Understanding Variables, Statistics, and Sample Considerations in a Psychiatric Continuity of Care QI Project

Understanding Variables, Statistics, and Sample Considerations in a Psychiatric Continuity of Care QI Project

I chose an article for the quality improvement (QI) project by Ojo et al. (2024) explored how to improve follow-up care for psychiatric patients after they are discharged from the hospital. The study was conducted in a family medicine outpatient clinic and focused on ensuring that patients received timely outpatient mental health follow-up, a critical step in preventing relapses and readmissions. The intervention included scheduling outpatient appointments before patients were discharged and improving care coordination to support better adherence to treatment.

Continuous and Categorical Variables

In this project, the researchers worked with both continuous and categorical demographic variables. One continuous variable was patient age, measured in years. This variable was useful in calculating averages and understanding the age range of the participants. For example, if the average age of participants was around 42, this helps give context to the population studied (Salkind & Frey, 2019).

A key categorical variable was follow-up attendance, recorded as either "yes" or "no." This simple variable allowed the researchers to measure the success of the intervention by comparing how many patients returned for follow-up appointments before and after the new process was introduced (Salkind & Frey, 2019).

Descriptive versus Inferential Statistics

The study used both descriptive and inferential statistics. Descriptive statistics were used to summarize and describe key data points, such as the percentage of patients who followed up with care. For instance, the study found that follow-up rates increased from 55% before the intervention to 78% afterward.

To go beyond basic observation, the study also used inferential statistics—specifically, a chi-square test—to determine if the difference in follow-up rates was statistically significant. This type of test helps answer whether the improvement was likely due to the intervention itself rather than chance (BMJ, 2019).

Sample Sizes and Type I vs. Type II Errors

The sample size in this QI project was about 80 patients. While this is a reasonable size for a practice-based improvement effort, it still poses a risk for Type II error, meaning that the study might not detect a true effect if the sample isn’t large enough. However, in this case, the sample was sufficient to show a significant improvement in follow-up rates.

In comparison, a traditional research study might include hundreds of participants to reduce both Type I error (detecting an effect when there is none) and Type II error. DNP projects, like QI efforts, often involve smaller samples (30–100 participants) and are more focused on practice improvement than statistical generalizability. Because of their size and scope, DNP projects must be especially careful about study design to avoid statistical errors (Dang et al., 2021).

Understanding the SIR Rate

Although not specifically mentioned in the article, it’s important to understand the Standardized Infection Ratio (SIR) as part of quality measurement. The SIR compares the number of observed infections in a healthcare setting to the number expected based on national data. The formula is: SIR= Predicted Infections divided by  Observed Infections  (CDC, 2022)

An SIR above 1.0 indicates more infections than expected, while an SIR below 1.0 means fewer infections occurred. Healthcare facilities use the SIR to evaluate their performance, improve infection control, and meet regulatory standards. In mental heal

Order Now