Communication, Continuity of Care, Electronic Health Records, Hospital Discharge Communication Processes, Patient Safety, Quality Indicators, Health Care
Conditions
Keywords
Discharge Summary, Patient Handover, Quality Improvement, Cluster Randomized Trial, Healthcare Outcomes, Patient Satisfaction, Quality Indicators, Real-Time Feedback, Electronic Health Records (EHR)
Brief summary
This study, titled Automated Indicator Feedback for Improving the Quality of Discharge Letters: A Cluster-Randomized Controlled Trial (FIAQ-LS), aims to evaluate whether continuous real-time feedback to hospital teams can improve the quality of discharge letters. Discharge letters are critical for ensuring continuity of care and reducing adverse events by providing detailed information about a patient's hospital stay to both the patient and their primary care physician. The study will be conducted at Grenoble Alpes University Hospital and involve 40 hospital services across three campuses. The trial design includes two parallel arms: an intervention group receiving monthly performance feedback through automated dashboards and a control group with no additional intervention. Services are randomized into these groups using a stratified cluster approach. The primary objective is to assess whether this intervention increases the proportion of discharge letters validated on the day of discharge compared to usual care. Secondary objectives include evaluating patient satisfaction, rates of unplanned 30-day readmissions, and completeness of discharge letter content. The study will include data from approximately 132,000 patient stays over two phases: a pre-implementation observational period (12 months) and an intervention phase (12 months). All data will be collected and analyzed anonymously, with findings expected to inform the broader implementation of quality improvement strategies in French hospitals.
Detailed description
Detailed Description Effective communication at hospital discharge is vital for continuity of care and patient safety. Discharge letters summarize the hospital stay, outlining diagnoses, treatments, and follow-up care. Despite national guidelines mandating that discharge letters be validated and provided to patients on the day of discharge, compliance remains suboptimal in France, with average performance scores well below targets. This study seeks to address this gap through an automated feedback mechanism. Using the hospital's electronic health record (EHR) system, the study will generate monthly dashboards for each participating service in the intervention group. These dashboards will provide a real-time view of performance metrics, including the proportion of discharge letters validated on the day of discharge and the completeness of required content fields. The trial employs a cluster-randomized controlled design with 40 hospital services as the unit of randomization. Services are stratified by activity type (medicine, surgery/obstetrics) and baseline performance. The study is divided into two phases: Pre-implementation Phase (January 2024 - January 2025): A 12-month observational period to collect baseline data and stratify services for randomization. Implementation Phase (February 2025 - February 2026): Intervention services receive monthly performance feedback, while control services continue with standard care practices. The primary endpoint is the proportion of hospital stays where discharge letters are validated on the day of discharge. Secondary outcomes include: Patient satisfaction, measured through the national e-Satis survey. Rates of unplanned readmissions within 30 days of discharge. Completeness of discharge letters, evaluated across mandated content fields (e.g., patient identification, discharge summary, follow-up plan). This study will enroll all eligible patient stays within the 40 participating services, excluding stays of less than 24 hours or cases where the patient died during hospitalization. The anticipated sample size is 132,000 stays. Data collection will rely on routine administrative data from the EHR system, anonymized at the patient level. Statistical analyses will adopt a difference-in-differences approach, comparing changes in outcomes between the intervention and control groups over time. A mixed-effects logistic regression model will account for intra-cluster correlations. The results of this study aim to demonstrate the effectiveness of automated feedback in driving quality improvements in hospital discharge processes. If successful, the approach could be scaled across other hospitals in France, contributing to better continuity of care and patient outcomes.
Interventions
OTHERMonthly Performance Feedback with Dashboards (Automated Audit and Feedback)
Hospital services in the intervention group will receive monthly automated dashboards that provide detailed performance metrics. These include: The proportion of patients with a discharge letter generated on the day of discharge, The proportion of discharge letters validated on the day of discharge, Median delays for generating discharge letters, Median delays for validating discharge letters. The dashboards are shared with all physicians, nurse managers, and secretarial staff in each service. A designated quality improvement officer is available to assist teams in interpreting the data and implementing organizational changes based on the feedback. The intervention uses real-time data extraction from the hospital's electronic health record system to generate these insights.
Sponsors
University Hospital, Grenoble
Study design
Allocation
RANDOMIZED
Intervention model
PARALLEL
Primary purpose
HEALTH_SERVICES_RESEARCH
Masking
SINGLE (Outcomes Assessor)
Masking description
The statistician in charge of the data analyses will be blinded to the allocation of hospital services to the intervention or control group to prevent bias in the statistical evaluation of outcomes. This includes the primary outcome (proportion of discharge letters validated on the day of discharge) and secondary outcomes. Outcomes assessors and the statistician will work with anonymized datasets without group allocation information. However, participants (hospital services), care providers, and investigators managing the intervention are not masked due to the need to deliver feedback in real-time and monitor its implementation.
