Impact of non-uniform correlation structure on sample size and power in multiple-period cluster randomised trials.

Stepped wedge and cluster randomised crossover trials are examples of cluster randomised designs conducted over multiple time periods that are being used with increasing frequency in health research. Recent systematic reviews of both of these designs indicate that the within-cluster correlation is typically taken account of in the analysis of data using a random intercept mixed model, implying a constant correlation between any two individuals in the same cluster no matter how far apart in time they are measured: within-period and between-period intra-cluster correlations are assumed to be identical. Recently proposed extensions allow the within- and between-period intra-cluster correlations to differ, although these methods require that all between-period intra-cluster correlations are identical, which may not be appropriate in all situations. Motivated by a proposed intensive care cluster randomised trial, we propose an alternative correlation structure for repeated cross-sectional multiple-period cluster randomised trials in which the between-period intra-cluster correlation is allowed to decay depending on the distance between measurements. We present results for the variance of treatment effect estimators for varying amounts of decay, investigating the consequences of the variation in decay on sample size planning for stepped wedge, cluster crossover and multiple-period parallel-arm cluster randomised trials. We also investigate the impact of assuming constant between-period intra-cluster correlations instead of decaying between-period intra-cluster correlations. Our results indicate that in certain design configurations, including the one corresponding to the proposed trial, a correlation decay can have an important impact on variances of treatment effect estimators, and hence on sample size and power. An R Shiny app allows readers to interactively explore the impact of correlation decay.

Correction to: Effectiveness of a new model of primary care management on knee pain and function in patients with knee osteoarthritis: Protocol for THE PARTNER STUDY.

After the publication of this protocol [1], our collaborator Prima Health solutions advised us of their intent to withdraw from the study.

Targeted anthelmintic treatment of parasitic gastroenteritis in first grazing season dairy calves using daily live weight gain as an indicator.

Control of parasitic gastroenteritis in cattle is typically based on group treatments with anthelmintics, complemented by grazing management, where feasible. However, the almost inevitable evolution of resistance in parasitic nematodes to anthelmintics over time necessitates a reappraisal of their use in order to reduce selection pressure. One such approach is targeted selective treatment (TST), in which only individual animals that will most benefit are treated, rather than whole groups of at-risk cattle. This study was designed to assess the feasibility of implementing TST on three commercial farms, two of which were organic. A total of 104 first-grazing season (FGS), weaned dairy calves were enrolled in the study; each was weighed at monthly intervals from the start of the grazing season using scales or weigh-bands. At the same time dung and blood samples were collected in order to measure faecal egg counts (FEC) and plasma pepsinogen, respectively. A pre-determined threshhold weight gain of 0.75kg/day was used to determine those animals that would be treated; the anthelmintic used was eprinomectin. No individual animal received more than one treatment during the grazing season and all treatments were given in July or August; five animals were not treated at all because their growth rates consistently exceeded the threshold. Mean daily live weight gain over the entire grazing season ranged between 0.69 and 0.82kg/day on the three farms. Neither FEC nor pepsinogen values were significantly associated with live weight gain. Implementation of TST at farm level requires regular (monthly) handling of the animals and the use of weigh scales or tape, but can be integrated into farm management practices. This study has shown that acceptable growth rates can be achieved in FGS cattle with modest levels of treatment and correspondingly less exposure of their nematode populations to anthelmintics, which should mitigate selection pressure for resistance by increasing the size of the refugia in both hosts and pasture.

Stepped wedge designs: insights from a design of experiments perspective.

Stepped wedge designs (SWDs) have received considerable attention recently, as they are potentially a useful way to assess new treatments in areas such as health services implementation. Because allocation is usually by cluster, SWDs are often viewed as a form of cluster-randomized trial. However, since the treatment within a cluster changes during the course of the study, they can also be viewed as a form of crossover design. This article explores SWDs from the perspective of crossover trials and designed experiments more generally. We show that the treatment effect estimator in a linear mixed effects model can be decomposed into a weighted mean of the estimators obtained from (1) regarding an SWD as a conventional row-column design and (2) a so-called vertical analysis, which is a row-column design with row effects omitted. This provides a precise representation of "horizontal" and "vertical" comparisons, respectively, which to date have appeared without formal description in the literature. This decomposition displays a sometimes surprising way the analysis corrects for the partial confounding between time and treatment effects. The approach also permits the quantification of the loss of efficiency caused by mis-specifying the correlation parameter in the mixed-effects model. Optimal extensions of the vertical analysis are obtained, and these are shown to be highly inefficient for values of the within-cluster dependence that are likely to be encountered in practice. Some recently described extensions to the classic SWD incorporating multiple treatments are also compared using the experimental design framework.

