Non-ignorable missing covariate data in survival analysis: a case-study of an International Breast Cancer Study Group trial

Summary.  Non-ignorable missing data, a serious problem in both clinical trials and observational studies, can lead to biased inferences. Quality-of-life measures have become increasingly popular in clinical trials. However, these measures are often incompletely observed, and investigators may suspect that missing quality-of-life data are likely to be non-ignorable. Although several recent references have addressed missing covariates in survival analysis, they all required the assumption that missingness is at random or that all covariates are discrete. We present a method for estimating the parameters in the Cox proportional hazards model when missing covariates may be non-ignorable and continuous or discrete. Our method is useful in reducing the bias and improving efficiency in the presence of missing data. The methodology clearly specifies assumptions about the missing data mechanism and, through sensitivity analysis, helps investigators to understand the potential effect of missing data on study results.     

Children and well-being: A comparison of employed and non-employed women

While employment improves health, having children at home does not. Indeed, the most likely work-family conflict for women is having children at home. In this study the effect of children on well-being was predicted to differ for employed and non-employed women. In addition, only amongst the employed women was locus of control hypothcsized to moderate this relationship. In a sample of women with partners (n = 116), using multiple-regression analysis, the results indicated that the number of children living at home, rather than the presence of young children, was associated with increased happiness in women. Employment was not predictive of women's well-being. However, it moderated the effect of number of children on their levels of happiness. In addition, non-employed women showed a U-shaped relationship between number of children and psychological distress. Finally, contrary to prediction, locus of control did not moderate the relationship between number of children and women's well-being. The findings are discussed in terms of the role expansion hypothesis.     

Early stopping in seamless phase I/II clinical trials

In recent years, seamless phase I/II clinical trials have drawn much attention, as they consider both toxicity and efficacy endpoints in finding an optimal dose (OD). Engaging an appropriate number of patients in a trial is a challenging task. This paper attempts a dynamic stopping rule to save resources in phase I/II trials. That is, the stopping rule aims to save patients from unnecessary toxic or subtherapeutic doses. We allow a trial to stop early when widths of the confidence intervals for the dose-response parameters become narrower or when the sample size is equal to a predefined size, whichever comes first. The simulation study of dose-response scenarios in various settings demonstrates that the proposed stopping rule can engage an appropriate number of patients. Therefore, we suggest its use in clinical trials.     

Rare and clustered population estimation using the adaptive cluster sampling with some robust measures

The use of robust measures helps to increase the precision of the estimators, especially for the estimation of extremely skewed distributions. In this article, a generalized ratio estimator is proposed by using some robust measures with single auxiliary variable under the adaptive cluster sampling (ACS) design. We have incorporated tri-mean (TM), mid-range (MR) and Hodges-Lehman (HL) of the auxiliary variable as robust measures together with some conventional measures. The expressions of bias and mean square error (MSE) of the proposed generalized ratio estimator are derived. Two types of numerical study have been conducted using artificial clustered population and real data application to examine the performance of the proposed estimator over the usual mean per unit estimator under simple random sampling (SRS). Related results of the simulation study show that the proposed estimators provide better estimation results on both real and artificial population over the competing estimators.     

Technical and vocational education and training in Uganda: A critical analysis

This article undertakes a diagnostic study of the Technical and Vocational Education and Training (TVET) sub‐sector in Uganda, with a view to characterizing the sub‐sector and identifying its potential strengths and weakness. We undertook a cross‐sectional pre‐survey of purposively selected key stakeholders in the TVET sub‐sector. We selected performance indicators following their importance in influencing the TVET reform process. Both qualitative and quantitative data was solicited from the stakeholders. Quantitative data was collected through stakeholder‐specific structured questionnaires, whilst qualitative data was collected through desk review and field visits, individual focused interviews and focus group discussions. Our findings indicate that financing and planning constraints have resulted in poor quality equipment, under‐ and ill‐trained staff, limited adoption of a competence‐based education and training (CBET) curriculum, not to mention the supervision inadequacies of TVET institutions. Besides, the limited TVET sub‐sector interaction with the private sector has incapacitated TVET curriculum development to nurture skills demanded by the private sector. Furthermore, backward technology use in the private sector has equally inhibited the success of student attachment programmes. Finally, legal ambiguities have perpetuated a qualification jungle and overlapping curricula.     

