Question: A health system is collecting data regarding comorbidities among the population within the catchment area. The analytic team is collecting self-reported data on chronic diseases using a population-based phone and mail survey. The team wants to validate whether their self-reported information is consistent with clinical data. Among 500 patients surveyed, 88 self-reported they had diabetes, 127 high blood pressure, 136 high cholesterol, and 219 chronic pain. When compared to medical records, the team found 84 had diagnosed diabetes, 132 high blood pressure, 129 high cholesterol and 43 chronic pain. The team began to investigate the discordance for chronic pain, and found that 39 people had both self-reported and diagnosed chronic pain based on clinical signs and symptoms.

1. What is the sensitivity and specificity of survey self-report as a means of detecting chronic pain?
2. What is the positive predictive value of the survey? What does this mean?
3. In this case, why is the positive predictive value low or high?
4. Speculate why chronic pain may be more difficult to collect via survey than other chronic conditions. How might the team improve data collection?

Price: $2.99

Solution: The downloadable solution consists of 2 pages
Deliverable: Word Document