Corri Black

Validity of the General Practice Research Database

The United Kingdom General Practice Research Database (GPRD) is an office‐based, computer‐generated, medical resource designed from its inception to be used for epidemiologic research. A distinct version of the GPRD is maintained by the Boston Collaborative Drug Surveillance Program and has been the source of more than 130 scientific articles primarily addressing drug safety issues. We reviewed evidence related to the validity of the GPRD. Specifically, with our extensive experience with this automated database, we evaluated the quality and completeness of the data that it contains.

RELATION OF CHILDHOOD GASTROINTESTINAL DISORDERS TO AUTISM: NESTED CASE-CONTROL STUDY USING DATA FROM THE UK GENERAL PRACTICE RESEARCH DATABASE

Abstract Objectives: To assess whether children with autism are more likely to have a history of gastrointestinal disorders than children without autism. Design: Nested case-control study. Setting: UK General Practice Research Database. Subjects: Children born after 1 January 1988 and registered with the General Practice Research Database within 6 months of birth. Outcome measures: Chronic inflammation of the gastrointestinal tract, coeliac disease, food intolerance, and recurrent gastrointestinal symptoms recorded by the general practitioner. Results: 9 of 96 (9%) children with a diagnosis of autism (cases) and 41 of 449 (9%) children without autism (matched controls) had a history of gastrointestinal disorders before the index date (the date of first recorded diagnosis of autism in the cases and the same date for controls). The estimated odds ratio for a history of gastrointestinal disorders among children with autism compared with children without autism was 1.0 (95% confidence interval 0.5 to 2.2). Conclusions: No evidence was found that children with autism were more likely than children without autism to have had defined gastrointestinal disorders at any time before their diagnosis of autism.

Associations of estimated glomerular filtration rate and albuminuria with mortality and renal failure by sex : a meta-analysis

Objective To assess for the presence of a sex interaction in the associations of estimated glomerular filtration rate and albuminuria with all-cause mortality, cardiovascular mortality, and end stage renal disease. Design Random effects meta-analysis using pooled individual participant data. Setting 46 cohorts from Europe, North and South America, Asia, and Australasia. Participants 2 051 158 participants (54% women) from general population cohorts (n=1 861 052), high risk cohorts (n=151 494), and chronic kidney disease cohorts (n=38 612). Eligible cohorts (except chronic kidney disease cohorts) had at least 1000 participants, outcomes of either mortality or end stage renal disease of ≥50 events, and baseline measurements of estimated glomerular filtration rate according to the Chronic Kidney Disease Epidemiology Collaboration equation (mL/min/1.73 m2) and urinary albumin-creatinine ratio (mg/g). Results Risks of all-cause mortality and cardiovascular mortality were higher in men at all levels of estimated glomerular filtration rate and albumin-creatinine ratio. While higher risk was associated with lower estimated glomerular filtration rate and higher albumin-creatinine ratio in both sexes, the slope of the risk relationship for all-cause mortality and for cardiovascular mortality were steeper in women than in men. Compared with an estimated glomerular filtration rate of 95, the adjusted hazard ratio for all-cause mortality at estimated glomerular filtration rate 45 was 1.32 (95% CI 1.08 to 1.61) in women and 1.22 (1.00 to 1.48) in men (Pinteraction<0.01). Compared with a urinary albumin-creatinine ratio of 5, the adjusted hazard ratio for all-cause mortality at urinary albumin-creatinine ratio 30 was 1.69 (1.54 to 1.84) in women and 1.43 (1.31 to 1.57) in men (Pinteraction<0.01). Conversely, there was no evidence of a sex difference in associations of estimated glomerular filtration rate and urinary albumin-creatinine ratio with end stage renal disease risk. Conclusions Both sexes face increased risk of all-cause mortality, cardiovascular mortality, and end stage renal disease with lower estimated glomerular filtration rates and higher albuminuria. These findings were robust across a large global consortium.

Measuring the population burden of chronic kidney disease: a systematic literature review of the estimated prevalence of impaired kidney function

BACKGROUND Internationally, there have been substantial efforts to improve the early identification of chronic kidney disease (CKD), with a view to improving survival, reducing progression and minimizing cardiovascular morbidity and mortality. In 2002, a new and globally adopted definition of CKD was introduced. The burden of kidney function impairment in the population is unclear and widely ranging prevalence estimates have been reported. METHODS We conducted a systematic literature review, searching databases to June 2009. We included all adult population screening studies and studies based on laboratory or clinical datasets where the denominator was clear. Studies reporting prevalence estimates based on at least one eGFR <60 mL/min/1.73m(2) or elevated creatinine above a stated threshold were included. Study design and quality were explored as potential factors leading to heterogeneity. RESULTS We identified 43 eligible studies (57 published reports) for inclusion. Substantial heterogeneity was observed with estimated prevalence (0.6-42.6%). The included studies demonstrated significant variation in methodology and quality that impacted on the comparability of their findings. From the higher quality studies, the six studies measuring impaired kidney function (iKF) using estimated glomerular filtration rate in community screening samples reported a prevalence ranging from 1.7% in a Chinese study to 8.1% in a US study, with four reporting an estimated prevalence of 3.2-5.6%. Heterogeneity was driven by the measure used, study design and study population. CONCLUSION In the general population, estimated iKF, particularly eGFR 30-59 mL/min/1.73m(2) was common with prevalence similar to diabetes mellitus. Appropriate care of patients poses a substantial global health care challenge.

