Despite decades of progress in global child health, neonatal mortality remains high, accounting for nearly half of all under-five deaths worldwide.1 Most of these deaths occur in low- and middle-income countries and are preventable with timely, high-quality care for small and sick newborns.2 The WHO has called for every newborn to receive essential, high-impact interventions,3 yet the challenge lies not only in knowing what works, but in implementing those interventions at scale, with quality, and within real-world health systems. Quality improvement (QI) and implementation science (IS) offer complementary strategies to address this challenge. QI focuses on local, iterative problem solving to adapt and improve evidence-based or locally generated care processes,4 5 while IS provides structured, theory-driven methods to promote their uptake and sustainability.6 7 Yet too often, these fields operate independently rather than in a...

The number of complaints received by healthcare organisations from patients and families is on an upward trajectory.1 For example, in 2023–2024, the NHS in England received 241 922 complaints,2 an increase of 5% on the previous year and 37% since 2013–2014. Moreover, while relatively few NHS patient encounters result in a formal complaint (approximately 0.4%), just 9% of patients who report poor healthcare experiences actually submit one.3

Although the motivation for complainants can vary—for instance, some patients seek redress, and others want resolution of ongoing problems—they nearly always request organisational learning.4 Furthermore, while complaints can be incorrect or ill-intentioned, leading to concerns about their validity,5 the collective scale of the information they provide is hard to dismiss. They are, in effect, a massive rolling compendium of ethnographies from patients and families at the sharp end of treatment delivery, revealing perceived...

Background

Improving small and sick newborn care (SSNC) is crucial in resource-limited settings. Newborn Essential Solutions and Technologies (NEST360), a multicountry alliance, aims to reduce newborn mortality through evidence-based interventions. NEST360 developed a multipronged approach to improving quality. We use implementation research (IR) to describe this approach and report emerging implementation outcomes.

Methods

The implementation research logic model (IRLM) was applied to link contextual factors, implementation strategies, mechanisms and implementation outcomes, capturing successes and challenges of the improving quality approach. Data sources included programme data, peer-reviewed publications and team input. Contextual factors were organised by the NEST360-UNICEF SSNC implementation toolkit. Strategies were grouped by the Expert Recommendations for Implementation Change list, and implementation outcomes were measured using Proctor’s implementation outcomes.

Results

We developed an IRLM to describe the implementation of NEST360’s improving quality model. This IRLM included 33 contextual factors; 42% were barriers, 42% were facilitators, and 15% were both a barrier and facilitator. Additionally, we identified 10 implementation strategies that NEST360 used. The logic model also describes the connections between the contextual factors, the strategies that address them, and the preliminary implementation outcomes. Examples of the outcomes measured include Reach with 100% of units logging into the NEST360-Implementation Tracker (NEST-IT) at least once (October 2023 to March 2024), Adoption with 100% of units conducting a quality improvement (QI) project (April 2024 to June 2024), and Feasibility with 93% of units reporting NEST-IT data in their QI project documentation (April 2024 to June 2024). Finally, this study identified sustainability strategies as a critical need.

Conclusions

Integrating IR and QI enhances SSNC in resource-limited settings. Addressing barriers, leveraging facilitators and using structured IR frameworks advanced QI efforts, thereby improving intervention reach, adoption and feasibility while building scalable systems for high-quality healthcare.

Background

Since 2015, the National Institute for Health and Care Excellence (NICE) guidelines have recommended antenatal magnesium sulfate (MgSO4) for mothers in preterm labour (<30 weeks’ gestation) to reduce the risk of cerebral palsy (CP) in the preterm baby. However, the implementation of this guideline in clinical practice was slow, and MgSO4 use varied between maternity units. In 2018, the PRrevention of Cerebral palsy in PreTerm labour (PReCePT) programme, an evidence-based quality improvement (QI) intervention to improve use of MgSO4, was rolled out across England. Earlier evaluation found this programme to be effective and cost-effective over the first 12 months. We extended the original evaluation to determine the programme’s longer-term impact over 4 years, its impact in later preterm births, the impact of the COVID-19 pandemic, and to compare MgSO4 use in England (where PReCePT was implemented) to Scotland and Wales (where it was not).

