Assessing causal treatment effect estimation when using large observational datasets.

BACKGROUND: Recently, there has been a heightened interest in developing and evaluating different methods for analysing observational data. This has been driven by the increased availability of large data resources such as Electronic Health Record (EHR) data alongside known limitations and changing characteristics of randomised controlled trials (RCTs). A wide range of methods are available for analysing observational data. However, various, sometimes strict, and often unverifiable assumptions must be made in order for the resulting effect estimates to have a causal interpretation. In this paper we will compare some common approaches to estimating treatment effects from observational data in order to highlight the importance of considering, and justifying, the relevant assumptions prior to conducting an observational analysis. METHODS: A simulation study was conducted based upon a small cohort of patients with chronic obstructive pulmonary disease. Two-stage least squares instrumental variables, propensity score, and linear regression models were compared under a range of different scenarios including different strengths of instrumental variable and unmeasured confounding. The effects of violating the assumptions of the instrumental variables analysis were also assessed. Sample sizes of up to 200,000 patients were considered. RESULTS: Two-stage least squares instrumental variable methods can yield unbiased treatment effect estimates in the presence of unmeasured confounding provided the sample size is sufficiently large. Adjusting for measured covariates in the analysis reduces the variability in the two-stage least squares estimates. In the simulation study, propensity score methods produced very similar results to linear regression for all scenarios. A weak instrument or strong unmeasured confounding led to an increase in uncertainty in the two-stage least squares instrumental variable effect estimates. A violation of the instrumental variable assumptions led to bias in the two-stage least squares effect estimates. Indeed, these were sometimes even more biased than those from a naïve linear regression model. CONCLUSIONS: Instrumental variable methods can perform better than naïve regression and propensity scores. However, the assumptions need to be carefully considered and justified prior to conducting an analysis or performance may be worse than if the problem of unmeasured confounding had been ignored altogether.

Completion of hand-written surgical consent forms is frequently suboptimal and could be improved by using electronically generated, procedure-specific forms.

INTRODUCTION: Completion of hand-written consent forms for surgical procedures may suffer from missing or inaccurate information, poor legibility and high variability. We audited the completion of hand-written consent forms and trialled a web-based application to generate modifiable, procedure-specific consent forms. METHODS: The investigation comprised two phases at separate UK hospitals. In phase one, the completion of individual responses in hand-written consent forms for a variety of procedures were prospectively audited. Responses were categorised into three domains (patient details, procedure details and patient sign-off) that were considered "failed" if a contained element was not correct and legible. Phase two was confined to a breast surgical unit where hand-written consent forms were assessed as for phase one and interrogated for missing complications by two independent experts. An electronic consent platform was introduced and electronically-produced consent forms assessed. RESULTS: In phase one, 99 hand-written consent forms were assessed and the domain failure rates were: patient details 10%; procedure details 30%; and patient sign-off 27%. Laparoscopic cholecystectomy was the most common procedure (7/99) but there was significant variability in the documentation of complications: 12 in total, a median of 6 and a range of 2-9. In phase two, 44% (27/61) of hand-written forms were missing essential complications. There were no domain failures amongst 29 electronically-produced consent forms and no variability in the documentation of potential complications. CONCLUSION: Completion of hand-written consent forms suffers from wide variation and is frequently suboptimal. Electronically-produced, procedure-specific consent forms can improve the quality and consistency of consent documentation.

High-quality CMV-specific CD4+ memory is enriched in the lung allograft and is associated with mucosal viral control.

The maintenance of CMV-specific T cell memory in lung transplant recipients (LTRs) is critical for host defense and allograft durability, particularly in donor(+) /recipient(-) (D(+) R(-) ) individuals who demonstrate increased mortality. We studied CD4(+) and CD8(+) CMV-specific memory responses to phosphoprotein 65 (pp65) in a prospective cohort of 18 D(+) R(-) LTRs, from bronchoalveolar lavage (BAL)-obtained lung mononuclear cells (LMNC) and PBMC. Unexpectedly, pp65-specific CD4(+) and CD8(+) IFN-γ memory responses from LMNC were similar, in contrast to persistent CD8(+) predominance in PBMC. Unlike the pulmonary CD8(+) predominance during acute primary infection, compartmental equalization occurred in the CMV-specific CD8(+) memory pool during chronic infection, whereas CMV-specific CD4(+) memory was enriched in the bronchoalveolar space. Moreover, CMV-specific CD4(+) memory T cells with multifunctional production of IFN-γ, TNF-α, IL-2 and MIP-1ß were significantly increased in LMNCs, in contrast to similar intercompartmental CD8(+) memory function. Moreover, the absolute number of CMV-specific CD4(+) IFN-γ(+) memory cells in BAL was significantly increased in LTRs exhibiting viral control compared to those with CMV early antigen positivity. Collectively, these data demonstrate both preferential distribution and functional quality of CMV-specific CD4(+) memory in the lung allograft during chronic infection, and show an important association with CMV mucosal immunity and viral control.

