Siaw-Teng Liaw - Academia.edu (original) (raw)

Papers by Siaw-Teng Liaw

Yearbook of medical informatics, Apr 25, 2019

Background: Artificial intelligence (AI) is heralded as an approach that might augment or substit... more Background: Artificial intelligence (AI) is heralded as an approach that might augment or substitute for the limited processing power of the human brain of primary health care (PHC) professionals. However, there are concerns that AI-mediated decisions may be hard to validate and challenge, or may result in rogue decisions. Objective: To form consensus about perceptions, issues, and challenges of AI in primary care. Method: A three-round Delphi study was conducted. Round 1 explored experts' viewpoints on AI in PHC (n=20). Round 2 rated the appropriateness of statements arising from round one (n=12). The third round was an online panel discussion of findings (n=8) with the members of both the International Medical

Research Square (Research Square), Jun 15, 2021

PubMed, Aug 21, 2019

Unstructured electronic health records are valuable resources for research. Before they are share... more Unstructured electronic health records are valuable resources for research. Before they are shared with researchers, protected health information needs to be removed from these unstructured documents to protect patient privacy. The main steps involved in removing protected health information are accurately identifying sensitive information in the documents and removing the identified information. To keep the documents as realistic as possible, the step of omitting sensitive information is often followed by replacement of identified sensitive information with surrogates. In this study, we present an algorithm to generate surrogates for unstructured electronic health records. We used this algorithm to generate realistic surrogates on a Health Science Alliance corpus, which is constructed specifically for the use of development of automated de-identification systems.

AJGP, Jun 1, 2022

Background and objective Watch Me Grow-Electronic (WMG-E) platform is an online resource to enhan... more Background and objective Watch Me Grow-Electronic (WMG-E) platform is an online resource to enhance the capacity of general practitioners (GPs) to involve parents in developmental surveillance. The aim of this study was to evaluate the acceptability and perceived utility of WMG-E. Methods Semi-structured interviews were conducted with GPs/paediatricians (n = 6) and parents (n = 6). Focus groups were conducted with child and family health nurses (n = 25). Transcripts were analysed thematically. Results Participants indicated that WMG-E could empower clinicians and parents by enhancing health literacy about child developmental issues, but that it could also be disempowering if not used carefully. Clinicians mentioned being strategic at health service and public policy levels. A final theme was that of the need to balance widespread promotion with its targeted use. Discussion This study established the face validity of WMG-E, and reveals key lessons to inform the ways in which it is promoted and used. MORE THAN 20% OF AUSTRALIAN CHILDREN are 'developmentally vulnerable' at school entry, with higher rates among disadvantaged groups. 1 Neurodevelopmental disorders are also common, with the childhood prevalence of autism spectrum disorder (ASD) now estimated to be one in 59. 2 These figures are of concern, both at individual and public health levels, given the known trajectories towards behavioural, psychosocial and educational problems throughout early childhood, and compromised health, education and social outcomes in adulthood. 3,4 Given the prevalence and negative outcomes of developmental and neurodevelopmental concerns, 1 and the known benefits of early intervention, childhood developmental surveillance is best practice. 5 Defined as a 'flexible, continuous process whereby knowledgeable professionals perform skilled observations of children during the provision of healthcare', 6 developmental surveillance integrates developmental screening test results within a broader picture of a child's life, including information from medical history, current physical examination, parent input and clinician observations. 7 In practice, however, developmental surveillance opportunities are often missed due to limited uptake of voluntary surveillance programs in primary care. 8 Children from culturally and linguistically diverse (CALD) and socioeconomically disadvantaged backgrounds are at highest risk, and often miss opportunities for early intervention. 9 General practitioners (GPs) are usually the first point of contact for families accessing services in the healthcare system, 10 and barriers to developmental surveillance in primary care can include constraints on time, knowledge and/or self-efficacy with child assessment. 11 Some international initiatives have increased developmental surveillance within primary care, 11,12 but there have been challenges with uptake, referral pathways, and workflow procedures. 8 There is a need to develop innovative approaches to promote developmental surveillance. Advances in information technology and internet accessibility provide new opportunities to improve healthcare in the child health