Methods

Quasi-experimental longitudinal study using data from the National Neonatal Research Database on babies born <30 weeks’ gestation and admitted to a National Health Service neonatal unit. Primary outcome was the percentage of eligible mothers receiving MgSO4, aggregated to the national level. Impact of PReCePT on MgSO4 use was estimated using multivariable linear regression. The net monetary benefit (NMB) of the programme was estimated.

Results

MgSO4 administration rose from 65.8% in 2017 to 85.5% in 2022 in England. PReCePT was associated with a 5.8 percentage points improvement in uptake (95% CI 2.69 to 8.86, p<0.001). Improvement was greater when including older preterm births (<34 weeks’ gestation, 8.67 percentage points, 95% CI 6.38 to 10.96, p<0.001). Most gains occurred in the first 2 years following implementation. PReCePT had a NMB of £597 000 with 89% probability of being cost-effective. Following implementation, English uptake appeared to accelerate compared with Scotland and Wales. There was some decline in use coinciding with the onset of the pandemic.

Conclusions

The PReCePT QI programme cost-effectively improved use of antenatal MgSO4, with anticipated benefits to the babies who have been protected from CP.

Background

The Child Opportunity Index (COI) is a multidimensional measure of US neighbourhood-level conditions needed for healthy development. COI is associated with healthcare delivery and outcomes. Formal quality improvement (QI) may influence the relationship between COI, quality of care and outcomes in children.

Objective

To assess the association between COI and paediatric sepsis care delivery and outcomes and determine if baseline disparities in care change over time among hospitals in the Improving Pediatric Sepsis Outcomes (IPSO) collaborative.

Methods

Retrospective cohort study of IPSO patients probabilistically linked to the Pediatric Health Information System database from 2017 to 2021. Primary exposure was COI. We estimated differences in the proportions of patients in each COI quintile identified via standardised sepsis recognition protocols (screening tool, huddle documentation and/or order set use) and who received a bundle of recommended care (standardised sepsis recognition, plus bolus <1 hour and antibiotic <3 hours). We further assessed the timeliness of each bundle component and mortality. We evaluated changes in standardised sepsis recognition over time using generalised linear models.

Results

31 260 sepsis cases from 24 hospitals were included. Cross-sectional analysis over the entire study period found patients in the Very High COI quintile were most likely to be identified via standardised recognition protocols and receive IPSO’s recommended care bundle (67.7% and 46%, respectively). Over time, standardised sepsis recognition improved for all; the greatest improvements were among inpatients in the Very Low COI quintile.

Conclusion

Disparities exist in paediatric sepsis care delivery by COI. Over the course of the IPSO collaborative, care improved most for children in the lowest COI quintile. QI collaboratives focused on standardisation and shared learning may reduce disparities.

Background

Regulation plays a central role in health and social care systems, particularly in ensuring quality, safety and accountability. However, there is limited understanding of how organisations effectively implement and adhere to these regulatory requirements. In particular, little is known about how providers of residential care facilities for people with disabilities (RCF-D) navigate and apply statutory care regulations.

Methods

We conducted semistructured interviews with managers of RCF-D. Participant recruitment followed a purposive maximum variation sampling approach. 19 participants were interviewed, representing 22 RCF-D and 16 provider organisations. Interview data were analysed using a mixed deductive–inductive approach.

Results

Most managers were supportive of regulatory goals, creating a more favourable environment for successful implementation. By making sense of regulatory requirements and sharing insights across their organisations, managers facilitated smoother implementation. Crucially, building strong internal and external networks played a pivotal role in driving success. Collaborative relationships with inspectors, centred on a shared commitment to improving residents’ lives, further strengthened the implementation process.