Stability of symptom patterns in Australian Gulf War veterans: 10-year longitudinal study.

OBJECTIVES: Previously we established that symptoms reported by 1990-1991 Gulf War veterans were correlated and exhibited a pattern with 3 factors (psychophysiological distress, somatic distress and arthroneuromuscular distress), and this pattern was similar to that observed in a military comparison group. In this follow-up study, we examined whether the patterns of symptomatology have changed over time. METHODS: Using data on 56 symptoms that was collected in 2000-2003 (wave 1) and 2011-2012 (wave 2) from an Australian cohort of Gulf War veterans (veterans) and a military comparison group, exploratory factor analysis was conducted and Tucker's Congruence Coefficient (TCC) was used to determine factor structure similarity across study groups and waves. RESULTS: The results showed that the 3 factors observed at wave 1 were still present at wave 2, and factor structures across study groups and study waves were fairly similar, with TCC ranging 0.86-0.92. Veterans consistently reported more symptoms across all 3 factors. Veterans' symptomatology specific to psychophysiological distress increased between waves 1 and 2 (ratio of means 1.15; 95% CI 1.07 to 1.25) but psychophysiological distress symptomatology was constant in the comparison group (ratio of means 0.97; 95% CI 0.89 to 1.06). Somatic and arthroneuromuscular distress symptomatology significantly increased over time for both study groups, although at a similar rate. CONCLUSIONS: While the symptom groupings (measured by the 3 factors) remained unchanged at 10 years of follow-up, and remained comparable between Gulf War and comparison group, symptomatology continued to be elevated in Gulf War veterans than in the comparison group, and was most evident for psychophysiological distress.

A telephone survey of factors affecting willingness to participate in health research surveys.

BACKGROUND: In recent years, reduced participation has been encountered across all epidemiological study designs, both in terms of non-response as well as refusal. A low response rate may reduce the statistical power but, more importantly, results may not be generalizable to the wider community. METHODS: In a telephone survey of 1413 randomly selected members of the Australian general population and of 690 participants sourced from previous studies, we examined factors affecting people's stated willingness to participate in health research. RESULTS: The majority of participants (61 %) expressed willingness to participate in health research in general but the percentage increased when provided with more specific information about the research. People were more willing if they have personal experience of the disease under study, and if the study was funded by government or charity rather than pharmaceutical companies. Participants from previous studies, older people and women were the groups most willing to participate. Younger men preferred online surveys, older people a written questionnaire, and few participants in any age and sex groups preferred a telephone questionnaire. CONCLUSION: Despite a trend toward reduced participation rates, most participants expressed their willingness to participate in health research. However, when seeking participants, researchers should be concrete and specific about the nature of the research they want to carry out. The preferred method of recommended contact varies with the demographic characteristics.

Tipping point analysis of atmospheric oxygen concentration.

We apply tipping point analysis to nine observational oxygen concentration records around the globe, analyse their dynamics and perform projections under possible future scenarios, leading to oxygen deficiency in the atmosphere. The analysis is based on statistical physics framework with stochastic modelling, where we represent the observed data as a composition of deterministic and stochastic components estimated from the observed data using Bayesian and wavelet techniques.

Depression in Gulf War veterans: a systematic review and meta-analysis.

BACKGROUND: Although post-traumatic stress disorder (PTSD) has been a focus of attention in 1990/1991 Gulf War veterans, the excess risk of depression has not been clearly identified. We investigated this through a systematic review and meta-analysis of studies comparing depression in Gulf War veterans to depression in a comparison group of non-deployed military personnel. METHOD: Multiple electronic databases and grey literature were searched from 1990 to 2012. Studies were assessed for eligibility and risk of bias according to established criteria. RESULTS: Of 14 098 titles and abstracts assessed, 14 studies met the inclusion criteria. Gulf War veterans had over twice the odds of experiencing depression [odds ratio (OR) 2.28, 95% confidence interval (CI) 1.88-2.76] and dysthymia or chronic dysphoria (OR 2.39, 95% CI 2.0-2.86) compared to non-deployed military personnel. This finding was robust in sensitivity analyses, and to differences in overall risk of bias and psychological measures used. CONCLUSIONS: Despite divergent methodologies between studies, depression and dysthymia were twice as common in Gulf War veterans and are important medical conditions for clinicians and policymakers to be aware of in managing Gulf War veterans' health.

