-
Computers in Biology and Medicine Oct 2023Uncertainty estimation in healthcare involves quantifying and understanding the inherent uncertainty or variability associated with medical predictions, diagnoses, and... (Review)
Review
Uncertainty estimation in healthcare involves quantifying and understanding the inherent uncertainty or variability associated with medical predictions, diagnoses, and treatment outcomes. In this era of Artificial Intelligence (AI) models, uncertainty estimation becomes vital to ensure safe decision-making in the medical field. Therefore, this review focuses on the application of uncertainty techniques to machine and deep learning models in healthcare. A systematic literature review was conducted using the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. Our analysis revealed that Bayesian methods were the predominant technique for uncertainty quantification in machine learning models, with Fuzzy systems being the second most used approach. Regarding deep learning models, Bayesian methods emerged as the most prevalent approach, finding application in nearly all aspects of medical imaging. Most of the studies reported in this paper focused on medical images, highlighting the prevalent application of uncertainty quantification techniques using deep learning models compared to machine learning models. Interestingly, we observed a scarcity of studies applying uncertainty quantification to physiological signals. Thus, future research on uncertainty quantification should prioritize investigating the application of these techniques to physiological signals. Overall, our review highlights the significance of integrating uncertainty techniques in healthcare applications of machine learning and deep learning models. This can provide valuable insights and practical solutions to manage uncertainty in real-world medical data, ultimately improving the accuracy and reliability of medical diagnoses and treatment recommendations.
Topics: Artificial Intelligence; Bayes Theorem; Reproducibility of Results; Uncertainty; Delivery of Health Care
PubMed: 37683529
DOI: 10.1016/j.compbiomed.2023.107441 -
Journal of Clinical Epidemiology Apr 2023To critically review and analyze evidence synthesis articles using health inequality/inequity guidance to support their review. (Review)
Review
OBJECTIVES
To critically review and analyze evidence synthesis articles using health inequality/inequity guidance to support their review.
STUDY DESIGN AND SETTING
A comprehensive, systematic search of six social science databases (1990 to May 2022) and grey literature sources was undertaken. A narrative approach to synthesis was adopted, describing and categorizing the characteristics of included articles. A comparison of the existing methodological guides was also conducted, discussing the similarities and differences between them.
RESULTS
From 205 identified reviews published between 2008 and 2022, 62 (30%) focusing on health inequality/inequity, met the criteria. The reviews were diverse in terms of methodology, populations, intervention level, and clinical areas. Only 19 (31%) reviews discussed the definition of inequality/inequity. Two methodological guides were identified: (i) the PROGRESS/Plus framework and (ii) the Preferred Reporting Items for Systematic Reviews and Meta-Analyses-Equity checklist.
CONCLUSION
A critique on the methodological guides reaffirms a lack of clarity or guidance on how health inequality/inequity should be considered. The PROGRESS/Plus framework narrowly focuses on dimensions of health inequality/inequity but rarely considers the pathways and interactions of these dimensions and their effect on outcomes. The Preferred Reporting Items for Systematic Reviews and Meta-Analyses-Equity checklist on the other hand provides guidance on report. A conceptual framework is needed to show the pathways and interactions of dimensions of health inequality/inequity.
Topics: Humans; Health Status Disparities; Consensus; Checklist; Research Report
PubMed: 36813002
DOI: 10.1016/j.jclinepi.2023.02.013 -
Frontiers in Endocrinology 2023The optimal first-line immune checkpoint inhibitor (ICI) treatment strategy for metastatic or early triple-negative breast cancer (TNBC) has not yet been determined as a... (Meta-Analysis)
Meta-Analysis
BACKGROUND
The optimal first-line immune checkpoint inhibitor (ICI) treatment strategy for metastatic or early triple-negative breast cancer (TNBC) has not yet been determined as a result of various randomized controlled trials (RCTs). The purpose of this study was to compare the efficacy and safety of ICIs in patients with metastatic or early TNBC.
