General Practice and Primary Care - Research Publications

Permanent URI for this collection

Search Results

Now showing 1 - 10 of 19
  • Item
    Thumbnail Image
    Artificial intelligence and machine learning algorithms for early detection of skin cancer in community and primary care settings: a systematic review
    Jones, OT ; Matin, RN ; van der Schaar, M ; Bhayankaram, KP ; Ranmuthu, CK ; Islam, MS ; Behiyat, D ; Boscott, R ; Calanzani, N ; Emery, J ; Williams, HC ; Walter, FM (ELSEVIER, 2022-06)
    Skin cancers occur commonly worldwide. The prognosis and disease burden are highly dependent on the cancer type and disease stage at diagnosis. We systematically reviewed studies on artificial intelligence and machine learning (AI/ML) algorithms that aim to facilitate the early diagnosis of skin cancers, focusing on their application in primary and community care settings. We searched MEDLINE, Embase, Scopus, and Web of Science (from Jan 1, 2000, to Aug 9, 2021) for all studies providing evidence on applying AI/ML algorithms to the early diagnosis of skin cancer, including all study designs and languages. The primary outcome was diagnostic accuracy of the algorithms for skin cancers. The secondary outcomes included an overview of AI/ML methods, evaluation approaches, cost-effectiveness, and acceptability to patients and clinicians. We identified 14 224 studies. Only two studies used data from clinical settings with a low prevalence of skin cancers. We reported data from all 272 studies that could be relevant in primary care. The primary outcomes showed reasonable mean diagnostic accuracy for melanoma (89·5% [range 59·7-100%]), squamous cell carcinoma (85·3% [71·0-97·8%]), and basal cell carcinoma (87·6% [70·0-99·7%]). The secondary outcomes showed a heterogeneity of AI/ML methods and study designs, with high amounts of incomplete reporting (eg, patient demographics and methods of data collection). Few studies used data on populations with a low prevalence of skin cancers to train and test their algorithms; therefore, the widespread adoption into community and primary care practice cannot currently be recommended until efficacy in these populations is shown. We did not identify any health economic, patient, or clinician acceptability data for any of the included studies. We propose a methodological checklist for use in the development of new AI/ML algorithms to detect skin cancer, to facilitate their design, evaluation, and implementation.
  • Item
    Thumbnail Image
    The SCRIPT trial: study protocol for a randomised controlled trial of a polygenic risk score to tailor colorectal cancer screening in primary care
    Saya, S ; Boyd, L ; Chondros, P ; McNamara, M ; King, M ; Milton, S ; Lourenco, RDA ; Clark, M ; Fishman, G ; Marker, J ; Ostroff, C ; Allman, R ; Walter, FM ; Buchanan, D ; Winship, I ; McIntosh, J ; Macrae, F ; Jenkins, M ; Emery, J (BMC, 2022-09-27)
    BACKGROUND: Polygenic risk scores (PRSs) can predict the risk of colorectal cancer (CRC) and target screening more precisely than current guidelines using age and family history alone. Primary care, as a far-reaching point of healthcare and routine provider of cancer screening and risk information, may be an ideal location for their widespread implementation. METHODS: This trial aims to determine whether the SCRIPT intervention results in more risk-appropriate CRC screening after 12 months in individuals attending general practice, compared with standard cancer risk reduction information. The SCRIPT intervention consists of a CRC PRS, tailored risk-specific screening recommendations and a risk report for participants and their GP, delivered in general practice. Patients aged between 45 and 70 inclusive, attending their GP, will be approached for participation. For those over 50, only those overdue for CRC screening will be eligible to participate. Two hundred and seventy-four participants will be randomised to the intervention or control arms, stratified by general practice, using a computer-generated allocation sequence. The primary outcome is risk-appropriate CRC screening after 12 months. For those in the intervention arm, risk-appropriate screening is defined using PRS-derived risk; for those in the control arm, it is defined using family history and national screening guidelines. Timing, type and results of the previous screening are considered in both arms. Objective health service data will capture screening behaviour. Secondary outcomes include cancer-specific worry, risk perception, predictors of CRC screening behaviour, screening intentions and health service use at 1, 6 and 12 months post-intervention delivery. DISCUSSION: This trial aims to determine whether a PRS-derived personalised CRC risk estimate delivered in primary care increases risk-appropriate CRC screening. A future population risk-stratified CRC screening programme could incorporate risk assessment within primary care while encouraging adherence to targeted screening recommendations. TRIAL REGISTRATION: Australian and New Zealand Clinical Trial Registry ACTRN12621000092897p. Registered on 1 February 2021.