Conclusion

Managers of RCF-D devised a range of strategies to manage compliance, balancing regulatory demands with problem-solving and relationship-building. These efforts were supported by a collaborative approach to working with inspectors, which fostered a shared commitment to improving residents’ lives. Our findings offer practical guidance for organisations seeking to improve regulatory compliance through effective relationship management and resource alignment. Future research could investigate how framing regulation as an adaptive intervention could further enhance implementation and sustain compliance.

Background

There is increasing interest in how patient complaint material can be used to highlight areas requiring quality improvement (QI) in healthcare. However, knowledge of using complaint material to initiate or monitor QI is limited.

Objectives

This review explored the use of complaint material in QI by identifying problems related to substandard care that were addressed by QI initiatives, exploring how complaint material was used before or after a QI initiative, and mapping changes in complaint material after QI initiatives.

Methods

This scoping review followed the Joanna Briggs Institute methodology and adhered to the Preferred Reporting Items for Systematic reviews and Meta-Analyses extension for Scoping Reviews reporting guideline.

Eligibility criteria

Studies were included if a QI initiative was initiated or monitored using complaint material. Eligible designs included observational studies, QI projects, pre-intervention and post-intervention studies and randomised controlled trials. Audio, online and symptom-based complaints were excluded.

Information sources

A systematic search was conducted on 10 December 2024 in Embase, Medline, CINAHL and Web of Science, and additional sources, with no language or date limitations.

Synthesis of results

Substandard problems targeted by QI initiatives were categorised using the Healthcare Complaints Analysis Tool by two independent coders. Findings were synthesised narratively and summarised using frequency analyses where applicable.

Results

We identified 58 QI initiatives, most frequently targeting safety (n=39). Before QI, complaint material was usually analysed through review (n=19), counts (n=17), content categorisation (n=9) or root cause analysis (n=2). After QI, analyses included counts (n=34), rates (n=20), content categorisation (n=7) and review (n=4). Reviewing or categorisation methods were often unspecified. Among studies using complaints as an outcome, most reported complaint reductions (n=43), while a few reported increases (n=2) or mixed results (n=4).

Discussion

The QI initiatives primarily targeted patient safety and applied simple quantitative analyses. Some studies relied on reviews or categorisations without reporting the validation or reliability of the used tools. Improved reporting standards are needed to strengthen learning. Furthermore, while QI initiatives appear to have the potential to change complaint patterns, this finding should be interpreted with caution, as this is based on a scoping review.

Other

Preregistered protocol: https://osf.io/6g4qw.

Introduction

Quality improvement (QI) and implementation science (IS) are distinct, yet related fields that aspire to improve the quality of healthcare for all people. QI is a systematic, data-driven approach to continuous problem solving in healthcare, with a focus on enhancing the efficiency, quality and safety of care delivery.1 Emerging in the mid-to-late twentieth century, QI identifies and analyses local problems, develops and implements targeted structural and process-focused solutions, and evaluates their outcomes. Separately, IS emerged in the late twentieth century due to growing awareness of the need to bridge the gap between research findings and their application in real-world settings. IS, defined as the study of methods to promote the systematic uptake of research findings and other evidence-based practices into routine healthcare and public health settings, emphasises improving healthcare processes and outcomes.2 While the disciplines share common goals, the fields diverge with respect...

More than 20 years since a landmark study1 documented hospitalised patients were more likely to die when their nurse cared for too many patients at a time, hundreds of rigorously conducted studies in over 30 countries have documented a relationship between nurse understaffing and poorer outcomes of all kinds, including preventable patient deaths and avoidable burnout of nurses.2–4 Despite the empirical evidence, chronic hospital nurse understaffing persists.

Why has the research evidence not substantively transformed hospital staffing practices? One possible explanation is that the benefits of eliminating nurse understaffing accrue to patients and nurses, while the costs of staffing more nurses accrue to hospitals.

Hospitals are the most expensive healthcare setting, largely because of the intensive nursing care that hospitalised patients require. Indeed, if patients can have procedures and treatments administered outside of hospitals (eg, outpatient offices, home care), they do,...

