Patient Safety in Hospital – Knowledge or Campaign? [Internet]

Background Eastern Norway Regional Health Authority asked The Norwegian Knowledge Centre for the Health Services to assess the documentation for six areas of intervention launched by the Institute for Healthcare Improvement (100K-Campaign). The six areas were: Deploy ’Rapid Response Team’ Improve Care for Acute Myocardial Infarction Prevent Adverse Drug Events Prevent Surgical Site Infection Prevent Ventilator-Associated Pneumonia Prevent Central Line-Associated Bloodstream Infection The objectives were: To identify and assess scientific documentation on the above mentioned six areas ofintervention and their effect on hospital mortality.To discuss the relevance for Norwegian hospitals.Working group A support group of clinicians has contributed to this work: Elisabeth Arntzen, Director of department, Helse Ost RHF Anders Baalsrud, Senior adviser Sosial-og helsedirektoratet, Head of department, Rikshospitalet-Radiumhospitalet HF Mads Gilbert, Avdelingsoverlege, professor, Akuttmedisinsk avdeling, Universitetssykehuset Nord-Norge, Tromso Stein Tore Nilsen, Fagdirektor, professor, Stavanger Universitetssjukehus, Helse Stavanger HF Bjarne Riis Strom, Medisinsk fagdirektor i den Norske Legeforening, Oslo (two meetings) From the Knowledge centre: Unni Krogstad, Senior researcher (prosjektleder) Liv Rygh, Senior adviser Sari Ormstad, Research librarian Inger Norderhaug, Research director We searched the Cochrane, Medline, Cinahl and Embase databases for all systematic reviews, guidelines and review articles on documentation of effect of the six areas. Separate searches were done for each issue. Assessment of the retrieved literature was done stepwise by two persons independently. Agreement on the inclusion of studies was reached through discussions.Method All six areas in the 100K-campaign should be included. Database searches were restricted to secondary literature defined as systematic reviews, health technology assessments and other review studies presenting themselves as systematic. Each area was treated separately with individual strategy for searches. Where we could not find systematic reviews we opened for assessment of primary studies. Selection of relevant studies and assessment of retrieved literature were done stepwise by two persons. Results We found two systematic reviews on patient safety in general. A total of 1411 abstracts were retrieved. 40 articles were read in full text and 19 were included in the documentation. The amount of research literature on each of the six areas varied substantially. Results are summarised under the separate issues: Rapid response team The intervention is not much studied. No systematic reviews were found. We found two review articles of poor to moderate quality which were positive, but not conclusive. One 13 new cluster-randomised trial of 23 Australian hospitals concluded that the intervention was not cost-effective. Evidence based treatment of Acute Myocardial Infarction The literature on Acute Myocardial Infarction is large. The 100K-campaign suggest seven interventions: Aspirin given as initial treatment, Aspirin at discharge, Betablocker given within 24 hours, Betablocker by discharge, ACE-inhibitors or angiotensin-blockers by discharge, Thrombolysis within 30 minutes after admittance or PCI within 2 hours, advice on smoking cessation. All interventions are recommended by the American and the European cardiology guidelines. We found two studies reporting on practice and outcomes of the recommended interventions. Regional variation in the practice were found by register studies in the US. Whether implementation saves lives on a large scale is not documented. Prevent Adverse Drug Events by reconciliation'' Medication reconciliation' is a complex process and this intervention is not much studied. One relevant review article was found that compared discrepancies between the medication history obtained by the physician and the comprehensive medication history at the time of admission. No conclusions can be drawn on this issue. Prevent Surgical Site Infection We found six Cochrane reviews and three relevant review articles on this issue. The main general conclusion is that Antibiotic prophylaxis is effective treatment in different kinds of surgery and should be recommended. We found no evidence of hair removal reducing surgical site infections. There is, however, documented that if hair removal is conducted clipping, not shaving should be used. Prevent Ventilator-Associated Pneumonia The field is scarcely studied due to difficulties with definition of the diagnosis. Two articles were included which support the elevation of head by 30-45 degrees, and daily assessment of possible extubation. Two other suggestions were not supported. Prevent Central Line-Associated Bloodstream Infection Two studies were included but none were conclusive on the suggested interventions. Both discuss the difficulty of deciding causal relationships of isolated intervention in complex patient situations. The six different areas and the suggested interventions vary considerably with regard to level of evidence. The seven advices for treatment of acute myocardial infarction are well documented as is the case for the recommendation of antibiotic prophylaxis for surgical site infections. There is not sufficient evidence in the cases of prevention of sepsis in relation to intra venous central lines catheters or ventilator associated pneumonia. Rapid response teams and reconciliation has not been much studied. These interventions also are deeply rooted in the organising of work, which may vary considerably across countries and health systems. Studies in these areas should probably 14 be related to the relevant health system setting. In general medical interventions are better documented than organisational interventions. The literature included in the review is from English speaking countries only. The studies are conducted in UK, Canada, USA and Australia. This may imply a bias which may be more related to financial, organisational or cultural aspects than to language in itself.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.