One-sided Confidence Interval Research Articles

A randomised, prospective double-blind, propiverine-controlled trial of imidafenacin in patients with overactive bladder

To assess the efficacy and safety of imidafenacin compared with propiverine for treatment of overactive bladder (OAB) in Korean patients. Patients with OAB symptoms were randomised to double-blind treatment with 0.1 mg of imidafenacin twice daily (group A) or propiverine 20 mg once daily (group B) for 12-week regimen, and assessed for efficacy and safety. The primary efficacy outcome was per cent change of weekly urgency urinary incontinence (UUI) episodes at week 12. The secondary efficacy outcomes were changes in the micturitions per day, urine volume voided per micturition, urgency episodes per day, complete disappearance of incontinence episodes and severity of urgency from baseline to week 12. Quality of life and safety profiles were also compared. Of 162 patients randomised, 140 completed the study protocol. The per cent change of weekly UUI episodes at week 12 was -69.1% in group A and -70.4% in group B (both p < 0.0001). The lower limit of 95% one-sided confidence interval of the difference between the groups was above the non-inferiority margin (-19.42%). Other voiding parameters and quality of life significantly improved at week 12 in both the groups. The discontinuation rates caused by adverse events were low in both the groups. While dry mouth was the most common adverse event (group A: 28.4% vs. B: 30.4%, p = 0.783), the severity of dry mouth was significantly less in the group A than B (p = 0.042) There were no significant differences in other safety profiles. After the 12-week treatment of imidafenacin 0.1 mg twice daily, all OAB symptoms and quality of life improved. Imidafenacin was not inferior to propiverine for the reduction of UUI episodes, and was better tolerated than propiverine in the safety profile. Our results indicate that imidafenacin is a safe and effective drug in Korean patients with OAB.

Results of a phase II pilot study of moderate dose radiotherapy for inoperable desmoid-type fibromatosis—an EORTC STBSG and ROG study (EORTC 62991–22998)

BackgroundTo determine the activity of radiotherapy in patients with inoperable desmoid-type fibromatosis (DF) a multicenter prospective phase II trial was carried out. Materials and methodsPatients with inoperable progressive disease of primary, recurrent or incompletely resected lesions received a dose of 56 Gy in 28 fractions. Follow-up MRI studies were carried out every 3 months for 2 years and thereafter every 6 months. The primary end point was local control rate at 3 years, estimated by a nonparametric method for interval-censored survival data. Secondary end points were objective tumor response, acute and late toxic effect. ResultsForty-four patients (27 F/17 M) were enrolled from 2001 to 2008. Median age was 39.5 years. Main tumor sites included trunk 15 (34.1%) and extremities 27 (61.3%). Median follow-up was 4.8 years. The 3-year local control rate was 81.5% (90% one-sided confidence interval 74% to 100%). Best overall response during the first 3 years was complete response (CR) 6 (13.6%), partial response (PR) 16 (36.4%), stable disease 18 (40.9%), progressive disease 3 (6.8%) and nonassessable 1 (2.3%). Five patients developed new lesions. After 3 years, the response further improved in three patients: (CR 2, PR 1). Acute grade 3 side-effects were limited to skin, mucosal membranes and pain. Late toxic effect consisted of mild edema in 10 patients. ConclusionsModerate dose radiotherapy is an effective treatment of patients with DF. Response after radiation therapy is slow with continuing regression seen even after 3 years.

Measuring governance at health facility level: developing and validation of simple governance tool in Zambia

