"˜Regression toward the mean" in replications of fundamental research studies on homeopathic high potencies?

Around 100 high-potency studies, each one replicating earlier studies, were located in a bibliometric study on fundamental research [1]. 70% of the replication studies yielded the same result as the original study, 10% the opposite result, and 20% found no difference between the test and control groups. This is to say, the final standing was 70 positive, 10 negative and 20 undecided results. Among the replication studies performed by the original laboratory, the outcomes were 83 positive and 5 negative (12 undecided); in the ones performed in laboratories of independent researchers invited to participate in a multicentric study the corresponding outcomes were 75 and 11 (14), and the ones of studies performed by external researchers 48 and 14 (38). Had the experiments been performed on purely random phenomena the most probable outcomes would have been 33 : 33 : 33. In a follow-up study we sought to establish whether the outcomes of the replication studies might be attributed to regression toward the means. As is known, this term describes the situation in which a random experiment yields an extreme result, while subsequent replications give average results quite independent from the initial one. By contrast, when the experiment is not entirely based entirely on chance, subsequent replications tend to yield results that deviate from the average in the same direction as the initial one. We thus grouped the studies into 3 clusters of at least 7 articles according to their date of publication. The relative frequencies of positive, undecided and negative results were graphically represented through by curves to allow for visual comparison of the initial result versus the results of the replication studies for each of the 3 clusters. Regression towards the mean was neither observed in the results of the replication studies taken together nor in the group of laboratory-internal replication studies. By contrast, the group of multicentric replication studies exhibited increase of the relative frequency of undecided results and decrease of the one of positive results. In the group of external replication studies the relative frequency of undecided results increased and the one of positive and of negative results decreased. The statistical probability of drawing a ‘positive result’ from a total of 3 equally probable options (positive, undecided, negative) is 33.3%. By contrast, within the entirety of replication studies the frequency of positive results was 77% on the first replication, 66.7% on the second and 66.7% on the third. Unlike the results from the group of laboratory-internal studies, the ones of multicentric and external replications are suggestive of regression towards the mean. However the magnitude of this effect is far weaker than would be expected from a purely random distribution of results. One question that remains is why there appears to have been practically no regression in the laboratory-internal replications. This should not be cheaply attributed to publication bias. Especially in experiments on homoeopathic high potencies, it could well be that numerous, in some part hidden factors of influence are involved, which internal laboratories – possibly unwittingly through unreflected working habits– are better at controlling than external laboratories, regardless of the effort invested to this end. Keywords: high dilution, replication experiments, regression to the mean References [1] Endler PC, Bellavite P, Bonamin L, Jäger T. Mazon S. Replications of fundamental research models in ultra high dilutions, 1994 and 2015: an update on a bibliometric study. Homeopathy 2015;104:235-245

Survey design to monitor drug efficacy for the control of soil-transmitted helminthiasis and schistosomiasis

Background Control of soil-transmitted helminthiasis and schistosomiasis relies heavily on regular preventive chemotherapy. Monitoring drug efficacy is crucial to provide early warning of treatment failures. The World Health Organization (WHO) recommends a survey design in which only egg-positive individuals are retested after treatment. Although this practice makes more efficient use of resources, it may lead to biased drug efficacy estimates. Methods We performed a simulation study to assess the potential for bias when evaluating drug efficacy using the WHO-recommended survey design, and to identify alternative designs for evaluating drug efficacy that are less affected by bias. These designs were also based on selection of egg-positive individuals, but involve retesting them a second time at baseline and up to two times at follow-up. The utility of the different designs was compared fairly by constraining them to the same budget. Results The standard procedure of selecting egg-positive individuals can introduce a substantial positive bias in drug efficacy due to regression towards the mean, particularly when infection levels or drug efficacy are low. This bias was completely eliminated by using a second baseline sample, conditionally on the first sample being excluded from analysis. Precision of estimates can be improved by increasing the number of thick smears and/or samples per person at follow-up, despite fewer individuals being tested within the same budget. Discussion We present optimised survey designs to monitor drug efficacy in field settings which are highly relevant for sustained control of soil-transmitted helminths and schistosomiasis, as well as onchocerciasis and lymphatic filariasis.

PP081 Relation Between Pain And Treatment/Activity Based On Mobile App Data

INTRODUCTION:Recently, a number of mobile apps to record symptoms and medication by patients themselves have been developed. These apps are expected to improve the patients' symptoms through self-management, and to enable a smooth decision making through effective communication between doctors and patients. “Itami Renrakucho” (Pain Diary, Welby Inc.) is one of these apps that records body pain, medication, physical conditions, and activity in life. We examined the relationship between pain and medication/activity based on its data.METHODS:Data between 25 December 2015 and 9 December 2016 were used. Medication and degree of pain (0-10, low < high) were recorded at morning, daytime, evening, and bedtime. Of ninteen activities, up to three were recorded about whether they could or could not do them. We compared the degree of pain among different frequency/timing of medication, or activities that they could or could not do.RESULTS:Data included 708 individuals. Among 561 individuals who answered about pain, the mean (Standard Deviation, SD) degree was 5.0 (2.3). The mean degree in individuals taking 0, 1, 2, 3, and 4 times medication a day were 4.6, 5.0, 5.4, 5.5, and 6.2, respectively. Regarding medication timing and degree of pain in two consecutive time points (t0, t1), regression towards the mean occurred for individuals without medication in both time points. The degree changed more for individuals taking medicine only at t0, but not for those taking at both time points. Weaker pain was reported when they could do hanging laundry and rising early than when they could not, but they could do shopping, strolling and light exercise even having stronger pain.CONCLUSIONS:We showed a tendency of relationship between pain and medication/activity based on the data from the app. More data and connecting to claims will help us to show characteristics of patients and diseases, select a treatment, and evaluate a medicine.

