CCCraig J. Currie

Craig J. Currie

*This is an algorithmically generated author record based on publicly available information.

Craig J. Currie has a total of 198 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    The descriptive epidemiology of type 2 diabetes in the United Kingdom from 2004 to 2021

    by Craig J. Currie, Jessica G. Currie, Sarah E. Holden, Christopher Ll. Morgan, Benjamin R. Heywood, John R. Peters, the Livingstone Development Team

    Purpose To provide contemporary estimates of the incidence and prevalence of type 2 diabetes in the UK. Methods Data from UK primary care (CPRD) were analyzed using an automated, analytical platform that produces validated, rapid analytics–Livingstone. Results We selected 1,125,028 people with type 2 diabetes (44.7% female). The crude incidence was stable. In 2004 the incidence rate was 4.18 cases per 1,000 people, reducing to 4.13/1,000 in 2021. There was a shift in age-specific incidence to earlier age-groups. The crude prevalence showed a marked increase, rising from 2.95% to 5.41%, levelling off slightly in later years. Increases in prevalence were observed in all groups. Northern Ireland consistently had the lowest prevalence, and Wales had a notably higher prevalence, increasing from 3.71% in 2004 to 6.37%. The mean age at diagnosis decreased in females by 5.5 years to 58.5 years, and by 2.5 years in men, to 58.8 years. Conclusions The incidence of type 2 diabetes was largely stable over 20 years but there were changes over time in age-specific groups resulting from a shift to earlier onset or earlier diagnosis. The shift to earlier diagnosis was far more pronounced in women than in men. Prevalence increased markedly, but the rate of increase appeared to be leveling-off a little. There were notable differences in the epidemiology of type 2 diabetes between the four constituent countries of the UK. The number of younger people diagnosed with type 2 diabetes remains a concern. Guarantor Craig Currie is the guarantor of this work and, as such, had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.


  • Open Access

    Predicting hepatic encephalopathy in patients with cirrhosis: A UK population–based study and validation of risk scores

    by Bethan I. Jones, Cerys A. Jenkins, Daniel Murphy, James Orr, Andrew Yeoman, Ellen R. Hubbuck, Ben R. Heywood, Craig J. Currie

    Background: HE is a common neurologic complication in cirrhosis associated with substantial disease and economic burden. HE symptoms are nonspecific and there are limited ways of identifying patients with cirrhosis at high risk of later developing HE. A risk score was previously developed to identify patients at risk of developing HE in a predominately male US cohort. Here, we evaluated the performance of the HE risk scores in a UK cohort study. Methods: Health care records from Clinical Practice Research Datalink and linked Hospital Episode Statistics were used to select patients with cirrhosis who were diagnosed with HE, confirmed by a diagnosis code for HE or a rifaximin-α prescription. The index date was the date of incident cirrhosis. The study period was from January 2003 to June 2019. Results: A total of 40,809 patients with cirrhosis were selected in the UK cohort, of whom 59% were male. A total of 1561 patients were diagnosed with HE. Applying the UK cohort to the baseline sensitivity risk cutoff (≥−11) from the US cohort provided a sensitivity of 92% and a negative predictive value of 99%. Within a longitudinal model, applying a sensitivity cutoff of ≥−3 to this cohort gave a sensitivity of 89% and a negative predictive value of 99%. Conclusions: Using data from the UK, the previously developed HE risk scores were found to be reliable for selecting those most likely to progress to HE in patients with liver cirrhosis. Despite the HE risk scores originally being estimated using the data from a predominately male US cohort, the scores were validated and found to be generalizable to female patients.


  • Open Access

    Real-world evidence from the first online healthcare analytics platform—Livingstone. Validation of its descriptive epidemiology module

    by Benjamin R. Heywood, Christopher Ll. Morgan, Thomas R. Berni, Darren R. Summers, Bethan I. Jones, Sara Jenkins-Jones, Sarah E. Holden, Lauren D. Riddick, Harry Fisher, James D. Bateman, et al.