BackgroundGovernance has been cited as a key determinant of economic growth, social advancement and overall development. Achievement of millennium development goals is partly dependant on governance practices. In 2007, Health Systems 20/20 conducted an Internet-based survey on the practice of good governance. The survey posed a set of good practices related to health governance and asked respondents to indicate whether their experience confirmed or disconfirmed those practices. We applied the 17 governance statements in rural health facilities of Zambia. The aim was to establish whether the statements were reliable and valid for assessing governance practices at primary care level.MethodsBoth quantitative and qualitative methods were used. We first applied the governance statements developed by the health system 20/20 and then conducted focus group discussion and In-depth interviews to explore some elements of governance including accountability and community participation. The target respondents were the health facility management team and community members. The sample size include 42 health facilities. Data was analyzed using SPSS version 17 and Nvivo version 9.ResultsThe 95% one-sided confidence interval for Cronbach’s alpha was between 0.69 and 0.74 for the 16 items.The mean score for most of the items was above 3. Factor analysis yielded five principle components: Transparency, community participation, Intelligence & vision, Accountability and Regulation & oversight. Most of the items (6) clustered around the transparency latent factor. Chongwe district performed poorly in overall mean governance score and across the five domains of governance. The overall scores in Chongwe ranged between 51 and 94% with the mean of 80%. Kafue and Luangwa districts had similar overall mean governance scores (88%). Community participation was generally low. Generally, it was noted that community members lacked capacity to hold health workers accountable for drugs and medical supplies.ConclusionsThe study successfully validated and applied the new tool for evaluating health system governance at health facility level. The results have shown that it is feasible to measure governance practices at health facility level and that the adapted tool is fairly reliable with the 95% one-sided confidence interval for Cronbach’s alpha laying between 0.69 and 0.74 for the 16 items. Caution should be taken when interpreting overall scores as they tended to mask domain specific variations.

The impact of drug-related QT prolongation on FDA regulatory decisions

The impact of drug-related QT prolongation on FDA regulatory decisions

Multiple comparisons with a control for exponential location parameters under heteroscedasticity

In this paper, a new design-oriented two-stage two-sided simultaneous confidence intervals, for comparing several exponential populations with control population in terms of location parameters under heteroscedasticity, are proposed. If there is a prior information that the location parameter of k exponential populations are not less than the location parameter of control population, one-sided simultaneous confidence intervals provide more inferential sensitivity than two-sided simultaneous confidence intervals. But the two-sided simultaneous confidence intervals have advantages over the one-sided simultaneous confidence intervals as they provide both lower and upper bounds for the parameters of interest. The proposed design-oriented two-stage two-sided simultaneous confidence intervals provide the benefits of both the two-stage one-sided and two-sided simultaneous confidence intervals. When the additional sample at the second stage may not be available due to the experimental budget shortage or other factors in an experiment, one-stage two-sided confidence intervals are proposed, which combine the advantages of one-stage one-sided and two-sided simultaneous confidence intervals. The critical constants are obtained using the techniques given in Lam [9,10]. These critical constant are compared with the critical constants obtained by Bonferroni inequality techniques and found that critical constant obtained by Lam [9,10] are less conservative than critical constants computed from the Bonferroni inequality technique. Implementation of the proposed simultaneous confidence intervals is demonstrated by a numerical example.

The Efficacy and Safety of FSK0808, Filgrastim Biosimilar: A Multicenter, Non-randomized Study in Japanese Patients with Breast Cancer

FSK0808 is a filgrastim biosimilar. This study assessed the efficacy and safety of FSK0808 in patients with breast cancer. One hundred and four breast cancer patients undergoing chemotherapy were enrolled in the study. FSK0808 was used to treat the neutropenia experienced by the patients in the course of their chemotherapy. Efficacy was evaluated by the recovery of absolute neutrophil count following FSK0808 administration based on the duration of neutropenia in patients who received pre- or postoperative chemotherapy containing fluorouracil, epirubicin and cyclophosphamide. Adverse events were evaluated in accordance with the Common Terminology Criteria for Adverse Events version 3.0. The incidence of febrile neutropenia and generation of an anti-granulocyte colony-stimulating factor antibody were also evaluated. The average duration of neutropenia in Cycle 2 was 2.2 days with a standard deviation of 1.5 days. The upper limit of the 97.5% one-sided confidence interval was 2.5 days and was confirmed not to exceed 3.0 days, which was defined as the threshold value of absolute neutrophil count recovery. The incidence of febrile neutropenia across all treatment cycles was 34.6%. Observed adverse drug reactions with an incidence of > 5% were back pain (60.6%), bone pain (9.6%), alanine aminotransferase increase (8.7%), aspartate aminotransferase increase (5.8%) and arthralgia (5.8%). Production of the anti-granulocyte colony-stimulating factor antibody was not observed in any patient during the study. FSK0808 was safe and well tolerated in breast cancer patients undergoing chemotherapy and effectively stimulated neutrophil recovery.