BMJ is concerned about the consent obtained from veterans to use their personal information in this research paper1 and broader Veterans' Medicines Advice and Therapeutics Education Services (MATES) program. BMJ was contacted by veterans who asked the journal to retract the content on this basis.

 The MATES program was operated by the Department for Veterans Affairs (DVA), and the research was conducted by the University of South Australia (UniSA).2 DVA was responsible for obtaining consent from the veterans to use their personal information in the MATES program; they were also responsible for managing opt-out requests.

 In 2018, a veteran lodged a complaint with the Australian Information and Privacy Commissioner about the validity of consent for DVA’s collection and use of their personal information in the MATES programme. In 2023, the Privacy Commissioner determined that DVA had breached an Australian privacy principle by using and disclosing...

Since the 18th century, bedside rounds have been a fundamental component of clinical care, serving as a setting where clinical information is gathered, processed and shared.1 This tradition highlights the importance of maintaining a high level of structure during clinical encounters. Over time, structured tools to guide care have been widely adopted across multiple specialties.2–6 Systematic checklists, in particular, have become the most used form of structured intervention in bedside wards to enhance patient care and safety.7–9 This intervention has been associated with improved non-clinical outcomes, such as communication and adherence to standard protocols.7–9 However, their impact on clinical outcomes remains a matter of debate.10

A 2014 systematic review found that safety checklists improved team communication, improved adherence to standards and reduced adverse...

Background

Understaffing by nursing staff in hospitals is linked to patients coming to harm and dying unnecessarily. There is a vicious cycle whereby poor work conditions, including understaffing, can lead to nursing vacancies, which in turn leads to further understaffing. Is hospital investment in nursing staff, to eliminate understaffing on wards, cost-effective?

Methods

This longitudinal observational study analysed data on 185 adult acute units in four hospital Trusts in England over a 5-year period. We modelled the association between a patient’s exposure to ward nurse understaffing (days where staffing was below the ward mean) over the first 5 days of stay and risk of death, risk of readmission and length of stay, using survival analysis and linear mixed models. We estimated the incremental cost-effectiveness of eliminating understaffing by registered nurses (RN) and nursing support (NS) staff, estimating net costs per quality-adjusted life year (QALY). We took a hospital cost perspective.

Findings

Exposure to RN understaffing is associated with increased hazard of death (adjusted HR (aHR) 1.079, 95% CI 1.070 to 1.089), increased chance of readmission (aHR 1.010, 95% CI 1.005 to 1.016) and increased length of stay (ratio 1.687, 95% CI 1.666 to 1.707), while exposure to NS understaffing is associated with smaller increases in hazard of death (aHR 1.072, 95% CI 1.062 to 1.081) and length of stay (ratio 1.608, 95% CI 1.589 to 1.627) but reduced readmissions (aHR 0.994, 95% CI 0.988 to 0.999). Eliminating both RN and NS understaffing is estimated to cost £2778 per QALY (staff costs only), £2685 (including benefits of reduced staff sickness and readmissions) or save £4728 (including benefits of reduced lengths of stay). Using agency staff to eliminate understaffing is less cost-effective and would save fewer lives than using permanent members of staff. Targeting specific patient groups with improved staffing would save fewer lives and, in the scenarios tested, cost more per QALY than eliminating all understaffing.

Interpretation

Rectifying understaffing on inpatient wards is crucial to reduce length of stay, readmissions and deaths. According to the National Institute for Health and Care Excellence £10 000 per QALY threshold, it is cost-effective to eliminate understaffing by nursing staff. This research points towards investing in RNs over NS staff and permanent over temporary workers. Targeting particular patient groups would benefit fewer patients and is less cost-effective.

Objectives

In the intensive care unit (ICU), antibiotics are often given longer than recommended in guidelines. A better understanding of the factors influencing antibiotic therapy duration is needed to develop improvement strategies to effectively address these drivers of excessive duration. This study aimed to explore the determinants of adherence to recommended antibiotic therapy durations among healthcare professionals involved in antibiotic decision-making within the ICU, focusing on multidisciplinary meetings (MDMs).