Efficacy of treatment of cattle for liver fluke at housing: influence of differences in flukicidal activity against juvenile Fasciola hepatica.

Flukicides are commonly administered at housing to cattle that have grazed fluke-infected pastures or that have been purchased from endemic areas. The choice of product is determined by numerous factors, one of which is the stages of Fasciola hepatica that are killed. Flukicides can be categorised into three main groups: (A) those that kill all juvenile stages and adults; (B) those that kill juveniles from six to eight weeks of age and adults and (C) those that kill adults only. This study was conducted on a commercial beef farm in Scotland and was designed to compare the efficacy of flukicides from each of these three classes in terms of their effects on faecal egg output, coproantigen and liveweight gain. The majority of animals in the untreated control group were positive for coproantigen, fluke eggs or both throughout the study duration of 16 weeks. Egg reappearance interval following housing treatment was eight weeks for clorsulon and 13 weeks for nitroxynil, though patent infections in both groups developed in only a small minority of animals; no fluke eggs were recovered from cattle treated with triclabendazole. Coproantigen was detected four weeks before the reappearance of fluke eggs in the dung. Animals treated with flukicides had significantly fewer faecal samples positive for eggs (P<0.006) and coproantigen (P<0.05) following treatment compared with the controls. Despite differences in the efficacy profiles among the flukicide-treated groups, there were no significant differences (P>0.05) in growth rates among any of the four treatment groups. There was, however, a significant negative association (P<0.001) between fluke positivity at housing and subsequent growth performance, irrespective of treatment group.

Targeted antibiotic treatment of lame sheep with footrot using either oxytetracycline or gamithromycin.

A study on parenteral antibiotic treatment in sheep footrot was conducted on 10 farms in southern Germany to obtain information on the efficacy of gamithromycin under practical use conditions compared with a positive control. On each farm, 20 (10 on one farm) lame sheep were clinically evaluated and divided into two groups. On day 1, sheep were treated once according to group with either long-acting oxytetracycline (OTC) at 20 mg/kg bodyweight or gamithromycin at 6 mg/kg; clinical responses were assessed 21 days later. When compared with day 1, both treatments reduced clinical lameness, as reflected in the reduction in the number of footrot-affected feet (OTC: 79.3 per cent; gamithromycin: 93.7 per cent) and in the severity of the lesions. The difference between the two treatments was significant (P<0.01) with an OR of 6.1 in favour of gamithromycin. Of the 33 sheep that were still lame on day 21, nine mildly affected animals were not re-treated and the remaining 24 sheep were re-treated with gamithromycin. On day 42, all but two (on the same farm) of the 33 sheep were cured, giving an overall response rate in this study to one or two parenteral antibiotic treatments of 99 per cent.

LongRange™ (eprinomectin 5%) extended-release injection parasiticide and the utility of extended-activity antiparasitics in cattle.

An extended-release injection, which is administered at a rate of 1 mg eprinomectin/kg body weight, has been developed to provide up to 150 days control of parasites of cattle. The product can facilitate the achievement of two of the fundamental aims of parasite control. The first is protection of the host against the negative impact of susceptible parasites in order to ensure control of disease and to enhance performance. The second is to reduce parasite transmission and hence the challenge to animals when grazing. In addition, farmers and veterinarians can benefit from high levels of convenience and hence compliance from a single administration, which also limits handling stress in the cattle. This introductory paper provides some perspective on the practical applications for this extended-release product under various husbandry systems and in different classes of cattle and discusses its role in sustainable parasite control.

Use of diagnostic markers to monitor fasciolosis and gastrointestinal nematodes on an organic dairy farm.

A 12-month study was conducted to assess and monitor gastrointestinal tract nematodes and liver fluke in cohorts of cattle on a Scottish organic dairy farm. Various diagnostic markers for helminth parasites of cattle from different age groups were assessed monthly from April 2007 to March 2008. First season grazing stock were subjected to significant challenge from Ostertagia ostertagi nematodes as reflected in serum pepsinogen concentrations, which rose markedly in the second half of the grazing season. In addition, plasma albumin concentrations decreased and faecal egg counts (FEC) increased moderately, indicating exposure to both O ostertagi and probably Cooperia oncophora. Second season grazing animals had a peak FEC early in the grazing period, suggestive of a potential carry-over of Ostertagia species infection ('Type 2') during housing. All classes of cattle showed evidence of fluke (Fasciola hepatica) infection. Adult cow exposure to O ostertagi and fluke was estimated via the use of ELISA testing to detect antibodies to O ostertagi and F hepatica and the high levels detected suggested a significant exposure response. Despite low stocking densities and sympathetic grazing management, there was a significant challenge to all grazing stock from gastrointestinal nematodes and liver fluke.

