Heterogeneity has a key role in meta-analysis methods and can greatly affect conclusions. However, true levels of heterogeneity are unknown and often researchers assume homogeneity. We aim to: a) investigate the prevalence of unobserved heterogeneity and the validity of the assumption of homogeneity; b) assess the performance of various meta-analysis methods; c) apply the findings to published meta-analyses.
\n \n\n \n \nOverweight and obesity are common and important causes of chronic disease. This should mean that primary care physicians feel tackling obesity is important, but it is uncommon for them to do so. Physicians perceive that this is not their job; they fear offending their patients and are unclear what may be effective. In this review we found two systematic reviews showing that motivational interviewing can lead to effective weight loss but it may not be practicable in this setting. Two trials show referral to specially trained nurses in primary care appears ineffective. Several randomized trials show referral to commercial weight management companies is effective. Observational data but no trials suggest that screening for and opportunistic brief interventions may motivate attempts to lose weight and lead to some weight loss. We conclude there is insufficient evidence to promote treatment opportunistically but sufficient evidence to refer patients wanting to lose weight to commercial weight management services.\r\n\r\nhttp://link.springer.com/article/10.1007/s13679-013-0073-8
\n \n\n \n \nAlthough screening and brief intervention is effective at reducing alcohol consumption \r\nin primary care and is recommended by guidelines, there are numerous barriers to its \r\ndelivery. Screening newly-registered patients for alcohol-use disorders provides an opportunity for systematic collection of alcohol consumption data.\r\n\r\n
\n \n\n \n \nThe assessment of patient-reported outcomes (PROs) in clinical trials poses a number of potential problems. What happens when a patient reports a severe symptom and no one is monitoring that information; for example, when questionnaires are not reviewed until the end of a study? Do hospitals or researchers face liability if a patient reports suicidal thoughts on a questionnaire?\r\n\r\nhttp://jama.jamanetwork.com/article.aspx?articleid=1741830
\n \n\n \n \nObjectives: We aimed to examine the temporal association between selective serotonin reuptake inhibitors (SSRI) and tricyclic antidepressant (TCA) prescriptions and suicide-related events in children and adolescents.\r\n\r\nDesign: Self-controlled case series.\r\n\r\nSetting: Electronic health records were used from 479 general practices in The Health Improvement Network (THIN) UK primary care database from 1995 to 2009.\r\n\r\nParticipants: 81 young people aged 10\u201318\u2005years with a record of completed suicide, 1496 who attempted suicide, 1178 with suicidal ideation and 2361 with intentional self-harm.\r\n\r\nMain outcome measures Incidence Rate Ratios (IRRs) for completed and attempted suicide, suicidal ideation and intentional self-harm.\r\n\r\nResults: For non-fatal suicide-related behaviour, IRRs were similar for the time the person was prescribed either SSRIs or TCAs: IRRs increased during pre-exposure, peaked on prescription day, were stable up to the fourth prescription-week, and decreased after the prescriptions were stopped. For both types of antidepressants, IRRs were lower or similar to pre-exposure levels during the period of prescription. For SSRIs, there was an increase in the IRR for completed suicide on the day of prescription (N=5; IRR=42.5, 95% CI 4.5 to 403.4), and during the fourth week of SSRI prescription (N=2; IRR=11.3, 95% CI 1.1 to 115.6).\r\n\r\nConclusions: We found that a very small number of young people were prescribed antidepressants and that there was an absence of a sustained increase in rates of suicide-related events in this group. There were no systematic differences between the association of TCAs and SSRIs and the incidence risk ratios for attempted suicide, suicidal ideation or intentional self-harm and, apart from the day of prescription, rates did not exceed pre-exposure levels. The pattern of IRR for suicide for SSRIs was similar to that found in non-fatal suicide-related events. Our results warrant a re-evaluation of the current prescription of SSRIs in young people. We recommend the creation of a pragmatic registry for active pharmacovigilance.
