Search

Background Epigenetic clocks are biomarkers of ageing derived from DNA methylation levels at a subset of CpG sites. The difference between age predicted by these clocks and chronological age, termed “epigenetic age acceleration”, has been shown to predict age-related disease and mortality. We aimed to assess the prognostic value of epigenetic age acceleration and a DNA methylation-based mortality risk score with all-cause mortality in a prospective clinical cohort of individuals with head and neck cancer: Head and Neck 5000. We investigated two markers of intrinsic epigenetic age acceleration (IEAAHorvath and IEAAHannum), one marker of extrinsic epigenetic age acceleration (EEAA), one optimised to predict physiological dysregulation (AgeAccelPheno), one optimised to predict lifespan (AgeAccelGrim) and a DNA methylation-based predictor of mortality (ZhangScore). Cox regression models were first used to estimate adjusted hazard ratios (HR) and 95% confidence intervals (CI) for associations of epigenetic age acceleration with all-cause mortality in people with oropharyngeal cancer (n = 408; 105 deaths). The added prognostic value of epigenetic markers compared to a clinical model including age, sex, TNM stage and HPV status was then evaluated. Results IEAAHannum and AgeAccelGrim were associated with mortality risk after adjustment for clinical and lifestyle factors (HRs per standard deviation [SD] increase in age acceleration = 1.30 [95% CI 1.07, 1.57; p = 0.007] and 1.40 [95% CI 1.06, 1.83; p = 0.016], respectively). There was weak evidence that the addition of AgeAccelGrim to the clinical model improved 3-year mortality prediction (area under the receiver operating characteristic curve: 0.80 vs. 0.77; p value for difference = 0.069). Conclusion In the setting of a large, clinical cohort of individuals with head and neck cancer, our study demonstrates the potential of epigenetic markers of ageing to enhance survival prediction in people with oropharyngeal cancer, beyond established prognostic factors. Our findings have potential uses in both clinical and non-clinical contexts: to aid treatment planning and improve patient stratification.

Supplemental Digital Content is available in the text, Introduction: HIV cohort data from high-income European countries were compared with the UNAIDS Spectrum modelling parameters for these same countries to validate mortality rates and excess mortality estimates for people living with HIV (PLHIV) on antiretroviral therapy (ART). Methods: Data from 2000 to 2015 were analysed from the Antiretroviral Therapy Cohort Collaboration (ART-CC) for Austria, Denmark, France, Italy, the Netherlands, Spain, and Switzerland. Flexible parametric models were used to compare all-cause mortality rates in the ART-CC and Spectrum. The percentage of AIDS-related deaths and excess mortality (both are the same within Spectrum) were compared, with excess mortality defined as that in excess of the general population mortality. Results: Analyses included 94 026 PLHIV with 585 784 person-years of follow-up, from which there were 5515 deaths. All-cause annual mortality rates in Spectrum for 2000–2003 were 0.0121, reducing to 0.0078 in 2012–2015, whilst the ART-CC's corresponding annual mortality rates were 0.0151 [95% confidence interval (95% CI): 0.0130–0.0171] reducing to 0.0049 (95% CI: 0.0039–0.0060). The percentage of AIDS-related deaths in Spectrum was 74.7% in 2000–2003, dropping to 43.6% in 2012–2015. In the ART-CC, AIDS-related mortality constitutes 45.3% (95% CI: 38.4–52.9%) of mortality in 2000–2003 and 26.7% (95% CI: 19–46%) between 2012 and 2015. Excess mortality in the ART-CC was broadly similar to the Spectrum estimates, dropping from 75.3% (95% CI: 60.3–95.2%) in 2000–2003 to 30.7% (95% CI: 25.5–63.7%) in 2012–2015. Conclusion: All-cause mortality assumptions for PLHIV on ART in high-income European settings should be adjusted in Spectrum to be higher in 2000–2003 and decline more quickly to levels currently captured for recent years.

In patient follow-up studies, events of interest may take place between periodic clinical assessments and so the exact time of onset is not observed. Such events are known as ‘bounded’ or ‘interval-censored’. Methods for handling such events can be categorised as either (i) applying multiple imputation (MI) strategies or (ii) taking a full likelihood-based (LB) approach. We focused on MI strategies, rather than LB methods, because of their flexibility. We evaluated MI strategies for bounded event times in a competing risks analysis, examining the extent to which interval boundaries, features of the data distribution and substantive analysis model are accounted for in the imputation model. Candidate imputation models were predictive mean matching (PMM); log-normal regression with postimputation back-transformation; normal regression with and without restrictions on the imputed values and Delord and Genin’s method1 based on sampling from the cumulative incidence function. We used a simulation study to compare MI methods and one LB method when data were missing at random and missing not at random, also varying the proportion of missing data, and then applied the methods to a haematopoietic stem cell transplantation dataset. We found that cumulative incidence and median event time estimation were sensitive to model misspecification. In a competing risks analysis, we found that it is more important to account for features of the data distribution than to restrict imputed values based on interval boundaries or to ensure compatibility with the substantive analysis by sampling from the cumulative incidence function. We recommend MI by type 1 PMM.

An amendment to this paper has been published and can be accessed via a link at the top of the paper.

Background and aims Recent research demonstrated that obesity and high dietary sodium intake, the two established risk factors for hypertension, were associated with each other. The objective was to investigate the potential indirect effect of sodium intake on blood pressure via body mass index (BMI). Methods and results Using ten years data from US NHANES (2007–2016), the study included adult participants (>20 years old) who were not taking antihypertensive medications and without baseline diseases (n = 12,262). BMI was modelled as the mediator of sodium intake on systolic and diastolic blood pressure, adjusted for age, sex, socioeconomic status, smoking, drinking, physical activity, calorie intake, fluid intake and potassium intake. Mediation analysis was performed to evaluate total effect, direct effect and indirect effect via BMI. Subgroup analyses based on three age subgroups (20–40, 41–60 and ≥61 years old) were performed. The mean age was 39.29 (13.4) years and 53.1 (0.45) % were males. The mean BMI was 27.8 (6.20) kg/m2. Overall, 1 g/d increase in sodium intake was associated with an increased systolic blood pressure by 0.36 (95% confidence interval 0.14 to 0.58) mmHg, with a direct effect (0.14 (0.09–0.19)) and an indirect effect via BMI (0.23 (0.02–0.44)). The indirect effect was mainly observed in participants ≤60 years old. Conclusion Sodium intake showed both direct effect and indirect effect (via BMI) on systolic blood pressure in US NHANES. The findings provide evidence for combining sodium restriction and weight reduction measures for prevention of hypertension. Cautions should be taken when generalizing the findings to other populations with lower average BMI.