METHODS
RCTs comparing the efficacy and safety of ICIs in patients with TNBC were included in the studies. Based on PRISMA guidelines, we estimated pooled hazard ratios (HRs) and odds ratios (ORs) using random-effects models of Bayesian network meta-analysis. Primary outcomes were progression-free survival (PFS) and overall survival (OS). Secondary outcomes included pathologic complete response rate (pCR), grade ≥ 3 treatment-related adverse events (trAEs), immune-related adverse events (irAEs), and grade ≥ 3 irAEs.
RESULTS
The criteria for eligibility were met by a total of eight RCTs involving 4,589 patients with TNBC. When ICIs were used in patients without programmed death-ligand 1 (PD-L1) selection, there was a trend toward improved PFS, OS, and pCR, without significant differences. Pembrolizumab plus chemotherapy is superior to other treatment regimens in terms of survival for TNBC patients based on Bayesian ranking profiles. Subgroup analysis by PD-L1 positive population indicated similar results, and atezolizumab plus chemotherapy provided better survival outcomes. Among grade ≥ 3 trAEs and any grade irAEs, there was no statistically significant difference among different ICI agents. The combination of ICIs with chemotherapy was associated with a higher incidence of grade ≥ 3 irAEs. Based on rank probability, the ICI plus chemotherapy group was more likely to be associated with grade ≥ 3 trAEs, any grade irAEs, and grade ≥ 3 irAEs. Hypothyroidism and hyperthyroidism were the most frequent irAEs in patients receiving ICI.
CONCLUSIONS
ICI regimens had relatively greater efficacy and safety profile. Pembrolizumab plus chemotherapy and atezolizumab plus chemotherapy seem to be superior first-line treatments for intention-to-treat and PD-L1-positive TNBC patients, respectively. It may be useful for making clinical decisions to evaluate the efficacy and safety of different ICIs based on our study.
SYSTEMATIC REVIEW REGISTRATION
https://www.crd.york.ac.uk/PROSPERO/, identifier CRD42022354643.
Topics: Humans; Immune Checkpoint Inhibitors; Triple Negative Breast Neoplasms; B7-H1 Antigen; Network Meta-Analysis; Decision Making
PubMed: 37229447
DOI: 10.3389/fendo.2023.1137464 -
Value in Health : the Journal of the... Oct 2022Access to medicines in Europe depends on a benefit-risk decision taken by regulators and a relative effectiveness assessment performed by health technology assessment...
OBJECTIVES
Access to medicines in Europe depends on a benefit-risk decision taken by regulators and a relative effectiveness assessment performed by health technology assessment bodies (HTABs) to inform, as one element, a reimbursement decision. Although various similarities in evidence needs exist, understanding of their needs is currently suboptimal and therefore the evidence generated does not always meet their needs. Subsequently, delays in decision making can be expected, negatively affecting access. To overcome this, this study reviewed the evidentiary needs of European regulators and HTABs at European level and analyzed how their collaboration can further facilitate optimal evidence generation plans, evidence use, and evidence presentation.
METHODS
Through systematic literature review, expert interviews, and pairwise comparison of assessment reports by the European Medicines Agency and European network for health technology assessment, respective clinical evidence requirements and impact of product-specific collaboration between European Medicines Agency and HTABs were established.
RESULTS
Clinical evidence needs are quite similar but differences exist in comparator choice, preferred efficacy endpoints, and target population. Results of the impact of collaboration to date were mixed: preapproval joint advice procedures were successful and highly valued by all stakeholders; information exchange at the time of regulatory decision is coming together, yet the European Public Assessment Report can be further optimized; and collaboration on postlicensing evidence generation requirements shows potential but needs solidifying.
CONCLUSIONS
These findings demonstrate the potential to further improve the evidence utilization across stakeholders to avoid duplication and streamline decision making, to ultimately improve access to medicines for European patients.