\n \n\n \n \nHeterogeneity has a key role in meta-analysis methods and can greatly affect conclusions. However, true levels of heterogeneity are unknown and often researchers assume homogeneity. We aim to: a) investigate the prevalence of unobserved heterogeneity and the validity of the assumption of homogeneity; b) assess the performance of various meta-analysis methods; c) apply the findings to published meta-analyses.\r\nMethods and findings\r\nWe accessed 57,397 meta-analyses, available in the Cochrane Library in August 2012. Using simulated data we assessed the performance of various meta-analysis methods in different scenarios. The prevalence of a zero heterogeneity estimate in the simulated scenarios was compared with that in the Cochrane data, to estimate the degree of unobserved heterogeneity in the latter. We re-analysed all meta-analyses using all methods and assessed the sensitivity of the statistical conclusions.\r\nLevels of unobserved heterogeneity in the Cochrane data appeared to be high, especially for small meta-analyses. A bootstrapped version of the standard DerSimonian-Laird approach performed better both in detecting heterogeneity and providing more accurate overall effect estimates. Re-analysing all meta-analyses with this method we found that 17-20% of the statistical conclusions changed, when heterogeneity was detected with the standard model and ignored. The rates were much lower when the standard method did not detect heterogeneity or took it into account, between 1% and 3%.\r\nConclusions:\r\nWhen evidence for heterogeneity is lacking, standard practice is to assume homogeneity and apply a simpler fixed-effect meta-analysis. We find that assuming homogeneity often results in a misleading analysis, since heterogeneity is very likely present but undetected. Our new method represents a small improvement but the problem largely remains, especially for very small meta-analyses. One solution is to test the sensitivity of the meta-analysis conclusions to assumed moderate and large degrees of heterogeneity. Equally, whenever heterogeneity is detected, it should not be ignored.\r\nKeywords: Meta-analysis; heterogeneity; homogeneity; DerSimonian-Laird; bootstrap; Cochrane library; Cochrane Database of Systematic Reviews; sensitivity analysis.
\n \n\n \n \nBackground: Symptomatic osteoarthritis poses a major challenge to primary health care but no studies have related accessing primary care (\u2018detection\u2019), receiving recommended treatments (\u2018treatment\u2019), and achieving adequate control (\u2018control\u2019). \r\nObjective: To provide estimates of detection, treatment, and control within a single population adapting the approach used to determine a Rule of Halves for other long-term conditions.\r\nSetting: General population\r\nParticipants: 400 adults aged 50+ years with prevalent symptomatic knee osteoarthritis.\r\nDesign: Prospective cohort with baseline questionnaire, clinical assessment, and plain radiographs, and questionnaire follow-up at 18 and 36 months and linkage to primary care medical records.\r\nOutcome measures: \u2018Detection\u2019 was defined as at least one musculoskeletal knee-related GP consultation between baseline and 36 months. \u2018Treatment\u2019 was self-reported use of at least one recommended treatment or physiotherapy/hospital specialist referral for their knee problem at all three measurement points. Pain was \u2018controlled\u2019 if characteristic pain intensity <5 out of 10 on at least two occasions. \r\nResults: In 221 cases (55.3%; 95%CI: 50.4, 60.1) there was evidence that the current problem had been detected in general practice. Of those detected, 164 (74.2% (68.4, 80.0)) were receiving one or more of the recommended treatments at all three measurement points. Of those detected and treated, 45 (27.4% (20.5, 34.3)) had symptoms under control on at least two occasions. Using narrower definitions resulted in substantially lower estimates. \r\nConclusion: Osteoarthritis care does not conform to a Rule of Halves. Symptom control is low among those accessing healthcare and receiving treatment. \r\n
\n \n\n \n \nQuantitative empirical analyses of a population of interest usually aim to estimate\r\nthe causal eect of one or more independent variables on a dependent variable. How-\r\never, only in rare instances is the whole population available for analysis. Researchers\r\ntend to estimate causal eects on a selected sample and generalize their conclusions to\r\nthe whole population. The validity of this approach rests on the assumption that the\r\nsample is representative of the population on certain key characteristics. A study using a\r\nnon-representative sample is lacking in external validity by failing to minimize population\r\nchoice bias. When the sample is large and non-response bias is not an issue, a random\r\nselection process is adequate to ensure external validity. If that is not the case, however,\r\nresearchers could follow a more deterministic approach to ensure representativeness on\r\nthe selected characteristics, provided these are known, or can be estimated, in the parent\r\npopulation. Although such approaches exist for matched sampling designs, research on\r\nrepresentative sampling and the similarity between the sample and the parent popula-\r\ntion seems to be lacking. In this article we propose a greedy algorithm for obtaining a\r\nrepresentative sample and quantifying representativeness in Stata.