Exploration of chest wall perforator vascular anatomy on standard breast MRI: A potential aid to chest wall perforator flap planning.

Local chest wall perforator flaps (CWPFs) are a volume replacement technique permitting breast-conserving surgery in patients who otherwise may require a mastectomy. These flaps are based on one or more perforating arteries arising from the lateral chest wall that travel through the soft tissue and into the sub-dermal plexus to perfuse the flap. Examples include the lateral intercostal and lateral thoracic artery perforators (LICAP and LTAP, respectively). Cross-sectional imaging of perforating vessels is not routinely performed, and vessels are mapped pre- and peri-operatively using a hand-held acoustic doppler device. As many breast cancer patients undergo pre-operative MRI scanning for oncological purposes, we investigated the role of MRI in mapping the vascular anatomy to aid with the surgical planning of CWPFs. We collated data retrospectively on a cohort of breast cancer patients who underwent breast MRI as part of routine pre-operative imaging. Axial 3D high-resolution dynamic contrast-enhanced MRI sequences with multiplanar reconstructions were analysed by a consultant radiologist. The presence and calibre of lateral chest wall perforator vessels were assessed. Fifty patients were suitable for inclusion. A consistent pattern of lateral chest wall vasculature was observed. Forty-eight patients (96%) demonstrated a bilateral lateral thoracic artery (LTA) descending inferiorly along the chest wall with two-thirds of these communicating with perforating intercostal vessels. True independent LICAP vessels were identified in six patients. From our observations, lateral CWPFs are dependent on an intricate intercommunication between intercostal vessels and the LTA which in turn supply perforators to the lateral chest wall donor site.

CD154 blockade abrogates allospecific responses and enhances CD4(+) regulatory T-cells in mouse orthotopic lung transplant.

Acute cellular rejection (ACR) is a common and important clinical complication following lung transplantation. While there is a clinical need for the development of novel therapies to prevent ACR, the regulation of allospecific effector T-cells in this process remains incompletely understood. Using the MHC-mismatched mouse orthotopic lung transplant model, we investigated the short-term role of anti-CD154 mAb therapy alone on allograft pathology and alloimmune T-cell effector responses. Untreated C57BL/6 recipients of BALB/c left lung allografts had high-grade rejection and diminished CD4(+) : CD8(+) graft ratios, marked by predominantly CD8(+) >CD4(+) IFN-γ(+) allospecific effector responses at day 10, compared to isograft controls. Anti-CD154 mAb therapy strikingly abrogated both CD8(+) and CD4(+) alloeffector responses and significantly increased lung allograft CD4(+) : CD8(+) ratios. Examination of graft CD4(+) T-cells revealed significantly increased frequencies of CD4(+) CD25(+) Foxp3(+) regulatory T-cells in the lung allografts of anti-CD154-treated mice and was associated with significant attenuation of ACR compared to untreated controls. Together, these data show that CD154/CD40 costimulation blockade alone is sufficient to abrogate allospecific effector T-cell responses and significantly shifts the lung allograft toward an environment predominated by CD4(+) T regulatory cells in association with an attenuation of ACR.

Contradiction of auditory and visual information by brain stimulation.

Cats with permanently implanted electrodes were trained to discriminate between trains of flashes or clicks at two different repetition frequencies. After substantial overtraining with these sensory stimuli, high levels of stimulus generalization were obtained to electrical stimulation of the reticular formation at either frequency stimultaneously with contradictory flicker or click stimulation at the opposite frequency resulted in control of the behavior by the reticualr stimulus. Lateral geniculate stimulation failed to show this effect.

Equipotentiality quantified: the anatomical distribution of the engram.

Sensory events and the representation of past experience cause distinctive changes in the electrical activity of widespread regions of the brain. These regions have similar roles in the engram in the sense that they all seem to participate in responses to external events and in subsequent representations of these events. However, the relative contribution of these processes to the activity of different brain regions is quantitatively different, in that some regions are much more strongly affected than others. These results may constitute the basis for reconciliation of localizationist and antilocalizationist views of brain function.

Effects of visual form on the evoked response.

The average visual evoked potentials elicited from relaxed human subjects are different for a blank visual field and one containing a geometric form, are different for different geometric forms of equal area, are similar for versions of the same geometric fortn of unequal area, and are different for two printed words equated for total letter area. These findings suggest that the waveform of evoked responses is not determined solely by the set of peripheral receptors which is stimulated, but it also reflects the perceptual content of the stimulus.