International Journal of Medical Informatics, Dec 1, 2020

Research Square (Research Square), Jan 5, 2023

International Journal of Cardiology, Sep 1, 2023

American Journal of Epidemiology, Feb 4, 2021

Ambitious World Health Organization targets for disease elimination require monitoring of epidemi... more Ambitious World Health Organization targets for disease elimination require monitoring of epidemics using routine health data in settings of decreasing and low incidence. We evaluated 2 methods commonly applied to routine testing results to estimate incidence rates that assume a uniform probability of infection between consecutive negative and positive tests based on 1) the midpoint of this interval and 2) a randomly selected point in this interval. We compared these with an approximation of the Poisson binomial distribution, which assigns partial incidence to time periods based on the uniform probability of occurrence in these intervals. We assessed bias, variance, and convergence of estimates using simulations of Weibull-distributed failure times with systematically varied baseline incidence and varying trend. We considered results for quarterly, half-yearly, and yearly incidence estimation frequencies. We applied the methods to assess human immunodeficiency virus (HIV) incidence in HIV-negative patients from the Treatment With Antiretrovirals and Their Impact on Positive and Negative Men (TAIPAN) Study, an Australian study of HIV incidence in men who have sex with men, between 2012 and 2018. The Poisson binomial method had reduced bias and variance at low levels of incidence and for increased estimation frequency, with increased consistency of estimation. Application of methods to real-world assessment of HIV incidence found decreased variance in Poisson binomial model estimates, with observed incidence declining to levels where simulation results had indicated bias in midpoint and random-point methods. disease incidence rates; longitudinal data; Poisson binomial distribution; routine diagnostic testing Abbreviations: ACCESS, Australian Collaboration for Coordinated Enhanced Sentinel Surveillance of Sexually Transmissible Infections and Blood Borne Viruses; HIV, human immunodeficiency virus; MSM, men who have sex with men; PrEP, preexposure prophylaxis; PY, person-years; TAIPAN, Treatment With Antiretrovirals and Their Impact on Positive and Negative Men.

BMC Infectious Diseases, Dec 1, 2016

Background: Australia has increased coverage of antiretroviral treatment (ART) over the past deca... more Background: Australia has increased coverage of antiretroviral treatment (ART) over the past decade, reaching 73% uptake in 2014. While ART reduces AIDS-related deaths, accumulating evidence suggests that it could also bolster prevention efforts by reducing the risk of HIV transmission ('treatment as prevention'). While promising, evidence of community-level impact of treatment as prevention on reducing HIV incidence among gay and bisexual men is limited. We describe a study protocol that aims to determine if scale up of testing and treatment for HIV leads to a reduction in community viraemia and, in turn, if this reduction is temporally associated with a reduction in HIV incidence among gay and bisexual men in Australia's two most populous states. Methods: Over the period 2009 to 2017, we will establish two cohorts making use of clinical and laboratory data electronically extracted retrospectively and prospectively from 73 health services and laboratories in the states of New South Wales and Victoria. The 'positive cohort' will consist of approximately 13,000 gay and bisexual men (>90% of all people living with HIV). The 'negative cohort' will consist of at least 40,000 HIV-negative gay and bisexual men (approximately half of the total population). Within the negative cohort we will use standard repeat-testing methods to calculate annual HIV incidence. Community prevalence of viraemia will be defined as the proportion of men with a viral load ≥200RNA copies/mm 3 , which will combine viral load data from the positive cohort and viraemia estimates among those with an undiagnosed HIV infection. Using regression analyses and adjusting for behavioural and demographic factors associated with infection, we will assess the temporal association between the community prevalence of viraemia and the incidence of HIV infection. Further analyses will make use of these cohorts to assess incidence and predictors of treatment initiation, repeat HIV testing, and viral suppression. Discussion: This study will provide important information on whether 'treatment as prevention' is associated with a reduction in HIV incidence at a community level among gay and bisexual men.

Journal of innovation in health informatics, Dec 20, 2016

Sharing health and social care data is essential to the delivery of high-quality health care as w... more Sharing health and social care data is essential to the delivery of high-quality health care as well as disease surveillance and public health and for conducting research. However, these societal benefits may be constrained by privacy and data protection principles. Hence, societies are striving to find a balance between the two competing public interests. Whilst the spread of IT advancements in recent decades has increased the demand for an increased privacy and data protection in many ways, health is a special case. The UK is adopting guidelines, codes of conduct and regulatory instruments aimed to implement privacy principles into practical settings and enhance public trust. Accordingly, in 2015, the UK National Data Guardian requested to conduct a further review of data protection, referred to as Caldicott 3. The scope of this review is to strengthen data security standards and confidentiality. It also proposes a consent system based on an 'opt-out' model rather than on 'opt-in'. Across Europe as well as internationally, the privacy health data sharing balance is not fixed. In Europe, the enactment of the new EU Data Protection Regulation in 2016 constitutes a major breakthrough, which is likely to have a profound effect on European countries and beyond. In Australia and across North America, different ways are being sought to balance out these twin requirements of a modern society-to preserve privacy alongside affording high-quality health care for an ageing population. Whilst in the UK privacy legal framework remains complex and fragmented into different layers of legislation, which may negatively impact on both the rights to privacy and health, the UK is at the forefront in the uptake of international and EU privacy and data protection principles. And, if the privacy regime was reorganised in a more comprehensive manner, it could be used as a sound implementation model for other countries.

