Religiosity was measured using a validated 11-item Likert scale containing statements about intrinsic spiritual beliefs and formal religious motivation [6]. In addition, subjects were asked if they, their mother, or father had been full article born in the United States (USA) or Mexico. If all three were US born, the subject was considered third generation. If the subject was born in the USA and either parent was born in Mexico they were considered second generation. Subjects born in Mexico were considered first generation. For purposes of analysis, the first and second generation subjects were combined and contrasted against those in the third generation.Depressive symptoms were assessed using a 15-item Geriatric Depression Scale and examined as a continuous variable [7, 8].

Instrumental Activities of Daily Living (IADL) function was dichotomized as needing no help with any activity versus needing help with at least one activity [9]. Self-reported health status was measured by asking subjects, ��in general, how do you see your health today?�� and coded using a 4-point Likert scale ranging from poor (1) to excellent (4). ��Poor�� was combined with ��Fair�� health and contrasted against the combination of ��good�� and ��excellent�� health.2.4. Statistical AnalysisLogistic regression analysis was used to test associations with predictor and outcome variables using SPSS version 17 (Chicago, IL). Variables were selected for the multiple variable model based on either their hypothesized importance or bivariate associations as analyzed by t-test and chi-square with ventilator attitudes and ethnicity.

3. ResultsThe mean age of the sample was 70.1 years. About 58% were female, 62.6% were married, and the majority reported greater than 10 years of education (63.0%). A significant proportion of the sample had been in a professional occupation (34%). English proficiency was high at 75.8% and over a third of the sample were 3rd generation American (34.4%). Health in the sample was good with 4.0% of the sample reporting poor self-reported health and 24% reporting one or more IADL disabilities. Overall depressive symptomatology was also low (mean GDS score = 1.64, SD = 1.95). Religiosity was high in our participants with a mean score of 31.9 (SD = 4.3) out of a possible maximum score of 37.

Table 1 shows the characteristics of those who agreed with ventilator support with 28% of the sample indicated agreement and/or strong agreement for its use at the end of life. Being male (OR = 2.54, CI: 1.02�C6.28) and first or second generation Mexican American (OR = 3.94, CI: 0.122�C12.66).Table Anacetrapib 1Description of the community dwelling older Mexican American sample.In the logistic regression analysis (Table 2) first and second generation Mexican Americans were more likely to agree with ventilation compared with third generation Mexican Americans, (OR = 5.56, CI: 1.51�C20.0).