    Incidence and prevalence are key epidemiological determinants characterizing the quantum of a disease. We compared incidence and prevalence estimates derived automatically from the first ever online, essentially real-time, healthcare analytics platform—Livingstone—against findings from comparable peer-reviewed studies in order to validate the descriptive epidemiology module. The source of routine NHS data for Livingstone was the Clinical Practice Research Datalink (CPRD). After applying a general search strategy looking for any disease or condition, 76 relevant studies were first retrieved, of which 10 met pre-specified inclusion and exclusion criteria. Findings reported in these studies were compared with estimates produced automatically by Livingstone. The published reports described elements of the epidemiology of 14 diseases or conditions. Lin’s concordance correlation coefficient (CCC) was used to evaluate the concordance between findings from Livingstone and those detailed in the published studies. The concordance of incidence values in the final year reported by each study versus Livingstone was 0.96 (95% CI: 0.89–0.98), whilst for all annual incidence values the concordance was 0.93 (0.91–0.94). For prevalence, concordance for the final annual prevalence reported in each study versus Livingstone was 1.00 (0.99–1.00) and for all reported annual prevalence values, the concordance was 0.93 (0.90–0.95). The concordance between Livingstone and the latest published findings was near perfect for prevalence and substantial for incidence. For the first time, it is now possible to automatically generate reliable descriptive epidemiology from routine health records, and in near-real time. Livingstone provides the first mechanism to rapidly generate standardised, descriptive epidemiology for all clinical events from real world data. Analyses of large datasets from patients’ care are now common. These studies are complex though. They normally require a team of people with a range of advanced skills. This is particularly true of UK NHS data. Because of these various hurdles, until now, automation of these types of healthcare study has never happened. Here we report the first automated analysis platform that does this, called Livingstone. Livingstone very quickly can carry out some types of health-related studies that has typically taken 12 to 24 months to complete. Here we report how we checked the output numbers generated by Livingstone versus corresponding numbers published in peer reviewed scientific studies. Livingstone performed extremely well.


  • The absolute risk of incident type 2 diabetes following exposure to systemic corticosteroids in selected steroid‐related and phenotypic groups

    by Philip Ambery, Samuel Adamsson Eryd, Sara Jenkins‐Jones, Ben Heywood, Ellen Berni, Mary N. Brown, Carol Astbury, Phillip Hunt, Craig J. Currie

    Aims: Exposure to corticosteroids is known to increase the risk of developing type 2 diabetes. We estimated the risk of incident type 2 diabetes in selected patient groups exposed to systemic corticosteroids. Materials and methods: In a retrospective, observational cohort study, using real‐world data from UK primary care, patients were selected who had at least one episode of exposure to oral or intravenous corticosteroids for any indication. Corticosteroid‐exposed patients were matched with non‐exposed patients. Relative dosage was estimated as a weight‐based, prednisolone‐equivalent dose. Crude rates of progression to type 2 diabetes were determined for patient groups defined by relevant steroid‐related and phenotypic characteristics present at corticosteroid exposure. Results: Overall, rates of incidence of type 2 diabetes were 12.5 and 6.7 events per thousand person‐years' (pkpy) exposure, respectively, in those who received at least one dose of corticosteroids versus those never exposed. This represented a rate ratio of 1.85 (95% CI 1.74‐1.97). The incidence of type 2 diabetes was found to be associated with several of the selected characteristics, both individually and multi‐dimensionally. The highest rate of incident type 2 diabetes was observed in very severely obese men aged 46‐55 years having had the longest corticosteroid exposure and highest corticosteroid dose (190 incident events pkpy exposure). Conclusions: Corticosteroid exposure increased the risk of incident type 2 diabetes, and there was evidence of both a dose‐response and a duration response. The impact of corticosteroid exposure upon the rate of incident type 2 diabetes appeared, however, to involve a complex, multi‐dimensional interaction between the selected characteristics, some of which might be impacted by reverse causality.


  • Comorbid diseases and conditions in people with HIV in the UK

    by Bethan I. Jones, Andrew Freedman, Melissa J. Thomas, Celia Villalba-Mendez, Leena Sathia, Deborah Flanagan, Scott Francis, Craig J. Currie

    This study aimed to characterize the risk of people living with HIV (PLHIV) in the UK progressing to pre-specified HIV-associated comorbidities, compared with matched, HIV-negative controls.Primary...


