and persistence of H. pylori infection in children from El Paso, Texas, and Juarez, Mexico, enrolled prenatally at maternal-child clinics. Children were screened for infection at target intervals of 6 months from 6-84 months of age by the 13C-urea breath test corrected for body-size-dependent variation in CO2 production. This dissertation aimed to estimate the risk of spontaneous clearance at the next test following an initial detected H. pylori infection (first detected clearance), estimate the effect of antibiotic exposure on the risk of first detected clearance (risk difference), and estimate the effect of antibiotic exposure on the rate of first detected infection (rate ratio). Data on infection status and medication history were available for 608 children followed for a mean of 3.5 years. Among 265 subjects with a first detected infection, 218 had a subsequent test, and among them, the risk of first detected clearance was 68% (95% CI: 61-74%). Children who took antibiotics during the interval between first detected infection and next test had an increased probability (risk difference of 10 percentage points) of a first detected clearance. However, there was also a similar effect of average antibiotic use >0 courses across all intervals preceding the next test. Average antibiotic exposure across all intervals preceding the first detected infection appeared to have a much stronger protective effect than interval/specific exposure when estimating incidence rate ratios (0.45 vs. 1.0). Incidental antibiotic exposure appears to influence the acquisition and duration of childhood H. pylori infection, however, given that many exposed children acquired the infection and many unexposed children cleared the infection, antibiotic exposure does not explain all infection events. ^

Estimating parameters in Markov models for longitudinal studies with missing data or surrogate outcomes

The discrete-time Markov chain is commonly used in describing changes of health states for chronic diseases in a longitudinal study. Statistical inferences on comparing treatment effects or on finding determinants of disease progression usually require estimation of transition probabilities. In many situations when the outcome data have some missing observations or the variable of interest (called a latent variable) can not be measured directly, the estimation of transition probabilities becomes more complicated. In the latter case, a surrogate variable that is easier to access and can gauge the characteristics of the latent one is usually used for data analysis. ^ This dissertation research proposes methods to analyze longitudinal data (1) that have categorical outcome with missing observations or (2) that use complete or incomplete surrogate observations to analyze the categorical latent outcome. For (1), different missing mechanisms were considered for empirical studies using methods that include EM algorithm, Monte Carlo EM and a procedure that is not a data augmentation method. For (2), the hidden Markov model with the forward-backward procedure was applied for parameter estimation. This method was also extended to cover the computation of standard errors. The proposed methods were demonstrated by the Schizophrenia example. The relevance of public health, the strength and limitations, and possible future research were also discussed. ^

Development and preliminary validation of the cancer family impact scale for colorectal cancer

This study aimed to develop and validate The Cancer Family Impact Scale (CFIS), an instrument for use in studies investigating relationships among family factors and colorectal cancer (CRC) screening when family history is a risk factor. We used existing data to develop the measure from 1,285 participants (637 families) across the United States who were in the Johns Hopkins Colon Cancer Genetic Testing study. Participants were 94% white with an average age of 50.1 years, and 60% were women. None had a personal CRC history, and eighty percent had 1 FDR with CRC and 20% had more than one FDR with CRC. The study had three aims: (1) to identify the latent factors underlying the CFIS via exploratory factor analysis (EFA); (2) to confirm the findings of the EFA via confirmatory factor analysis (CFA); and (3) to assess the reliability of the scale via Cronbach's alpha. Exploratory analyses were performed on a split half of the sample, and the final model was confirmed on the other half. The EFA suggested the CFIS was an 18-item measure with 5 latent constructs: (1) NEGATIVE: negative effects of cancer on the family; (2) POSITIVE: positive effects of cancer on the family; (3) COMMUNICATE: how families communicate about cancer; (4) FLOW: how information about cancer is conveyed in families; and (5) NORM: how individuals react to family norms about cancer. CFA on the holdout sample showed the CFIS to have a reasonably good fit (Chi-square = 389.977, df = 122, RMSEA= 0.058 (.052-.065), CFI=.902, TLI=.877, GF1=.939). The overall reliability of the scale was α=0.65. The reliability of the subscales was: (1) NEGATIVE α = 0.682; (2) POSITIVE α = 0.686; (3) COMMUNICATE α = 0.723; (4) FLOW α = 0.467; and (5) NORM α = 0.732. ^ We concluded the CFIS to be a good measure with most fit levels over 0.90. The CFIS could be used to compare theoretically driven hypotheses about the pathways through which family factors could influence health behavior among unaffected individuals at risk due to family history, and also aid in the development and evaluation of cancer prevention interventions including a family component. ^