Topics: Decision Making; Europe; Humans; Technology Assessment, Biomedical
PubMed: 35370077
DOI: 10.1016/j.jval.2022.01.026 -
JAMA Network Open Sep 2023Testing new medical devices or procedures in terms of safety, effectiveness, and durability should follow the strictest methodological rigor before implementation. (Meta-Analysis)
Meta-Analysis
IMPORTANCE
Testing new medical devices or procedures in terms of safety, effectiveness, and durability should follow the strictest methodological rigor before implementation.
OBJECTIVES
To review and analyze studies investigating devices and procedures used in intracranial aneurysm (IA) treatment for methods and completeness of reporting and to compare the results of studies with positive, uncertain, and negative conclusions.
DATA SOURCES
Embase, MEDLINE, Web of Science, and The Cochrane Central Register of Clinical Trials were searched for studies on IA treatment published between January 1, 1995, and the October 1, 2022. Grey literature was retrieved from Google Scholar.
STUDY SELECTION
All studies making any kind of claims of safety, effectiveness, or durability in the field of IA treatment were included.
DATA EXTRACTION AND SYNTHESIS
Using a predefined data dictionary and analysis plan, variables ranging from patient and aneurysm characteristics to the results of treatment were extracted, as were details pertaining to study methods and completeness of reporting. Extraction was performed by 10 independent reviewers. A blinded academic neuro-linguist without involvement in IA research evaluated the conclusion of each study as either positive, uncertain, or negative. The study followed Preferring Reporting Items for Systematic Reviews and Meta-Analyses guidelines.
MAIN OUTCOMES AND MEASURES
The incidence of domain-specific outcomes between studies with positive, uncertain, or negative conclusions regarding safety, effectiveness, or durability were compared. The number of studies that provided a definition of safety, effectiveness, or durability and the incidence of incomplete reporting of domain-specific outcomes were evaluated.
RESULTS
Overall, 12 954 studies were screened, and 1356 studies were included, comprising a total of 410 993 treated patients. There was no difference in the proportion of patients with poor outcome or in-hospital mortality between studies claiming a technique was safe, uncertain, or not safe. Similarly, there was no difference in the proportion of IAs completely occluded at last follow-up between studies claiming a technique was effective, uncertain, or noneffective. Less than 2% of studies provided any definition of safety, effectiveness, or durability, and only 1 of the 1356 studies provided a threshold under which the technique would be considered unsafe. Incomplete reporting was found in 546 reports (40%).
CONCLUSIONS AND RELEVANCE
In this systematic review and meta-analysis of IA treatment literature, studies claiming safety, effectiveness, or durability of IA treatment had methodological flaws and incomplete reporting of relevant outcomes supporting these claims.
Topics: Humans; Intracranial Aneurysm; Hospital Mortality; Neurology; Uncertainty
PubMed: 37656458
DOI: 10.1001/jamanetworkopen.2023.31798 -
Epilepsy & Behavior : E&B Jan 2024Guidance documents play a pivotal role in shaping the management of status epilepticus (SE). However, the methodological quality of these documents remains uncertain. In... (Review)
Review
Guidance documents play a pivotal role in shaping the management of status epilepticus (SE). However, the methodological quality of these documents remains uncertain. In this systematic review, we comprehensively searched 12 literature and guideline databases to assess the quality of clinical practice guidelines and consensus statements related to SE management using the AGREE II methodology. Additionally, we summarized the associated recommendations. We identified a total of 14 clinical practice guidelines and 11 consensus statements spanning the period from 1993 to 2022. The median score for clarity of presentation was 71.8% (ranging from 15.3% to 91.7%), indicating generally good clarity. However, the aspect of editorial independence received poor ratings, with a median score of 32.1% (ranging from 0% to 83.3%). Notably, the 2016 guideline published by the American Epilepsy Society in Epilepsy (AES) received the highest overall scores. Across these guidance documents, there was consistency in the definition and diagnosis of SE. However, significant variability was observed in therapeutic recommendations, particularly in terms of the timing for adding or changing medications. The methodological approaches used in most SE guidance documents require improvement, and the disparities in recommendations highlight existing gaps in evidence. Enhanced methodological rigor results in increased standardization of the guideline, consequently augmenting its reference value. Given the urgency of SE as an emergency condition, it is imperative that these documents also address relevant management strategies before admission.
