Adopting Logic Model to Predict Ovarian Cancer

The Logic model was primarily used in educational programs and then to evaluate tuberculosis control, cervical cancer prevention programs, and cardiovascular disease in health. Unlike cervical cancer, there is a gap in screening for ovarian cancer. However, clinical services exist. Thus, the Logic model has been used to evaluate the service standards for the secondary prevention of ovarian cancer.

This is the multi-centric service evaluation research adopted from the Logic Model. There are four domains namely utility, feasibility, propriety, and accuracy standards in the Logic model that includes 53 question items altogether for each participant. For each item, the participants responded on a Likert scale to assess their satisfaction with the service provided to the patients. There are 5-point satisfaction levels from strongly disagree to agree strongly. The internal consistency of items was calculated and the factor analysis was performed. Software used were Microsoft Excel, SPSS, SPSS Amos, and R.

The agreement level of all specialist participants was satisfactory for the current prediction and management approach to ovarian cancer with a median value of 73.5% towards positive sentiment. Cronbach's alfa was at an acceptable level of more than 0.8 for utility, feasibility, and accuracy domains. The propriety domain had poor yield. Chi-squared test-based model fit is good (Baseline and Factor Models <0.001) and Barlott's test of sphericity is likely to work (X2=5460.242, df=1378, and p<0.001). Other confirmatory factors were not at an acceptable level.

The logic model may work to predict ovarian cancer with an acceptable level of reliability, however for the perfect fit it requires a larger sample size.