Study Sample Clause Samples

Study Sample. Due to resource limitations, we extracted hospitals in 12 states (CA, IL, IN, MA, MI, MN, MO, NY, OH, PA, TX, and WA) from the 2008 AHA EHR Dataset. All acute care hospitals in these states with a Medicare provider ID were included in our analysis. In the CMS dataset, hospitals were excluded if all of the measures they reported were based on fewer than 25 patients in the given year. After merging the aforementioned three datasets, the sample available for this analysis was 969 hospitals. It encompasses approximately 24% of the non-federal, acute care hospitals in the United States.

Study Sample. Six participants took part in this research. They provided answers to the semi-structured interview questions as shown in the interview protocol (Appendix 1). The questions are open- ended thus allowing the researcher to probe for further clarification where necessary. All participants were secondary school principals within Kisumu County, who had worked in the same position for more than five years. They consisted of two females and four males. All the interviews were conducted on a one-on-one basis within the school compound.

Study Sample. A total of seven participants participated in the present study and responded to all questions outlined in the interview protocol (Appendix 1) as well as to emerging probes. All respondents were representatives of NGOs in two large urban cities in Kazakhstan. Four participants were interviewed through the face-to-face method and three respondents were interviewed via Skype video call. Two participants were mothers of children with special needs, two others were individuals with disabilities, and the remainders were NGO management representatives who had neither a disability nor a child with special needs. Each respondent had at least two years of experience working in an NGO that works toward the inclusion of people with disabilities.

Study Sample. The couples sub-sample was obtained by matching male heads of households with their spouses in a secondary data analysis exercise (details Table 1). Overall at baseline, 5,232 women and 5,547 men completed interviews within households selected for a male interview. We excluded: 2,251 women because they were not designated as the spouse of a head of the household, 214 because they were not legally married or cohabitating, and 7 were not full time residents of the home*. A similar exclusion criteria was used among the 5,547 men surveyed. Overall, 2,760 women and 2,510 men were considered eligible to be matched. During the matching process, we could not identify partners for 576 women and 399 men, so these individuals were excluded from final analysis. Thus, the final matched sample includes 2,184 couples (2,184 women and 2,111 men since some men had multiple wives with whom they matched). The MLE project obtained ethical clearance from the University of North Carolina at Chapel Hill Institutional Review Board (UNC IRB) and the National Health Research Ethics Committee of Nigeria to conduct the surveys. This secondary data analysis was also approved by the UNC IRB.

Study Sample. We included all women aged 19-64 diagnosed with breast cancer who are eligible for BCCPTA with those in the same age spectrum diagnosed with any one of five other cancers and enrolled in Medicaid in or after the month of their cancer diagnoses. We excluded those who enrolled in Medicaid over 65 since we were unable to observe their medical claims once they were into Medicare. We also excluded those who were enrolled prior to diagnosis because they would not be affected by the new eligibility rules under BCCPTA. These exclusions resulted in 3,238 observations, among whom 2,502 were breast cancer patients and 736 were control cancer patients. Those for which stage was missing in the GCCR and/or Medicaid claims data and those who had more than one primary cancer site over their lifetime were also omitted (N=826). The latter exclusion was made since the regional and distant codes in claims data could easily represent regional or distant disease progression from a different primary cancer than the one we observed. The final sample size was 2,412 women diagnosed with breast (N=1,898) and control cancers (N=514).

