The development and psychometric properties of the bipolar disorders knowledge scale.

PURPOSE

Bipolar Disorder (BD) presents in 1-4% of the world's population, carrying significant financial and functional consequences. Beyond the direct disease burden, patients with BD experience a high degree of both internal and external stigma. Additionally, medication adherence tends to be poor in patients with BD. Knowledge appears to play a role in mitigating both stigma and non-adherence, but these relationships have not been fully elucidated. The Bipolar Disorder Knowledge Scale (BDKS) was designed to explore the role of knowledge and better define such relationships. This research provides the evidence for the reliability and validity of the scale.

METHODS

Forty-seven items were developed to assess knowledge of BD. The 47-item survey was sent out to two groups: first a group of 43 pharmacists with BCPP credentials from the College of Psychiatric and Neurologic Pharmacists (CPNP) who were recruited from the CPNP directory, and second a group of 250 members of the general public who were recruited using Qualtrics Online Sample service. Participants were surveyed on their education status, health literacy, BD diagnostic status, and exposure to patients with BD. Participants then completed the 47-item scale. After 48 h 100 members from the original general public group were sent the same survey to assess test-retest reliability. For each item a difficulty index to evaluate how well participants performed on the item and a discrimination index to determine how well each item performed in high-scorers versus low scorers were calculated. Additionally, Cronbach's alpha was calculated to determine internal consistency validity and a Pearson correlation was run to determine test-retest reliability. Items were removed based on the results from the difficulty index, discrimination index, and Cronbach's alpha. Finally the pharmacist final scores were compared to the general public using an unpaired t-test to assess whether content experts were more likely to perform better on the scale. Following item removal, the scale was finalized at 25-items.

RESULTS

The mean score for the scale was 34.48 (71.83%; SD: 5.50) and the Cronbach's alpha was 0.773 before item analysis. Following item analysis, 22 items were dropped leaving 25 items on the final version of the scale. The remaining items retained a difficulty index below 90% and a discrimination index above 20%. The mean of the 25-item scale was 18.40 (73.6%; SD: 4.13) for the general public and 23.20 (92.8%; SD: 1.36) for the pharmacists group (p < 0.001). The Cronbach's alpha for the finalized scale was 0.760, indicating a high-degree of internal consistency. While this is lower than the original alpha, this may be explained by the reduced number of scale items. A 25-item scale is much more practical and the items on the scale retain stronger item analysis statistics. Finally, the Pearson Correlation for the group who underwent the test-retest procedure was 0.841 (p < 0.001) indicating strong test-retest reliability.

CONCLUSION

The BDKS is a 25-item true-false scale that takes approximately 5-10 min to complete. The scale assesses knowledge of BD with items targeting diagnosis, etiology, disease course, symptoms, treatment, and life impact. The scale has shown strong internal consistency and test-retest reliability in a general population and will be useful for evaluating knowledge of BD as it relates to stigma, non-adherence, and other variables.