Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Ashley Elizabeth Muller has a total of 66 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Overdosevarslingssystemer – en kartleggingsoversikt med maskinlæring

    by Tiril C. Borge, Ashley E. Muller

    Sammendrag Bakgrunn: Helsedirektoratet ønsker å vurdere muligheten for å etablere et nasjonalt overdosevarslingssystem (OVS) for å kunne varsle brukere om økt overdosefare. OVS som system forstås her som et sett med sammenhengende elementer som fungerer som en helhet, der hensikten er tidlig oppdagelse av økt overdoserisiko kombinert med rask varsling til de som står i fare for overdoser. Metode: Vi utførte en kartleggingsoversikt over forskning om varslingssystemer for økt overdosefare. Mål 1 var å kartlegge eksisterende forskning om effekten av et OVS (antall brukere nådd, antall overdoser, varslingshurtighet). Mål 2 var å kartlegge eksisterende forskning som kunne belyse brukeres erfaringer med slike systemer. Resultater: Vi brukte flere maskinlæringsfunksjoner for å identifisere og vurdere totalt 4102 referanser fra litteratursøk gjennomført i mars/april 2022, hvorav 11 studier oppfylte inklusjonskriteriene. Effekt av OVS på antall overdoser var ikke mulig å skille fra eksisterende overdoseforebyggende tiltak. OVS’er varslet skadereduksjonsgrupper eller erfaringskonsulenter så tidlig som fire timer etter innmeldt/oppdaget fare. Antall brukere varslet ble i stor grad ikke rapportert, antageligvis fordi data om dette ikke ble innsamlet. Fra syv kvalitative studier ble mønstre i hovedfunn sett på tvers, som omhandlet preferanser for språkbruk og innhold, hvor varslinger burde komme fra, og prinsippene bak dem. Konklusjon: Vi identifiserte lite empirisk forskning om eksisterende OVS’er, særlig relatert til effekt av slike systemer. Uten tilstrekkelig forskningsgrunnlag om effekt må vi være forsiktige med å trekke konklusjoner om hvilke typer OVS’er som bør og ikke bør innføres i Norge. Opprettelse av et OVS trenger ikke nødvendigvis innebære etablering av en ny infrastruktur, men heller være snakk om at to godt utviklede, eksisterende systemer – systemer for overvåking og skadereduksjonstiltak – kan kobles sammen.


  • Open Access

    The effect of machine learning tools for evidence synthesis on resource use and time-to-completion: protocol for a retrospective pilot study

    by Ashley Elizabeth Muller, Rigor C Berg, Jose Francisco Meneses-Eschavez, Heather M. R. Ames, Tiril C. Borge, Patricia Sofia Jacobsen Jardim, Chris Cooper, Christopher James Rose

    Background Machine learning (ML) tools exist that can reduce or replace human activities in repetitive or complex tasks. Yet ML is underutilized within evidence synthesis, despite the steadily growing rate of primary study publication and need to periodically update reviews to reflect new evidence. Underutilization may be partially explained by a paucity of evidence on how ML tools can reduce resource use and time-to-completion of reviews. Methods This protocol describes how we will answer two research questions using a retrospective study design: Is there a difference in resources used to produce reviews using recommended ML versus not using ML, and is there a difference in time-to-completion? We will also compare recommended ML use to non-recommended ML use. We will retrospectively include all reviews conducted at our institute from 1 August 2020, corresponding to the commission of the first review in our institute that used ML. We will use the results from this study to design a rigorous, multi-institutional, prospective study that will additionally explore review quality. Conclusion We invite other evidence synthesis groups to adopt and adapt this protocol and to collaborate with us.