Study Sample. The Women’s Interagency HIV Study (WIHS) is a multisite, prospective study designed to investigate the impact and progression of HIV among HIV-infected women and among HIV-uninfected women who are at high risk of HIV infection in the US.35,36 This cross-sectional analysis utilizes screening data from women who were enrolled at the newly added WIHS clinical research sites in Alabama, Georgia, Mississippi, Florida, and North Carolina between October 2013 and September 2015. WIHS participants were women between 25-60 years old. Eligible HIV-infected women were antiretroviral therapy (ART) naïve or started highly active antiretroviral therapy (HAART) after December 31, 2004; had never used didanosine, zalcitabine, or stavudine (unless during pregnancy or for pre- or post-exposure HIV prophylaxis); had never been on non- HAART ART, and had documented pre-HAART CD4 counts and HIV viral load. Eligible HIV-uninfected women reported at least one personal characteristic or male sex partner characteristic associated with increased risk of HIV acquisition within past 5 years (e.g., clinical STI diagnosis). Participants were identified using diverse recruitment strategies, including physician referrals and health fair contacts. Institutional Review Board Approval was obtained at each of the collaborating institutions and written informed consent was obtained from each participant prior to initiation of study procedures. Methods are described in more detail elsewhere.35-37 The analyses described herein are restricted to WIHS participants who provided written informed consent to collect and geocode their home address.

Study Sample. The overall study sample, outlined in Table 2.1a, contains 35 patients with early psychosis, 44 patients with chronic psychosis, 69 unaffected first-degree relatives, 40 ARMS individuals and 76 unrelated controls with no history of psychosis. From this sample, I selected sub-samples for analysis, based on the availability of the EEG data. An overview of this process and the various samples investigated in the course of this thesis is presented in Figure 2.1a. The vast majority of patients included in this study had schizophrenia, schizoaffective disorder or another psychotic disorder (Table 2.1a), however a small number of bipolar disorder patients were also included, if the patient had a lifetime DSM-IV diagnosis of bipolar affective disorder type-1 with clear psychotic features (experiencing hallucinations and/or delusions at some point during their symptom exacerbation). All but 6 psychosis patients were medicated, the majority with an antipsychotic or a combination of antipsychotic + mood stabilizer/antidepressant (Figure 2.1b). Most participants (controls, ARMS and early psychosis patients) were recruited individually, but part of the chronic patients and relatives groups were recruited as part of a family study. Of the 264 participants, 181 (68.56%) were singletons, 58 (21.97%) were part of families with two members in the study, 21 (0.08%) were in three-person families, and 4 (0.015%) were part of one family with four members participating. COMBINED STUDY SAMPLE N=264 79 CONTROLS, 69 RELATIVES, 35 EARLY PSYCHOSIS PATIENTS, 44 CHRONIC PSYCHOSIS PATIENTS & 40 ARMS SUBEJCTS Oddball task Passive oddball paradigm Paired-click paradigm CHAPTER THREE CHAPTER FOUR CHAPTER FIVE N= 261 N=256 N=252 76 CONTROLS, 68 RELATIVES, 74 CONTROLS, 68 RELATIVES, 69 CONTROLS, 67 RELATIVES, 35 EARLY PSYCHOSIS 34 EARLY PSYCHOSIS 35 EARLY PSYCHOSIS PATIENTS, 43 CHRONIC PATIENTS, 43 CHRONIC PATIENTS, 43 CHRONIC PSYCHOSIS PATIENTS & 39 PSYCHOSIS PATIENTS & 39 PSYCHOSIS PATIENTS & 38 ARMS SUBJECTS ARMS SUBJECTS ARMS SUBJECTS CHAPTER SIX COMBINED PARADIGMS EROS SAMPLE N=246 69 CONTROLS, 66 RELATIVES, 34 EARLY PSYCHOSIS PATIENTS, 40 CHRONIC PSYCHOSIS PATIENTS & 37 ARMS SUBJECTS Figure 2.1a Breakdown of samples under investigation N ♀ : ♂ (% Male)a Mean Age (SD)b Tobacco2 – Non-Smokers : Smokers (% Smokers)c DSM-IV Diagnosis Positive Positive (SD) And Negative Negative (SD) Symptoms General (SD) Scale3 Total (SD) 76 69 79 40 34 : 35 (51%) 41 : 28 (41%) 20 : 59 (75%)** 16 : 24 (60%...