ObjectivesTo confirm the association of previously reported prognostic factors with future progression of localised prostate cancer using primary care data and identify new potential prognostic factors for further assessment in prognostic model development and validation.DesignRetrospective cohort study, employing Cox proportional hazards regression controlling for age, prostate specific antigen (PSA), and Gleason score, was stratified by diagnostic stage.SettingPrimary care in England.ParticipantsMales with localised prostate cancer diagnosedbetween 01/01/1987 and 31/12/2016 within the Clinical Practice ResearchDatalink database, with linked data from the National Cancer Registration andAnalysis Service and Office for National Statistics.Primary and secondary outcomesPrimary outcome measure was prostate cancer mortality. Secondary outcome measures were all-cause mortality and commencing systemic therapy. Up-staging after diagnosis was not used as a secondary outcome owing to significant missing data.Results10 901 men (mean age 74.38±9.03 years) with localised prostate cancer were followed up for a mean of 14.12 (±6.36) years. 2331 (21.38%) men underwent systemic therapy and 3450 (31.65%) died, including 1250 (11.47%) from prostate cancer. Factors associated with an increased risk of prostate cancer mortality included age; high PSA; current or ex-smoker; ischaemic heart disease; high C reactive protein; high ferritin; low haemoglobin; high blood glucose and low albumin.ConclusionsThis study identified several new potential prognostic factors for prostate cancer progression, as well as confirming some known prognostic factors, in an independent primary care data set. Further research is needed to develop and validate a prognostic model for prostate cancer progression.

Background Chronic fatigue syndrome (CFS) is relatively common and disabling. Over 8000 patients attend adult services each year, yet little is known about the outcome of patients attending NHS services. Aim Investigate the outcome of patients with CFS and what factors predict outcome. Design Longitudinal patient cohort. Methods We used data from six CFS/ME (myalgic encephalomyelitis) specialist services to measure changes in fatigue (Chalder Fatigue Scale), physical function (SF-36), anxiety and depression (Hospital Anxiety and Depression Scale) and pain (visual analogue pain rating scale) between clinical assessment and 8–20 months of follow-up. We used multivariable linear regression to investigate baseline factors associated with outcomes at follow-up. Results Baseline data obtained at clinical assessment were available for 1643 patients, of whom 834 (51%) had complete follow-up data. There were improvements in fatigue [mean difference from assessment to outcome: −6.8; 95% confidence interval (CI) −7.4 to −6.2; P < 0.001]; physical function (4.4; 95% CI 3.0–5.8; P < 0.001), anxiety (−0.6; 95% CI −0.9 to −0.3; P < 0.001), depression (−1.6; 95% CI −1.9 to −1.4; P < 0.001) and pain (−5.3; 95% CI −7.0 to −3.6; P < 0.001). Worse fatigue, physical function and pain at clinical assessment predicted a worse outcome for fatigue at follow-up. Older age, increased pain and physical function at assessment were associated with poorer physical function at follow-up. Conclusions Patients who attend NHS specialist CFS/ME services can expect similar improvements in fatigue, anxiety and depression to participants receiving cognitive behavioural therapy and graded exercise therapy in a recent trial, but are likely to experience less improvement in physical function. Outcomes were predicted by fatigue, disability and pain at assessment.

ObjectiveTo estimate the costs and outcomes associated with treating non-asthmatic adults (nor suffering from other lung-disease) presenting to primary care with acute lower respiratory tract infection (ALRTI) with oral corticosteroids compared with placebo.DesignCost-consequence analysis alongside a randomised controlled trial. Perspectives included the healthcare provider, patients and productivity losses associated with time off work.SettingFifty-four National Health Service (NHS) general practices in England.Participants398 adults attending NHS primary practices with ALRTI but no asthma or other chronic lung disease, followed up for 28 days.Interventions2× 20 mg oral prednisolone per day for 5 days versus matching placebo tablets.Outcome measuresQuality-adjusted life years using the 5-level EuroQol-5D version measured weekly; duration and severity of symptom. Direct and indirect resources related to the disease and its treatment were also collected. Outcomes were measured for the 28-day follow-up.Results198 (50%) patients received the intervention (prednisolone) and 200 (50%) received placebo. NHS costs were dominated by primary care contacts, higher with placebo than with prednisolone (£13.11 vs £10.38) but without evidence of a difference (95% CI £3.05 to £8.52). The trial medication cost of £1.96 per patient would have been recouped in prescription charges of £4.30 per patient overall (55% participants would have paid £7.85), giving an overall mean ‘profit’ to the NHS of £7.00 (95% CI £0.50 to £17.08) per patient. There was a quality adjusted life years gain of 0.03 (95% CI 0.01 to 0.05) equating to half a day of perfect health favouring the prednisolone patients; there was no difference in duration of cough or severity of symptoms.ConclusionsThe use of prednisolone for non-asthmatic adults with ALRTI, provided small gains in quality of life and cost savings driven by prescription charges. Considering the results of the economic evaluation and possible side effects of corticosteroids, the short-term benefits may not outweigh the long-term harms.Trial registration numbersEudraCT 2012-000851-15 and ISRCTN57309858; Pre-results.

People living with HIV (PLHIV) are more likely than the general population to develop AIDS‐defining malignancies (ADMs) and several non‐ADMs (NADMs). Information is lacking on survival outcomes and cause‐specific mortality after cancer diagnosis among PLHIV. We investigated causes of death within 5 years of cancer diagnosis in PLHIV enrolled in European and North American HIV cohorts starting antiretroviral therapy (ART) 1996–2015, aged ≥16 years, and subsequently diagnosed with cancer. Cancers were grouped: ADMs, viral NADMs and nonviral NADMs. We calculated cause‐specific mortality rates (MR) after diagnosis of specific cancers and compared 5‐year survival with the UK and France general populations. Among 83,856 PLHIV there were 4,436 cancer diagnoses. Of 603 deaths after ADM diagnosis, 292 (48%) were due to an ADM. There were 467/847 (55%) and 74/189 (39%) deaths that were due to an NADM after nonviral and viral NADM diagnoses, respectively. MR were higher for diagnoses between 1996 and 2005 versus 2006–2015: ADMs 102 (95% CI 92–113) per 1,000 years versus 88 (78–100), viral NADMs 134 (106–169) versus 111 (93–133) and nonviral NADMs 264 (232–300) versus 226 (206–248). Estimated 5‐year survival for PLHIV diagnosed with liver (29% [19–39%]), lung (18% [13–23%]) and cervical (75% [63–84%]) cancer was similar to general populations. Survival after Hodgkin's lymphoma diagnosis was lower in PLHIV (75% [67–81%]). Among ART‐treated PLHIV diagnosed with cancer, MR and causes of death varied by cancer type, with mortality highest for liver and lung cancers. Deaths within 5 years of NADM diagnoses were more likely to be from cancer than AIDS., What's new? People with HIV live longer than they used to, thanks to advances in antiretroviral therapy. These improvements reduced the incidence of AIDS‐defining malignancies, such as Kaposi's sarcoma, but the increased life expectancy has led to more diagnoses of cancers not traditionally associated with HIV. Here, the authors studied cause‐specific mortality among people with HIV diagnosed with cancer. For those people, within 5 years after a cancer diagnosis, cause of death was more likely to be cancer than AIDS. Survival rates after diagnosis varied by cancer type, but were similar to rates among the general population.

