What is the aim of this review?
Clinical pathways (CPW) are document-based tools that provide a link between the best available evidence and clinical practice. They provide recommendations, processes and time frames for the management of specific medical conditions or interventions. This review update aimed to summarize the evidence and assess the effect of clinical pathways on patient outcomes (inhospital mortality, mortality (up to 6 months), inhospital complications, and hospital readmissions (up to 6 months)), length of hospital stay, hospital costs and charges, and professional practice (i.e. healthcare professionals adhering to recommended practice), compared to hospital care as usual. Also, we identified and compared different implementation strategies. We included patients in hospitals that were treated according to (1) the recommendations of a CPW, or (2) a CPW that has been implemented together with other interventions, such as a case manager or quality improvement initiatives. We analyzed 58 studies (24,841 patients and 2027 healthcare professionals), of which 27 were included in a previously published review (Rotter 2010) and 31 were retrieved for this update. This is the first update of the previous review.
Key messages
CPWs might have the potential to improve patient outcomes, and they may reduce length of hospital stay, hospital costs, and improve adherence to recommended practice. But we still need more high-quality studies that report on implementation strategies used during the pathway development and implementation.
What was studied in the review?
Decision-making in hospitals has evolved from being opinion-based to being based on sound scientific evidence (i.e. evidence-based practice). Hospitals incorporate evidence into clinical pathways for health professionals to follow. They have been implemented worldwide but the evidence about their impact from single trials is contradictory. Perpetual publication of new evidence combined with the demands of everyday practice makes it difficult for health professionals to keep up to date.
Included study designs were individual and cluster-randomized studies, non-randomized studies, controlled before-after studies (CBA), and interrupted time-series studies (ITS). In individual randomized trials, study participants were allocated to the CPW or usual care group by chance, called random allocation. Cluster-randomized trials divided all study participants into smaller groups known as clusters. These clusters were then allocated by chance to the CPW or usual care group. For non-randomized trials, participants were allocated to different groups by investigators in a quasi-random fashion. Quasi-random allocation means that study participants were allocated to the CPW or usual care group based on criteria such as their date of birth or the day of the week. CBA studies are experimental studies without a random or quasi-random allocation process. Data are collected from the CPW and usual care group before the CPW was implemented, and then further data were collected after the CPW was introduced. ITS studies represent a robust method of measuring the effect of a CPW as a trend over time.
All included studies tested the impact of clinical pathways used in hospitals on one or more of the prespecified outcomes: inhospital mortality, mortality (up to 6 months), inhospital complications, hospital readmissions (up to 6 months), length of hospital stay, hospital costs and charges, and adherence to recommended practice. Studies focused on the change in outcome measures following the implementation of a stand-alone clinical pathway or a multifaceted clinical pathway combined with other interventions compared to usual care.
What are the main results of the review?
We found 58 studies that measured the effects of clinical pathways on included outcomes. The main results are that, compared to usual care, it is uncertain if the implementation of a stand-alone clinical pathway has any effect on in-hospital mortality and mortality (up to 6 months) (low certainty). Stand-alone CPWs are likely to reduce inhospital complications (moderate certainty) but it is very uncertain if they make any difference to hospital readmissions (up to 6 months) (very low certainty). Stand-alone CPWs are likely to reduce length of hospital stay (moderate certainty). Costs and charges were generally lower in CPWs in nine out of ten studies included in this comparison (very low certainty). Stand-alone CPWs may also slightly increase adherence to recommended practice (low certainty).
For multifaceted clinical pathways compared to usual care, it is uncertain whether there is a reduction in inhospital mortality (low certainty) and they may make little or no difference to mortality (up to 6 months) (low certainty). It is uncertain whether CPWs that have been combined with other interventions reduce inhospital complications (low certainty) or hospital readmissions (up to 6 months) (low certainty). It is also uncertain if they reduce length of hospital stay (low certainty), hospital costs and charges (very low certainty), and adherence to recommended practice (low certainty), compared to usual care.
How up-to-date is this review?
This review update searched for new studies up to July 26, 2024.
Stand-alone CPWs are likely to reduce inhospital complications and length of hospital stay and may slightly increase adherence to recommended practice. There was little conclusive evidence for multifaceted CPWs due to mixed results from a limited number of included studies. It is uncertain whether stand-alone CPWs or CPWs, as part of a multifaceted approach, reduce inhospital mortality, mortality (up to 6 months), hospital readmission (up to 6 months) or costs and charges.
Clinical pathways (CPWs) are structured multidisciplinary care plans. They aim to translate evidence into practice and optimize clinical outcomes. This is the first update of the previous systematic review (Rotter 2010).
To investigate the effect of CPWs on patient outcomes, length of stay, costs and charges, adherence to recommended practice, and to measure the impact of different approaches to implementation of CPWs.
For this update, CENTRAL, MEDLINE, and Embase were searched on 25 July 2024. Two trial registries were searched on 26 July 2024, along with reference checking, citation searching and contacting authors to identify additional studies.
