Primary Outcomes

Description: A primary outcome of this study will be the degree of DIF in NIH PROMIS questionnaires observed across adequate versus low health literacy. The outcome will be measured by a McFadden pseudo-R-square, which captures the degree to which health-literacy group determines probability of response type for each item. PROMIS has the following subscales: Physical function, Anxiety, Depression, Fatigue, Sleep disturbance, Ability to participate in social roles and activities, Pain interference, Pain intensity. For DIF analysis, each scale is normalized to a theta metric (mean = 0, standard dev. = 1 by definition). PROMIS is scaled by T-scores, referenced against the mean of the US population. Items are aggregated using item response theory, but aggregation is not relevant to this study because the McFadden pseudo-R-square is evaluated for each item. Pain intensity will not be included because it is only 1 item and is not amenable to DIF.

Description: A primary outcome of this study will be the degree of DIF observed in the Berlin Questionnaire across adequate versus low health. The outcome will be measured by a McFadden pseudo-R-square, which captures the degree to which health-literacy group determines probability of response type for each item. For DIF analysis, each scale is normalized to a theta metric (mean = 0, standard dev. = 1 by definition). Thus, the unit of measure is one standard deviation. Normally items are aggregated by a sum score, but the aggregation is not relevant to this study because the McFadden pseudo-R-square is evaluated for each item, not for aggregate scores.

Description: A primary outcome of this study will be the degree of DIF observed in the NIH Toolbox - Meaning and Purpose questionnaire across adequate versus low health literacy, determined by Health LiTT. The outcome will be measured by a McFadden pseudo-R-square, which captures the degree to which health-literacy group determines probability of response type for each item. For DIF analysis, each scale is normalized to a theta metric (mean = 0, standard dev. = 1 by definition). Thus, the unit of measure is one standard deviation. Normally items are aggregated by a score determined by an item response theory scoring algorithm using a two-parameter model. It should be noted that the aggregation is not relevant to this study because the McFadden pseudo-R-square is evaluated for each item, not for aggregate scores. By convention NIH Toolbox scores are rescaled to T scores (mean = 50, Standard Deviation (SD) = 10 by definition), referenced against the mean of the US population.

Description: A primary outcome of this study will be the degree of DIF observed in the PCL-5 (PTSD checklist) across adequate versus low health literacy, determined by the Health Literacy Assessment Using Talking Touchscreen Technology (Health LiTT). The PCL-5 is a questionnaire used to determine the presence of PTSD and severity of PTSD. The outcome will be measured by a McFadden pseudo-R-square, which captures the degree to which health-literacy group determines probability of response type for each item. For DIF analysis, each scale is normalized to a theta metric (mean = 0, standard dev. = 1 by definition). Thus, the unit of measure is one standard deviation. Normally items are aggregated by a sum score, but the aggregation is not relevant to this study because the McFadden pseudo-R-square is evaluated for each item, not for aggregate scores.

Description: A primary outcome of this study will be the degree of DIF observed in the RRS across adequate versus low health literacy, determined by Health LiTT. The RRS is a questionnaire used to quantify the degree to which the participant ruminates. The outcome will be measured by a McFadden pseudo-R-square, which captures the degree to which health-literacy group determines probability of response type for each item. For DIF analysis, each scale is normalized to a theta metric (mean = 0, standard dev. = 1 by definition). Thus, the unit of measure is one standard deviation. Normally items are aggregated by a sum score, but the aggregation is not relevant to this study because the McFadden pseudo-R-square is evaluated for each item, not for aggregate scores.

Description: A primary outcome of this study will be the degree of DIF observed in the Patient Health Questionnaire (PHQ-9), a questionnaire used to measure depression, across adequate versus low health literacy. The outcome will be measured by a McFadden pseudo-R-square, which captures the degree to which health-literacy group determines probability of response type for each item. For DIF analysis, each scale is normalized to a theta metric (mean = 0, standard dev. = 1 by definition). Thus, the unit of measure is one standard deviation. Normally items are aggregated by a sum score, but the aggregation is not relevant to this study because the McFadden pseudo-R-square is evaluated for each item, not for aggregate scores.

Description: At the end of the 6-month study, each participant will provide information about their experience during testing. There are about 1) Comfort, 2) Difficulty, and 3) Nervousness. Each items is rated on a 5-point scale. Investigators will treat these data as ordinal and thus no unit of measure is assumed. Investigators will evaluate this outcome using non-parametric statistics, that is, a Spearman correlation between rating and whether the participant received the talking touchscreen (coded 0 or 1). These items will be administered verbally by a research coordinator. Investigators predict that participant's receiving the talking touchscreen will report high levels of comfort, less difficulty, and less nervousness. Investigators will evaluate statistical significance using a 99% confidence interval, rejecting the null hypothesis if this interval does not contain zero. The items will not be aggregated; each item will be analyzed separately.