Methods

Semistructured interviews were held with healthcare professionals involved in antibiotic decision-making during MDMs in four Dutch ICUs. Participants included intensivists, clinical microbiologists and ICU residents. Transcripts were analysed using deductive and inductive content analysis methods.

Results

A total of 20 participants were interviewed. The interviews revealed that decision-making regarding antibiotic therapy duration is a complex process, primarily centred around professional interactions during MDMs and involving a broad range of determinants. These determinants were categorised into the following four steps: (1) the introduction of duration as a topic for discussion in the MDM (eg, lack of priority to discuss antibiotic therapy duration); (2) the discussion of antibiotic therapy duration itself (eg, lack of core members during MDM); (3) the establishment of a concrete decision (eg, lack of documentation of the decisions made); (4) the execution of the decision (eg, forgetting to stop antibiotics).

Conclusions

Our study identified numerous factors that influence decisions about the duration of antibiotic therapy during MDMs in the ICU. By describing these factors throughout the decision-making process, we provided valuable insights into barriers that commonly arise in specific steps, highlighting critical areas for improvement. Daily MDMs were deemed essential for informed decision-making regarding antibiotic therapy duration by the interviewees. Strategies to improve appropriate duration in the ICU should prioritise strengthening interdisciplinary communication between healthcare professionals and adding structure to these meetings.

Objectives

The aim of this article is to provide an estimate of the proportion of the general public reporting healthcare-related harm in Great Britain, its location, impact, responses post-harm and desired reactions from healthcare providers.

Design

We used a cross-sectional survey, using quota sampling.

Setting

This research was conducted in Great Britain.

Participants

The survey had 10 064 participants (weighted analysis).

Results

In our survey 9.7% participants reported harm caused by the National Health Service (NHS) in the last 3 years through treatment or care (6.2%) or the lack of access to care (3.5%). The main location where the harm first occurred was hospitals. A total of 37.6% of participants reported a moderate impact and 44.8% a severe impact of harm. The most common response to harm was to share their experience with others (67.1%). Almost 60% sought professional advice and support, with 11.6% contacting the Patient Advice and Liaison Service (PALS). Only 17% submitted a formal complaint, and 2.1% made a claim for financial compensation. People wanted treatment or care to redress the harm (44.4%) and an explanation (34.8%). Two-thirds of those making a complaint felt it was not handled well and approximately half were satisfied with PALS. Experiences and responses differed according to sex and age (eg, women reported more harm). People with long-term illness or disability, those in lower social grades, and people in other disadvantaged groups reported higher rates and more severe impact of harm.

Conclusions

We found that 9.7% of the British general population reported harm by the NHS, a higher rate than reported in two previous surveys. Our study used a broader and more inclusive definition of harm and was conducted during the COVID-19 pandemic, making comparison to previous surveys challenging. People responded to harm in different ways, such as sharing experiences with others and seeking professional advice and support. Mostly, people who were harmed wanted help to redress the harm or to gain access to the care needed. Low satisfaction with PALS and complaints services may reflect that these services do not always deliver the required support. There is a need to better understand the patient perspective following harm and for further consideration of what a person-centred approach to resolution and recovery might look like.

Background

Evaluation of neck trauma is a common reason for emergency department (ED) visits. There are several validated clinical decision rules, such as the National Emergency X-Radiography Utilization Study (NEXUS) Cervical Spine (C-spine) Rule, that can be used to risk stratify these patients and identify low-risk patients who do not require CT imaging. Overutilisation of CT imaging exposes patients to unnecessary radiation, impairs hospital throughput and increases healthcare costs. Various audit-and-feedback strategies have been described in other settings, but it is not known whether these strategies are effective for reducing imaging overutilisation in the ED. Additionally, the effectiveness of face-to-face feedback strategies as compared with digital feedback strategies for addressing this problem has not been previously evaluated. The aim of this study was to compare audit-and-feedback strategies to reduce CT overutilisation in the ED.

