Evaluating the Effectiveness of QR-Code–Based Microlearning in Enhancing Medical Equipment Use and Maintenance among Nurses in Decentralized Health Facilities in Sri Lanka.

Arm 1: QR Video + Text Microlearning (T1)

QR codes mounted on or near each device link to short, brand-neutral modules containing:
• a video demonstration of safe operation
• step-by-step text instructions
• pre-use safety checks
• preventive care
• first-line troubleshooting
• escalation guidance

Arm 2: QR Voice + Text Microlearning (T2)

Same content structure as T1, but multimedia is voice-instruction + text (no video).
Designed as a lightweight, low-bandwidth alternative.

Control

Usual practice during the trial period.
Hospitals receive full access to all QR modules after primary outcomes are completed.

This study evaluates two formats of QR-code–based microlearning modules designed to improve nurses’ knowledge of medical equipment operation and preventive maintenance, and to enhance system-level equipment readiness in decentralized hospitals. The trial is a three-arm cluster randomized controlled trial implemented across all 88 divisional hospitals in the Central Province of Sri Lanka, with the hospital serving as the unit of randomization and intervention.
1. Microlearning Intervention Overview
Each intervention hospital receives QR codes affixed directly on, or adjacent to, six commonly used medical devices:
Defibrillator
Syringe pump
ECG machine
Bedside patient monitor
Suction apparatus
Tabletop autoclave
Each QR code links to a device-specific microlearning page containing:
• Step-by-step instructions for safe device operation
• Pre-use safety checks
• Post-use preventive maintenance (cleaning, cable checks, battery care, alarm tests)
• First-line troubleshooting for common malfunctions
• Guidance on escalation to the biomedical engineer or charge nurse
In line with the concepts of microlearning and just-in-time clinical guidance, all content is brand-neutral and intended for less than three minutes of engagement per module.
2. Intervention Arms
Arm 1: QR Video + Text Microlearning (T1)
Hospitals in this arm receive QR codes linking to microlearning pages that feature:
• A short device-specific video demonstrating operation and preventive care
• A text summary reinforcing key steps
• Icons and checklists to support rapid bedside recall
These modules require moderate bandwidth but provide a richer multimedia learning environment.
Arm 2: QR Voice + Text Microlearning (T2)
Hospitals receive QR codes linking to microlearning pages with:
• A short audio (voice) narration of key steps
• The same text-based guidance and checklists as T1
This format is designed for low-connectivity environments and may be more scalable than video.
The comparison T1 vs T2 isolates the added value of video content relative to audio + text.
Control Arm
Control hospitals continue usual practice with no QR codes during the trial period.
Following the final outcome assessment, they receive full access to all microlearning modules.
3. Implementation Procedures
After baseline data collection, QR codes are installed in intervention hospitals, and nurses receive a brief orientation on how to scan and access content. No in-person training or demonstrations are provided, preserving the trial’s focus on microlearning as a low-burden, point-of-care tool.
Exposure to QR content is voluntary and unmonitored by staff.
Assessments occur at:
• Baseline (T0): Knowledge test (Form A) and Equipment Functionality Audit
• 3 months (T1): Knowledge test (Form B) and functionality audit
• 6 months (T2): Knowledge test (Form C) and functionality audit
The knowledge instruments use parallel forms to minimize recall bias.
Device audits classify each unit as “Ready” or “Not Ready,” with any critical failure recorded as “Not Ready”.
4. Randomization and Allocation
Hospitals were randomized in a 1:1:1 ratio to T1, T2, or Control using a stratified randomization procedure (stratified by district Kandy, Matale, Nuwara Eliya and hospital type A/B/C). Randomization was conducted in Excel using computer-generated random numbers within strata. Assignment was concealed until completion of baseline data collection.
5. Target Population and Sample
The target population consists of registered nurses working in wards where the six devices are routinely used. Approximately 500 nurses (5–7 per hospital) are enrolled. Inclusion criteria: nurses currently assigned to active clinical wards and expected to remain for the trial duration. Nurses on extended leave or temporary transfer are excluded.
6. Rationale for Intervention Content
The intervention responds to documented challenges in LMIC hospital settings, including limited access to training, device misuse, insufficient preventive maintenance, and prolonged device downtime. Microlearning with short, focused modules has demonstrated improved knowledge retention and engagement in prior studies. QR codes allow instant access to operational and maintenance guidance at the point of care, bridging knowledge gaps that traditional training methods do not address.
The comparison between video-based (T1) and voice-based (T2) microlearning identifies whether high-resource multimedia is necessary or whether a lower-bandwidth format yields similar improvements.
7. Expected Mechanisms of Effect
At the nurse level, the intervention is expected to:
• Improve operational knowledge and performance
• Increase retention across the 6-month period
• Enhance confidence in handling equipment
At the hospital system level, improvements in device readiness are expected through:
• More consistent pre-use checks
• Better preventive maintenance
• Early identification and reporting of failures
These mechanisms are grounded in cognitive load theory, multimedia learning theory, and prior empirical studies on QR-based training.