In or out? Poverty dynamics among older individuals in the UK

Using the largest household panel surveyUnderstanding Society, this paper investigates low-income dynamics among pensioner households in the UK controlling for biases due to initial conditions and non-random survey attrition. Estimation results indicate there is a correlation between initial and conditional poverty status, specifically, there is regression towards the mean. The results find no evidence of a correlation between initial poverty status, conditional poverty status and survey attrition. The findings show the importance of benefit income in determining poverty status, suggesting that a dichotomous measure such as poverty status may not suitably reflect actual pensioner living standards. Aside from benefit income, receipt of employer and occupational pension, health, education and subjective financial situation are important in determining initial and conditional poverty status. Stylised examples highlight the significant differences in the ‘poverty experience’ which arise due to differences individual and household characteristics.

The Influence of Clinical Experience and Photographic Presentation on Age Assessment of Women

Background: Epidemiological studies have reported that a higher perceived age is associated with poor health and higher mortality. However, the method used for the assessment of perceived age differs between studies with regard to age, gender, the number and occupation of assessors as well as the presentation of participants. Objective: It is not known whether the clinical experience of the assessor or photographic presentation have an influence on the assessment of perceived age, which the present study aimed to investigate. Methods: In a cross-sectional study of 460 women aged 25-93 years, 10 consultants and 10 residents were asked to estimate the age of each participant using three different photographic presentations: facial photograph, whole-body photograph, and combined facial and whole-body photographs. Data were analyzed by means of summary statistics and linear mixed models. Results: The inter-class correlation coefficient within each assessor group and photographic presentation varied from 0.66 to 0.75. Limits of agreement were in a broad range but were similar in the two assessor groups. The best inter-assessor agreement was obtained from photographs of both the face and the whole body. Intra- and inter-assessor agreements between photographic presentations were similar among both assessor groups. The accuracy in age assessment was significantly influenced by the photographic presentation but not by the clinical experience of the assessor. The difference in the mean perceived age of a participant of average age was estimated as +0.40 years (95% CI: -1.80; 2.59) for consultants versus residents, -2.05 years (95% CI: -2.90; -1.19) for facial photographs versus both facial and whole-body photographs, and -1.44 years (95% CI: -2.30; -0.58) for whole-body photographs versus both facial and whole-body photographs. A regression towards the mean age was seen. Conclusion: The assessment of perceived age was influenced by the photographic presentation but not by the clinical experience of the assessor.

Statistical cures and other fallacies

Relapse rates and thus the impact of therapies have been decreasing. Why they decline and the impact on our ability to understand which treatments are better require more than simple math. The objective of this review is to discuss the impact of regression to the mean, changes in outcome effects and how we compare outcomes over time and among studies. This paper provides discourse on the topics of regression to the mean, some examples of the pitfalls of changes and some difficulties in the interpretation of the common percentage change in outcomes. The results show that we can often be deceived by what we think we see and they also demonstrate how such confusion evolves in the literature. This article aims to caution against the over-interpretation of changes from baseline, which are helped along by regression towards the mean and other factors. Furthermore, how we interpret changes from baseline requires care and not wishful thinking, coupled with careful digestion of seemingly reasonable explications of results.

Musical training generalises across modalities and reveals efficient and adaptive mechanisms for judging temporal intervals

Expert musicians are able to accurately and consistently time their actions during a musical performance. We investigated how musical expertise influences the ability to reproduce auditory intervals and how this generalises to vision in a ‘ready-set-go’ paradigm. Subjects reproduced time intervals drawn from distributions varying in total length (176, 352 or 704 ms) or in the number of discrete intervals within the total length (3, 5, 11 or 21 discrete intervals). Overall musicians performed more veridically than non-musicians, and all subjects reproduced auditory-defined intervals more accurately than visually-defined intervals. However non-musicians, particularly with visual intervals, consistently exhibited a substantial and systematic regression towards the mean of the interval. When subjects judged intervals from distributions of longer total length they tended to exhibit more regression towards the mean, while the ability to discriminate between discrete intervals within the distribution had little influence on subject error. These results are consistent with a Bayesian model which minimizes reproduction errors by incorporating a central tendency prior weighted by the subject’s own temporal precision relative to the current intervals distribution (Cicchini et al., 2012; Jazayeri and Shadlen, 2010). Finally a strong correlation was observed between all durations of formal musical training and total reproduction errors in both modalities (accounting for 30% of the variance). Taken together these results demonstrate that formal musical training improves temporal reproduction, and that this improvement transfers from audition to vision. They further demonstrate the flexibility of sensorimotor mechanisms in adapting to different task conditions to minimise temporal estimation errors.