Design and Analysis of Cross-Over Trials

Jones, Byron; Kenward, Michael G.

doi:10.4324/9780203009277

Cited by 649 publications

(453 citation statements)

References 0 publications

Supporting

Mentioning

440

Contrasting

Unclassified

Order By: Relevance

“…The response to treatment was analysed as a standard two-phase crossover trial with binary data. Estimates of true treatment success rates were derived using a log-linear model [23]. The exact p-value was calculated from Prescott's test.…”

Section: Discussionmentioning

confidence: 99%

Managing migraine headaches experienced by patients who self–report with menstrually related migraine: a prospective, placebo–controlled study with oral sumatriptan

2005

View full text Add to dashboard Cite

IntroductionMigraine is equally common in both genders until puberty, when it becomes approximately three times more prevalent in women than in men [1,2]. This greater prevalence in women is due, in part, to the influence of female sexual hormones. More than 50% of women relate a periodicity of their migraine attacks to their menstrual cycles [3,4], although in most cases they also have migraine attacks outside the menstrual period. It has been proposed that menstrual migraine is due to oestrogen withdrawal in the late luteal phase of the normal menstrual cycle acting as a migraine trigger in predisposed individuals [5], but other factors such as prostaglandin release have also been implicated [6]. However, the true prevalence of menstrual migraine attacks (those occurring during the peri-menstrual period) is unclear due to retrospective reports and variable definitions of menstrual migraine [7].Menstrual migraine attacks are treated with standard acute therapies and with specific prophylactic treatments such as oestrogen supplements taken during the peri-menstrual period [6]. It has been widely considered that menstrual migraine attacks are more severe and less responsive to drug treatments and non-pharmacological approaches than non-menstrual attacks [6,8], but the data to support this J Headache Pain (2005) 6:81-87 DOI 10.1007 Managing migraine headaches experienced by patients who self-report with menstrually related migraine: a prospective, placebo-controlled study with oral sumatriptan Abstract The objective was to evaluate the efficacy and tolerability of oral sumatriptan (100 mg) in patients who self-reported with menstrually related migraine. A prospective, multicentre, randomised, double-blind, placebocontrolled, two-group crossover study was carried out in 20 UK primary and secondary care surgeries. Of 115 patients with a self-reported history of menstrually related migraine that entered the study, 93 patients completed it. Patients treated all migraine attacks for 2 months with sumatriptan (100 mg) and for 2 months with placebo. The primary endpoint was the proportion of patients reporting headache relief at 4 hours for the first treated attack.Only 11% of patients fulfilled the protocol definition of menstrually related migraine. Patients reported a variable pattern of migraine attacks occurring inside and outside the menstrual window. For the first attack, significantly more patients receiving sumatriptan than placebo reported headache relief for attacks occurring inside (67% vs. 33%, p=0.007) and outside (79% vs. 31%, p<0.001) the menstrual period. Sumatriptan was generally well tolerated. Oral sumatriptan (100 mg) is an effective and well tolerated acute treatment for patients who report menstrually related migraine.

show abstract

Section: Discussionmentioning

confidence: 99%

Managing migraine headaches experienced by patients who self–report with menstrually related migraine: a prospective, placebo–controlled study with oral sumatriptan

2005

View full text Add to dashboard Cite

show abstract

“…When applicable, regression models benefitted from the robust standard error approach [29], so that the participant's series of repeated measurements were considered as individual clusters. Absence of a carryover effect was checked before treatment-effect analysis, following a method described elsewhere [30]. Statistical comparisons were two-tailed, and all testing was conducted at a ¼ 0.05, on per protocol data.…”

Section: Methodsmentioning

confidence: 99%

Development of a cross-over randomized trial method to determine the acceptability and safety of novel ready-to-use therapeutic foods

Dibari

Bahwere

Huerga³

et al. 2013

Nutrition

View full text Add to dashboard Cite

Development of a cross-over randomized trial method to determine the acceptability and safety of novel ready-to-use therapeutic foods Objective: To develop a method for determining the acceptability and safety of ready-to-use therapeutic foods (RUTF) before clinical trialing. Acceptability was defined using a combination of three consumption, nine safety, and six preference criteria. These were used to compare a soy/ maize/sorghum RUTF (SMS-RUTFh), designed for the rehabilitation of human immunodeficiency virus/tuberculosis (HIV/TB) wasted adults, with a peanut-butter/milk-powder paste (P-RUTF; brand: Plumpy'nut) designed for pediatric treatment. Methods: A cross-over, randomized, controlled trial was conducted in Kenya. Ten days of repeated measures of product intake by 41 HIV/TB patients, >18 y old, body mass index (BMI) 18-24 kg$m À2 , 250 g were offered daily under direct observation as a replacement lunch meal. Consumption, comorbidity, and preferences were recorded. Results: The study arms had similar age, sex, marital status, initial BMI, and middle upper-arm circumference. No carryover effect or serious adverse events were found. SMS-RUTFh energy intake was not statistically different from the control, when adjusted for BMI on day 1, and the presence of throat sores. General preference, taste, and sweetness scores were higher for SMS-RUTFh compared to the control (P < 0.05). Most consumption, safety, and preference criteria for SMS-RUTFh were satisfied except for the average number of days of nausea (0.16 versus 0.09 d) and vomiting (0.04 versus 0.02 d), which occurred with a higher frequency (P < 0.05). Conclusion: SMS-RUTFh appears to be acceptable and can be safely clinically trialed, if close monitoring of vomiting and nausea is included. The method reported here is a useful and feasible approach for testing the acceptability of ready-to-use foods in low income countries. MSF Field Research

show abstract

“…Jones and Keward [14] suggest the use of the ordinary least squares (OLS) covariance estimate is in keeping with the spirit of this approach. However, for data which are unbalanced or have missing values, so that the OLS and restricted maximum likelihood (REML) estimates do not coincide, it may be more practical to simply adopt the unstructured REML estimate, which is widely implemented in existing software.…”

Section: Box's Correctionmentioning

confidence: 98%

The analysis of very small samples of repeated measurements II: A modified Box correction

Skene

Kenward

2010

Statistics in Medicine

Self Cite

View full text Add to dashboard Cite

There is a need for appropriate methods for the analysis of very small samples of continuous repeated measurements. A key feature of such analyses is the role played by the covariance matrix of the repeated observations. When subjects are few it can be difficult to assess the fit of parsimonious structures for this matrix, while the use of an unstructured form may lead to a serious lack of power. The Kenward-Roger adjustment is now widely adopted as a means of providing an appropriate inferences in small samples, but does not perform adequately in very small samples. Adjusted tests based on the empirical sandwich estimator can be constructed that have good nominal properties, but are seriously underpowered. Further, when such data are incomplete, or unbalanced, or nonsaturated mean models are used, exact distributional results do not exist that justify analyses with any sample size. In this paper, a modification of Box's correction applied to a linear model based F -statistic is developed for such small sample settings and is shown to have both the required nominal properties and acceptable power across a range of settings for repeated measurements.

show abstract

Design and Analysis of Cross-Over Trials

Cited by 649 publications

References 0 publications

Managing migraine headaches experienced by patients who self–report with menstrually related migraine: a prospective, placebo–controlled study with oral sumatriptan

Managing migraine headaches experienced by patients who self–report with menstrually related migraine: a prospective, placebo–controlled study with oral sumatriptan

Development of a cross-over randomized trial method to determine the acceptability and safety of novel ready-to-use therapeutic foods

The analysis of very small samples of repeated measurements II: A modified Box correction

Contact Info

Product

Resources

About