\n \n\n \n \nObjectives:\r\nTo investigate the relationship between performance on the UK Quality and Outcomes Framework pay-for-performance scheme and choice of clinical computer system.\r\nDesign:\r\nRetrospective longitudinal study.\r\nSetting:\r\nData for 2007-8 to 2010-11, extracted from the clinical computer systems of general practices in England.\r\nParticipants:\r\nAll English practices participating in the pay-for-performance scheme: average 8257 each year, covering over 99% of the English population registered with a general practice.\r\nMain outcome measures:\r\nLevels of achievement on 62 quality of care indicators, measured as: reported achievement (levels of care after excluding inappropriate patients); population achievement (levels of care for all patients with the relevant condition); and percentage of available quality points attained. Multilevel mixed effects multiple linear regression models were used to identify population, practice, and clinical computing system predictors of achievement.\r\nResults:\r\nSeven clinical computer systems were consistently active in the study period, collectively holding approximately 99% of the market share. Of all population and practice characteristics assessed, choice of clinical computing system was the strongest predictor of performance across all three outcome measures. Differences between systems were greatest for intermediate outcomes indicators (for example, control of cholesterol levels).\r\nConclusions:\r\nUnder the UK\u2019s pay-for-performance scheme, differences in practice performance were associated with choice of clinical computing system. This raises the question of whether particular system characteristics facilitate higher quality of care, better data recording, or both. Inconsistencies across systems need to be understood and addressed, and researchers need to be cautious when generalising findings from samples of providers using a single computing system.
\n \n\n \n \nObjective To evaluate the utilisation of paper-based track and trigger (T&T) charts in a UK emergency department (ED).\r\n\r\nMethods A single-centre prospective observational cohort study was conducted in the ED of a medium-sized teaching hospital. Charted vital-sign data were collected from adults attending the resuscitation room, majors or observation ward. These data were examined in parallel with clinical notes to identify \u2018escalation\u2019 events. For each set of vital signs, the authors calculated the T&T score retrospectively.\r\n\r\nResults Data from 472 patient episodes (2965 sets of vital signs) were examined. 85.8% of patients had at least one full set of observations (CEM standard) and 60.6% had at least one T&T score documented. However, only 34.5% of observation sets had a corresponding T&T score. 20.6% of T&T score totals (1024) were incorrect, potentially preventing a \u2018trigger\u2019 from being recognised. 204 patient episodes had at least one recorded escalation. Physiological escalations were associated with vital-sign scores that met the triggering thresholds (98/104), while patients who had non-physiological escalations or no escalations were more likely to have scores below the triggering thresholds (88/100). Only 26.9% of physiological escalations were associated with a documented T&T score above the triggering threshold. Retrospective completion of the charts increased that figure to 94.2%.\r\n\r\nConclusion T&T in the ED is challenged by poor completion rates and numerical errors made during score calculation. However the potential for recognition of a deteriorating patient should not be ignored. The future work of the authors intends to evaluate an electronic system for automatically calculating T&T scores within the ED environment.
\n \n\n \n \nRespiratory rate, an important antecedent of patient deterioration, is inadequately recorded by hospital staff, partially due to the absence of a reliable automated technique for measuring it. The ECG has been proposed by several authors in recent years as a source of reliable respiratory information. Most algorithms proposed use either respiratory sinus arrhythmia (RSA) or the R-peak amplitude (RPA) modulation of the ECG. In this paper, we propose a novel method for estimating respiratory rate from the ECG which fuses frequency information from the two methods. The method was evaluated on data from 40 young and elderly subjects and validated against a \u201cgold standard\u201d respiratory rate obtained from simultaneously recorded respiration data. The fusion method outperformed the RSA and RPA methods, giving a mean absolute error of 0.81 bpm for the young subject population and 0.84 bpm for the elderly, using 1-min windows of data. Unlike other algorithms, the technique does not underperform at the lower or higher respiratory rates.