Effect of eprinomectin treatment on milk yield and quality in dairy cows in South Tyrol, Italy.

The effect of treatment with eprinomectin on milk yield, milk composition and somatic cell counts (SCCs) was studied in 105 dairy cows located on seven farms in South Tyrol, Italy. On each farm, half of the animals were treated with eprinomectin and the other half were used as an untreated control group. Three test day records per animal were obtained before treatment (days -117, -75 and -33) and another three test day records were obtained after treatment (days 22, 62 and 131). Test day records comprised milk yield, milk composition, SCC and days in milk. On the day of treatment, blood samples and faecal samples were taken for parasitological analysis. Cows with positive faecal egg counts yielded less milk. A significant effect of eprinomectin on milk yield was observed after treatment and was most pronounced on the second and the third test days after treatment (+1.90 kg [P=0.002] and +2.63 kg [P<0.001], respectively). Furthermore, a significant decrease in SCC was observed on the second test day after treatment.

Determination of the duration of antibacterial efficacy following administration of gamithromycin using a bovine Mannheimia haemolytica challenge model.

The antibacterial efficacy of gamithromycin administered once 1, 5, or 10 days prior to a challenge infection with Mannheimia haemolytica serotype A1 was evaluated. Forty calves were randomly allocated on day -11, restricted by body weight, to one of three treatment groups given gamithromycin at 6 mg/kg of body weight 10, 5, or 1 days before challenge or to an untreated control group. M. haemolytica A1 challenge infections were induced on day 0 by depositing 7.4 × 10(7) CFU at the bifurcation of the main bronchus using a bronchoscope. Clinical observations were made daily from the day of allocation to day 10, when necropsy was scheduled; three calves died or were euthanized in extremis on welfare grounds prior to scheduled necropsy. At necropsy the lungs were removed, pneumonic lesions were scored, and samples of lung tissue were cultured for M. haemolytica. The three groups of animals treated with gamithromycin before challenge had significantly lower lung M. haemolytica counts and fewer clinical signs of respiratory disease than did the saline-treated group. For most of the clinical parameters, the pattern of responses differed significantly (P < 0.05) between the gamithromycin-treated groups and the control group. There were no statistically significant differences between groups in the mean lung lesion scores, partly as a result of high individual variability, particularly within the control group. The administration of gamithromycin 1, 5, and 10 days prior to M. haemolytica A1 challenge resulted in a reduction in bacterial isolation from the lungs and a reduction in the severity of clinical disease.

Temporal studies on Fasciola hepatica in Galba truncatula in the west of Ireland.

The population dynamics and prevalence of Fasciola hepatica in Galba truncatula were investigated on the Teagasc hill sheep farm in Country Mayo, Ireland from August 2006 until March 2008. Galba truncatula (n=974) were collected at fortnightly intervals in 4 snail habitats on the farm, their relative density and mean monthly rainfall and temperature were recorded. Snail abundance was associated with rainfall and temperature. G. truncatula abundance was generally low during winter 2006/2007 (December 2006-February 2007), mid spring and summer 2007 (April-July) and late autumn and winter 2007/2008 (October 2007-February 2008). Overall, two peaks in G. truncatula abundance were evident during the study period, an early spring (March) and late summer/autumn (late August-October) peak. Seasonal differences were observed in the prevalence of F. hepatica in G. truncatula, the greatest prevalence was evident in 2007 during the summer (25%) and autumn (16%). Two seasonal transmission peaks, as evident by the presence of mature infections in snails, one in summer/autumn and the other in late winter/early spring. Not unsurprising, cercariae (28%) and rediae (29%) were the predominant developmental stages observed in mature and pre-adult snails respectively. Worthy of note in this study, cercariae were also evident in pre-adult snails. The present study highlights the influence of milder temperatures and wetter conditions on both snail and fluke development. If the trend in mild and wet weather conditions continues, there is the potential for an increase in the number of G. truncatula habitats and ultimately the prevalence of fasciolosis in grazing livestock in Ireland.