Introduction : Syringes with attached needles (low dead space syringes [LDSS]) retain far less blood following injection than syringes with detachable needles (high dead space syringes [HDSS]). People who inject drugs (PWID) who share needles/syringes may be less likely to acquire Hepatitis C virus (HCV) infection using LDSS, compared with HDSS, but data are limited.Methods : Utilizing drug behavior and HCV antibody testing data from the UK 2014/2015 Unlinked Anonymous Monitoring Survey of PWID, we calculated the percentage of syringes used in the past month that were LDSS. We investigated which injecting characteristics and demographic factors were associated with 100% LDSS (against 0–99%) usage, and whether 100% LDSS use was associated with antibody HCV-status, after adjusting for confounders.Result : Of 2174 participants, 55% always used LDSS, 27% always used HDSS, and 17% used both LDSS and HDSS. PWID that had injected into their groin during the past month were unlikely to use LDSS, adjusted odds ratio (aOR) 0.14 (95% confidence interval 0.11–0.17), compared to those not using the groin. Those injecting crack were less likely to use LDSS than those not, aOR 0.79 (0.63–0.98). Polydrug use was negatively associated with LDSS use, aOR 0.88 (0.79-0.98) per additional drug. LDSS use was associated with lower prevalent HCV among all PWID (aOR 0.77, [0.64–0.93]), which was stronger among recent initiates (aOR 0.53 [0.30–0.94]) than among experienced PWID (aOR 0.81 [0.66–0.99]).Discussion : People who inject into their groin were less likely to use LDSS. Exclusive LDSS use was associated with lower prevalence of HCV amongst PWID that started injecting recently, suggesting LDSS use is protective against HCV.

Background The risk of acute kidney injury (AKI) attributable to renin angiotensin aldosterone (RAAS) inhibitors and diuretics remains unclear. Methods We conducted a prospective cohort study using the Clinical Practice Research Datalink (2008–2015) linked to Hospital Episode Statistics – Admitted Patient Care and Office for National Statistics mortality data. Patients were included if they had one or more chronic diagnoses requiring medication. Exposed patients had a first ever prescription for RAAS inhibitors/diuretics during the study period. AKI risk associated with exposure was determined by multivariable Cox regression, propensity score-adjusted Cox regression and a prior event rate ratio (PERR) analysis. Results One hundred forty thousand nine hundred fifty-two individuals were included. Increased AKI risk in the exposed group was demonstrated in both the multivariable and propensity score-adjusted cox regressions (HR 1.23 (95% CI 1.04–1.45) and HR 1.24 (1.05–1.47) respectively). The PERR analysis provided a similar overall hazard ratio with a wider confidence interval (HR 1.29 (0.94–1.63)). The increased AKI risk in the exposed group was present only in those receiving two or more antihypertensives. Absolute AKI risk was small. Conclusions RAAS inhibitors/diuretics result in an increased risk of AKI. The absolute increase in AKI risk is small, however, and needs to be considered in the context of any potential benefits.

Tobacco smoking and alcohol consumption are well-established risk factors for head and neck cancer. The prognostic role of smoking and alcohol intake at diagnosis have been less well studied. We analysed 1,393 people prospectively enrolled into the Head and Neck 5000 study (oral cavity cancer, n=403; oropharyngeal cancer, n=660; laryngeal cancer, n=330) and followed up for a median of 3.5 years. The primary outcome was all-cause mortality. We used Cox proportional hazard models to derive minimally adjusted (age and gender) and fully adjusted (age, gender, ethnicity, stage, comorbidity, body mass index, HPV status, treatment, education, deprivation index, income, marital status, and either smoking or alcohol use) mortality hazard ratios (HR) for the effects of smoking status and alcohol intake at diagnosis. Models were stratified by cancer site, stage and HPV status. The fully-adjusted HR for current versus never-smokers was 1.7 overall (95% confidence interval [CI] 1.1, 2.6). In stratified analyses, associations of smoking with mortality were observed for oropharyngeal and laryngeal cancers (fully adjusted HRs for current smokers: 1.8 (95% CI=0.9, 3.40 and 2.3 (95% CI=0.8, 6.4)). We found no evidence that people who drank hazardous to harmful amounts of alcohol at diagnosis had a higher mortality risk compared to non-drinkers (HR=1.2 (95% CI=0.9, 1.6)). There was no strong evidence that HPV status or tumour stage modified the association of smoking with survival. Smoking status at the time of a head and neck cancer diagnosis influenced all-cause mortality in models adjusted for important prognostic factors.

BackgroundThe World Health Organization (WHO) has developed a global health strategy to eliminate viral hepatitis. We project the treatment and prevention requirements to achieve the WHO HCV elimination target of reducing HCV incidence by 80% and HCV-related mortality by 65% by 2030 in Pakistan, which has the second largest HCV burden worldwide.MethodsWe developed an HCV transmission model for Pakistan, and calibrated it to epidemiological data from a national survey (2007), surveys among people who inject drugs (PWID), and blood donor data. Current treatment coverage data came from expert opinion and published reports. The model projected the HCV burden, including incidence, prevalence and deaths through 2030, and estimated the impact of varying prevention and direct-acting antiviral (DAA) treatment interventions necessary for achieving the WHO HCV elimination targets.ResultsWith no further treatment (currently ∼150 000 treated annually) during 2016–30, chronic HCV prevalence will increase from 3.9% to 5.1%, estimated annual incident infections will increase from 700 000 to 1 100 000, and 1 400 000 HCV-associated deaths will occur. To reach the WHO HCV elimination targets by 2030, 880 000 annual DAA treatments are required if prevention is not scaled up and no treatment prioritization occurs. By targeting treatment toward persons with cirrhosis (80% treated annually) and PWIDs (double the treatment rate of non-PWIDs), the required annual treatment number decreases to 750 000. If prevention activities also halve transmission risk, this treatment number reduces to 525 000 annually.ConclusionsSubstantial HCV prevention and treatment interventions are required to reach the WHO HCV elimination targets in Pakistan, without which Pakistan’s HCV burden will increase markedly.