  • Retinal thickness fluctuations in patients receiving fluocinolone acetonide implant for diabetic macular edema

    by Sarah E. Holden, Maged Habib, Craig J. Currie

    To evaluate central foveal thickness (CFT) variability and accompanying changes in visual acuity (VA) 12 months before and after treatment with the 190 mcg fluocinolone acetonide (FAc) intravitreal...


  • Evaluation of the epidemiology of peanut allergy in the United Kingdom

    by Laura A. Scott, Bethan I. Jones, Thomas R. Berni, Ellen R. Berni, Jane De Vries, Craig J. Currie

    Aims: To describe the epidemiology of peanut allergy (PA) in the UK over the last three decades. Methods: PA patients were identified from the Clinical Practice Research Datalink between 1987 and 2015. Incidence and prevalence of PA were compared between 2000 and 2015. Prevalence and relative risk (RR) of atopic comorbidities, anaphylaxis, adrenaline prescriptions versus matched controls were calculated. Results: Point prevalence of PA in the entire population and those <18 years increased from 31 to 202 and 116 to 635 per 100,000, respectively, between 2000 and 2015. Incidence increased from 8.6 to 18.2 per 100,000. Incidence in 2015 was 105 cases per 100,000 aged 0–4 years versus 13.4 per 100,000 aged 5+ years. Anaphylactic events affected 1.2% of the cases and 0.007% of the controls. The rate of adrenaline prescriptions was 5,910 per 100,000 person-years for PA patients. RRs for asthma, eczema and allergic rhinitis in PA patients versus controls were 4.5 (95% CI 4.2–4.8), 3.2 (3.1–3.4) and 2.6 (2.4–3.0), respectively. Conclusions: The prevalence and incidence of PA increased markedly over the study period. PA was associated with atopic conditions and anaphylaxis. PA patients had increased adrenaline prescriptions.


  • Cough presentation in primary care and the identification of chronic cough: a need for diagnostic clarity?

    by Sarah E. Holden, Alyn Morice, Surinder S. Birring, Sara Jenkins-Jones, Haya Langerman, Jessica Weaver, Craig J. Currie

    Objective: To investigate patterns of presentation of cough in primary care and develop an algorithm to identify probable and possible chronic cough (CC). Methods: This retrospective observational study used routine English primary care data and linked hospital data. Patients with ≥1 cough event in the study period (March 2014–February 2015) were selected. Index date was that of the earliest cough event in this period. Adults (aged ≥18 years) were classified as having probable CC if they had an explicit CC diagnosis; as having possible CC if they had ≥3 cough events recorded over 8–26 weeks; or, otherwise, as having acute cough. Underlying conditions associated with CC were identified. Results: 198,151 people were identified. 56.5% were female; median age was 47.0 years. The prevalence of cough in the study year was 17.6%. Of the 150,213 identified adults, 1,600 (1.1%), 10,913 (7.3%) and 137,718 (91.7%) were classified as having probable CC, possible CC or acute cough, respectively. Compared with probable CC and acute cough, a higher percentage of possible CC cases had a post-index record indicative of chronic obstructive pulmonary disease (30.6% versus 10.1% and 9.7%), gastro-esophageal reflux disease (32.6% versus 24.9% and 21.1%) or asthma (45.9% versus 27.6% and 27.9%). Prevalences of probable and possible CC were 0.18% and 1.2%, respectively. Conclusions: The prevalence of CC was lower than reported in previous studies. People with possible CC had higher rates of underlying conditions associated with CC. These observations may suggest poor recognition and/or under-recording of CC in primary care.