\n \n\n \n \nThis article describes a new individual patient data (IPD) meta-analysis post-estimation command, ipdforest. The command produces a forest plot, following an one-stage meta-analysis with xtmixed or xtmelogit. The over-all effect is obtained from the preceding mixed-effects regression and the study effects from linear or logistic regressions on each study which are executed within ipdforest. IPD meta-analysis models with Stata are discussed.
\n \n\n \n \nBackground: Comorbidity is increasingly common in primary care. The cost implications for patient care and budgetary management are unclear.\r\nAim: To investigate whether caring for patients with specific disease combinations increases or decreases primary care costs compared with treating separate patients with one condition each.\r\nDesign: Retrospective observational study using data on 86 100 patients in the General Practice Research Database.\r\nMethod: Annual primary care cost was estimated for each patient including consultations, medication, and investigations. Patients with comorbidity were defined as those with a current diagnosis of more than one chronic condition in the Quality and Outcomes\r\nFramework. Multiple regression modelling was used to identify, for three age groups, disease combinations that increase (cost-increasing) or decrease (cost-limiting) cost compared with treating each condition separately.\r\nResults: Twenty per cent of patients had at least two chronic conditions. All conditions were found to be both cost-increasing and cost-limiting when co-occurring with other conditions except dementia, which is only cost-limiting. Depression is the most important costincreasing condition when co-occurring with a range of conditions. Hypertension is costlimiting, particularly when co-occurring with other cardiovascular conditions.\r\nConclusion: Three categories of comorbidity emerge, those that are: cost-increasing, mainly due to a combination of depression with physical comorbidity; cost-limiting because treatment for the conditions overlap; and cost-limiting for no apparent reason but possibly because of inadequate care. These results can contribute to efficient and effective management of chronic conditions in primary care.\r\nKeywords: comorbidity; costs and cost analysis; delivery of health care; depression; family practice; resource allocation.
\n \n\n \n \nAbstract: Objective: To explore associations between low muscle mass and a wide range of lifestyle, dietary and cardiovascular risk factors in older men including metabolic risk factors, markers of inflammation, endothelial dysfunction and coagulation. Design: Cross-sectional study. Setting: British Regional Heart Study. Participants: 4252 men aged 60-79 years. Measurements: Participants attended a physical examination in 1998-2000, and completed a general questionnaire and a food frequency questionnaire. Low muscle mass was assessed by two measures: midarm muscle circumference (MAMC) and fat-free mass index (FFMI). Associations between risk factors and low muscle mass were analysed using logistic regression. Results: Physical inactivity, insulin resistance, C-reactive protein, von Willebrand factor and fibrinogen were associated with significantly increased odds of low MAMC and FFMI after adjustment for body mass index, lifestyle characteristics and morbidity. Those with higher percent energy intake from carbohydrates showed decreased odds of low MAMC (OR: 0.73, 95% CI: 0.55-0.96) and FFMI (OR: 0.76, 95% CI: 0.58-0.99). Other dietary variables, smoking, alcohol intake, D-dimer, interleukin 6 and homocysteine showed no important associations with MAMC and FFMI. Conclusion: Increasing physical activity, consuming a diet with a high proportion of energy from carbohydrates, and taking steps to prevent insulin resistance and reduce inflammation and endothelial dysfunction may help to reduce the risk of low muscle mass in older men.
