Show simple item record

dc.contributor.authorMund, Jaime
dc.date.accessioned2016-04-05T19:34:08Z
dc.date.available2016-04-05T19:34:08Z
dc.date.issued2014
dc.identifier.isbn9781321264012
dc.identifier.other1625768548
dc.identifier.urihttp://hdl.handle.net/10477/51057
dc.description.abstractIntroduction. Sleep disturbances are prevalent and often go undiagnosed. The Centers for Disease Control and Prevention (CDC) currently uses 5 screening questions in their Behavior Risk Factor Surveillance System (BRFSS) to screen for sleep disturbances. In order to validate these 5 questions we compared question responses with the gold standard measures for the construct each question was hypothesized to measure. It is important to validate these questions being used in the largest national survey on population health to verify that the data being collected is accurate and is measuring what it intends to measure. Methods. A sample of 299 subjects participated in the CDC-funded validation study from the Buffalo and Rochester areas. The study consisted of data collection at multiple time points. At screening, eligibility was determined and initial responses to the 5 BRFSS questions were collected. At baseline, demographics and sleep-related information was collected using the Insomnia Severity Index (ISI) total score, PROMIS- 57 profile, Epworth Sleepiness Scale (ESS), and the Sleep Disorders Screening Questionnaire (SDQ). Descriptive statistics were estimated to define the sample and t-tests were used to analyze group differences based on sex and age. Results. The mean age of the sample was approximately 40 years (SD=17.2) and the sample was 34% male and 79.9% white. A majority of the sample had at least some college education (48.2% with 2-4 years and 32.8% with greater than 4 years). The analysis of BRFSS question 1, which asked about number of days an individual did not get enough rest or sleep, showed a correlation of 0.57 (p<0.0001) with the Insomnia Severity Index total score. BRFSS question 2, which asked about average total sleep time, resulted in a Pearson correlation of 0.16 (p=0.01) when correlated with total sleep time measured via actigraphy. Analysis of BRFSS question 3, which measures reported snoring prevalence, resulted in a sensitivity of 66%, a specificity of 66%, a positive predictive value of 42%, and a negative predictive value of 84%. BRFSS question 4, which measures number of days an individual unintentionally fell asleep, analysis resulted in a correlation of 0.54 (p<0.0001) with the Epworth Sleepiness scale total score. BRFSS question 5, which assessed falling asleep while driving, analysis resulted in a sensitivity of 25%, a specificity of 96%, a positive predictive value of 69%, and a negative predictive value of 77% when compared with excessive daytime sleepiness diagnosis defined as a total Epworth Sleepiness Scale score of 10 or greater. Conclusions. Responses to the BRFSS questions 1, 2, and 4 all had statistically significant correlations with the gold standard clinical criteria measures, which indicated validity in assessing the hypothesized construct. However, BRFSS question 2 had a weak correlation with the objective measure of total sleep time. For BRFSS question 3, the 66% sensitivity and specificity reveal the question to only have moderate validity. Question 5 had low sensitivity (25%) and high specificity (96%). Therefore, the question was able to identify those that truly did not have EDS well but was not sensitive enough to correctly identify a majority of the cases. Evidence of validity and reliability of these questions support the utility of some of these questions to be used to survey the prevalence of sleep disturbance in the population but it is apparent that some revisions to the questions also need to be made. Accurate knowledge on the prevalence of sleep-related disorders can lead to an improved focus on prevention, screening, assessment, and treatment as deemed necessary. Therefore, having valid and reliable questions to ask on a national survey such as the BRFSS is important to overall public health. (Abstract shortened by UMI.)
dc.languageEnglish
dc.sourceDissertations & Theses @ SUNY Buffalo,ProQuest Dissertations & Theses Global
dc.subjectBiological sciences
dc.subjectHealth and environmental sciences
dc.subjectBehavior Risk Factor Surveillance System
dc.subjectSleep disorders
dc.subjectValidation
dc.titleValidation of CDC BRFSS sleep questions to screen for sleep disturbances
dc.typeDissertation/Thesis


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record