One stage multiple comparisons with the average for exponential location parameters under heteroscedasticity

Two-stage multiple comparisons with the average for location parameters of two-parameter exponential distributions under heteroscedasticity are proposed by Wu and Wu [Wu, S.F., Wu, C.C., 2005. Two stage multiple comparisons with the average for exponential location parameters under heteroscedasticity. Journal of Statistical Planning and Inference 134, 392–408]. When the additional sample for the second stage may not be available, one-stage procedures including one-sided and two-sided confidence intervals are proposed in this paper. These intervals can be used to identify a subset which includes all no-worse-than-the-average treatments in an experimental design and to identify better-than-the-average, worse-than-the-average and not-much-different-from-the-average products in agriculture, the stock market, pharmaceutical industries. Tables of upper limits of critical values are obtained using the technique given in Lam [Lam, K., 1987. Subset selection of normal populations under heteroscedasticity. In: Proceedings of the Second International Advanced Seminar/Workshop on Inference Procedures Associated with Statistical Ranking and Selection. Sydney, Australia. August 1987. Lam, K., 1988. An improved two-stage selection procedure. Communications in Statistics—Simulation and Computation 17 (3), 995–1006]. An example of comparing four drugs in the treatment of leukemia is given to demonstrate the proposed procedures. The relationship between the one-stage and the two-stage procedures is also elaborated in this paper.

A Test of Homogeneity Against Umbrella Scale Alternative Based on Gini’s Mean Difference

In this paper, a test procedure based on Gini's mean difference for testing homogeneity of scale parameters against umbrella ordered alternative with at least one strict inequality, when peak of the umbrella is known, is proposed. The exact critical points through simulation are computed for the proposed test in case of standard exponential, standard logistic and standard uniform distributions; however the proposed test can be applied even in case of other distributions like Laplace, Pareto, Weibull, etc. Construction of Simultaneous one-sided confidence intervals (SOCIs) is proposed along with an illustration. Power of the proposed test is computed and some power comparisons are also made.

Oral flucloxacillin and phenoxymethylpenicillin versus flucloxacillin alone for the emergency department outpatient treatment of cellulitis: study protocol for a randomised controlled trial

BackgroundOral flucloxacillin, either alone or in combination with phenoxymethylpenicillin, is a commonly prescribed antibiotic for the treatment of cellulitis, particularly in Ireland and the United Kingdom. This study aims to establish the non-inferiority of oral monotherapy (flucloxacillin alone) to dual therapy (flucloxacillin and phenoxymethylpenicillin) for the outpatient treatment of cellulitis in adults.Methods/designThis study is a multicentre, randomised, double-blind, placebo-controlled trial of adults who present to the emergency department (ED) with cellulitis that is deemed treatable on an outpatient basis with oral antibiotics. After fulfilling specified inclusion and exclusion criteria, informed consent will be taken. Patients will be given a treatment pack containing 7 days of treatment with flucloxacillin 500 mg four times daily and placebo or flucloxacillin 500 mg four times daily and phenoxymethylpenicillin 500 mg four times daily. The primary outcome measure under study is the proportion of patients in each group in which there is greater than or equal to a 50% reduction in the area of diameter of infection from the area measured at enrolment at the end-of-treatment visit (7 to 10 days). Secondary endpoints include a health-related quality of life measurement as rated by the SF-36 score and the Extremity Soft Tissue Infection Score (not validated), compliance and adverse events. Patients will be followed up by telephone call at 3 days, end-of-treatment visit (EOT) at 7 to 10 days and test-of-cure (TOC) visit at 30 days. To achieve 90% power, a sample size of 172 patients per treatment arm is needed. This assumes a treatment success rate of 85% with oral flucloxacillin and phenoxymethylpenicillin, an equivalence threshold Δ = 12.5% and an α = 0.025. Non-inferiority will be assessed using a one-sided confidence interval on the difference of proportions between the two groups. Standard analysis including per-protocol and intention-to-treat will be performed.DiscussionThis trial aims to establish the non-inferiority of flucloxacillin monotherapy to dual therapy in the treatment of uncomplicated cellulitis among ED patients. In doing so, this trial will bridge a knowledge gap in this understudied and common condition and will be relevant to clinicians across several different disciplines.Trial registrationEudraCT Number 2008-006151-42