\n \n\n \n \nObjective: To assess how initial severity of depression affects the benefit derived from low intensity interventions for depression. \r\nDesign: Meta-analysis of individual patient data from 16 datasets comparing low intensity interventions with usual care. \r\nSetting: Primary care and community settings. \r\nParticipants: 2470 patients with depression. \r\nInterventions: Low intensity interventions for depression (such as guided self help by means of written materials and limited professional support, and internet delivered interventions). \r\nMain outcome measures: Depression outcomes (measured with the Beck Depression Inventory or Center for Epidemiologic Studies Depression Scale), and the effect of initial depression severity on the\r\neffects of low intensity interventions. \r\nResults: Although patients were referred for low intensity interventions, many had moderate to severe depression at baseline. We found a significant interaction between baseline severity and treatment effect (coefficient \u22120.1 (95% CI \u22120.19 to \u22120.002)), suggesting that patients who are more severely depressed at baseline demonstrate larger treatment effects than those who are less severely depressed. However, the magnitude of the interaction (equivalent to an additional drop of around one point on the Beck Depression Inventory for a one standard deviation increase in initial severity) was small and may not be clinically significant. \r\nConclusions: The data suggest that patients with more severe depression at baseline show at least as much clinical benefit from low intensity interventions as less severely depressed patients and could usefully be offered these interventions as part of a stepped care model.\r\n\r\n\r\n
\n \n\n \n \nOpioid prescribing for chronic noncancer pain is increasing, but there is limited knowledge about longerterm\r\noutcomes of people receiving opioids for conditions such as back pain. This study aimed to explore\r\nthe relationship between prescribed opioids and disability among patients consulting in primary care\r\nwith back pain. A total of 715 participants from a prospective cohort study, who gave consent for review\r\nof medical and prescribing records and completed baseline and 6 month follow-up questionnaires, were\r\nincluded. Opioid prescription data were obtained from electronic prescribing records, and morphine\r\nequivalent doses were calculated. The primary outcome was disability (Roland-Morris Disability Questionnaire [RMDQ]) at 6 months. Multivariable linear regression was used to examine the association\r\nbetween opioid prescription at baseline and RMDQ score at 6 months. Analyses were adjusted for potential\r\nconfounders using propensity scores reflecting the probability of opioid prescription given baseline\r\ncharacteristics. In the baseline period, 234 participants (32.7%) were prescribed opioids. In the final multivariable analysis, opioid prescription at baseline was significantly associated with higher disability at 6- month follow-up (P < .022), but the magnitude of this effect was small, with a mean RMDQ score of 1.18\r\n(95% confidence interval: 0.17 to 2.19) points higher among those prescribed opioids compared to those\r\nwho were not. Our findings indicate that even after adjusting for a substantial number of potential confounders, opioids were associated with slightly worse functioning in back pain patients at 6-month follow-up. Further research may help us to understand the mechanisms underlying these findings and\r\ninform clinical decisions regarding the usefulness of opioids for back pain.
\n \n\n \n \nObjectives: To describe the number and types of problems discussed in general practice (GP) consultations, differences between problems raised by patients or doctors, and between problems discussed and recorded in medical records.\r\nDesign: Cross-sectional study based on video-recordings of consultations. \r\nSetting: 30 representative GPs from 22 general practices in Bristol and North Somerset \r\nParticipants: Adults making a pre-booked day-time appointment. \r\nMain outcome measures: Number and types of problems and issues discussed; who raised each problem/ issue; consultation duration; whether problems were recorded and coded. \r\nResults: Of 318 eligible patients, 229 (72.0%) participated. On average, 2.5 (95%CI 2.3 to 2.6) problems were discussed per consultation, with 41% (55/229) of consultations involving at least 3 problems. 72% (165/229) of consultations involved problems in multiple disease areas. Mean consultation duration was 11.9 minutes (95% CI 11.2 to 12.6). Consultation duration increased by 2 minutes per additional problem. Most problems discussed were raised by patients, but 43% (99/229) of consultations involved problem(s) raised by doctors. Patients and doctors raised different types of problems and issues. Of 562 problems discussed, 81% (n=455) were recorded in notes, but only 37% (n=206) were Read-coded, with differences between types of problems which were coded or not. \r\nConclusions: Consultations in general practice are complex encounters addressing multiple problems across a wide range of disease areas in a short time. Additional problems are dealt with very briefly. GPs, like patients, bring an agenda to consultations. There is systematic bias in the types of problems coded in electronic medical records databases. \r\n
\n \n\n \n \nAim: To map the availability and types of depression and anxiety groups, to examine men\u2019s experiences and perception of this support as well as the role of health professionals in accessing support. Background: The best ways to support men with depression and anxiety in primary care are not well understood. Group-based interventions are sometimes offered but it is unknown whether this type of support is acceptable to men. Methods: Interviews with 17 men experiencing depression or anxiety. A further 12 interviews were conducted with staff who worked with depressed men (half of whom also experienced depression or anxiety themselves). There were detailed observations of four mental health groups and a mapping exercise of groups in a single English city (Bristol). Findings: Some men attend groups for support with depression and anxiety. There was a strong theme of isolated men, some reluctant to discuss problems with their close family and friends but attending groups. Peer\r\nsupport, reduced stigma and opportunities for leadership were some of the identified\r\nbenefits of groups. The different types of groups may relate to different potential member audiences. For example, unemployed men with greater mental health and support needs attended a professionally led group whereas men with milder mental health problems attended peer-led groups. Barriers to help seeking were commonly reported, many of which related to cultural norms about how men should behave. General practitioners played a key role in helping men to acknowledge their experiences of depression and anxiety, listening and providing information on the range of support options, including groups. Men with depression and anxiety do go to groups and appear to be well supported by them. Groups may potentially be low cost and offer additional advantages for some men. Health professionals could do more to\r\nidentify and promote local groups.