Methods

This was a prospective randomised controlled trial, in which emergency medicine clinicians were randomised into three arms to receive digital feedback, hybrid face-to-face/digital feedback or no feedback. Each clinician received three rounds of feedback on patient encounters in which they ordered a CT of the C-spine. Patient encounters were retrospectively reviewed to determine each clinician’s overutilisation rate, defined as the percentage of patients who underwent CT of the C-spine despite being classified as low risk by NEXUS criteria.

Results

A total of 78 emergency medicine clinicians were randomised into three arms. Baseline overutilisation rates for each group were 46%–47% of CT of the C-spine studies. After three rounds of audit-and-feedback strategy, the clinicians in the digital feedback group had an overutilisation rate of 33%, compared with 44% in the control group (p=0.020). The hybrid feedback group had an overutilisation rate of 36% (p=0.055 vs control; p=0.577 vs digital feedback). Over the study period, the digital group saw a reduction of 1.26 CT of the C-spine studies per provider per month (p=0.049), and the hybrid feedback group saw a reduction of 1.43 CTs per provider per month (p=0.044).

Conclusion

A digital audit-and-feedback strategy is effective for reducing overutilisation of CT imaging of the C-spine in the ED, while the effectiveness of a hybrid strategy requires further investigation.

Background

Ward rounds are an essential activity occurring in hospital settings. Despite their fundamental role in guiding patient care, they have no standardised approach. Implementation of structured interventions during ward rounds was shown to improve outcomes such as efficiency, documentation and communication. Whether these improvements have an impact on clinical outcomes is unclear. Our systematic review assessed whether structured interventions to guide ward rounds affect patient outcomes.

Methods

A systematic search was carried out in May 2023 on Embase, Medline, CINAHL, ERIC, Web of Science Core Collection, the Cochrane Library (Wiley) and Google Scholar, and a backward and forward citation search in January 2024. We included peer-reviewed, original studies assessing the use of structured interventions during bedside ward rounds (BWRs) on clinical outcomes. All inpatient hospital settings where BWRs are performed were included. We excluded papers looking at board, teaching or medication rounds.

Results

Our search strategy yielded 29 studies. Two were randomised controlled trials (RCTs) and 27 were quasi-experimental interventional studies. The majority (79%) were conducted in intensive care units. The main clinical outcomes reported were mortality, infectious complications, length of stay (LOS) and duration of mechanical ventilation (DoMV). Mortality, LOS and rates of urinary tract and central-line associated bloodstream infections did not seem to be affected, positively or negatively, by interventions structuring BWRs, while evidence was conflicting regarding their effects on rates of ventilator-associated pneumonia and DoMV, with a signal towards improved outcomes. Studies were generally of low-to-moderate quality.

Conclusion

The impact of structured interventions during BWRs on clinical outcomes remains inconclusive. Higher quality research focusing on multicentric RCTs or on prospective pre–post trials with concurrent cohorts, matched for key characteristics, is needed.

PROSPERO registration number

CRD42023412637.

Background

Hospital incident reporting and patient concerns systems are widely used to detect and respond to patient harm. Despite increasing recognition of the link between equity and safety, equity remains poorly integrated into the design and function of these systems. Consequently, these systems risk obscuring or reproducing inequities rather than revealing and attending to them.

Objective

To examine how issues of equity are currently considered in research about hospital incident reporting and patient concerns systems and identify opportunities to more systematically include equity in how patient safety is addressed.

Methods

A critical interpretive synthesis was conducted to develop a theoretical understanding of the topic through inductive analysis and interpretation. The databases CINAHL, EMBASE, MEDLINE and PsycINFO were searched from database inception to 6 February 2024. Select social science, patient safety and health services literature supported the interpretive process.

Results

After screening 6508 abstracts and conducting hand searches, we included 30 articles in our review. Our analysis identified four equity-related themes. The first theme describes how knowledge injustices in ‘what counts as a safety event or contributor’ shape what patient issues are recognised, recorded and addressed. The second theme examines how individual bias and systemic discrimination affect which safety events and concerns get reported. The third theme explores both opportunities and limitations of stratifying data to uncover equity-related patterns of harm. The fourth theme presents alternate frameworks, including restorative and human rights approaches, as ways to address inequities and humanise harm.