Pakistan has a high prevalence of hepatitis C virus (HCV) infection, estimated at 4.9% (2,290/46,843) in the 2007 national HCV seroprevalence survey. We used data from this survey to assess the importance of risk factor associations with HCV prevalence in Pakistan. Exposures were grouped as community (going to the barbers, sharing smoking equipment, having an ear/nose piercing, tattoo, or acupuncture), healthcare (ever having hemodialysis, blood transfusion, or ≥ 5 injections in the last year), demographic (marital status and age), and socio-economic (illiterate or laborer). We used mutually adjusted multivariable regression analysis, stratified by sex, to determine associations with HCV infection, their population attributable fraction, and how risk of infection accumulates with multiple exposures. Strength of associations was assessed using adjusted odds ratios (aOR). Community [aOR females 1.5 (95% confidence interval [CI]: 1.2, 1.8); males 1.2 (1.1, 1.4)] and healthcare [females 1.4 (1.2, 1.6); males 1.2 (1.1, 1.4)] exposures, low socio-economic status [females 1.6 (1.3, 1.80); males 1.3 (1.2, 1.5)], and marriage [females 1.5 (1.2, 1.9); males 1.4 (1.1, 1.8)] were associated with increased HCV infection. Among married women, the number of children was associated with an increase in HCV infection; linear trend aOR per child 1.06 (1.01, 1.11). Fewer infections could be attributed to healthcare exposures (females 13%; males 6%) than to community exposures (females 25%; males 9%). Prevalence increased from 3% to 10% when cumulative exposures increased from 1 to ≥ 4 [aOR per additional exposure for females 1.5 (1.4, 1.6); males 1.2 (1.2, 1.3)]. A combination of community, healthcare, and other factors appear to drive the Pakistan HCV epidemic, highlighting the need for a comprehensive array of prevention strategies.

BackgroundHighly sensitive, commercial nucleic acid amplification tests (NAAT) for Trichomonas vaginalis have only recently been recommended for use in the UK. While testing for T. vaginalis is routine in symptomatic women attending genitourinary medicine (GUM) clinics, it is rare in asymptomatic women or those attending primary care. The aim of this study was to evaluate the positivity of T. vaginalis using a commercial NAAT, in symptomatic and asymptomatic women undergoing testing for chlamydia and gonorrhoea in GUM and primary care settings.MethodsSamples from 9186 women undergoing chlamydia and gonorrhoea testing in South West England between May 2013 and Jan 2015 were also tested for T. vaginalis by NAAT alongside existing tests.ResultsT. vaginalis positivity using NAAT was as follows: in GUM 4.5% (24/530, symptomatic) and 1.7% (27/1584, asymptomatic); in primary care 2.7% (94/3499, symptomatic) and 1.2% (41/3573, asymptomatic). Multivariable regression found that in GUM older age, black ethnicity and deprivation were independent risk factors for T. vaginalis infection. Older age and deprivation were also risk factors in primary care. Testing women presenting with symptoms in GUM and primary care using TV NAATs is estimated to cost £260 per positive case diagnosed compared with £716 using current microbiological tests.ConclusionsAptima TV outperforms existing testing methods used to identify T. vaginalis infection in this population. An NAAT should be used when testing for T. vaginalis in women who present for testing with symptoms in primary care and GUM, based on test performance and cost.

ObjectivesTo estimate trends in HIV testing, positivity and prevalence in UK primary care for 2000–2015 as part of a wider investigation into reasons for late diagnosis of HIV.DesignRetrospective cohort study using the Clinical Practice Research Datalink (CPRD) which is derived from computerised clinical records produced during consultations in primary care.Setting404 general practices in England.Participants5 979 598 adults aged ≥16 years registered between 2000 and 2015 with 45 093 761 person years of observation.OutcomesAnnual HIV testing rates, proportion of positive tests and prevalence of HIV-infected people recorded in primary care 2000–2015.ResultsHIV testing in primary care increased from 2000 to 2010, but then declined. Testing was higher in females than in males and in those aged 16–44 years compared with older adults. Rates per 100 000 in women aged 16–44 years were 177 (95% CI 167 to 188); 1309 (95% CI 1282 to 1336); 1789 (95% CI 1757 to 1821) and 839 (95% CI 817 to 862) in 2000, 2005, 2010 and 2015, respectively, and for non-pregnant women: 22.5 (95% CI 19 to 26); 134 (95% CI 125 to 143); 262 (95% CI 250 to 275); 190 (95% CI 179 to 201). For men aged 16–44 years rates were: 26 (95% CI 22 to 29); 107 (95% CI 100 to 115); 196 (95% CI 185 to 206); 137 (95% CI 127 to 146). Over the study period, there were approximately two positive results per 1000 HIV tests. Men were eightfold more likely to test positive than women. The percentage of HIV diagnoses among adults recorded in CPRD may be as low as 55% in London and 67% in the rest of the UK.ConclusionsHIV testing rates in primary care peaked in 2010 and subsequently declined. Access to testing was higher for women despite the prevalence of HIV being higher in men.Implications and further research neededOpportunities remain in primary care for increasing HIV testing to prevent costly late diagnoses and decrease HIV transmission. Interventions to improve targeting of tests and increase adherence to HIV testing guidelines are needed in primary care.

Background Acute lower respiratory tract infection (LRTI) is one of the most common conditions managed internationally and is costly to health services and patients. Despite good evidence that antibiotics are not effective for improving the symptoms of uncomplicated LRTI, they are widely prescribed, contributing to antimicrobial resistance. Many of the symptoms observed in LRTI are mediated by inflammatory processes also observed in exacerbations of asthma, for which there is strong evidence of corticosteroid effectiveness. The primary aim of the OSAC (Oral Steroids for Acute Cough) Trial is to determine whether oral prednisolone (40 mg daily for 5 days) can reduce the duration of moderately bad (or worse) cough and the severity of all its associated symptoms on days 2 to 4 post-randomisation (day 1 is trial entry) by at least 20% in adults ≥18 years with acute LRTI presenting to primary care. Methods/design OSAC is a two-arm, multi-centre, placebo-controlled, randomised superiority trial. The target sample size is 436 patients, which allows for a 20% dropout rate. Patients will be recruited from primary care sites (General Practitioner surgeries) across England and followed up until symptom resolution. The two primary clinical outcomes are the duration of moderately bad (or worse) cough, and the severity of all its associated symptoms on days 2 to 4 post-randomisation. Secondary outcomes include: antibiotic consumption; symptom burden; adverse events; participant satisfaction with treatment and intention to consult for future similar illnesses. A parallel economic evaluation will investigate the cost-effectiveness of the intervention. Discussion Results from the OSAC trial will increase knowledge regarding the clinical and cost-effectiveness of corticosteroids for LRTI, and will establish the potential of a new treatment option that could substantially improve patient health. We have chosen a relatively high ‘efficacy dose’ as this will enable us to decide on the potential for further research into lower dose oral and/or inhaled corticosteroids. This trial will also contribute to a growing body of research investigating the natural course of this very common illness, as well as the effects of steroids on the undesirable inflammatory symptoms associated with infection. Trial registration Current Controlled Trials ISRCTN57309858 (31 January 2013). Electronic supplementary material The online version of this article (doi:10.1186/s13063-015-0569-5) contains supplementary material, which is available to authorized users.