We considered two groups of participants: health professionals involved in CPW utilization, including (but not limited to) physicians, nurses, physiotherapists, pharmacists, occupational therapists and social workers; and patients managed using a CPW. We included randomized trials, non-randomized trials, controlled before-after (CBA) studies, and interrupted time-series (ITS) studies comparing (1) stand-alone clinical pathways with usual care, and (2) clinical pathways as part of a multifaceted intervention with usual care.
Two authors independently screened all titles, abstracts and full-text manuscripts to assess eligibility and the methodological quality of included studies using the Cochrane Effective Practice and Organization of Care 'Risk of Bias' tool. Certainty of evidence was assessed by two authors independently. Interventions were scored as 'high', 'moderate' or 'low' for the evidence-based implementation process.
The update provided 31 additional studies for a total of 58 included studies (24,841 patients and 2027 healthcare professionals). Forty-one (71%) were randomized trials, four (7%) non-randomized trials, four (7%) CBA studies and nine (16%) ITS studies. Forty-nine studies compared stand-alone CPWs to usual care and nine compared multifaceted interventions including a CPW to usual care. Collectively, the risk of bias was high due to potential contamination by healthcare professionals, lack of blinding of patients and personnel, lack of allocation concealment and selective reporting in ITS studies.
Stand-alone clinical pathway interventions
It is uncertain whether stand-alone CPWs reduce inhospital mortality (13% v 16%: OR 0.79, 95% CI 0.53 to 1.20; P = 0.27; I² = 65%; 7 randomized trials; n = 4603; low-certainty evidence due to serious imprecision and inconsistency) or mortality (up to 6 months) (4% v 3%: OR 1.37, 95% CI 0.72 to 2.60; P = 0.34; I² = 20%; 3 randomized trials, n = 805; low-certainty evidence due to serious risk of bias and imprecision). Stand-alone CPWs likely reduce inhospital complications (10% v 17%: OR 0.57, 95% CI 0.41 to 0.80; P = 0.001; I² = 52%; 11 randomized trials, n = 3668; moderate-certainty evidence due to serious risk of bias). It is very uncertain whether stand-alone CPWs reduce hospital readmissions (up to 6 months) (9% v 13%: OR 0.67, 95% CI 0.44 to 1.03; P = 0.07; I² = 11%; 9 randomized trials, n = 1578; very low-certainty evidence due to serious risk of bias and very serious imprecision). Stand-alone CPWs likely reduce the length of hospital stay compared to usual care (MD -1.12 days, 95% CI -1.60 to -0.65; P < 0.00001; I² = 64%; 21 studies; n = 5201; moderate-certainty evidence due to serious inconsistency). Costs and charges were generally lower in CPWs as indicated by negative MDs in nine studies (10 studies, n = 2113, data not pooled; very low-certainty evidence due to serious indirectness and very serious inconsistency). Stand-alone CPWs may slightly increase adherence to recommended practice compared with usual care (3 randomized studies, n = 573; data not pooled; low-certainty evidence due to serious risk of bias and serious inconsistency).
Multifaceted clinical pathway interventions
It is uncertain whether multifaceted CPWs reduce inhospital mortality (2 randomized studies, n = 6304, data not pooled; low-certainty evidence due to very serious inconsistency). Multifaceted CPWs may make little or no difference to mortality (up to 6 months) (9% v 8%: OR 1.05, 95% CI 0.88 to 1.25; P = 0.61; I² = 0%; 3 randomized studies; n = 6531; low-certainty evidence due to serious imprecision and serious risk of bias). It is uncertain whether multifaceted CPWs reduce inhospital complications (9% v 23%: OR 0.32, 95% CI 0.12 to 0.87; 1 study, n = 140; low-certainty evidence due to very serious imprecision).
It is uncertain whether multifaceted CPWs reduce hospital readmission (up to 6 months) (2 randomized studies, n =1569, data not pooled; low-certainty evidence due to very serious inconsistency), or length of stay (4 randomized studies, n = 1936, data not pooled; low-certainty evidence due to very serious inconsistency), or hospital costs and charges (4 randomized studies, n = 2015, data not pooled; very low-certainty evidence due to very serious imprecision and serious indirectness in outcome measures). It is uncertain whether multifaceted CPWs increase adherence to recommended practice (2 randomized studies, n = 6304, data not pooled, low-certainty evidence due to very serious inconsistency).
Key study characteristics
The highest proportion of included studies were from the USA (36%), followed by Australia (10%), China (10%), Japan (5%), the UK (5%), Canada (5%), Italy (5%), and Germany (5%). More than half of the included studies tested CPW in general acute wards (53%), followed by emergency departments (17%), intensive care (14%), and extended-stay facilities (10%). The most common clinical conditions were asthma (16%), stroke (10%), mechanical ventilation (9%) and myocardial infarction (7%).