2025-11-18

2025-11-28

Nurse knowledge score (0–100)

Measured via 15–20 item multiple-choice Knowledge Assessment Questionnaire (KAQ).
Parallel forms A, B, C administered at baseline, 3 months, and 6 months.
Outcome metric: continuous (0–100).

Nurse Knowledge Score (0–100)
The knowledge outcome is constructed from the Knowledge Assessment Questionnaire (KAQ) consisting of 15–20 multiple-choice questions, each scored as:
1 = correct
0 = incorrect
For each nurse at each time point:
Sum all correct responses
Divide by the total possible score
Multiply by 100


Knowledge Score = (Total Correct / Total Questions) × 100
Parallel test forms A, B, and C are used at baseline, 3 months, and 6 months to reduce recall bias, but all forms contain the same number of items and difficulty structure, allowing consistent scoring.
Thus the outcome is a continuous score (0–100).

Equipment Functionality (Ready / Not Ready)

Binary classification using the Equipment Functionality Audit (EFA).
Percentage Ready per hospital will also be analyzed.

Equipment Functionality (Ready / Not Ready)
This outcome is constructed using the Equipment Functionality Audit (EFA).
Each target device (defibrillator, syringe pump, ECG, autoclave, bedside monitor, suction machine) is assessed for:
Power/battery
Cables and accessories
Alarms and self-test
Cleanliness
Labeling
Critical components specific to the device


Each device is classified as:
Ready = 1 (all essential components functioning)
Not Ready = 0 (any critical failure)


At the hospital level, two metrics will be constructed:
Binary outcome per device:
Ready (1) vs Not Ready (0).
Hospital-level percentage Ready:
% Ready = (Number of Ready devices / Total assessed devices) × 100.
The binary variable will be used for the mixed-effects logistic regression (primary), while the percentage score is used for descriptive summaries and robustness checks.

This study is a three-arm cluster randomized controlled trial evaluating whether QR-code–based microlearning improves nurses’ knowledge and medical equipment functionality in decentralized health facilities in Sri Lanka. The trial is conducted in 88 divisional hospitals in the Central Province. Hospitals serve as the unit of randomization, and all nurses working in clinical wards that routinely use the target devices are eligible to participate.

Hospitals are randomly assigned in equal proportions to one of three groups:

Treatment 1: QR codes linking to short video-based microlearning modules with accompanying text

Treatment 2: QR codes linking to audio-based microlearning modules with accompanying text

Control: Usual practice, with no QR codes provided during the study period

Outcomes are measured at baseline, 3 months, and 6 months.
The primary outcome is nurses’ knowledge, assessed using a standardized multiple-choice questionnaire scored on a 0–100 scale.
The secondary outcome is medical equipment functionality, measured using a structured audit that classifies devices as Ready or Not Ready.

The trial is designed to compare each microlearning approach to usual practice, and to compare the two microlearning formats with each other. All analyses will follow an intention-to-treat framework and will account for the clustered design.

This study uses a three-arm cluster randomized controlled trial (cRCT) to evaluate the effectiveness of QR-code–based microlearning for improving nurses’ operational knowledge of medical equipment and the functional readiness of key devices in decentralized health facilities. The trial is implemented across all 88 divisional hospitals in the Central Province of Sri Lanka. The hospital serves as the unit of randomization, intervention delivery, and clustering.
1. Study Arms
Treatment 1 (T1): QR Video + Text Microlearning
Hospitals in this arm receive QR codes affixed on or near six key medical devices (defibrillator, syringe pump, ECG machine, bedside monitor, suction apparatus, autoclave). Each QR code links to a microlearning page containing:
A short instructional video,
Text-based operational steps,
Pre-use safety checks,
Preventive maintenance actions,
First-line troubleshooting instructions,
Escalation guidance to technical staff.