Incidence of thrombotic thrombocytopenic purpura/hemolytic uremic syndrome.

Background: Thrombotic thrombocytopenic purpura and hemolytic uremic syndrome are rare disorders characterized by platelet aggregation, microthrombi, and resulting tissue damage. We studied the incidence and possible risk factors for these diseases in 3 large populations in the United States, United Kingdom, and Canada. Methods: Data were derived from a large health insurer in the United States, general practices in the United Kingdom, and the Province of Saskatchewan. We identified potential cases of thrombotic thrombocytopenia purpura and hemolytic uremic syndrome in computerized data and verified them by medical record review. We estimated incidence rates for thrombotic thrombocytopenia purpura and hemolytic uremic syndrome together and separately, and we conducted a case-control study to evaluate potential risk factors. Results: The age–sex standardized incidence of thrombotic thrombocytopenia purpura and hemolytic uremic syndrome was higher than previously reported (6.5, 2.2, and 3.2 per million per year in the United States, United Kingdom, and Saskatchewan, respectively), but there was no secular trend. The incidence of thrombotic thrombocytopenia purpura and hemolytic uremic syndrome was higher in women than men. Most cases of hemolytic uremic syndrome occurred before 20 years of age. We confirmed several known risk factors for thrombotic thrombocytopenia purpura and hemolytic uremic syndrome (cancer, bone marrow transplantation, pregnancy). Conclusion: The incidence of thrombotic thrombocytopenia purpura and hemolytic uremic syndrome is higher than previously reported but does not appear to be rising. Apparent international differences in incidence could be the result of imprecision in identifying thrombotic thrombocytopenia purpura and hemolytic uremic syndrome in large research databases.

Long-term prognosis after acute kidney injury (AKI): what is the role of baseline kidney function and recovery? A systematic review

Objectives To summarise the evidence from studies of acute kidney injury (AKI) with regard to the effect of pre-AKI renal function and post-AKI renal function recovery on long-term mortality and renal outcomes, and to assess whether these factors should be taken into account in future prognostic studies. Design/Setting A systematic review of observational studies listed in Medline and EMBASE from 1990 to October 2012. Participants All AKI studies in adults with data on baseline kidney function to identify AKI; with outcomes either stratified by pre-AKI and/or post-AKI kidney function, or described by the timing of the outcomes. Outcomes Long-term mortality and worsening chronic kidney disease (CKD). Results Of 7385 citations, few studies met inclusion criteria, reported baseline kidney function and stratified by pre-AKI or post-AKI function. For mortality outcomes, three studies compared patients by pre-AKI renal function and six by post-AKI function. For CKD outcomes, two studies compared patients by pre-AKI function and two by post-AKI function. The presence of CKD pre-AKI (compared with AKI alone) was associated with doubling of mortality and a fourfold to fivefold increase in CKD outcomes. Non-recovery of kidney function was associated with greater mortality and CKD outcomes in some studies, but findings were inconsistent varying with study design. Two studies also reported that risk of poor outcome reduced over time post-AKI. Meta-analysis was precluded by variations in definitions for AKI, CKD and recovery. Conclusions The long-term prognosis after AKI varies depending on cause and clinical setting, but it may also, in part, be explained by underlying pre-AKI and post-AKI renal function rather than the AKI episode itself. While carefully considered in clinical practice, few studies address these factors and with inconsistent study design. Future AKI studies should report pre-AKI and post-AKI function consistently as additional factors that may modify AKI prognosis.

Cesarean delivery in the United Kingdom : Time trends in the general practice research database

Objective: To estimate recent temporal trends in delivery by cesarean during the past decade and the proportion of vaginal deliveries after prior caesarean in the United Kingdom. Methods: We conducted a cohort study using information from the General Practice Research Database. We identified all women with 1 or more deliveries between January 1990 and December 1999 and determined the method(s) of delivery. We estimated the proportion of women with vaginal delivery after cesarean in a subcohort who had at least 3 years of follow-up. Results: We identified 39,938 cesareans among 271,663 deliveries (14.7%), with an increase from 12.5% in 1990 to 18.3% in 1999. The proportion of cesarean deliveries increased with age and increased over time in all age groups except women aged younger than 20 years. Among 26,480 women with a caesarean delivery between 1990 and 1996, 7,649 (28.9%) had a subsequent delivery. The proportion of vaginal delivery after prior cesarean decreased from 45% in 1991 to 37% in 1999. Conclusion: Cesarean deliveries increased as a proportion of all deliveries in the United Kingdom during the past decade, and the proportion of vaginal delivery after prior cesarean decreased. Still, the proportion of cesarean deliveries is lower and the proportion of vaginal deliveries after prior cesarean is higher in the United Kingdom than in the United States. Level of Evidence: II-2