  • Item
    No Preview Available
    The SYMPTOM-upper gastrointestinal study: A mixed methods study exploring symptom appraisal and help-seeking in Australian upper gastrointestinal cancer patients
    Karnchanachari, N ; Milton, S ; Muhlen-Schulte, T ; Scarborough, R ; Holland, JF ; Walter, FM ; Zalcberg, J ; Emery, J (WILEY, 2022-09)
    OBJECTIVE: There is limited evidence on the development of pancreatic and oesophagogastric cancer, how patients decide to seek help and the factors impacting help-seeking. Our study, the first in Australia, aimed to explore symptom appraisal and diagnostic pathways in these patients. A secondary aim was to examine the potential to recruit cancer patients through a cancer quality registry. METHODS: Patients diagnosed with pancreatic or oesophagogastric cancer were recruited through Monash University's Upper-Gastrointestinal Cancer Registry. Data collected through general practitioners (GP) and patient questionnaires included symptoms and their onset, whereas patient interviews focused on the patient's decision-making in seeking help from healthcare pracitioners. Data collection and analysis was informed by the Aarhus statement. Coding was inductive, and themes were mapped onto the Model of Pathways to Treatment. RESULTS: Between November 2018 and March 2020, 27 patient questionnaires and 13 phone interviews were completed. Prior to diagnosis, patients lacked awareness of pancreatic and oesophagogastric cancer symptoms, leading to the normalisation, dismissal and misattribution of the symptoms. Patients initially self-managed symptoms, but worsening of symptoms and jaundice triggered help-seeking. Competing priorities, beliefs about illnesses and difficulties accessing healthcare delayed help-seeking. CONCLUSION: Increased awareness of insidious pancreatic and oesophagogastric cancer symptoms in patients and general practitioners may prompt more urgent investigations and lead to earlier diagnosis.
  • Item
    Thumbnail Image
    Patients' Experiences of Using Skin Self-monitoring Apps With People at Higher Risk of Melanoma: Qualitative Study.
    Habgood, E ; McCormack, C ; Walter, FM ; Emery, JD (JMIR Publications Inc., 2021-08-13)
    BACKGROUND: Melanoma is the fourth most commonly diagnosed cancer in Australia. Up to 75% of melanomas are first detected by patients or their family or friends. Many mobile apps for melanoma exist, including apps to encourage skin self-monitoring to improve the likelihood of early detection. Previous research in this area has focused on their development, diagnostic accuracy, or validation. Little is known about patients' views and experiences of using these apps. OBJECTIVE: This study aims to understand patients' views and experiences of using commercially available melanoma skin self-monitoring mobile apps for a period of 3 months. METHODS: This qualitative study was conducted in two populations: primary care (where the MelatoolsQ tool was used to identify patients who were at increased risk of melanoma) and secondary care (where patients had a previous diagnosis of melanoma, stages T0-T3a). Participants downloaded 2 of the 4 mobile apps for skin self-monitoring (SkinVision, UMSkinCheck, Mole Monitor, or MySkinPal) and were encouraged to use them for 3 months. After 3 months, a semistructured interview was conducted with participants to discuss their experiences of using the skin self-monitoring mobile apps. RESULTS: A total of 54 participants were recruited in the study, with 37% (20) of participants from primary care and 62% (34) from secondary care. Interviews were conducted with 34 participants when data saturation was reached. Most participants did not use the apps at all (n=12) or tried them once but did not continue (n=14). Only 8 participants used the apps to assist with skin self-monitoring for the entire duration of the study. Patients discussed the apps in the context of the importance of early detection and their current skin self-monitoring behaviors. A range of features of perceived quality of each app affected engagement to support skin self-monitoring. Participants described their skin self-monitoring routines and potential mismatches with the app reminders. They also described the technical and practical difficulties experienced when using the apps for skin self-monitoring. The app's positioning within existing relationships with health care providers was crucial to understand the use of the apps. CONCLUSIONS: This study of patients at increased risk of melanoma highlights several barriers to engagement with apps to support skin self-monitoring. The results highlight the wide-ranging and dynamic influences on engagement with mobile apps, which extend beyond app design and relate to broader contextual factors about skin self-monitoring routines and relationships with health care providers.