The Texas driver responsibility program: A preliminary analysis of the impact on impaired driving and trauma system funding

Objective. In 2003, the State of Texas instituted the Driver Responsibility Program (TDRP), a program consisting of a driving infraction point system coupled with a series of graded fines and annual surcharges for specific traffic violations such as driving while intoxicated (DWI). Approximately half of the revenues generated are earmarked to be disbursed to the state's trauma system to cover uncompensated trauma care costs. This study examined initial program implementation, the impact of trauma system funding, and initial impact on impaired driving knowledge, attitudes and behaviors. A model for targeted media campaigns to improve the program's deterrence effects was developed. ^ Methods. Data from two independent driver survey samples (conducted in 1999 and 2005), department of public safety records, state health department data and a state auditor's report were used to evaluate the program's initial implementation, impact and outcome with respect to drivers' impaired driving knowledge, attitudes and behavior (based on constructs of social cognitive theory) and hospital uncompensated trauma care funding. Survey results were used to develop a regression model of high risk drivers who should be targeted to improve program outcome with respect to deterring impaired driving. ^ Results. Low driver compliance with fee payment (28%) and program implementation problems were associated with lower surcharge revenues in the first two years ($59.5 million versus $525 million predicted). Program revenue distribution to trauma hospitals was associated with a 16% increase in designated trauma centers. Survey data demonstrated that only 28% of drivers are aware of the TDRP and that there has been no initial impact on impaired driving behavior. Logistical regression modeling suggested that target media campaigns highlighting the likelihood of DWI detection by law enforcement and the increased surcharges associated with the TDRP are required to deter impaired driving. ^ Conclusions. Although the TDRP raised nearly $60 million in surcharge revenue for the Texas trauma system over the first two years, this study did not find evidence of a change in impaired driving knowledge, attitudes or behaviors from 1999 to 2005. Further research is required to measure whether the program is associated with decreased alcohol-related traffic fatalities. ^

A cross sectional examination of risky sexual behaviors among a population of drug users

This study is a secondary data analysis that assesses the relationship between risky sexual behaviors and sexually transmitted infections (STIs) among drug users. This study analyzes data collected from drug users in the Houston Metropolitan area during 2004 and through August 2005, by researchers with the DASH (Drugs, AIDS, STDs and Hepatitis) project at The University of Texas at Houston School of Public Health. Specifically, the sexually transmitted infections that will be of interest in this proposed study are Human Immunodeficiency Virus (HIV) and Hepatitis B Virus (HBV). Risky sexual behaviors that will be examined include lack of condom use, sexual orientation, trading sex for drugs, trading sex for money, and number of male and female sexual partners in the last 4 weeks. ^ Unadjusted, gender, sexual orientation, number of recent male and female sex partners, and a history of injection drug use were all found to be significant independent variables that increased the odds of STI status. When included in an overall model, these variables significantly increased the odds of STI status, including HBV infection, HIV infection, and HBV/HIV co-infection. History of injection drug use was significant for both HBV and HBV/HIV co-infection, whereas a gay sexual orientation was significant for both HIV and HBV/HIV co-infection. Additionally, having excessive female sex partners was significant for HIV infection. This significant association increases the need for implementation of stronger intervention programs tailored to suit this population's needs such as a combination of drug and sexually transmitted disease (STD) treatment. ^ The importance of these findings is that they establish the strength of associations between the previously mentioned risky sexual behaviors and STI status among drug users. This is crucial for assessing future risk of infection as well as for serving as a necessary component in intervention and treatment programs both for drug use and STIs. ^

Oncologic outcomes in diabetic patients with colon cancer

Type II diabetes mellitus is a growing problem worldwide and although its association with increased cardiovascular morbidity and mortality is well known, its role in the development of cancer is now being further elucidated. Recently, there has been increasing evidence that not only are diabetics more susceptible towards development of particular types of cancer, but also have worse oncologic outcomes. This retrospective chart review investigates whether diabetics with colon cancer have a poorer prognosis than their nondiabetic counterparts. Patients with high risk Stage II and Stage III colon cancer who were diagnosed and/or treated at The University of Texas M.D. Anderson Cancer Center from 1/1/2000 till 12/1/2004 were included in our study. We carried out a survival analysis using Kaplan-Meier method and multivariable analysis to assess differences in outcomes of the two population groups. We found that the decreased overall survival in diabetics did not reach statistical significance but this could be due to a lower event rate in our study. Larger studies are required to investigate this further. ^