Topics: Humans; Consensus; Epilepsy; Hospitalization; Status Epilepticus; United States; Practice Guidelines as Topic
PubMed: 38128315
DOI: 10.1016/j.yebeh.2023.109555 -
Osteoarthritis and Cartilage Oct 2021To determine the effectiveness of patient decision aids (PtDAs) compared to alternative interventions (including usual care) on decision quality and quality of the...
OBJECTIVES
To determine the effectiveness of patient decision aids (PtDAs) compared to alternative interventions (including usual care) on decision quality and quality of the decision-making process for adults with hip and knee osteoarthritis considering primary elective total joint arthroplasty.
METHODS
A systematic review guided by Cochrane methods and PRISMA reporting guidelines. Studies were searched in five databases. Included studies were RCTs evaluating the effect of PtDAs on total joint arthroplasty decision-making. Study quality was appraised with Cochrane's risk of bias tool. Quality and strength of recommendations were appraised with GRADE.
RESULTS
Ten included studies were conducted in North American using the same PtDA. Compared to usual care, PtDA groups demonstrated increased decision quality (e.g., higher knowledge, more informed values-based choices) and quality of the decision making process (e.g., decreased decisional conflict) (6 trials). Secondary outcomes showed increased surgeon satisfaction within the consultation and no difference in patient satisfaction or uptake of the chosen option (surgery: RR 1.03, 95% CI = 0.84 to 1.25; I = 66%; 4 trials). When PtDAs formtats were compared, there were similar effects but no difference between PtDAs (4 trials).
CONCLUSIONS
There was low to very low GRADE certainty of evidence for the effect of PtDAs on decision quality and quality of the decision-making process compared to usual care. No differences were found when different formats of PtDAs were compared (moderate to very low GRADE certainty of evidence).
Topics: Arthroplasty, Replacement, Hip; Arthroplasty, Replacement, Knee; Decision Making; Decision Support Techniques; Humans
PubMed: 34302958
DOI: 10.1016/j.joca.2021.07.006 -
BMC Geriatrics Feb 2021The aim of this study was to describe barriers and facilitators for shared decision making (SDM) as experienced by older patients with multiple chronic conditions... (Review)
Review
BACKGROUND
The aim of this study was to describe barriers and facilitators for shared decision making (SDM) as experienced by older patients with multiple chronic conditions (MCCs), informal caregivers and health professionals.
METHODS
A structured literature search was conducted with 5 databases. Two reviewers independently assessed studies for eligibility and performed a quality assessment. The results from the included studies were summarized using a predefined taxonomy.
RESULTS
Our search yielded 3838 articles. Twenty-eight studies, listing 149 perceived barriers and 67 perceived facilitators for SDM, were included. Due to poor health and cognitive and/or physical impairments, older patients with MCCs participate less in SDM. Poor interpersonal skills of health professionals are perceived as hampering SDM, as do organizational barriers, such as pressure for time and high turnover of patients. However, among older patients with MCCs, SDM could be facilitated when patients share information about personal values, priorities and preferences, as well as information about quality of life and functional status. Informal caregivers may facilitate SDM by assisting patients with decision support, although informal caregivers can also complicate the SDM process, for example, when they have different views on treatment or the patient's capability to be involved. Coordination of care when multiple health professionals are involved is perceived as important.
CONCLUSIONS
Although poor health is perceived as a barrier to participate in SDM, the personal experience of living with MCCs is considered valuable input in SDM. An explicit invitation to participate in SDM is important to older adults. Health professionals need a supporting organizational context and good communication skills to devise an individualized approach for patient care.
