Substituting physicians for nurses in hospital interventions does not affect patient safety, says review

In general terms, the methodological basis is solid, for several reasons:

• It is a Cochrane systematic review, which implies rigorous methodology, exhaustive searching, and critical assessment of risk of bias.
• It includes 82 randomised clinical trials (RCTs) with more than 28,000 patients in 20 countries.
• RCTs are the gold standard for comparing interventions.
• It assesses clinically relevant outcomes: mortality, adverse events, quality of life, self-efficacy, and satisfaction.

However, there are aspects that require cautious interpretation:

• Significant heterogeneity:
• Different types of nurses (APNs, CNSs, registered nurses).
• Different levels of autonomy and supervision.
• Different medical specialities.
• Different organisational models.
• Variability in the measurement of outcomes.
• Limited and inconsistent economic evidence.
• Predominance of high-income countries (39% United Kingdom).

The methodological conclusion is that the evidence is robust but heterogeneous. The results are reliable at a global level, although they cannot always be extrapolated uniformly to each context.

The results are consistent with previous studies showing that:

• Advanced practice nurses (APN) can manage chronic conditions with similar outcomes to doctors.
• In primary care, safety in the management of mild conditions and follow-up of chronic conditions had already been demonstrated.
• Previous studies also reported good patient satisfaction.

Therefore, it does not contradict previous evidence, but rather reinforces it.

In terms of new developments, this study:

• Focuses specifically on substitution in hospital settings and specialised outpatient clinics, not just in primary care.
• Extends the evidence to multiple specialties (oncology, diabetes, cardiology, dermatology, gynaecology, etc.).
• Identifies areas where nursing care could be even superior (diabetes control, oncology follow-up).
• Clearly points out that substitution is not a simple “one-for-one” replacement, but depends on the organisational model.
• The main novelty [consists of] broad and cross-cutting evidence that consolidates the role of advanced nursing in specialised hospital settings as well.

[In terms of limitations:]

1. Structural heterogeneity: There is no single “replacement” model. Results may depend on training level, protocolisation, actual autonomy, medical supervision and healthcare system organisation.
2. Limited generalisation: Most studies were conducted in the United Kingdom, with little representation from countries with different healthcare systems and little evidence from low- and middle-income countries.
3. Unclear costs and inconsistent economic results. Some studies show savings, others show higher costs due to longer consultation times, more referrals and differences in prescribing.
4. Possible implementation bias. In some cases, nurses offered more time per patient, additional education, and more frequent follow-up. Therefore, the better results could be due not only to the professional category, but also to the more intensive care model.

[The relevance to clinical practice in Spain is] very high, especially in the current context.

The topic is profound and the conclusions are striking, as it concludes that there is no difference in such important areas as mortality or patient safety. It is necessary to take a closer look at the entire study. As a preliminary comment, the study is of great interest, given that the demand for healthcare professionals is growing, and we often do not optimise the tasks we assign to our professionals—bearing in mind that the study has focused on countries with healthcare models different from ours. In Spain in particular, healthcare professionals and, often, family doctors have to spend a large part of their time on administrative processes that could be reorganised with administrative staff and by taking advantage of new technologies. On the other hand, there are healthcare tasks that can be delegated to other healthcare professionals, not only nurses, but also physiotherapists, psychologists, pharmacists, etc. Some of these are indicated in the study, but others do not appear here, such as minor symptoms, screening, or specific follow-ups. In any case, it is important to clearly establish the roles of delegation and coordination for this to be effective, who assumes responsibility, whether there is supervision, etc., while being aware that there are roles that cannot be transferred from the doctor, such as diagnosis or surgery.

Healthcare systems around the world are undergoing a profound transformation marked by an ageing population, an increase in chronic conditions, growing complexity of care and a shortage of professionals. This scenario requires a reorganisation of how services are provided and optimisation of the role of all healthcare professionals, especially nurses. The latest scientific evidence shows that when nurses take on certain functions traditionally associated with the medical field (such as clinical assessment, monitoring of pathologies, requesting tests or therapeutic adjustments within a regulated framework), the results for patients are equivalent in terms of safety, quality and effectiveness, and even better in some contexts, without an increase in adverse events.