Study Sample. State-years are included in this analysis based on the presence (or absence) of a contraceptive coverage mandate and the years of availability in PRAMS. The study sample includes 11 treatment states that implemented contraceptive coverage mandates 2000-2008 and 13 control states that did not implement contraceptive coverage mandates (Table 1). The remaining 26 states and the District of Columbia are excluded from the study due to a lack of participation in PRAMS or missing data years surrounding mandate implementation (Table 2). The state of Texas chose not to release their PRAMS data to be used in this study. First, individual-level analysis uses a quasi-experimental study design exploiting variation in the year of implementation of state prescription contraception coverage mandates using a two-way fixed-effect method. I create a dummy variable indicating the presence or absence of a state contraception coverage mandate for each state-year based on the year of mandate implementation. I then use logistic and multinomial logistic analysis of pooled PRAMS data to estimate the effect of a mandate on the likelihood of each dichotomous measure of the three outcomes of interest: pregnancy prevention efforts, problems getting birth control, and unintended birth. These models compare the treatment group of privately insured recent mothers in state-years with mandates to privately-insured recent mothers in state-years without mandates. All models include state and year fixed effects, robust standard errors clustered at the state level, and PRAMS survey-weights. The base version of these models is presented below: 𝑃(𝑌)𝑖𝑠𝑡 = 𝛽0+ 𝛽1𝑀𝑎𝑛𝑑𝑎𝑡𝑒𝑖𝑠𝑡 + 𝛽2𝑋𝑖𝑠𝑡 + 𝛽3𝑍𝑠𝑡 + 𝑈𝑠 + 𝑇𝑡 + 𝜖 Where 𝑃(𝑌)𝑖𝑠𝑡 is the probability of an outcome variable for a mother (i) in state (s) during year (t); 𝑀𝑎𝑛𝑑𝑎𝑡𝑒𝑠𝑡 is a dummy variable indicating the presence of a contraception coverage mandate in state (s) during year (t); 𝑋𝑖 is a vector of individual characteristics for mother (i); 𝑍𝑠𝑡 is a measure of the percentage of employees insured through self-insured firms in state (s) during year (t); 𝑈𝑠 is a state effect; 𝑇𝑡 is a year effect; and 𝜖 is an unobserved error term.

Study Sample. Recruitment took place at the participating GDPH clinics. Aside from the eligibility conditions previously outlined for entrance into the MFHP, the primary inclusion criterion for this research was randomization into the intervention arm of the parent study. Individuals allocated to the education arm of the study were excluded. The projected recruitment rate for the MFHP was 15 persons per month from all five sites; therefore, it was anticipated that 180 people would enter the study in a one-year timeframe. Using ▇▇▇▇▇▇▇ et al.’s (2013) previous music-based research as a guide, a sample size of 149 was calculated using an attrition rate of 17%. However, this sub-study only focused on the intervention subjects, which was expected to yield a sample size of 75 subjects. For the three aims stated above, all of which were to have involved multivariate regression of one to three variables tested after controlling for one to two covariates, moderate effect sizes were expected for Δr2 (change in r-squared) of 0.13 for a sample size of 75 at 80% power and 5% level of significance. Power analyses were completed using PASS, Version 13 (▇▇▇▇▇▇, 2014). Between March 2015 and February 2016, 34 participants were randomized into the intervention arm of the MFHP. Recruitment rates were inconsistent across all sites despite active screening efforts by study personnel and clinic liaisons, with some going weeks with few or no new participants. Factors contributing to the low recruitment rate are detailed in Chapter 5. Because the sample size was less than half of the expected 75 participants and data results had limited variability, adjustments were made to the initial statistical analysis plan. Specifically, hypothesis testing was conducted using nonparametric statistics, as opposed to the originally planned parametric tests. The Statistical Analysis section of this chapter gives a more thorough accounting of which tests were employed.