Crystal L Chapman Lambert1, Andres Azuero2, Comfort C Enah2, Susan C McMillan3. 1. School of Nursing, University of Alabama at Birmingham, Birmingham, AL 35487, USA. Electronic address: cclamber@uab.edu. 2. School of Nursing, University of Alabama at Birmingham, Birmingham, AL 35487, USA. 3. College of Nursing, University of South Florida, Tampa, FL 33620, USA.
Abstract
PURPOSE: The purpose of this paper was to examine the psychometric properties of Champion's Health Belief Model Scales for cervical cancer and screening among women living with HIV. METHODS: A secondary data analysis was conducted using data from an exploratory cross-sectional study with a convenience sample of 300 women living with HIV receiving care at two HIV ambulatory care clinics in Florida. A 39-item adaptation of the Champion's Health Belief Model Scales was administered via paper and pencil. RESULTS: The authors used internal consistency measures, confirmatory factor analysis, and ordinal item response theory (IRT) techniques to examine the psychometric properties of the instrument. The 39-item instrument had adequate internal consistency and factor structure. However, the IRT analyses suggested that the instrument could be reduced to 24-items (61.5%), without loss of relevant information. CONCLUSION: A shortened 24-item instrument demonstrated good internal consistency among women living with HIV. Future work should include validating the properties of the reduced instrument in diverse samples of patients and conditions.
PURPOSE: The purpose of this paper was to examine the psychometric properties of Champion's Health Belief Model Scales for cervical cancer and screening among women living with HIV. METHODS: A secondary data analysis was conducted using data from an exploratory cross-sectional study with a convenience sample of 300 women living with HIV receiving care at two HIV ambulatory care clinics in Florida. A 39-item adaptation of the Champion's Health Belief Model Scales was administered via paper and pencil. RESULTS: The authors used internal consistency measures, confirmatory factor analysis, and ordinal item response theory (IRT) techniques to examine the psychometric properties of the instrument. The 39-item instrument had adequate internal consistency and factor structure. However, the IRT analyses suggested that the instrument could be reduced to 24-items (61.5%), without loss of relevant information. CONCLUSION: A shortened 24-item instrument demonstrated good internal consistency among women living with HIV. Future work should include validating the properties of the reduced instrument in diverse samples of patients and conditions.