It is important to emphasise that this evolution should not be interpreted as a substitution or competition between professions, but rather as an intelligent redistribution of responsibilities to reinforce a collaborative and efficient care model. Nursing in the 21st century involves advanced university training, specialisation and expanded clinical competencies, and in many countries, advanced practice models have already proven to be effective tools for improving access, continuity of care and person-centred care. Therefore, the debate should not focus on ‘whether’ nurses can take on new roles, but on how to develop educational, organisational and regulatory frameworks that allow their potential to be fully exploited to build more sustainable, equitable and future-proof healthcare systems.

The study is based on a robust methodology, as it is a Cochrane systematic review with meta-analysis of clinical trials, developed under international standards such as EPOC/MECIR, with risk of bias assessment and evidence grading using GRADE. Overall, the results indicate that replacing doctors with nurses in hospitals produces small or non-existent differences in mortality and patient safety, and may provide benefits in some areas, such as the management of chronic diseases, physical and psychological function, and certain dermatological problems.

A significant limitation is the high heterogeneity among the studies included, both in terms of patient types and interventions, as well as healthcare systems, training levels, and degrees of professional autonomy. Furthermore, the study focuses exclusively on the hospital setting, so its conclusions cannot be directly extrapolated to primary care. All of this means that the certainty of the evidence is moderate or low in some outcomes and that the results should be interpreted with caution.

In the Spanish context, the study provides conclusions of the highest level of evidence on an ongoing process of professional reorganisation, driven by ageing, chronicity and the high complexity of care. However, the comparison must be made with caution, as in Spain many of the activities analysed, such as patient education, clinical assessment and some forms of prescribing, are part of standard nursing practice. Even so, the results reinforce the need to modernise and make professional boundaries more flexible, as the growing demand for care cannot be met by a model focused exclusively on doctors, and nurses have demonstrated that they have the necessary skills to take on new roles while ensuring quality and safety.

This review is highly reliable because it analyses 82 clinical trials involving more than 28,000 patients, and is the first to focus solely on hospitals. It concludes that nurses can safely perform medical tasks while maintaining quality and patient safety, although it warns that economic data varies greatly and most studies come from wealthy countries.

For Spain, this supports the creation of more advanced practice nursing consultations, primarily for chronic diseases, to streamline care. The key recommendation is that this change be made with advanced training and clear protocols, always ensuring that it does not place an excessive burden on nurses that affects their well-being or the quality of service.

The shared study provides somewhat surprising results, both in terms of the topic analysed and the possible conclusions.

The review shows that, in most of the hospital settings evaluated, replacing doctors with nurses offers equivalent clinical and safety outcomes, with specific improvements in chronic diseases and processes where health education is key.

These results may open up a line of research on this topic, including different aspects of interest such as an analysis of the implementation of different actions, protocols, and interventions.

An appropriate methodology is described, although some limitations could justify a careful interpretation of the results provided. This does not mean that we should not trust the results, but rather that we should interpret them within their context. 

As indicated, the heterogeneity, moderate risk of bias and enormous diversity of interventions require the results to be interpreted within each clinical context.

Part of the observed equivalence can be explained by the protocolisation of care and the filtering of complex patients. Many of the studies include stable or chronic patients, not critical patients, and therefore few patients in highly complex medical situations are included. It is possible that certain benefits may be due to characteristics specific to nursing care (more time, more support).

In summary, we believe it is appropriate to interpret the results derived from a review of heterogeneous studies cautiously, as there appears to be a general lack of blinding and loss to follow-up. In addition, there is likely to be variability in quality of life measures, regulatory differences, and possible selection of reported outcomes. Despite the above, it seems an interesting topic to expand new lines of research to clarify all possible uncertainties.