  • Item
    Thumbnail Image
    Role of primary care physician factors on diagnostic testing and referral decisions for symptoms of possible cancer: a systematic review
    Hardy, V ; Yue, A ; Archer, S ; Merriel, SWD ; Thompson, M ; Emery, J ; Usher-Smith, J ; Walter, FM (BMJ PUBLISHING GROUP, 2022-01)
    BACKGROUND: Missed opportunities for diagnosing cancer cause patients harm and have been attributed to suboptimal use of tests and referral pathways in primary care. Primary care physician (PCP) factors have been suggested to affect decisions to investigate cancer, but their influence is poorly understood. OBJECTIVE: To synthesise evidence evaluating the influence of PCP factors on decisions to investigate symptoms of possible cancer. METHODS: We searched MEDLINE, Embase, Scopus, CINAHL and PsycINFO between January 1990 and March 2021 for relevant citations. Studies examining the effect or perceptions and experiences of PCP factors on use of tests and referrals for symptomatic patients with any cancer were included. PCP factors comprised personal characteristics and attributes of physicians in clinical practice. DATA EXTRACTION AND SYNTHESIS: Critical appraisal and data extraction were undertaken independently by two authors. Due to study heterogeneity, data could not be statistically pooled. We, therefore, performed a narrative synthesis. RESULTS: 29 studies were included. Most studies were conducted in European countries. A total of 11 PCP factors were identified comprising modifiable and non-modifiable factors. Clinical judgement of symptoms as suspicious or 'alarm' prompted more investigations than non-alarm symptoms. 'Gut feeling' predicted a subsequent cancer diagnosis and was perceived to facilitate decisions to investigate non-specific symptoms as PCP experience increased. Female PCPs investigated cancer more than male PCPs. The effect of PCP age and years of experience on testing and referral decisions was inconclusive. CONCLUSIONS: PCP interpretation of symptoms as higher risk facilitated testing and referral decisions for possible cancer. However, in the absence of 'alarm' symptoms or 'gut feeling', PCPs may not investigate cancer. PCPs require strategies for identifying patients with non-alarm and non-specific symptoms who need testing or referral. PROSPERO REGISTRATION NUMBER: CRD420191560515.
  • Item
    Thumbnail Image
    Identifying Novel Biomarkers Ready for Evaluation in Low-Prevalence Populations for the Early Detection of Lower Gastrointestinal Cancers: A Systematic Review and Meta-Analysis
    Druce, P ; Calanzani, N ; Snudden, C ; Milley, K ; Boscott, R ; Behiyat, D ; Martinez-Gutierrez, J ; Saji, S ; Oberoi, J ; Funston, G ; Messenger, M ; Walter, FM ; Emery, J (SPRINGER, 2021-06)
    INTRODUCTION: Lower gastrointestinal (GI) cancers are a major cause of cancer deaths worldwide. Prognosis improves with earlier diagnosis, and non-invasive biomarkers have the potential to aid with early detection. Substantial investment has been made into the development of biomarkers; however, studies are often carried out in specialist settings and few have been evaluated for low-prevalence populations. METHODS: We aimed to identify novel biomarkers for the detection of lower GI cancers that have the potential to be evaluated for use in primary care. MEDLINE, Embase, Emcare and Web of Science were systematically searched for studies published in English from January 2000 to October 2019. Reference lists of included studies were also assessed. Studies had to report on measures of diagnostic performance for biomarkers (single or in panels) used to detect colorectal or anal cancers. We included all designs and excluded studies with fewer than 50 cases/controls. Data were extracted from published studies on types of biomarkers, populations and outcomes. Narrative synthesis was used, and measures of specificity and sensitivity were meta-analysed where possible. RESULTS: We identified 142 studies reporting on biomarkers for lower GI cancers, for 24,844 cases and 45,374 controls. A total of 378 unique biomarkers were identified. Heterogeneity of study design, population type and sample source precluded meta-analysis for all markers except methylated septin 9 (mSEPT9) and pyruvate kinase type tumour M2 (TuM2-PK). The estimated sensitivity and specificity of mSEPT9 was 80.6% (95% CI 76.6-84.0%) and 88.0% (95% CI 79.1-93.4%) respectively; TuM2-PK had an estimated sensitivity of 81.6% (95% CI 75.2-86.6%) and specificity of 80.1% (95% CI 76.7-83.0%). CONCLUSION: Two novel biomarkers (mSEPT9 and TuM2-PK) were identified from the literature with potential for use in lower-prevalence populations. Further research is needed to validate these biomarkers in primary care for screening and assessment of symptomatic patients.