Topics: Aged; Caregivers; Decision Making; Decision Making, Shared; Humans; Multiple Chronic Conditions; Patient Participation; Quality of Life
PubMed: 33549059
DOI: 10.1186/s12877-021-02050-y -
Breast Cancer Research and Treatment Nov 2019Decision aids (DAs) support patients in shared decision-making by providing balanced evidence-based treatment information and eliciting patients' preferences. The... (Review)
Review
PURPOSE
Decision aids (DAs) support patients in shared decision-making by providing balanced evidence-based treatment information and eliciting patients' preferences. The purpose of this systematic review was to assess the quality and communicative aspects of DAs for women diagnosed with early-stage breast cancer.
METHODS
Twenty-one currently available patient DAs were identified through both published literature (MEDLINE, Embase, CINAHL, CENTRAL, and PsycINFO) and online sources. The DAs were reviewed for their quality by using the International Patient Decision Aid Standards (IPDAS) checklist, and subsequently assessed to what extent they paid attention to various communicative aspects, including (i) information presentation, (ii) personalization, (iii) interaction, (iv) information control, (v) accessibility, (vi) suitability, and (vii) source of information.
RESULTS
The quality of the DAs varied substantially, with many failing to comply with all components of the IPDAS criteria (mean IPDAS score = 64%, range 31-92%). Five aids (24%) did not include any probability information, 10 (48%) presented multimodal descriptions of outcome probabilities (combining words, numbers, and visual aids), and only 2 (10%) provided personalized treatment outcomes based on patients and tumor characteristics. About half (12; 57%) used interaction methods for eliciting patients' preferences, 16 (76%) were too lengthy, and 5 (24%) were not fully accessible.
CONCLUSIONS
In addition to the limited adherence to the IPDAS checklist, our findings suggest that communicative aspects receive even less attention. Future patient DA developments for breast cancer treatment should include communicative aspects that could influence the uptake of DAs in daily clinical practice.
Topics: Breast Neoplasms; Decision Making, Shared; Decision Support Techniques; Evidence-Based Medicine; Female; Humans; Neoplasm Staging; Patient Participation; Patient Preference; Physician-Patient Relations
PubMed: 31342311
DOI: 10.1007/s10549-019-05351-4 -
Journal of Comparative Effectiveness... Oct 2020Breast cancer is a leading cause of cancer among women. Because guidelines on screening for breast cancer for certain ages are controversial, many experts advocate the...
Breast cancer is a leading cause of cancer among women. Because guidelines on screening for breast cancer for certain ages are controversial, many experts advocate the use of shared decision making (SDM) using validated decision aids (DAs). Recent studies have concluded that DAs are beneficial; however, the results have great heterogeneity. Therefore, further studies are needed to improve understanding of these tools. This systematic review and meta-analysis aimed to investigate the impact of using web-based DAs in women aged 50 years and below facing the decision to be screened for breast cancer in comparison with usual care. PubMed, Web of Science, Embase and the Cochrane CENTRAL databases were searched up to February 2020 for studies assessing web-based DAs for women making a breast cancer screening decision and reported quality of decision-making outcomes. Using a random-effects model or a fixed-effects model, meta-analyses were conducted pooling results using mean differences (MD), standardized mean differences (SMD) and relative risks (RR). Of 1097 unique citations, three randomized controlled trials and two before-after studies met the study eligibility criteria. Compared with usual care, web-based DAs increased knowledge (SMD = 0.69; 95% CI: 0.57-0.80; p < 0.00001), reduced decision conflict and increased the proportion of women who made an informed choice (RR = 1.86; 95% CI: 1.38 to 2.50; p < 0.0001), but did not change the intention of women deciding to be screened or affect decision regret. This analysis showed the positive effect of web-based DAs on patient-centered outcomes in breast cancer screening. In the future, more internet devices and free or larger discount WI-FI should be established to ensure more women can benefit from this effective tool.
Topics: Adult; Breast Neoplasms; Decision Making; Decision Support Techniques; Early Detection of Cancer; Female; Health Knowledge, Attitudes, Practice; Humans; Internet; Mammography; Mass Screening; Middle Aged; Patient Education as Topic; Patient Participation
PubMed: 33025800
DOI: 10.2217/cer-2020-0052