Evoked response correlate of symbol and significate.

Changing the source and intensity of the auditory signal to six trained cats responding to meaningful auditory stimuli permits exogenous and endogenous processes in the auditory evoked potential to be separated. For short-latency exogenous processes, latency and amplitude depend on the parameters of the physical stimulus. However, the amplitude and shape of longer-latency endogenous processes are essentially independent of the location and intensity of the signal source and seem to be invariant concomitants of the significance of the signal.

Evoked response correlate of symbol and significate.

Changing the source and intensity of the auditory signal to six trained cats responding to meaningful auditory stimuli permits exogenous and endogenous processes in the auditory evoked potential to be separated. For short-latency exogenous processes, latency and amplitude depend on the parameters of the physical stimulus. However, the amplitude and shape of longer-latency endogenous processes are essentially independent of the location and intensity of the signal source and seem to be invariant concomitants of the significance of the signal.

Neural readout from memory during generalization.

Eight cats with implanted electrodes were trained to obtain food on presentation of one flicker frequency and to avoid shock on presentation of a second flicker frequency. A third flicker frequency, midway between the first and the second, was then presented. Differential generalization ensued, in which either the food response or the avoidance response was performed. Average evoked potentials from generalization trials with different outcomes were significantly different. The wave shape elicited by the stimulus for generalization closely resembled the usual response to the appropriate signal for the behavior which was displayed. This constitutes evidence for release of a neural process representing previous experience. The release of this process begins about 35 milliseconds after stimulation.

Stable and plastic unit discharge patterns during behavioral generalization.

A movable microelectrode was implanted in adult cats trained to respond differentially to two different frequencies of light flicker. Unit responses were recorded along cortical and thalamic trajectories. The late components of the poststimulus response of 29% of the cells examined showed statistically significant differences when data from different behavioral outcomes to the same neutral generalization stimulus were compared.

Evoked potential correlates of generalization.

Late components of evoked potentials recorded during the occurrence and nonoccurrence of generalization are different. During generalization the evoked response waveshapes resembled those elicited by the conditioned stimulus during correct performance. The differences are statistically significant.

Neurometrics: computer-assisted differential diagnosis of brain dysfunctions.

Normative developmental equations provide reliable descriptors of brain electrical activity in people 6 to 90 years old. Healthy persons display only chance deviations beyond predicted ranges. Patients with neurological impairment, subtle cognitive dysfunctions, or psychiatric disorders (including dementia and primary depression) show a high incidence of abnormal values. The magnitude of the deviations increases with clinical severity. Different disorders are characterized by distinctive profiles of abnormal values of brain electrical features. Computerized differential classification of some of these disorders can be achieved with high accuracy. Such classification, providing objective corroboration of brain dysfunctions, may be a useful adjunct to psychiatric diagnosis, which relies primarily on subjective clinical impressions. These methods may provide independent criteria for diagnostic validity, evaluations of treatment efficacy, and more individualized therapy.

Information delivery and the sensory evoked potential.

The waveform of evoked responses recorded from human scalp is not determined solely by the physical eliciting stimulus, but also varies as a function of the effective information provided by the stimulus. There is a positive component whose latency is determined by the point in time at which ambiguity is reduced, and whose shape and amplitude are influenced by whether it is the presence or absence of an external event which delivers the information.

Observation learning in cats.

In two experiments cats acquired a stimulus-controlled approach or avoidance response by observational or conventional shaping procedures. Observer cats acquired the avoidance response (hurdle jumping in response to a buzzer stimulus) significantly faster and made fewer errors than cats that were conventionally trained. Observer cats acquired the approach response (lever pressing for food in response to a light stimulus) with significantly fewer errors than cats that were conventionally trained. In some cases, observer cats committed one or no errors while reaching criterion.

Developmental equations for the electroencephalogram.

Thirty-two linear regression equations predict the frequency composition of the electroencephalogram within four frequency bands, for four bilateral regions of the brain, as a function of age. Equations based on such data from large groups of healthy children in the United States and Sweden are closely similar. These equations describe the development of the electrical activity of the normal human brain, independent of cultural, ethnic, socioeconomic, or sex factors.

Developmental equations reflect brain dysfunctions.

Developmental equations, which predict 32 parameters of the electroencephalogram recorded from the healthy human as a function of age, were tested in diverse groups of children. Few significant deviations were found in normal children, even in a culture different from the one on which the equations were based. A high incidence of significant deviations was found in children with learning disabilities and those at risk for various neurological disorders.