Malaysian inpatient safety engagement tracks with better patient-reported outcomes

A multicenter survey in Malaysian public tertiary hospitals found that inpatients who were more engaged in safety-related behaviors also reported better safety outcomes. The association was positive but modest, and it centered on medication-related participation rather than a broad, abstract sense of empowerment. In a hospital setting where patients often defer to staff authority, even a measured increase in self-initiated safety behavior may carry practical weight.

The study enrolled 986 adult inpatients from medical, surgical, orthopaedic, and obstetrics-gynaecology wards across four public tertiary hospitals in Peninsular Malaysia. Investigators used validated Malay-language instruments to measure general self-efficacy, medication-safety involvement, and patient-reported safety outcomes. The population was not small, and the setting was not hypothetical; it was drawn from real inpatient wards where medication handling, handoffs, and escalation of concerns all depend on timely patient participation.

One result stood out immediately. 61.1% of participants were classified as having high self-efficacy for safety participation, while only 1.9% fell into the low category.

Self-efficacy was common, but not evenly distributed

The mean self-efficacy score was 3.27 on a four-point scale, suggesting that most patients believed they could participate meaningfully in actions meant to protect their own safety. The distribution matters because self-efficacy is not simply a personality trait; it shapes whether a patient feels able to question a medication name, ask about a dose, or speak up when something seems inconsistent. In busy wards, confidence can determine whether a safety behavior occurs at all.

The investigators framed self-efficacy as a contextual descriptor and not a causal claim, which fits the cross-sectional design. Still, the pattern is clinically intuitive. Patients who trust their own ability to act are more likely to engage in visible safety behaviors, and those behaviors may improve the quality of information shared with clinicians, reduce missed discrepancies, or prompt earlier correction of errors. In practice, this can matter most at points where the hospital process becomes vulnerable to miscommunication.

The cohort came from four public tertiary hospitals selected to represent the Northern, East Coast, Central, and Southern regions of Peninsular Malaysia. Hospitals in East Malaysia were not included because of logistical constraints, and the authors explicitly confined generalizability to the public tertiary setting. Even so, the sample spanned major inpatient disciplines and offered a useful window into how safety participation functions in a diverse hospital population.

Medication-focused participation correlated with safer perceptions

Greater patient safety engagement was positively associated with more favorable patient-reported safety outcomes, with Pearson correlation r = 0.38 and a 95% confidence interval of 0.32 to 0.43. Robust linear regression showed a similar direction and magnitude: patient safety engagement significantly predicted patient-reported safety outcomes, with B = 0.315, β = 0.38, and p < 0.001. The model explained 14% of the variance in the outcome measure.

That last figure is the most important constraint on interpretation. The signal is real, yet it is not large enough to suggest that patient engagement alone accounts for the complexity of inpatient safety. Hospital safety remains shaped by staffing, communication, workflow design, medication systems, ward culture, and the patient’s own clinical condition. Self-efficacy and engagement are part of the mechanism, not the entire machinery.

Even so, a 14% explanation of variance is meaningful in a behavioral and patient-reported outcome domain. Safety is rarely governed by one dominant variable. Small-to-moderate effects can be operationally relevant when they influence whether patients ask questions, confirm medication details, or feel able to challenge ambiguity. Those actions do not replace staff vigilance; they complement it.

Why confidence in safety participation may matter at the bedside

The biological logic here is less about physiology than about system reliability. Medication harm often begins with a small communication failure, then compounds through silence, uncertainty, or deference. A patient with higher self-efficacy is more likely to verify a drug list, flag a mismatch, or request clarification before an error reaches administration. In that sense, self-efficacy functions as a psychological gateway to a practical safety defense.

This is especially relevant in low- and middle-income settings, where hierarchical relationships and information asymmetry can limit patient voice. The study’s introduction makes clear that such environments may reduce the likelihood that patients will see themselves as active partners in safety. Malaysian public tertiary hospitals, by virtue of their size and complexity, offer a difficult but important test of whether engagement can work outside high-resource, highly participatory systems.

The results suggest that it can, though unevenly. Most patients already reported moderate-to-high confidence, leaving a small low-self-efficacy subgroup that may need tailored support. Those patients may not simply need information; they may need permission, coaching, and repeated reinforcement to speak up. A generic safety leaflet is unlikely to change behavior if a patient still believes questioning staff is inappropriate.

A modest effect, a clear operational target

The study’s strength lies in its combination of scale, multicenter sampling, and use of validated Malay-language instruments. It also examined a specific behavioral pathway rather than treating patient engagement as a vague ideal. By focusing on medication-related safety participation, the analysis connects directly to one of the most common and preventable sources of inpatient harm.

The design also limits what can be claimed. Cross-sectional data cannot determine whether higher engagement leads to better safety outcomes, or whether patients who already perceive care more favorably are simply more willing to report confidence and participation. Self-reported outcomes can also be influenced by recall, social desirability, and ward-specific expectations. The study does not resolve those issues, and it does not need to in order to be useful.

What it does establish is a measurable association in a large inpatient sample from Malaysian public hospitals. That alone gives hospitals a concrete starting point: identify patients who are reluctant to participate, build ward routines that normalize questions about medication, and design safety communication around behavior rather than information alone. The evidence base for patient engagement in safety remains limited in LMIC settings, and this study adds a large, locally grounded data point.

The remaining question is not whether patients can matter in safety work; it is how much hospitals are prepared to structure care around that participation, especially for the small group of patients whose confidence is lowest and whose voices are least likely to be heard.


Reporting by Dr. A.R. Athira, MBBS for Medicine World Foundation, with additional data provided by PLOS Medicine.