\n \n\n \n \nBackground\r\nPlacebos are widely used in clinical practice in spite of ethical restrictions. Whether such use is justified depends in part on the relative benefit of placebos compared to \u2018active\u2019 treatments. A direct test for differences between placebo and \u2018active\u2019 treatment effects has not been conducted.\r\n\r\nObjectives\r\nWe aimed to test for differences between treatment and placebo effects within similar trial populations.\r\n\r\nData Sources\r\nA Cochrane Review compared placebos with no treatment in three-armed trials (no treatment, placebo, and treatment). We added an analysis of treatment and placebo differences within the same trials.\r\n\r\nSynthesis Methods\r\nFor continuous outcomes we compared mean differences between placebo and no treatment with mean differences between treatment and placebo. For binary outcomes we compared the risk ratio for treatment benefit (versus placebo) with the risk ratio for placebo benefit (versus no treatment). We conducted several preplanned subgroup analyses: objective versus subjective outcomes, conditions tested in three or more trials, and trials with varying degrees of bias.\r\n\r\nResults\r\nIn trials with continuous outcomes (n = 115) we found no difference between treatment and placebo effects (MD = \u22120.29, 95% CI \u22120.62 to 0.05, P = 0.10). In trials with binary outcomes (n = 37) treatments were significantly more effective than placebos (RRR = 0.72, 95%CI = 0.61 to 0.86, P = 0.0003). Treatment and placebo effects were not different in 22 out of 28 predefined subgroup analyses. Of the six subgroups with differences treatments were more effective than placebos in five. However when all criteria for reducing bias were ruled out (continuous outcomes) placebos were more effective than treatments (MD = 1.59, 95% CI = 0.40 to 2.77, P = 0.009).
\n \n\n \n \nObjective. A lack of agreement between clinician and patient priorities can impact the clinician\u2013patient relationship, treatment concordance, and potential health outcomes. Studies have suggested that patients with osteoarthritis (OA) may prioritize comorbidities over their OA, but as yet no explicit systematic exploration of OA patients\u2019 priorities in relation to comorbidities exists. This study aims to explore how patients prioritize their OA among their conditions, which factors underlie this prioritization, and whether and why these priorities change over time.\r\nMethods. A secondary analysis of qualitative data was conducted utilizing 4 existing data sets collated from the 3 research centers involved. Purposive sampling provided a sample of 30 participants who all had OA and comorbidities. The research team collectively coded and analyzed the data thematically. Results. Three groups of patients emerged from the analysis. The 2 smaller groups had stable priorities (where OA was or was not prioritized) and illustrated the importance of factors, such as personal social context and the specific nature of the comorbid conditions. The third and largest group reported priorities that shifted over time. Shifting appeared to be influenced by the participants\u2019 perceptions of control and/or interactions with clinical professionals, and could have important consequences for self-management behavior. Conclusion. The various factors underlying patients\u2019 priorities among their conditions, and the fluctuating nature of these priorities, highlight the importance of regular assessments during clinician\u2013patient consultations to allow better communication and treatment planning, and ultimately optimize patient outcomes.
\n \n\n \n \n