  • Open Access

    Patient’s and healthcare provider’s experiences with Opioid Maintenance Treatment (OMT): A qualitative evidence synthesis

    by Asbjorn Steiro, Christine Hillestad Hestevik, Ashley (Ley) Muller

    Background Opioid Maintenance Treatment (OMT) is the gold standard for people with opioid dependence but drop-out can be high. Understanding both patients and health personnel’s experiences with the treatment can provide valuable information to improve the quality of OMT and to increase acceptability and accessibility of services. The aim of this systematic review is to explore the knowledge base of experiences of patients and health care providers, in order to inform the Norwegian Directorate of Health’s national guideline process. Methods We conducted a qualitative evidence synthesis. We first conducted a systematic literature search in electronic databases. Among the 56 studies that met inclusion criteria, we extracted data about study contexts and populations to assess relevance to our research question and on data richness, then purposively sampled to arrive at a manageable sample. Two researchers independently extracted and coded data in Nvivo, and we used the Andersen’s healthcare utilization model to organize and develop codes into four main domains. We assessed the methodological quality of the included studies, and our confidence in findings using CERQual. Results After a literature search retrieved 56 references, we included and analyzed 24 qualitative studies of patients’ and health providers’ experiences with OMT services. We found that stigma was a cross-cutting theme that linked the four domains of Andersen’s healthcare utilization model. Stigma from people outside OMT was a barrier to seeking out and remaining in treatment. Both patients and health personnel perceived that the OMT system contributed to further stigma. OMT services itself was a source of stigma and communication and staff-patient relations were either facilitators or barriers to treatment compliance. Inadequate knowledge and competence among health care providers was a barrier as reported by both patients and providers, which affected the availability and quality of OMT. Patients also had expectations related to non-medical treatment outcomes of OMT such as employment, housing and social relationships. Patients continuously balanced positive expectations of OMT, and negative outcomes related to stigma. Conclusion These results show the need for increased competence in relational work, among OMT personnel, to improve the quality and to avoid stigma and negative attitudes among health providers. OMT should also have a holistic approach to meet patients' non-health-related needs, as these seem to be crucial for treatment compliance and outcomes.


  • Open Access

    The effect of machine learning tools for evidence synthesis on resource use and time-to-completion: protocol for a retrospective pilot study

    by Ashley Elizabeth Muller, Rigmor C. Berg, Jose Francisco Meneses-Echavez, Heather M. R. Ames, Tiril C. Borge, Patricia Sofia Jacobsen Jardim, Chris Cooper, Christopher James Rose

    Background: Machine learning (ML) tools exist that can reduce or replace human activities in repetitive or complex tasks. Yet, ML is underutilized within evidence synthesis, despite the steadily growing rate of primary study publication and the need to periodically update reviews to reflect new evidence. Underutilization may be partially explained by a paucity of evidence on how ML tools can reduce resource use and time-to-completion of reviews. Methods: This protocol describes how we will answer two research questions using a retrospective study design: Is there a difference in resources used to produce reviews using recommended ML versus not using ML, and is there a difference in time-to-completion? We will also compare recommended ML use to non-recommended ML use that merely adds ML use to existing procedures. We will retrospectively include all reviews conducted at our institute from 1 August 2020, corresponding to the commission of the first review in our institute that used ML. Conclusion: The results of this study will allow us to quantitatively estimate the effect of ML adoption on resource use and time-to-completion, providing our organization and others with better information to make high-level organizational decisions about ML.


  • Can Remote Patient Monitoring Be the New Standard in Primary Care of Chronic Diseases, Post-COVID-19?

    by Ashley Elizabeth Muller, Rigmor C. Berg, Patricia Sofia Jacobsen Jardim, Trine Bjerke Johansen, Sari Susanna Ormstad