  • Item
    Thumbnail Image
    An RCT of a decision aid to support informed choices about taking aspirin to prevent colorectal cancer and other chronic diseases: a study protocol for the SITA (Should I Take Aspirin?) trial
    Milton, S ; McIntosh, J ; Macrae, F ; Chondros, P ; Trevena, L ; Jenkins, M ; Walter, FM ; Taylor, N ; Boyd, L ; Saya, S ; Karnchanachari, N ; Novy, K ; Forbes, C ; Gutierrez, JM ; Broun, K ; Whitburn, S ; McGill, S ; Fishman, G ; Marker, J ; Shub, M ; Emery, J (BMC, 2021-07-15)
    BACKGROUND: Australian guidelines recommend that all people aged 50-70 years old actively consider taking daily low-dose aspirin (100-300 mg per day) for 2.5 to 5 years to reduce their risk of colorectal cancer (CRC). Despite the change of national CRC prevention guidelines, there has been no active implementation of the guidelines into clinical practice. We aim to test the efficacy of a health consultation and decision aid, using a novel expected frequency tree (EFT) to present the benefits and harms of low dose aspirin prior to a general practice consultation with patients aged 50-70 years, on informed decision-making and uptake of aspirin. METHODS: Approximately five to seven general practices in Victoria, Australia, will be recruited to participate. Patients 50-70 years old, attending an appointment with their general practitioner (GP) for any reason, will be invited to participate in the trial. Two hundred fifty-eight eligible participants will be randomly allocated 1:1 to intervention or active control arms using a computer-generated allocation sequence stratified by general practice, sex, and mode of trial delivery (face-to-face or teletrial). There are two co-primary outcomes: informed decision-making at 1-month post randomisation, measured by the Multi-dimensional Measure of Informed Choice (MMIC), and self-reported daily use of aspirin at 6 months. Secondary outcomes include decisional conflict at 1-month and other behavioural changes to reduce CRC risk at both time points. DISCUSSION: This trial will test the efficacy of novel methods for implementing national guidelines to support informed decision-making about taking aspirin in 50-70-year-olds to reduce the risk of CRC and other chronic diseases. TRIAL REGISTRATION: The Australian New Zealand Clinical Trials Registry (ANZCTR) ACTRN12620001003965 . Registered on 10 October 2020.
  • Item
    Thumbnail Image
    Using an electronic self-completion tool to identify patients at increased risk of melanoma in Australian primary care
    Habgood, E ; Walter, FM ; O'Hare, E ; McIntosh, J ; McCormack, C ; Emery, JD (WILEY, 2020-08)
    BACKGROUND/OBJECTIVES: Some international guidelines recommend a risk-based approach to screening for melanoma, but few suggest how to account for multiple risk factors or how to implement risk-based screening in practice. This study investigated the acceptability and feasibility of identifying patients at increased risk of melanoma in Australian general practice using a self-completed risk assessment tool. Stratification of risk was based on the validated Williams melanoma risk prediction model. METHODS: Patients and companions aged 18 or older in Australian general practices were approached in the waiting room and invited to enter information about their melanoma risk factors into the tool using an iPad. Acceptability was measured by the proportion of people willing to participate from those invited and feasibility by the number of people able to complete the tool unaided. Risk of developing melanoma was stratified into four risk categories using the Williams model. RESULTS: 1535 (90.4%) participants were recruited from two general practices. Only 200 participants (13%) needed assistance to complete the tool. The mean risk score for participants was 15.2 (±SD 9.8). The Williams model estimated between 5% and 19% of the sample were at increased risk accounting for an estimated 30% to 60% of future incident melanomas. CONCLUSIONS: A risk-stratified tool using the Williams model was acceptable and feasible for patients to self-complete in general practice clinics. This could be an effective way to identify people in primary care for implementing risk-based targeted melanoma screening and prevention.