Choice of Baseline in Parallel Thorough QT Studies

Although the ICH E14 Guidance [1] was adopted in 2005 there is still much debate on the appropriate baseline needed to adjust for the variable nature of the QT interval and its heart rate corrected value (QTc) in the parallel Thorough QT (TQT) study. Numerous reports have documented the influence of circadian rhythm [2], food ingestion [3], sleep [4], and autonomic tone [5] on the QT/QTc interval. In the absence of baseline measurements, these factors can make it difficult to interpret QT effects of the study drug. In this issue of Drug Safety Dr. Zhang and colleagues [6] advocate the use of a time-matched baseline in parallel TQT studies to maximize the precision and accuracy of point estimates for QTc effects. Their recommendation is based on analyses of commonly used baseline correction methods using a large data set of parallel TQT studies submitted to the FDA for statistical review by sponsors. One assumption in the choice of baseline is that it should have no influence on the magnitude of QT effects (high accuracy) and it should minimize the uncertainty of the effects (high precision). Both accuracy and precision are important for the interpretation of a TQT trial. As a result, there is much focus on the implications of using different baseline adjustments in the parallel TQT study and understanding the role each baseline definition can have in determining the outcome of the trial as either positive or negative. The distinction between a positive or negative TQT study is based on the upper bound of the 95 % one-sided confidence interval (CI) for the largest time-matched mean difference in QTc between drug and placebo (baseline adjusted). If these (double-delta) DDQTc values exclude 10 ms at all study times the result is a negative TQT study [1]. Intensive ECG monitoring is then hardly ever required in subsequent trials (phase II/III). Conversely, drugs for which an effect exceeding 10 ms cannot be excluded almost always require additional QT interval monitoring in target patients including dose-concentration effects, outlier analysis, changes in mean QTc values, analysis of QTc in subgroups of interest, and identification of individuals who develop a markedly prolonged QTc. The need for additional ECG monitoring in late phase trials can add millions to drug development and may ultimately pose financial limitations on the number of medications which can be developed [7]. The importance of correctly adjusting for baseline in the TQT study is therefore evident.

New method to determine the decision threshold for low-level radioactivity measurements

We discuss a new method to determine the value of a decision threshold that can be used to decide whether to choose a one-sided confidence interval or a two-sided confidence interval. The method is based on the Feldman–Cousins unified approach providing a unique confidence region for estimated parameters. We apply this method to a net count rate measurand in low-level radioactivity measurements which is physically restricted to nonnegative values. We tabulate the values of the decision threshold and detection limit of the measurand for some typical coverage probabilities. The decision threshold in this method does indeed enable a decision on whether or not the physical effect quantified by the measurand is present.

Velaglucerase alfa enzyme replacement therapy compared with imiglucerase in patients with Gaucher disease

Enzyme replacement therapy for Gaucher disease (GD) has been available since 1991. This study compared the efficacy and safety of velaglucerase alfa with imiglucerase, the previous standard of care. A 9-month, global, randomized, double-blind, non-inferiority study compared velaglucerase alfa with imiglucerase (60 U/kg every other week) in treatment-naïve patients aged 3-73 years with anemia and either thrombocytopenia or organomegaly. The primary endpoint was the difference between groups in mean change from baseline to 9 months in hemoglobin concentration. 35 patients were randomized: 34 received study drug (intent-to-treat: 17 per arm), 20 were splenectomized. Baseline characteristics were similar in the two groups. The per-protocol population included 15 patients per arm. The mean treatment difference for hemoglobin concentration from baseline to 9 months (velaglucerase alfa minus imiglucerase) was 0.14 and 0.16 g/dL in the intent-to-treat and per-protocol populations, respectively. The lower bound of the 97.5% one-sided confidence interval in both populations lay within the pre-defined non-inferiority margin of -1.0 g/dL, confirming that velaglucerase alfa is non-inferior to imiglucerase. There were no statistically significant differences in the secondary endpoints. Most adverse events were mild to moderate. No patient receiving velaglucerase alfa developed antibodies to either drug, whereas four patients (23.5%) receiving imiglucerase developed IgG antibodies to imiglucerase, which were cross-reactive with velaglucerase alfa in one patient. This study demonstrates the efficacy and safety of velaglucerase alfa compared with imiglucerase in adult and pediatric patients with GD clinically characterized as Type 1. Differences in immunogenicity were also observed.