    Background:One lesson from the current COVID-19 pandemic is the need to optimize health care provision outside of traditional settings, and potentially over longer periods of time. An important strategy is remote patient monitoring (RPM), allowing patients to remain at home, while they transmit health data and receive follow-up services. Materials and Methods:We conducted an overview of the latest systematic reviews that had included randomized controlled trials with adult patients with chronic diseases. We summarized results and displayed these in forest plots, and used GRADE (Grading of Recommendations Assessment, Development, and Evaluation) to assess our certainty of the evidence. Results:We included 4 systematic reviews that together reported on 11 trials that met our definition of RPM, each including patients with diabetes and/or hypertension. RPM probably makes little to no difference on HbA1c levels. RPM probably leads to a slight reduction in systolic blood pressure, with questionable clinical meaningfulness. RPM probably has a small negative effect on the physical component of health-related quality of life, but the clinical significance of this reduction is uncertain. We have low confidence in the finding that RPM makes no difference to the remaining five primary outcomes. Conclusion:Most of our findings are consistent with reviews of other, broader definitions of RPM. The type of RPM examined in this review is as effective as standard treatment for patients with diabetes/hypertension. If this or other types of RPM are to be used for “long covid” patients or for other chronic disease groups post-pandemic, we need to understand why RPM may negatively affect quality of life.


  • Open Access

    A protocol to evaluate unsupervised text clustering to screen and categorize studies in systematic reviews

    by Ashley Elizabeth Muller, Heather Melanie R Ames, Tiril C Borge, Christine Hillestad Hestevik, Jose Francisco Meneses-Echavez, Christopher James Rose

    Background: The evidence base is steadily increasing regarding the time savings of using machine learning (ML) within evidence synthesis, particularly supervised methods such as classification. Unsupervised methods such as clustering have been less explored. Yet clustering – a method to uncover groups of similar data, from a large and heterogenous dataset – may be a particularly relevant tool within the study identification and data extraction processes of reviews, during which researchers must often read thousands of studies to identify the handful relevant to their review. This protocol is for a mixed methods evaluation of clustering implemented within systematic reviews.Research questions: We will answer the following questions: 1) Can clustering while screening at title/abstract level provide useful automatic categorizations of studies that help reviewers extra data to categorize studies? 2) What are reviewers’ perceptions of and experiences with clustering in relation to acceptability and feasibility in future reviews? Methods: We will identify one or two systematic reviews as models to answer these research questions. Reviewers will use the clustering algorithm Lingo3G built into EPPI Reviewer during the screening process. The main outcome to answer research question 1 is content validity, as an indicator of usefulness, defined by similarities between automatic clusters and reviewer-generated categories. We will display multi-class confusion matrices and use these to estimate precision, recall, and F1-score for each useful cluster. Research question 2 will be answered qualitatively, through semi-structured interviews with participating reviewers before and after having been trained on clustering and implementing it in the model review(s), followed by a focus group presenting results and exploring reviewer trust. Conclusion: The planned evaluation will provide important information as to the feasibility of an unsupervised ML method, clustering, in bridging and aligning two currently linear steps in evidence synthesis. We invite interested evidence synthesis organizations to follow this protocol and subsequently share data with us, or to tailor it to produce a more relevant evaluation.

  • Open Access

    Machine learning in systematic reviews: Comparing automated text clustering with Lingo3G and human researcher categorization in a rapid review

    by Ashley Elizabeth Muller, Heather Melanie R. Ames, Patricia Sofia Jacobsen Jardim, Christopher James Rose

    Systematic reviews are resource-intensive. The machine learning tools being developed mostly focus on the study identification process, but tools to assist in analysis and categorization are also needed. One possibility is to use unsupervised automatic text clustering, in which each study is automatically assigned to one or more meaningful clusters. Our main aim was to assess the usefulness of an automated clustering method, Lingo3G, in categorizing studies in a simplified rapid review, then compare performance (precision and recall) of this method compared to manual categorization. We randomly assigned all 128 studies in a review to be coded by a human researcher blinded to cluster assignment (mimicking two independent researchers) or by a human researcher non-blinded to cluster assignment (mimicking one researcher checking another's work). We compared time use, precision and recall of manual categorization versus automated clustering. Automated clustering and manual categorization organized studies by population and intervention/context. Automated clustering failed to identify two manually identified categories but identified one additional category not identified by the human researcher. We estimate that automated clustering has similar precision to both blinded and non-blinded researchers (e.g., 88% vs 89%), but higher recall (e.g., 89% vs 84%). Manual categorization required 33% more time than automated clustering. Using a specific clustering algorithm, automated clustering can be helpful with categorization of and identifying patterns across studies in simpler systematic reviews. We found that the clustering was sensitive enough to group studies according to linguistic differences that often corresponded to the manual categories.