Public Health Research & Practice, Dec 1, 2017

PubMed, 2020

Multi-center observational studies require recognition and reconciliation of differences in patie... more Multi-center observational studies require recognition and reconciliation of differences in patient representations arising from underlying populations, disparate coding practices and specifics of data capture. This leads to different granularity or detail of concepts representing the clinical facts. For researchers studying certain populations of interest, it is important to ensure that concepts at the right level are used for the definition of these populations. We studied the granularity of concepts within 22 data sources in the OHDSI network and calculated a composite granularity score for each dataset. Three alternative SNOMED-based approaches for such score showed consistency in classifying data sources into three levels of granularity (low, moderate and high), which correlated with the provenance of data and country of origin. However, they performed unsatisfactorily in ordering data sources within these groups and showed inconsistency for small data sources. Further studies on examining approaches to data source granularity are needed.

The Lancet HIV, Jun 1, 2023

International Journal of Integrated Care, Jul 24, 2019

Introduction Well-child Care refers to the provision of preventative primary healthcare services ... more Introduction Well-child Care refers to the provision of preventative primary healthcare services for children and their families. Different countries use, however, different terminology to describe the delivery of preventative health care services for children. These include, for example, child health promotion programs in the United Kingdom, child health surveillance programs in European Union countries, and Well-child Care in North America. As a significant proportion of the literature on preventive programs emanates from North America, we have consistently used here the term Well-Child Care. While there are differences in the use of the terminology in these programs, there are certain key components that are consistent, including: health supervision, anticipatory guidance, developmental surveillance, child and family psychosocial assessment, care coordination , immunisation, physical examination, and specific screening activities [1]. It is further conceptualised in the literature as a sub-component of primary health care for children that includes: Well-child Care, acute and chronic care for childhood health conditions, and coordination and follow-up for developmental problems. Thus, Well-child Care incorporates any program that targets child health promotion and child health surveillance activities with focus on prevention and early identification of problems. Currently there is no international consensus on how Well-child Care activities are achieved within primary health sectors of developed countries. There are international differences in the provision of Well-child

Journal of the American Medical Informatics Association

Background:Significant challenges remain in the early identification of child developmental condi... more Background:Significant challenges remain in the early identification of child developmental conditions in the community. Implementing supports and services early in the life course has been shown to promote positive developmental outcomes for children at high likelihood of developmental conditions including autism. As part of a cluster randomised controlled trial, this study seeks to examine and compare the perspectives and experiences of Australian general practitioners (GPs) in relation to a digital developmental surveillance program for autism and usual care pathway, in general practice clinics. Methods:A qualitative research methodology (semi-structured interviews and thematic analysis) was utilised in this study. All GPs from South Western Sydney (NSW) and Melbourne (Victoria) who participated in the main program (“GP Surveillance for Autism”) were invited to the interview. GPs who provided consent were interviewed either over online or in-person meeting. Interviews were audio-...

Journal of the American Medical Informatics Association, 2019

Objective Electronic health records are increasingly utilized for observational and clinical rese... more Objective Electronic health records are increasingly utilized for observational and clinical research. Identification of cohorts using electronic health records is an important step in this process. Previous studies largely focused on the methods of cohort selection, but there is little evidence on the impact of underlying vocabularies and mappings between vocabularies used for cohort selection. We aim to compare the cohort selection performance using Australian Medicines Terminology to Anatomical Therapeutic Chemical (ATC) mappings from 2 different sources. These mappings were taken from the Observational Medical Outcomes Partnership Common Data Model (OMOP-CDM) and the Pharmaceutical Benefits Scheme (PBS) schedule. Materials and Methods We retrieved patients from the electronic Practice Based Research Network data repository using 3 ATC classification groups (A10, N02A, N06A). The retrieved patients were further verified manually and pooled to form a reference standard which was u...

Frontiers in Pediatrics, 2021

Background: The early detection of developmental conditions such as autism is vital to ensure chi... more Background: The early detection of developmental conditions such as autism is vital to ensure children can access appropriate and timely evidence-based supports, services, and interventions. Children who have undetected developmental conditions early in life are more likely to develop later health, developmental, learning, and behavioral issues, which in turn can have a cumulative effect over the life course.Methods: The current protocol describes a multi-site, cluster randomized control trial comparing a developmental surveillance pathway for autism to usual care, using opportunistic visits to general practitioners (GPs). Units of randomization are GP clinics across two Australian states (New South Wales and Victoria), with thirty clinics within each state, each of which will aim to recruit approximately forty children aged between ~18- and 24-months, for a total of ~2,400 participants. Children will be randomized to two clusters; namely, an autism surveillance pathway (ASP) or sur...