Lesson 3. Study Methods and Outcomes

This lesson explains how the study was designed and how its outcomes were measured, using evidence from Qualitativearticle_82373639.

1. Study design: observational preimplementation versus postimplementation

The study used an observational design, not a randomized trial. The article describes it as a single-center observational design, which means the researchers observed what happened in one health system before and after a practice change rather than randomly assigning patients to intervention and control groups (Qualitativearticle_82373639).

More specifically, the study compared two time periods:

• Preimplementation period: patient outcomes before the hospitalwide pharmacy practice model change.
• Postimplementation period: patient outcomes after the new pharmacy team model was put into place.

This kind of before-and-after comparison helps show whether outcomes changed after implementation. However, because it is observational, it cannot prove causation as strongly as a randomized controlled trial. The article notes that this design introduced possible bias and confounding factors, and that several outside influences could also have affected healthcare utilization, such as patients’ ability to obtain medications after discharge, readiness for discharge, and readmissions unrelated to the original admission (Qualitativearticle_82373639).

2. How patients were grouped

The study also separated patients into high-risk and non-high-risk groups. High-risk patients included those with conditions or discharge situations associated with greater risk, such as acute myocardial infarction, congestive heart failure, pneumonia, chronic obstructive pulmonary disease, or discharge on oral anticoagulation therapy (Qualitativearticle_82373639).

This risk grouping mattered because the researchers analyzed whether the pharmacy practice change had different effects in these subgroups.

3. Primary outcome versus secondary outcome

The study had both a primary outcome and a secondary outcome.

Primary outcome

The primary outcome focused on post-discharge healthcare utilization within 30 days. The article indicates that the main measure was a composite outcome involving:

• 30-day readmissions, and
• 30-day returns to the emergency department (ED).

A composite outcome combines more than one event into a single measure. In this study, that means a patient met the primary outcome if they had either a readmission, an ED return, or both within 30 days after discharge (Qualitativearticle_82373639).

Secondary outcome

The secondary outcome was health system cost savings associated with differences in readmission and ED return rates before and after implementation. So, after examining whether utilization changed, the researchers also estimated the financial impact of those changes (Qualitativearticle_82373639).

4. How readmissions and ED returns were analyzed

The researchers examined outcomes at 30 days after discharge. Their analysis included:

• the composite outcome of readmission and/or ED return,
• readmissions alone, and
• ED returns alone.

They also broke the results down by patient risk category and by discharge medical service to compare healthcare utilization before and after the program was implemented (Qualitativearticle_82373639).

The results show why it is important to distinguish the composite outcome from its separate parts:

• Among high-risk patients, the composite outcome decreased from 23.1% to 17.3%, but this change was not statistically significant (p = 0.061).
• The 30-day ED return rate among high-risk patients decreased from 9.5% to 6.1%, which was also not statistically significant (p = 0.098).
• The 30-day readmission rate among high-risk patients decreased from 17.8% to 12.3%, and this change was statistically significant (p = 0.042).
• Among non-high-risk patients, no significant changes were observed in the composite outcome, readmissions, or ED returns (Qualitativearticle_82373639).

This means the intervention appeared to have its clearest effect on readmissions in high-risk patients, even though the broader composite measure did not reach statistical significance overall.

5. How cost savings were analyzed

For the secondary outcome, the researchers estimated cost savings based on changes in readmissions and ED visits. According to Qualitativearticle_82373639, they used average direct costs and total costs for patient readmissions and ED encounters at the University of Kansas Hospital during the study period.

These cost data came from the University HealthSystem Consortium Clinical Data Base/Resource Manager. The article explains the difference between the two cost approaches:

• Direct cost: estimated costs for items such as medications, laboratory testing, imaging, and other resource charges.
• Total cost: a broader estimate that also attempts to include overhead expenses such as facility costs and labor.

In other words, the investigators first identified differences in utilization, then applied average cost figures to estimate how much money the health system may have saved when fewer readmissions or ED returns occurred (Qualitativearticle_82373639).

6. Important limitations for interpreting the outcomes

When understanding the findings, it is important to remember the study’s limitations. The article states that:

• the study was conducted at a single center,
• the design was observational,
• possible readmissions outside the health system were not captured, and
• the study did not account for multiple readmissions or multiple ED returns after the initial discharge (Qualitativearticle_82373639).

These limitations matter because they affect how confidently we can interpret the results and how broadly we can apply them to other settings.

7. Key takeaways

• The study used an observational before-and-after design, comparing preimplementation and postimplementation periods.
• The primary outcome was 30-day healthcare utilization, especially a composite of readmissions and ED returns.
• The secondary outcome was cost savings related to changes in those utilization rates.
• Readmissions and ED returns were analyzed both together as a composite and separately as individual outcomes.
• The most notable result was a significant reduction in 30-day readmissions among high-risk patients, while the composite outcome and ED returns did not show statistically significant improvement.
• Cost savings were estimated using average direct and total costs of readmissions and ED encounters from the health system database.