ObjectiveTo evaluate the effects of four drug (quadruple) versus three drug (triple) combination antiretroviral therapies in treatment naive people with HIV, and explore the implications of existing trials for clinical practice and research.DesignSystematic review and meta-analysis of randomised controlled trials.Data sourcesPubMed, EMBASE, CENTRAL, Web of Science, and the Cumulative Index to Nursing and Allied Health Literature from March 2001 to December 2016 (updated search in PubMed and EMBASE up to June 2018); and reference lists of eligible studies and related reviews.Study selectionRandomised controlled trials comparing quadruple with triple combination antiretroviral therapies in treatment naive people with HIV and evaluating at least one effectiveness or safety outcome.Review methodsOutcomes of interest included undetectable HIV-1 RNA, CD4 T cell count, virological failure, new AIDS defining events, death, and severe adverse effects. Random effects meta-analyses were conducted.ResultsTwelve trials (including 4251 people with HIV) were eligible. Quadruple and triple combination antiretroviral therapies had similar effects on all relevant effectiveness and safety outcomes, with no point estimates favouring quadruple therapy. With the triple therapy as the reference group, the risk ratio was 0.99 (95% confidence interval 0.93 to 1.05) for undetectable HIV-1 RNA, 1.00 (0.90 to 1.11) for virological failure, 1.17 (0.84 to 1.63) for new AIDS defining events, 1.23 (0.74 to 2.05) for death, and 1.09 (0.89 to 1.33) for severe adverse effects. The mean difference in CD4 T cell count increase between the two groups was −19.55 cells/μL (−43.02 to 3.92). In general, the results were similar, regardless of the specific regimens of combination antiretroviral therapies, and were robust in all subgroup and sensitivity analyses.ConclusionIn this study, effects of quadruple combination antiretroviral therapy were not better than triple combination antiretroviral therapy in treatment naive people with HIV. This finding lends support to current guidelines recommending the triple regimen as first line treatment. Further trials on this topic should be conducted only when new research is justified by adequate systematic reviews of the existing evidence. However, this study cannot exclude the possibility that quadruple cART would be better than triple cART when new classes of antiretroviral drugs are made available.

Background There is currently considerable interest the complex relationship between PrEP and STI incidence. Our Challenges and Opportunities of PrEP (CHOP) study uses qualitative and quantitative methods to ascertain the changes in behaviour and STI rates in high risk men/transgender/nonbinary persons who have sex with men (MSM/TPSM) before and after commencement of the PrEP Impact trial (‘Impact’) in our clinic (01/12/2017). Methods We performed retrospective data analysis of clinic and laboratory records for MSM/TPSM attendees who attended a British sexual health clinic in a large urban area between 01/12/2016 and 28/02/2019. We used logistic regression to compare STI diagnoses (chlamydia, gonorrhoea and syphilis at any site) and sexual risk score in two time periods (1: 01/12/2016 – 31/11/17 and 2: 01/12/2017 to 28/02/2019). Results Of 72,667 attendances, 10,918 were MSM/TPSM and, excluding duplicate entries (164) and those who were already HIV positive (236), this left 3,407 individuals with 10,518 attendances: 4,378 (period 1) and 6,140 (period 2). Mean HIV acquisition risk score at first visit in period 1 was 5.8 and 5.4 in period 2. In period 1 at first visit 13.5% (266/1,975) had one or more STI compared with 20.2% (289/1143) in period 2, a rise of 6.7%. Logistic regression showed OR1.6 (p STI diagnoses for subsequent visits in period 1, 6.3% (151/2403) attendances had one or more STIs diagnosed, and in period 2 10.9% (513/4708). This is a rise of 4.6% (OR 1.8 (p Conclusion Although risk score was similar during both time periods, STI rates were significantly higher both at first visit and at subsequent visits. The qualitative aspect of this study will explore whether this reflects attracting more high risk MSM/TPSM into the service or whether use of PrEP increases risk-taking behaviour. Disclosure No significant relationships.

High systolic blood pressure (SBP) causes cardiovascular disease (CVD) and is associated with mortality from other causes, but conventional multivariably-adjusted results may be confounded. Here we used a son’s SBP (>1 million Swedish men) as an instrumental variable for parental SBP and examined associations with parents’ cause-specific mortality, avoiding reverse causation. The hazard ratio for CVD mortality per SD (10.80 mmHg) of SBP was 1.49 (95% CI: 1.43, 1.56); SBP was positively associated with coronary heart disease and stroke. SBP was also associated positively with all-cause, diabetes and kidney cancer mortality, and negatively with external causes. Negative associations with respiratory-related mortality were probably confounded by smoking. Hazard ratios for other causes were imprecise or null. Diastolic blood pressure gave similar results to SBP. CVD hazard ratios were intermediate between those from conventional multivariable studies and Mendelian randomization and stronger than those from clinical trials, approximately consistent with an effect of exposure duration on effect sizes. Plots of parental mortality against offspring SBP were approximately linear, supporting calls for lower SBP targets. Results suggest that conventional multivariable analyses of mortality and SBP are not substantially confounded by reverse causation and confirm positive effects of SBP on all-cause, CVD and diabetes mortality.

Background\ud The WHO elimination strategy for hepatitis C virus advocates scaling up screening and treatment to reduce global hepatitis C incidence by 80% by 2030, but little is known about how this reduction could be achieved and the costs of doing so. We aimed to evaluate the effects and cost of different strategies to scale up screening and treatment of hepatitis C in Pakistan and determine what is required to meet WHO elimination targets for incidence.\ud \ud Methods\ud We adapted a previous model of hepatitis C virus transmission, treatment, and disease progression for Pakistan, calibrating using available data to incorporate a detailed cascade of care for hepatitis C with cost data on diagnostics and hepatitis C treatment. We modelled the effect on various outcomes and costs of alternative scenarios for scaling up screening and hepatitis C treatment in 2018–30. We calibrated the model to country-level demographic data for 1960–2015 (including population growth) and to hepatitis C seroprevalence data from a national survey in 2007–08, surveys among people who inject drugs (PWID), and hepatitis C seroprevalence trends among blood donors. The cascade of care in our model begins with diagnosis of hepatitis C infection through antibody screening and RNA confirmation. Diagnosed individuals are then referred to care and started on treatment, which can result in a sustained virological response (effective cure). We report the median and 95% uncertainty interval (UI) from 1151 modelled runs.\ud \ud Findings\ud One-time screening of 90% of the 2018 population by 2030, with 80% referral to treatment, was projected to lead to 13·8 million (95% UI 13·4–14·1) individuals being screened and 350 000 (315 000–385 000) treatments started annually, decreasing hepatitis C incidence by 26·5% (22·5–30·7) over 2018–30. Prioritised screening of high prevalence groups (PWID and adults aged ≥30 years) and rescreening (annually for PWID, otherwise every 10 years) are likely to increase the number screened and treated by 46·8% and decrease incidence by 50·8% (95% UI 46·1–55·0). Decreasing hepatitis C incidence by 80% is estimated to require a doubling of the primary screening rate, increasing referral to 90%, rescreening the general population every 5 years, and re-engaging those lost to follow-up every 5 years. This approach could cost US$8·1 billion, reducing to $3·9 billion with lowest costs for diagnostic tests and drugs, including health-care savings, and implementing a simplified treatment algorithm.\ud \ud Interpretation\ud Pakistan will need to invest about 9·0% of its yearly health expenditure to enable sufficient scale up in screening and treatment to achieve the WHO hepatitis C elimination target of an 80% reduction in incidence by 2030.\ud \ud Funding\ud UNITAID.