Modules are made to facilitate just-in-time learning and take less than three minutes to finish.
Treatment 2 (T2): QR Voice + Text Microlearning
The same QR placement and topics are used, but instead of a video, nurses access a voice narration accompanying the text-based instructions. This arm tests whether audio-based modules (lower cost and bandwidth) produce comparable outcomes to video-enhanced modules.
Control
Control hospitals continue with usual practice during the trial. No QR codes are provided until the study is completed.
2. Participants and Recruitment
Eligible participants are registered nurses working in clinical areas where the six target devices are used. Approximately 500 nurses (about 5–7 per hospital) are enrolled. Nurses on extended leave or temporary transfer are excluded. After institutional approvals, nurses are briefed and provide written informed consent in Sinhala, Tamil, or English.
3. Randomization
Randomization occurs after baseline data collection to preserve allocation concealment. Hospitals are randomized in a 1:1:1 ratio using computer-generated random numbers in Excel, stratified by:
District (Kandy, Matale, Nuwara Eliya)
Hospital Type (A, B, C)


This ensures balanced distribution of structural characteristics and service profiles across trial arms.
4. Intervention Delivery
After randomization, QR codes are physically installed on each target device by the research team. Nurses receive a short orientation on how to scan and access materials but no additional training is provided to preserve focus on the microlearning intervention itself.
5. Outcome Measurement
Outcomes are measured at:
Baseline
3 months
6 months
Primary Outcome: Nurse Knowledge (0–100 score)
Constructed from a 15–20-item multiple-choice Knowledge Assessment Questionnaire (KAQ).
Parallel forms (A, B, C) are administered at baseline, 3 months, 6 months to reduce recall bias.
Each correct answer = 1; incorrect = 0.
Scores are summed and converted to percentages.
Secondary Outcome: Equipment Functionality (Ready / Not Ready)
Assessed using the Equipment Functionality Audit (EFA) for each target device.
A device is classified “Ready” if all essential functions pass; any critical failure yields “Not Ready.”
Hospital-level metrics include the proportion of devices Ready.
6. Analytical Approach
The primary analysis follows intention-to-treat (ITT) principles. Mixed-effects models account for:
Clustering at the hospital level,
Repeated measurements at the nurse level,
Stratification factors (district and hospital type).


For knowledge (continuous): linear mixed-effects regression.
For functionality (binary): mixed-effects logistic regression with complementary LPM robustness checks.
Primary contrasts:
T1 vs Control
T2 vs Control
T1 vs T2 (incremental value of video vs audio)
ICC and effective sample size will be reported.
7. Mechanisms and Implementation Logic
The intervention aims to:
Improve nurses’ access to operational guidance at the point of care,
Enhance retention through microlearning principles (short, structured modules),
Reduce misuse and preventable downtime through better safety checks and preventive maintenance,
Strengthen system-level readiness of equipment.


The comparison of T1 and T2 isolates whether richer multimedia (video) has added value beyond audio-supported microlearning.
8. Risks
Risks to participants are minimal and limited to potential time spent reviewing QR materials. Device audits are observational and nonintrusive. Control hospitals receive the intervention package after follow-up assessments.
9. Timeline
Baseline: November 2025
Randomization + QR installation: Late November 2025
Follow-up 1 (3 months): February 2026
Follow-up 2 (6 months): May 2026
Data lock and analysis: June 2026

Randomization was performed in the office using computer-generated random numbers. Hospitals were randomized in a 1:1:1 allocation to the three arms, with stratification by district and hospital type to ensure balanced distribution across study groups.

The hospital is the unit of randomization. All 88 divisional hospitals are randomized as clusters into one of the three study arms. Individual nurses are not randomized.

Yes

88 divisional hospitals

Approximately 500 nurses

28 divisional hospitals in Treatment 1 (QR video + text)
28 divisional hospitals in Treatment 2 (QR voice + text)
32 divisional hospitals in Control (usual practice)

Power calculation: Minimum Detectable Effect Size for Main Outcomes For the primary outcome (nurse knowledge, 0–100 scale), with 88 hospitals (average cluster size ≈ 6; total N ≈ 522 nurses, ≈174 per arm) and an assumed intracluster correlation ICC in the range 0.03–0.05, the design effect is approximately 1.15–1.25. Under these assumptions, the trial has 80% power at α = 0.05 (two-sided) to detect a standardized mean difference of about d = 0.35–0.40. On the 0–100 knowledge scale, assuming a standard deviation ≈ 20 points, this corresponds to a minimum detectable effect of roughly 7–8 points, i.e. a 7–8 percentage-point increase in knowledge scores between treatment and control arms. For the secondary main outcome (device functionality: Ready vs Not Ready), with the same cluster structure, the study is powered to detect an improvement of roughly 10–15 percentage points in the proportion of devices classified as Ready at 6 months, depending on the true ICC for device readiness.