  • Open Access

    The Model of Mortality with Incident Cirrhosis (MoMIC) and the model of Long-term Outlook of Mortality in Cirrhosis (LOMiC)

    by Ellen R Berni, Bethan I Jones, Thomas R Berni, James Whitehouse, Mark Hudson, James Orr, Pete Conway, Bharat Amlani, Craig J. Currie

    The purpose of this study was to produce two statistical survival models in those with cirrhosis utilising only routine parameters, including non-liver-related clinical factors that influence survival. The first model identified and utilised factors impacting short-term survival to 90-days post incident diagnosis, and a further model characterised factors that impacted survival following this acute phase. Data were from the Clinical Practice Research Datalink linked with Hospital Episode Statistics. Incident cases in patients ≥18 years were identified between 1998 and 2014. Patients that had prior history of cancer or had received liver transplants prior were excluded. Model-1 used a logistic regression model to predict mortality. Model-2 used data from those patients who survived 90 days, and used an extension of the Cox regression model, adjusting for time-dependent covariables. At 90 days, 23% of patients had died. Overall median survival was 3.7 years. Model-1: numerous predictors, prior comorbidities and decompensating events were incorporated. All comorbidities contributed to increased odds of death, with renal disease having the largest adjusted odds ratio (OR = 3.35, 95%CI 2.97–3.77). Model-2: covariables included cumulative admissions for liver disease-related events and admissions for infections. Significant covariates were renal disease (adjusted hazard ratio (HR = 2.89, 2.47–3.38)), elevated bilirubin levels (aHR = 1.38, 1.26–1.51) and low sodium levels (aHR = 2.26, 1.84–2.78). An internal validation demonstrated reliability of both models. In conclusion: two survival models that included parameters commonly recorded in routine clinical practice were generated that reliably forecast the risk of death in patients with cirrhosis: in the acute, post diagnosis phase, and following this critical, 90 day phase. This has implications for practice and helps better forecast the risk of mortality from cirrhosis using routinely recorded parameters without inputs from specialists.


  • Open Access

    Major adverse cardiovascular events in people with chronic kidney disease in relation to disease severity and diabetes status

    by Craig J. Currie, Ellen R. Berni, Thomas R. Berni, Sara Jenkins-Jones, Marvin Sinsakul, Lutz Jermutus, Philip Ambery, Meena Jain

    Diabetes plays an important role in the complex relationship between chronic kidney disease (CKD) and cardiovascular disease. This retrospective observational study compared the influence of estimated glomerular filtration rate (eGFR) and proteinuria on the risk of major adverse cardiovascular event (MACE; myocardial infarction or stroke) in CKD patients with and without diabetes. Data were from a linked database of UK electronic health records. Individuals with CKD and no prior MACE were classified as type 1 diabetes (T1DM; n = 164), type 2 diabetes (T2DM; n = 9,711), and non-diabetes (non-DM; n = 75,789). Monthly updated time-dependent Cox proportional hazard models were constructed to calculate adjusted hazard ratios (aHRs) for progression to MACE from first record of abnormal eGFR or proteinuria (index date). In non-DM, aHRs (95% CIs) by baseline eGFR category (referent G2) were G1: 0.70 (0.55–0.90), G3a: 1.28 (1.20–1.35), G3b: 1.64 (1.52–1.76), G4: 2.19 (1.98–2.43), and G5: 3.12 (2.44–3.99), and by proteinuria category (referent A1) were A2: 1.13 (1.00–1.28), A2/3 (severity indeterminable): 1.58 (1.28–1.95), and A3: 1.64 (1.38–1.95). In T2DM, aHRs were G1: 0.98 (0.72–1.32), G3a: 1.18 (1.03–1.34), G3b: 1.31 (1.12–1.54), G4: 1.87 (1.53–2.29), G5: 2.87 (1.82–4.52), A2: 1.22 (1.04–1.42), A2/3: 1.45 (1.17–1.79), and A3: 1.82 (1.53–2.16). Low numbers in T1DM precluded analysis. Modelling T2DM and non-DM together, aHRs were, respectively, G1: 3.23 (2.38–4.40) and 0.70 (0.55–0.89); G2: 3.18 (2.73–3.70) and 1.00 (referent); G3a: 3.65 (3.13–4.25) and 1.28 (1.21–1.36); G3b: 4.01 (3.40–4.74) and 1.65 (1.54–1.77); G4: 5.78 (4.70–7.10) and 2.21 (2.00–2.45); G5: 9.00 (5.71–14.18) and 3.14 (2.46–4.00). In conclusion, reduced eGFR and proteinuria were independently associated with increased risk of MACE regardless of diabetes status. However, the risk of MACE in the same eGFR state was 4.6–2.4 times higher in T2DM than in non-DM.