Intervention model description
This study is a cluster-randomized controlled trial with two parallel arms. The unit of randomization is the hospital service, with 40 services participating across three campuses of Grenoble Alpes University Hospital. Services are stratified by activity type (medicine vs. surgery/obstetrics) and baseline performance on the primary outcome. The intervention arm will receive monthly feedback via automated performance dashboards, highlighting the proportion of discharge letters validated on the day of discharge and the completeness of required content fields. The control arm will not receive any feedback but will continue routine care practices. Outcomes will be assessed over two phases: a 12-month pre-implementation observational phase and a 12-month intervention phase. The analysis uses a difference-in-differences approach to compare changes in outcomes between the two groups, accounting for intra-cluster correlations using mixed-effects logistic regression models.
Eligibility
Sex/Gender
ALL
Healthy volunteers
No
Inclusion criteria
* Patients hospitalized for at least 24 hours in participating services. * Patients discharged alive directly from participating services.
Exclusion criteria
* Patients hospitalized for less than 24 hours. * Patients who died during hospitalization. * Stays in services not meeting inclusion criteria (e.g., psychiatry, long-term care, emergency services with rare direct discharges, or critical care units).
Design outcomes
Primary
| Measure | Time frame | Description |
|---|---|---|
| Proportion of Discharge Letters Generated on the Day of Discharge | Measured monthly over the study period (January 2024 to February 2026), comparing a 12-month pre-implementation period to a 12-month intervention period. | The proportion of hospital stays where discharge letters are generated electronically on the same day as the patient's discharge. This measure evaluates the timeliness of generating discharge communication, a critical factor for continuity of care and patient engagement. Data will be extracted from the hospital's electronic health record system (EHR) and aggregated at the service level for analysis. |
Secondary
| Measure | Time frame | Description |
|---|---|---|
| Proportion of Discharge Letters Validated on the Day of Discharge | Measured monthly over the study period (January 2024 to February 2026), comparing a 12-month pre-implementation period to a 12-month intervention period. | The proportion of hospital stays where discharge letters are validated electronically on the day of discharge. This measure assesses the quality and timeliness of the validation process, ensuring that discharge letters are ready for patient handover and communication with primary care providers. Data will be extracted from the hospital's electronic health record system and analyzed at the service level. |
| Median Time to Generate Discharge Letters | Measured monthly over the study period (January 2024 to February 2026), comparing a 12-month pre-implementation period to a 12-month intervention period. | The median time (in hours) from the patient's discharge to the generation of the discharge letter. This measure evaluates process efficiency and timeliness, critical for improving discharge workflows and patient communication. Data will be extracted from the hospital's electronic health record system. |
| Median Time to Validate Discharge Letters | Measured monthly over the study period (January 2024 to February 2026), comparing a 12-month pre-implementation period to a 12-month intervention period. | The median time (in hours) from the generation of a discharge letter to its validation. This outcome assesses the efficiency of the validation process, a key step in finalizing discharge communication for patients and primary care providers. |
| Patient Satisfaction with Discharge Process (e-Satis Survey) | Collected monthly during the 12-month intervention period (February 2025 to February 2026). | Scores from the national e-Satis survey evaluating patient satisfaction with their hospital discharge process. Aggregate scores and specific sub-scores for organization of discharge and interaction with physicians will be compared between intervention and control groups. |
| Rate of Unplanned 30-Day Readmissions | Measured monthly over the study period (January 2024 to February 2026), comparing a 12-month pre-implementation period to a 12-month intervention period. | The proportion of hospital stays followed by unplanned readmissions within 30 days of discharge, measured via emergency admissions. This measure evaluates the impact of improved discharge communication on post-hospitalization outcomes. |
| Time from Patient Discharge to Electronic Submission of Discharge Letter to Primary Care Physicians | Measured monthly during the study period (January 2024 to February 2026), comparing the 12-month pre-implementation period to the 12-month intervention period. | The median time (in days) from the patient's discharge to the electronic transmission of the discharge letter to the external primary care physician (e.g., general practitioner). This outcome assesses the timeliness of communication between hospital services and external care providers, a key factor in ensuring continuity of care after hospitalization. |
Other
| Measure | Time frame | Description |
|---|---|---|
| Completeness of Required Fields in Discharge Letters | Measured monthly during the 12-month intervention period (February 2025 to February 2026), comparing intervention and control groups. | The proportion of discharge letters that include all mandatory fields as per national guidelines. These fields include patient identification, hospital stay dates, discharge summary, follow-up plan, and other regulatory elements. Completeness is critical to ensuring comprehensive communication for continuity of care. |
| Documentation of Organizational Changes in Intervention Services | Ongoing during the 12-month intervention period (February 2025 to February 2026). | Tracking and analysis of organizational adjustments reported by services in the intervention group in response to performance feedback. These changes will be documented through service logs and interviews, providing qualitative insights into the mechanisms driving performance improvement. |
Countries
France
Contacts
Primary ContactBastien Boussat, MD PhD
Outcome results
None listed