Hard thresholding regression

In this paper, we propose the hard thresholding regression (HTR) for estimating high‐dimensional sparse linear regression models. HTR uses a two‐stage convex algorithm to approximate the ?₀‐penalized regression: The first stage calculates a coarse initial estimator, and the second stage identifies the oracle estimator by borrowing information from the first one. Theoretically, the HTR estimator achieves the strong oracle property over a wide range of regularization parameters. Numerical examples and a real data example lend further support to our proposed methodology.     

Challenges of implementing restorative justice for intimate partner violence: An Islamic perspective

ABSTRACT

Intimate partner violence (IPV) is currently recognized as a critical public health concern and a human rights issue. Not surprisingly, Muslims – a religiously and socio-culturally diverse faith-based populace – are not an exception. To address this complex area of criminal justice and social policy, some scholars advocate implementing restorative justice (RJ) approaches. While RJ approaches have been traditionally used in Islamic cultures for conflict resolutions, to date, few studies have investigated how RJ operates in dealing with IPV in Muslim countries and communities. This article explores how RJ approaches towards IPV operate in some Muslim countries/communities, and offers insights into developing culturally and religiously appropriate ways of implementing RJ in IPV situations among Muslims. Given the prevalence of IPV among Muslims, the question is particularly important and timely. Taking exclusively limited examples of RJ approaches that have been used to mediate IPV cases in Muslim countries and communities, this paper found one significant challenge in the RJ approaches among Muslims: community acceptance of IPV. To reduce IPV in Muslim society, it is necessary to develop treatment models and techniques that meet cultural and religious needs. This paper found that since RJ is not alien to Islamic teachings, RJ approaches can be implemented effectively in dealing with IPV among Muslims by ensuring justice and equity of the abused woman. The findings of this paper will assist policymakers, practitioners, and service providers in providing religiously and culturally appropriate care when addressing IPV issues among Muslims.     

Age and total and free prostate-specific antigen levels for predicting prostate volume in patients with benign prostatic hyperplasia

Objectives: To investigate the predictive values of free prostate-specific antigen (fPSA), total PSA (tPSA) and age on the prostate volume.

Methods: The data of 2148 patients with lower urinary tract symptoms were analyzed retrospectively. The patients who had transrectal ultrasonography guided 10 core biopsies owing to the findings obtained on digital rectal examination and presence of high PSA levels (PSA?=?2.5–10?ng/dl), and proven to have BPH histopathologically were included in the study. Age, tPSA, fPSA and the prostate volumes (PV) of the patients were noted.

Results: One thousand patients that fulfilled the inclusion criteria were included in the study. The PV of the patients were significantly correlated with age, tPSA and fPSA (p?r?=?0.307, p?r?=?0.382, p?r?=?0.296, respectively). On linear regression model, fPSA was found as a stronger predictive for PV (AUC?=?0.75, p?p?p?=?0.013).

Conclusions: Although tPSA is an important prognostic factor for predicting PV, the predictive value of fPSA is higher. PV can easily be predicted by using age, and serum tPSA and fPSA levels.     

Control‐based imputation for sensitivity analyses in informative censoring for recurrent event data

In clinical trials, missing data commonly arise through nonadherence to the randomized treatment or to study procedure. For trials in which recurrent event endpoints are of interests, conventional analyses using the proportional intensity model or the count model assume that the data are missing at random, which cannot be tested using the observed data alone. Thus, sensitivity analyses are recommended. We implement the control‐based multiple imputation as sensitivity analyses for the recurrent event data. We model the recurrent event using a piecewise exponential proportional intensity model with frailty and sample the parameters from the posterior distribution. We impute the number of events after dropped out and correct the variance estimation using a bootstrap procedure. We apply the method to an application of sitagliptin study.