Population screening for lung cancer using computed tomography, is there evidence of clinical effectiveness? A systematic review of the literature

Lung cancer is the leading cause of death among all cancer types in the UK, killing approximately 34 000 people per year. By the time symptoms develop, the tumour is often at an advanced stage and the prognosis is bleak. Treatment at a less advanced stage of disease by surgical resection has been shown to substantially reduce mortality. Screening would be attractive if it could detect presymptomatic lung cancer at a stage when surgical intervention is feasible but has been the subject of scientific debate for the past three decades. The aim of this review was to examine the current evidence on the clinical effectiveness of screening for lung cancer using computed tomography. A systematic literature review searching 15 electronic databases and Internet resources from 1994 until December 2004/January 2005 was carried out. Information was summarised narratively. A total of 12 studies of computed tomography screening for lung cancer were identified including two RCTs and 10 studies of screening without comparator groups. The two RCTs were of short duration (1 year). None examined the effect of screening on mortality compared with no screening. The proportion of people with abnormal computed tomography findings varied widely between studies (5–51%). The prevalence of lung cancer detected was between 0.4% and 3.2% (number needed to screen to detect one lung cancer  = 31 to 249). Incidence rates of lung cancer were lower (0.1–1%). Among the detected tumours, a high proportion were stage I or resectable tumours, 100% in some studies. Currently, there is insufficient evidence that computed tomography screening is clinically effective in reducing mortality from lung cancer.

Acute kidney injury—how does automated detection perform?

Background Early detection of acute kidney injury (AKI) is important for safe clinical practice. NHS England is implementing a nationwide automated AKI detection system based on changes in blood creatinine. Little has been reported on the similarities and differences of AKI patients detected by this algorithm and other definitions of AKI in the literature. Methods We assessed the NHS England AKI algorithm and other definitions using routine biochemistry in our own health authority in Scotland in 2003 (adult population 438 332). Linked hospital episode codes (ICD-10) were used to identify patients where AKI was a major clinical diagnosis. We compared how well the algorithm detected this subset of AKI patients in comparison to other definitions of AKI. We also evaluated the potential ‘alert burden’ from using the NHS England algorithm in comparison to other AKI definitions. Results Of 127 851 patients with at least one blood test in 2003, the NHS England AKI algorithm identified 5565 patients. The combined NHS England algorithm criteria detected 91.2% (87.6–94.0) of patients who had an ICD-10 AKI code and this was better than any individual AKI definition. Some of those not captured could be identified by algorithm modifications to identify AKI in retrospect after recovery, but this would not be practical in real-time. Any modifications also increased the number of alerted patients (2-fold in the most sensitive model). Conclusions The NHS England AKI algorithm performs well as a diagnostic adjunct in clinical practice. In those without baseline data, AKI may only be seen in biochemistry in retrospect, therefore proactive clinical care remains essential. An alternative algorithm could increase the diagnostic sensitivity, but this would also produce a much greater burden of patient alerts.

KDIGO-based acute kidney injury criteria operate differently in hospitals and the community—findings from a large population cohort

Background Early recognition of acute kidney injury (AKI) is important. It frequently develops first in the community. KDIGO-based AKI e-alert criteria may help clinicians recognize AKI in hospitals, but their suitability for application in the community is unknown. Methods In a large renal cohort (n = 50 835) in one UK health authority, we applied the NHS England AKI ‘e-alert’ criteria to identify and follow three AKI groups: hospital-acquired AKI (HA-AKI), community-acquired AKI admitted to hospital within 7 days (CAA-AKI) and community-acquired AKI not admitted within 7 days (CANA-AKI). We assessed how AKI criteria operated in each group, based on prior blood tests (number and time lag). We compared 30-day, 1- and 5-year mortality, 90-day renal recovery and chronic renal replacement therapy (RRT). Results In total, 4550 patients met AKI e-alert criteria, 61.1% (2779/4550) with HA-AKI, 22.9% (1042/4550) with CAA-AKI and 16.0% (729/4550) with CANA-AKI. The median number of days since last blood test differed between groups (1, 52 and 69 days, respectively). Thirty-day mortality was similar for HA-AKI and CAA-AKI, but significantly lower for CANA-AKI (24.2, 20.2 and 2.6%, respectively). Five-year mortality was high in all groups, but followed a similar pattern (67.1, 64.7 and 46.2%). Differences in 5-year mortality among those not admitted could be explained by adjusting for comorbidities and restricting to 30-day survivors (hazard ratio 0.91, 95% confidence interval 0.80–1.04, versus hospital AKI). Those with CANA-AKI (versus CAA-AKI) had greater non-recovery at 90 days (11.8 versus 3.5%, P < 0.001) and chronic RRT at 5 years (3.7 versus 1.2%, P < 0.001). Conclusions KDIGO-based AKI criteria operate differently in hospitals and in the community. Some patients may not require immediate admission but are at substantial risk of a poor long-term outcome.