The World Health Organization (WHO) recently produced guidelines advising a treat-all policy for HCV to encourage widespread treatment scale-up for achieving HCV elimination. We modelled the prevention impact achieved (HCV infections averted [IA]) from initiating this policy compared with treating different subgroups at country, regional and global levels. We assessed what country-level factors affect impact. A dynamic, deterministic HCV transmission model was calibrated to data from global systematic reviews and UN data sets to simulate country-level HCV epidemics with ongoing levels of treatment. For each country, the model projected the prevention impact (in HCV IA per treatment undertaken) of initiating four treatment strategies; either selected randomly (treat-all) or targeted among people who inject drugs (PWID), people aged ≥35, or those with cirrhosis. The IA was assessed over 20 years. Linear regression was used to identify associations between IA per treatment and demographic factors. Eighty-eight countries (85% of the global population) were modelled. Globally, the model estimated 0.35 (95% credibility interval [95%CrI]: 0.16-0.61) IA over 20 years for every randomly allocated treatment, 0.30 (95%CrI: 0.12-0.53) from treating those aged ≥35 and 0.28 (95%CrI: 0.12-0.49) for those with cirrhosis. Globally, treating PWID achieved 1.27 (95%CrI: 0.68-2.04) IA per treatment. The IA per randomly allocated treatment was positively associated with a country's population growth rate and negatively associated with higher HCV prevalence among PWID. In conclusion, appreciable prevention benefits could be achieved from WHO’s treat-all strategy, although greater benefits per treatment can be achieved through targeting PWID. Higher impact will be achieved in countries with high population growth.

BackgroundUK guidelines recommend a ‘routine offer of HIV testing’ in primary care where HIV diagnosed prevalence exceeds 2 in 1000. However, current primary care HIV testing rates are low. Efforts to increase primary care HIV testing are needed. To examine how an educational intervention to increase HIV testing in general practice was experienced by healthcare professionals (HCPs) and to understand the perceived impacts on HIV testing. MethodQualitative interviews with general practitioners (GPs) and nurses 3-months after receiving an educational intervention developed from an adapted version of the Medical Foundation for HIV and Sexual Health (MEDFASH) HIV Testing In Practice (TIPs) online educational tool which included training on HIV associated clinical indicator conditions, why, who, and how to test. The intervention was delivered in 19 high-HIV prevalence general practices in Bristol. 27 semi-structured interviews were conducted across 13 practices with 16 GPs, 10 nurses and the sexual health clinician who delivered the intervention. Transcripts were analysed thematically informed by Normalisation Process Theory. ResultsHCPs welcomed the opportunity to update their HIV knowledge through a tailored, interactive session. Post-training, HCPs reported increased awareness of HIV indicator conditions, confidence to offer HIV tests and consideration of HIV tests. Continued testing barriers include perceived lack of opportunity. ConclusionsThis qualitative study found that HIV education is perceived as valuable in relation to perceived awareness, confidence, and consideration of HIV testing. However, repetition and support from other strategies are needed to encourage HCPs to offer HIV tests. Future interventions should consider using behaviour change theory to develop a complex intervention that addresses not only HCP capability to offer an HIV test, but also issues of opportunity and motivation.

To better understand hepatitis C virus (HCV) epidemiology in Punjab state, India, we estimated the distribution of HCV antibody positivity (anti-HCV+) using a 2013–2014 HCV household seroprevalence survey. Household anti-HCV+ clustering was investigated (a) by individual-level multivariable logistic regression, and (b) comparing the observed frequency of households with multiple anti-HCV+ persons against the expected, simulated frequency assuming anti-HCV+ persons are randomly distributed. Village/ward-level clustering was investigated similarly. We estimated household-level associations between exposures and the number of anti-HCV+ members in a household (N= 1593 households) using multivariable ordered logistic regression. Anti-HCV+ prevalence was 3.6% (95% confidence interval 3.0–4.2%). Individual-level regression (N= 5543 participants) found an odds ratio of 3.19 (2.25–4.50) for someone being anti-HCV+ if another household member was anti-HCV+. Thirty households surveyed had ⩾2 anti-HCV+ members, whereas 0/1000 (P< 0.001) simulations had ⩾30 such households. Excess village-level clustering was evident: 10 villages had ⩾6 anti-HCV+ members, occurring in 31/1000 simulations (P= 0.031). The household-level model indicated the number of household members, living in southern Punjab, lower socio-economic score, and a higher proportion having ever used opium/bhuki were associated with a household's number of anti-HCV+ members. Anti-HCV+ clusters within households and villages in Punjab, India. These data should be used to inform screening efforts.

Shared parameter joint models provide a framework under which a longitudinal response and a time to event can be modelled simultaneously. A common assumption in shared parameter joint models has been to assume that the longitudinal response is normally distributed. In this paper, we instead propose a joint model that incorporates a two-part 'hurdle' model for the longitudinal response, motivated in part by longitudinal response data that is subject to a detection limit. The first part of the hurdle model estimates the probability that the longitudinal response is observed above the detection limit, whilst the second part of the hurdle model estimates the mean of the response conditional on having exceeded the detection limit. The time-to-event outcome is modelled using a parametric proportional hazards model, assuming a Weibull baseline hazard. We propose a novel association structure whereby the current hazard of the event is assumed to be associated with the current combined (expected) outcome from the two parts of the hurdle model. We estimate our joint model under a Bayesian framework and provide code for fitting the model using the Bayesian software Stan. We use our model to estimate the association between HIV RNA viral load, which is subject to a lower detection limit, and the hazard of stopping or modifying treatment in patients with HIV initiating antiretroviral therapy. Copyright © 2016 John Wiley & Sons, Ltd.