  • Open Access

    Participant Engagement in Supported Employment: A Systematic Scoping Review

    by Mariya Khoronzhevych, Tatiana Maximova-Mentzoni, Erika Gubrium, Ashley Elizabeth Muller

    Purpose This study aimed to synthesise the available knowledge on how participant engagement in supported employment (SE) interventions is presented, defined, and conceptualised. We also aimed to develop a working definition of participant engagement in SE based on the results of our study. Methods This systematic scoping review was conducted following the PRISMA extension for scoping reviews. The following databases were systematically searched: EBSCO, SCOPUS, Social Care Online, and JSTOR. We included peer-reviewed publications in English based on empirical studies. Results Sixteen articles met the inclusion criteria and were included in the final analysis. Thematic framework analysis resulted in three themes conveying the concept of participant engagement: self-determined choice, empowerment, and collaboration/working alliance. We suggest that participant engagement in SE is an active multifaceted process that involves the empowerment of participants, participants’ exercise of self-determined informed choice, and their collaboration with SE practitioners in a working alliance. Conclusions Participant empowerment, self-determined choice, and collaboration are important aspects of participant engagement in SE. The study results will appeal to SE practitioners and make significant contributions to the broader field of other vocational services supporting people in (re-)entering the competitive labour market.


  • Dårlig forskning hindrer effektiv respons på covid-19

    by Ashley Elizabeth Muller, Erika Gubrium, Jørgen Dahlberg, Ariana Guilherme Fernandes

    25.05.2021: Kommentar - Denne artikkelen er et illustrerende eksempel på at innvandrere blir beskyldt for smittespredning av covid-19, noe som dels forankres i en bred (og stigmatiserende) oppfatning av spesifikke etniske grupper.

  • Open Access

    Instruments to measure fear of COVID-19: a diagnostic systematic review

    by Ashley Elizabeth Muller, Jan Peter William Himmels, Stijn Van de Velde

    Background: The COVID-19 pandemic has become a source of fear across the world. Measuring the level or significance of fear in different populations may help identify populations and areas in need of public health and education campaigns. We were interested in diagnostic tests developed to assess or diagnose COVID-19-related fear or phobia. Methods: We performed a systematic review of studies that examined instruments diagnosing or assessing fear or phobia of COVID-19 (PROSPERO registration: CRD42020197100). We utilized the Norwegian Institute of Public Health’s Live map of covid-19 evidence, a database of pre-screened and pre-categorized studies. The Live map of covid-19 evidence identified references published since 1 December 2019 in MEDLINE, Embase, and the Centers for Disease Control and Prevention. Following biweekly searches, two researchers independently categorized all studies according to topic (seven main topics, 52 subordinate topics), population (41 available groups), study design, and publication type. For this review, we assessed for eligibility all studies that had been categorized to the topic “Experiences and perceptions, consequences; social, political, economic aspects” as of 25 September 2020, in addition to hand-searching included studies’ reference lists. We meta-analyzed correlation coefficients of fear scores to the most common reference tests (self-reports of anxiety, depression, and stress), and reported additional concurrent validity to other reference tests such as specific phobias. We assessed study quality using the QUADAS-2 for the minority of studies that presented diagnostic accuracy statistics. Results: We found 18 studies that validated fear instruments. Fifteen validated the Fear of COVID-19 scale (FCV-19S). We found no studies that proposed a diagnosis of fear of COVID-19 or a threshold of significant/clinical versus non-significant/subclinical fear. Study quality was low, with the most common potential biases related to sampling strategy and un-blinded data analysis. The FSV-19S total score correlated strongly with severe phobia (r = 0.703, 95%CI 0.634–0.761) in one study, and moderately with anxiety in a meta-analysis. Conclusions: The accuracy of the FSV-19S needs to be measured further using fear-related reference instruments, and future studies need to provide cut-off scores and normative values. Further evaluation of the remaining three instruments is required.