Conclusion

The findings provide direction for changes within incident reporting and patient concerns practices (eg, expanding definitions of harms; creating accessible and culturally safe patient concerns systems). They also affirm the opportunity to learn from, and build on, initiatives such as taking a restorative approach that moves beyond a customer service and risk management framing.

Since the Institute of Medicine released its landmark report in 19991 highlighting serious concerns about patient safety (PS) and the quality of healthcare in the USA, training programmes around the world have made quality improvement (QI) and PS key components of resident education and experience. In recognising that physicians must learn to interact with and adapt to a constantly changing healthcare environment, the Accreditation Council for Graduate Medical Education (ACGME) and ACGME International (ACGME-I) identified QI and PS as two of the six ‘Pathways of Excellence’ expected in an optimal clinical learning environment.2 In this journal, Molina et al present an exemplary effort to train the next generation of healthcare professionals in improvement science.3 This 13-year QI initiative in a large paediatric residency programme demonstrated how applying QI methodology to the educational programme itself can strengthen resident training. Through phased interventions—including structured didactics,...

Medication-related problems remain one of the leading causes of patient harm.1 Studies show that advances in electronic health records (EHR) and computerised prescribing systems with clinical decision support (CDS) have reduced prescription errors, improved physician performance and patient outcomes. However, these effects are not universally experienced and are influenced by usability, perceived usefulness, relevance and efficiency.2 3

Medication-related CDS can be classified as basic (eg, drug-drug interaction (DDI) or drug-allergy checks) or advanced, whereby patient-specific information in the EHR is used (eg, drug-disease contraindication or drug-laboratory test checks).4 However, a key persistent issue with CDS is alert fatigue, where clinically important alerts are ignored alongside alerts that are not clinically important or relevant. This is exacerbated when healthcare providers are presented with excessive and unimportant alerts, resulting in high alert override rates.5 Indeed, DDI software is often associated with...

Background

Drug–drug interaction (DDI) alerts target the co-prescription of two potentially interacting medications and are a frequent feature of electronic medical records (EMRs). There have been few controlled studies evaluating the effectiveness of DDI alerts. This study aimed to determine the impact of DDI alerts on rates of DDIs and on associated patient harms.

Methods

Quasi-experimental controlled pre–post study in five Australian hospitals. Three hospitals acted as control hospitals (EMR with no DDI alerts) and two as intervention (EMR with DDI alerts). Only DDI alerts at the highest severity level (defined as ‘major contraindicated’) were switched on at intervention hospitals. These alerts were not tailored to clinical context (ie, patient, drug). A total of 2078 patients were randomly selected from all patients (adult and paediatric) admitted to hospitals 6 months before and 6 months after EMR implementation. A retrospective chart review was performed by study pharmacists. The primary outcome was the proportion of admissions with a clinically relevant DDI. Secondary outcomes included the proportions of admissions with a potential DDI and with DDI-related harm.

Results

Potential DDIs were identified in the majority of admissions (n=1574, 74.7%) and clinically relevant DDIs identified in half (n=1026, 48.7%). DDI alerts were associated with a reduction in the proportion of admissions with potential DDIs (adjusted OR (AOR)=0.38 (0.19, 0.78)) but no change in clinically relevant DDIs (AOR=1.12 (0.68, 1.84)) or in DDI-related harm (AOR=2.42 (0.47,12.31)). 199 DDIs (76 at control and 123 at intervention hospitals) for 35 patient admissions were associated with patient harm, and 2 patients experienced severe DDI-related harm pre-EMR implementation.

Discussion

Implementation of DDI alerts, without tailoring alerts to clinical context, is unlikely to reduce patient harms from DDIs. Organisations should reconsider implementation of DDI alerts in EMRs where significant tailoring of alerts is not possible. Future research should focus on identifying safe, efficient and cost-effective ways of refining DDI alerts, so expected clinical benefits are achieved, and negative consequences of excessive alerting are minimised.