Short communication: Genetic and nongenetic factors influencing Ostertagia ostertagi antibodies in UK Holstein-Friesian cattle.

The purpose of this study was to estimate and discuss the genetic variation, heritability, and effects of nongenetic factors on the ability of Holstein-Friesian cows to produce an immune response by producing IgG antibodies to Ostertagia ostertagi. Total IgG (IgG(1) and IgG(2)) antibody levels were determined using an ELISA and measured using optical density ratio (ODR=OD(sample) - OD(negative control)/OD(positive control) - OD(negative control)) from milk samples collected from 1,276 Holstein-Friesian cattle in 229 commercial dairy farms from 2002 to 2004 during their first (82%) and other (2 to 12) lactations. A sire (n=461) model was fitted to the ODR data using ASREML software, and variance components were estimated. The ability to produce O. ostertagi antibodies as measured by ODR had a heritability of 0.13+/-0.12, and both season of sample and herd had a significant effect on total IgG levels. To conclude, this study has ascertained that genetic variation is present in the ability of dairy cows to mount an immune response to the parasite O. ostertagi. Inasmuch as evidence exists that IgG is linked to protective immunity against the parasite via a reduction in its reproductive ability, this trait may be of potential interest to genetic selection programs as an aid to reduce the effect of O. ostertagi in dairy herds.

Preliminary study on the use of a Teladorsagia circumcincta bulk milk ELISA test in dairy goats under experimental conditions.

An ELISA test for the detection of anti-Teladorsagia circumcincta antibodies in goat milk samples (both individual and bulk milk samples) has been developed in dairy goats experimentally infected with a trickle infection. The results observed were compared with those obtained when some other standard methods for the detection of animals infected with gastrointestinal nematodes (GIN), such as faecal egg counts (FEC) and serum pepsinogen levels. Some factors that could affect the final results of the test (dilution and nature of samples as well as their preservation at 4 degrees C for 0-10 days) were also monitored. The results show that the proposed test could allow the detection of infected herds using bulk milk samples. The levels of anti-T. circumcincta antibodies in milk and serum samples and the concentrations of serum pepsinogen showed a similar pattern over time. Refrigeration of milk samples for less than 5 days or its preservation with potassium dichromate for 10 days did not significantly affect the ability of the test to detect infected animals. Finally, when a commercial Ostertagia ostertagi ELISA kit was used to assay milk samples from goats infected with T. circumcincta, the results suggest that it may be possible to use a crude O. ostertagi antigen-based ELISA for the detection of goats infected with T. circumcincta.

Associations between blood gastrin, ghrelin, leptin, pepsinogen and Ostertagia ostertagi antibody concentrations and voluntary feed intake in calves exposed to a trickle infection with O. ostertagi.

Twenty-five, castrated male Holstein-cross calves, between 4 and 5 months of age, weighing 156.5+/-12.2 kg and reared under conditions designed to minimise the risk of parasitic infection, were allocated to one of the five treatment groups on the basis of initial bodyweight. The groups were (1) ad libitum (ad lib) fed controls (ALC); (2) ad lib fed infected (INF) and treated with topical eprinomectin on Day 56; (3) controls pair-fed with the INF group (PFC); (4) ad lib fed controls treated with eprinomectin on Days 0 and 56 (E-ALC) and (5) ad lib fed, infected and treated with eprinomectin on Days 0 and 56 (E-INF). Infection comprised a trickle infection with the equivalent of 10,000 larvae of Ostertagia ostertagi per day from Day 0 to Day 56 and the study concluded on Day 77. Parameters measured throughout the study included: liveweight, feed intake, faecal egg counts; plasma pepsinogen, gastrin, ghrelin and leptin; plasma antibodies to adult O. ostertagi. No significant differences in feed intake or liveweight gain were observed between any of the different groups, a finding thought to result from the high quality of feed offered. Significant differences between the INF and control groups however were observed in faecal egg counts, plasma pepsinogen, gastrin and O. ostertagi antibodies, which were all elevated, and leptin, which was reduced. Values of these parameters for the E-INF group were intermediate between the INF and ALC groups. Plasma ghrelin showed no association with either feed intake or parasitism. Further studies are needed to fully elucidate the roles of various biochemical and neuroendocrine mediators for inappetence in ruminants with parasitic gastroenteritis.