Summary Background WHO aims to eliminate the hepatitis C virus (HCV) as a public health threat by 2030. Injection drug use is an important risk factor for HCV transmission, but its contribution to country-level and global epidemics is unknown. We estimated the contribution of injection drug use to risk for HCV epidemics globally, regionally, and at country level. Methods We developed a dynamic deterministic HCV transmission model to simulate country-level HCV epidemics among people who inject drugs and the general population. Each country's model was calibrated using country-specific data from UN datasets and systematic reviews on the prevalence of HCV and injection drug use. The population attributable fraction of HCV transmission associated with injection drug use was estimated—defined here as the percentage of HCV infections prevented if additional HCV transmission due to injection drug use was removed between 2018 and 2030. Findings The model included 88 countries (85% of the global population). The model predicted 0·23% (95% credibility interval [CrI] 0·16–0·31) of the global population were injection drug users in 2017, and 8% (5–12) of prevalent HCV infections were among people who currently inject drugs. Globally, if the increased risk for HCV transmission among people who inject drugs was removed, an estimated 43% (95% CrI 25–67) of incident HCV infections would be prevented from 2018 to 2030, varying regionally. This population attributable fraction was higher in high-income countries (79%, 95% CrI 57–97) than in countries of low and middle income (38%, 24–64) and was associated with the percentage of a country's prevalent HCV infections that are among people who inject drugs. Interpretation Unsafe injecting practices among people who inject drugs contribute substantially to incident HCV infections globally. Any intervention that can reduce HCV transmission among people who inject drugs will have a pronounced effect on country-level incidence of HCV. Funding None.

Supplemental Digital Content is Available in the Text., Background: An increasing number of HIV-positive individuals now start antiretroviral therapy (ART) with high CD4 cell counts. We investigated whether this makes restoration of CD4 and CD8 cell counts and the CD4:CD8 ratio during virologically suppressive ART to median levels seen in HIV-uninfected individuals more likely and whether restoration depends on gender, age, and other individual characteristics. Methods: We determined median and quartile reference values for CD4 and CD8 cell counts and their ratio using cross-sectional data from 2309 HIV-negative individuals. We used longitudinal measurements of 60,997 HIV-positive individuals from the Antiretroviral Therapy Cohort Collaboration in linear mixed-effects models. Results: When baseline CD4 cell counts were higher, higher long-term CD4 cell counts and CD4:CD8 ratios were reached. Highest long-term CD4 cell counts were observed in middle-aged individuals. During the first 2 years, median CD8 cell counts converged toward median reference values. However, changes were small thereafter and long-term CD8 cell count levels were higher than median reference values. Median 8-year CD8 cell counts were higher when ART was started with

Estimating velocity and acceleration trajectories allows novel inferences in the field of longitudinal data analysis, such as estimating change regions rather than change points, and testing group effects on nonlinear change in an outcome (ie, a nonlinear interaction). In this article, we develop derivative estimation for 2 standard approaches—polynomial mixed models and spline mixed models. We compare their performance with an established method—principal component analysis through conditional expectation through a simulation study. We then apply the methods to repeated blood pressure (BP) measurements in a UK cohort of pregnant women, where the goals of analysis are to (i) identify and estimate regions of BP change for each individual and (ii) investigate the association between parity and BP change at the population level. The penalized spline mixed model had the lowest bias in our simulation study, and we identified evidence for BP change regions in over 75% of pregnant women. Using mean velocity difference revealed differences in BP change between women in their first pregnancy compared with those who had at least 1 previous pregnancy. We recommend the use of penalized spline mixed models for derivative estimation in longitudinal data analysis.

Background HIV-infected patients often present to primary care several times with HIV-indicator conditions before diagnosis but the opportunity to test by healthcare professionals (HCPs) is frequently missed. Current HIV testing rates in primary care are low and educational interventions to facilitate HCPs to increase testing and awareness of HIV are needed. Method We implemented a pilot feasibility stepped-wedged randomised controlled trial of an educational intervention in high HIV prevalence practices in Bristol. The training delivered to HCPs including General Practitioners (GP) aimed to increase HIV testing and included why, who, and how to test. The intervention was adapted from the Medical Foundation for HIV and Sexual Health HIV Testing in Practice (MEDFASH) educational tool. Questionnaires assessed HCP feedback and perceived impacts of the intervention. HIV testing rates were compared between control and intervention practices using 12 monthly laboratory totals. Results 169 HCPs (from 19 practices) received the educational intervention. 127 (75%) questionnaires were completed. Delivery of the intervention was received positively and was perceived as valuable for increasing awareness, confidence and consideration of testing, with HCPs gaining more awareness of HIV testing guidelines. The main pre-training HIV testing barrier reported by GPs was the patient not considering themselves at risk, whilst for nurses it was a concern about embarrassing or offending the patient. Most HCPs reported the intervention addressed these barriers. The HIV testing rate increased more in the control than in the intervention practices: mean difference 2.6 (95% CI 0.5,4.7) compared with 1.9 (− 0.5,4.3) per 1000 patients, respectively. The number of HIV tests across all practices increased from 1154 in the first 6 months to 1299 in the second 6 months, an annual increase in testing rate of 2.0 (0.7,3.4) from 16.3 to 18.3 per 1000 patients. Conclusion There was a small increase in HIV testing rates over the study period, but this could not be attributed to the educational intervention. More effective and sustainable programmes tailored to each practice context are needed to change testing culture and HCP behaviour. Repeated training, supported by additional measures, such as testing prompts, may be needed to influence primary care HIV testing. Electronic supplementary material The online version of this article (10.1186/s12875-018-0880-9) contains supplementary material, which is available to authorized users.

Objectives A measure used for assessing the effectiveness of HIV care and comparing clinical centres is the proportion of people starting antiretroviral therapy (ART) with viral suppression (VS) after 1 year. We propose a method that adjusts for patients’ demographic characteristics, and visually compares this measure between different sites accounting for centre size.Methods We analysed viral load measurements for UK Collaborative HIV Cohort (UK CHIC) patients starting ART between 2006 and 2013. We used logistic regression to estimate the proportion with VS after 1 year of ART adjusted for patient mix (in terms of age and a combined gender/ethnicity/acquisition mode variable) and calendar year. We compared outcomes between centres using funnel plots which account for centre size. Results The overall proportion of the cohort with VS 1 year after starting ART was 90% and increased from 83% to 93% between 2006 and 2013. VS was lower in younger individuals. White men who have sex with men (MSM) had the highest (94%), and black African (81%) and white (82%) heterosexual women the lowest proportions achieving VS. Comparing the unadjusted funnel plot with the adjusted, there were movements of some centres from outside to inside the 95% contour limits, which was largely explained by the patient mix of these centres. ConclusionsVS 1 year after ART start was associated with demographic characteristics and centre size; therefore, to compare the performances of centres, adjustment for these factors is required. Adjusted funnel plot is an effective tool which accounts for both the demographic characteristics and the centre size. Social factors, rather than treatment decisions within the control of the centres, may drive differences in outcomes.