  • Item
    Thumbnail Image
    Artificial Intelligence Techniques That May Be Applied to Primary Care Data to Facilitate Earlier Diagnosis of Cancer: Systematic Review
    Jones, OT ; Calanzani, N ; Saji, S ; Duffy, SW ; Emery, J ; Hamilton, W ; Singh, H ; de Wit, NJ ; Walter, FM (JMIR PUBLICATIONS, INC, 2021-03-03)
    BACKGROUND: More than 17 million people worldwide, including 360,000 people in the United Kingdom, were diagnosed with cancer in 2018. Cancer prognosis and disease burden are highly dependent on the disease stage at diagnosis. Most people diagnosed with cancer first present in primary care settings, where improved assessment of the (often vague) presenting symptoms of cancer could lead to earlier detection and improved outcomes for patients. There is accumulating evidence that artificial intelligence (AI) can assist clinicians in making better clinical decisions in some areas of health care. OBJECTIVE: This study aimed to systematically review AI techniques that may facilitate earlier diagnosis of cancer and could be applied to primary care electronic health record (EHR) data. The quality of the evidence, the phase of development the AI techniques have reached, the gaps that exist in the evidence, and the potential for use in primary care were evaluated. METHODS: We searched MEDLINE, Embase, SCOPUS, and Web of Science databases from January 01, 2000, to June 11, 2019, and included all studies providing evidence for the accuracy or effectiveness of applying AI techniques for the early detection of cancer, which may be applicable to primary care EHRs. We included all study designs in all settings and languages. These searches were extended through a scoping review of AI-based commercial technologies. The main outcomes assessed were measures of diagnostic accuracy for cancer. RESULTS: We identified 10,456 studies; 16 studies met the inclusion criteria, representing the data of 3,862,910 patients. A total of 13 studies described the initial development and testing of AI algorithms, and 3 studies described the validation of an AI algorithm in independent data sets. One study was based on prospectively collected data; only 3 studies were based on primary care data. We found no data on implementation barriers or cost-effectiveness. Risk of bias assessment highlighted a wide range of study quality. The additional scoping review of commercial AI technologies identified 21 technologies, only 1 meeting our inclusion criteria. Meta-analysis was not undertaken because of the heterogeneity of AI modalities, data set characteristics, and outcome measures. CONCLUSIONS: AI techniques have been applied to EHR-type data to facilitate early diagnosis of cancer, but their use in primary care settings is still at an early stage of maturity. Further evidence is needed on their performance using primary care data, implementation barriers, and cost-effectiveness before widespread adoption into routine primary care clinical practice can be recommended.
  • Item
    Thumbnail Image
    Measuring patient experience of diagnostic care and acceptability of testing
    Forster, AS ; Rubin, G ; Emery, JD ; Thompson, M ; Sutton, S ; de Wit, N ; Walter, FM ; Lyratzopoulos, G (WALTER DE GRUYTER GMBH, 2021-08)
    A positive patient experience has been long recognised as a key feature of a high-quality health service, however, often assessment of patient experience excludes diagnostic care. Experience of diagnostic services and the acceptability of diagnostic tests are often conflated, with lack of clarity about when and how either should be measured. These problems contrast with the growth in the development and marketing of new tests and investigation strategies. Building on the appraisal of current practice, we propose that the experience of diagnostic services and the acceptability of tests should be assessed separately, and describe distinct components of each. Such evaluations will enhance the delivery of patient-centred care, and facilitate patient choice.