Introduction Prostate cancer is the most common cancer in men in the UK, with nearly 40 000 diagnosed in 2014; and it is the second most common cause of male cancer-related mortality. The clinical conundrum is that most men live with prostate cancer rather than die from it, while existing treatments have significant associated morbidity. Recent studies have shown very low mortality rates (1% after a median of 10-year follow-up) and no treatment-related reductions in mortality, in men with localised prostate cancer. This study will identify prognostic factors associated with prostate cancer progression to help differentiate aggressive from more indolent tumours in men with localised disease at diagnosis, and so inform the decision to adopt conservative (active surveillance) or radical (surgery or radiotherapy) management strategies. Methods and analysis The Clinical Practice Research Datalink (CPRD) contains 57 318 men who were diagnosed with prostate cancer between 1 January 1987 and 31 December 2016. These men will be linked to the Office for National Statistics (ONS) and the National Cancer Registration and Analysis Service registry databases for mortality, TNM stage, Gleason grade and treatment data. Men with a diagnosis date prior to 1 January 1987 and men with lymph node or distant metastases at diagnosis will be excluded. A priori determined prognostic factors potentially associated with prostate cancer mortality, the end point of cancer progression, will be measured at baseline, and the participants followed through to development of cancer progression, death or the end of the follow-up period (31 December 2016). Cox proportional hazards regression will be used to estimate crude and mutually adjusted HRs. Mortality risk will be predicted using flexible parametric survival models that can accurately fit the shape of the hazard function. Ethics and dissemination This study protocol has approval from the Independent Scientific Advisory Committee for the UK Medicines and Healthcare products Regulatory Agency Database Research (protocol 17_041). The findings will be presented in peer-reviewed journals and local CPRD researcher meetings.

With the aim to expand the randomized controlled trial evidence of cinacalcet treatment to the unselected, general chronic kidney disease (CKD) population we analysed a large inception cohort of CKD patients in the region of Stockholm, Sweden 2006–2012 (both non-dialysis, dialysis and transplanted) with evidence of secondary hyperparathyroidism (SHPT). We used marginal structural models to account for both confounding by indication and time-dependent confounding. Over 37 months, 435/3,526 (12%) initiated cinacalcet de novo. Before cinacalcet initiation, parathyroid hormone (PTH) had increased progressively to a median of 636ng/L. After cinacalcet initiation, PTH declined, as did serum calcium and phosphate. In total, 42% of patients experienced a fatal/non-fatal cardiovascular event, 32% died and 9% had a new fracture. The unadjusted cardiovascular odds ratio (OR) associated with cinacalcet treatment was 1.01 (95% confidence interval: 0.83, 1.22). In the fully weighted model, the cardiovascular odds was lower in cinacalcet treated patients (OR 0.67: 0.48, 0.93). The adjusted ORs for all-cause mortality and for fractures were 0.79 (0.56, 1.11) and 1.08 (0.59, 1.98) respectively. Our study suggests cinacalcet treatment improves biochemical abnormalities in the wider CKD population, and adds real-world support that treating SHPT with cinacalcet may have beneficial effects on cardiovascular outcomes.

INTRODUCTION: HIV-1 infection leads to chronic inflammation and to an increased risk of non-AIDS mortality. Our objective was to determine whether AIDS-defining events (ADEs) were associated with increased overall and cause-specific non-AIDS related mortality after antiretroviral therapy (ART) initiation. METHODS: We included HIV treatment-naïve adults from the Antiretroviral Therapy Cohort Collaboration (ART-CC) who initiated ART from 1996 to 2014. Causes of death were assigned using the Coding Causes of Death in HIV (CoDe) protocol. The adjusted hazard ratio (aHR) for overall and cause-specific non-AIDS mortality among those with an ADE (all ADEs, tuberculosis (TB), Pneumocystis jiroveci pneumonia (PJP), and non-Hodgkin's lymphoma (NHL)) compared to those without an ADE was estimated using a marginal structural model. RESULTS: The adjusted hazard of overall non-AIDS mortality was higher among those with any ADE compared to those without any ADE (aHR 2.21, 95% confidence interval (CI) 2.00 to 2.43). The adjusted hazard of each of the cause-specific non-AIDS related deaths were higher among those with any ADE compared to those without, except metabolic deaths (malignancy aHR 2.59 (95% CI 2.13 to 3.14), accident/suicide/overdose aHR 1.37 (95% CI 1.05 to 1.79), cardiovascular aHR 1.95 (95% CI 1.54 to 2.48), infection aHR (95% CI 1.68 to 2.81), hepatic aHR 2.09 (95% CI 1.61 to 2.72), respiratory aHR 4.28 (95% CI 2.67 to 6.88), renal aHR 5.81 (95% CI 2.69 to 12.56) and central nervous aHR 1.53 (95% CI 1.18 to 5.44)). The risk of overall and cause-specific non-AIDS mortality differed depending on the specific ADE of interest (TB, PJP, NHL). CONCLUSIONS: In this large multi-centre cohort collaboration with standardized assignment of causes of death, non-AIDS mortality was twice as high among patients with an ADE compared to without an ADE. However, non-AIDS related mortality after an ADE depended on the ADE of interest. Although there may be unmeasured confounders, these findings suggest that a common pathway may be independently driving both ADEs and NADE mortality. While prevention of ADEs may reduce subsequent death due to NADEs following ART initiation, modification of risk factors for NADE mortality remains important after ADE survival. We would like to thank the following cohorts for participating in this study; the AIDS Therapy Evaluation Project Netherlands (ATHENA); ANRS CO4 French Hospital Database on HIV (FHDH); ANRS CO3 Aquitaine Cohort, France; Departments of Internal Medicine at University of Cologne and Bonn, Germany; K€oln/Bonn Cohort, Germany; Italian Cohort of Antiretroviral-Na€ıve Patients (ICONA); CORIS, Spain; Proyecto para la Informatizaci on del Seguimiento Cl ınico-epidemiol ogico de la Infecci on por HIV y SIDA (PISCIS), Spain; Royal Free Hospital Cohort, London UK; Swiss HIV Cohort Study (SHCS); The Multicenter Study Group on EuroSIDA; Southern Alberta Clinic, Canada; HIV Atlanta Veterans affairs Cohort Study (HAVACS), USA; UAB 1917 Clinic Cohort, Birmingham, Alabama, USA; Veterans Ageing Cohort Study (VACS8),USA; Vanderbilt- Meharry and Tennessee Center for AIDS Research, Nashville, Tennessee, USA; University of Washington Harborview Medical Center, USA; € Osterreichische HIV-Kohortenstudie (OEHIVKOS), Austria; Danish HIV Cohort Study, Denmark; VACH, (Spain); and HAART Observational Medical Evaluation and Research (HOMER), British Columbia, Canada. We would also like to thank the ART-CC Steering group. Sí