Random Weighting Estimation of Poisson Distributions

This paper presents a new random weighting method for estimation of Poisson distributions. A theory is established for random weighting estimation of the population parameters of two Poisson distributions with partially missing data. The strong convergence of the random weighting estimation is rigorously proved under the condition of . The random weighting estimation of one-sided confidence intervals in Poisson distribution is also constructed, and its coverage probability is rigorously proved by using the Edgeworth expansion under certain conditions. DOI: http://dx.doi.org/10.11591/telkomnika.v11i1.1891

Sample size calculations for pilot randomized trials: a confidence interval approach

ObjectivesTo describe a method using confidence intervals (CIs) to estimate the sample size for a pilot randomized trial. Study DesignUsing one-sided CIs and the estimated effect size that would be sought in a large trial, we calculated the sample size needed for pilot trials. ResultsUsing an 80% one-sided CI, we estimated that a pilot trial should have at least 9% of the sample size of the main planned trial. ConclusionUsing the estimated effect size difference for the main trial and using a one-sided CI, this allows us to calculate a sample size for a pilot trial, which will make its results more useful than at present.

Effects of bazedoxifene/conjugated estrogens on endometrial safety and bone in postmenopausal women

ABSTRACTObjectives Bazedoxifene/conjugated estrogens (BZA/CE) has demonstrated efficacy in improving vasomotor and vulvar/vaginal atrophy symptoms in postmenopausal women. This study evaluated the endometrial safety of BZA/CE and effects on bone mineral density (BMD) compared with CE/medroxyprogesterone acetate (MPA) and placebo.Methods The Selective estrogens, Menopause, And Response to Therapy (SMART)-4 trial was a 1-year, multicenter, double-blind, randomized, placebo– and active-controlled, phase-3 study in non-hysterectomized, postmenopausal women (n = 1061; aged 40 −< 65 years). Subjects received BZA 20 mg/CE 0.45 or 0.625 mg, CE 0.45 mg/MPA 1.5 mg, or placebo daily. Primary endpoints were the incidence of endometrial hyperplasia and the change in lumbar spine BMD at 1 year. Secondary endpoints included the change in total hip BMD and rates of amenorrhea and breast pain.Results At 1 year, no cases of endometrial hyperplasia were identified in the BZA 20-mg/CE 0.45-mg group, while three cases (1.1%) were confirmed for the BZA 20-mg/CE 0.625-mg group (95% one-sided confidence interval upper limit < 4%). Both BZA/CE doses significantly increased lumbar spine and total hip BMD versus placebo (p ≤ 0.001) and showed low incidences of bleeding and breast tenderness, similar to placebo and significantly lower than for CE 0.45 mg/MPA 1.5 mg (p < 0.05). BZA/CE treatment was generally safe and well tolerated.Conclusions BZA 20 mg/CE 0.45 and 0.625 mg significantly improved BMD while maintaining endometrial safety and showed a favorable safety/tolerability profile over 1 year. BZA/CE may be a promising therapy for treating menopausal symptoms and preventing osteoporosis in non-hysterectomized, postmenopausal women.

Constrained Parameters in Applications: Review of Issues and Approaches

This review article provides an introduction to statistical issues that arise when some statistical model parameters are constrained. This often happens in applications, in particular in testing for variance components (e.g., genomics) and construction of one-sided confidence intervals (e.g., environmental risk analysis). Heuristic explanations are provided, and a number of general and recent statistical results that appeared in statistical literature are summarized for use in applications. Simulation results are shown for illustration of consequences of ignoring parameters on the boundary. Special attention is paid to likelihood ratio tests, but other approaches to confidence interval construction, such as Wald, bootstrap, and Bayesian are also briefly discussed. This paper presents examples from the risk assessment field and genomics, but all conclusions apply to whenever one-sided testing is conducted. Recommendations are provided for dealing with parameters on the boundary for a range of situations.

A Thorough QT Study of Teduglutide in Healthy Subjects.

Teduglutide, an analog of the endogenous hormone glucagon-like peptide-2, is currently being developed for the treatment of short bowel syndrome. This study investigated the potential effects of teduglutide on cardiac conduction and repolarization. Seventy-two healthy volunteers underwent 4 treatment periods in randomized order with a single subcutaneous injection of placebo, 5 and 20 mg teduglutide, and a single oral 400-mg dose of moxifloxacin. The primary variable to investigate the effect on cardiac repolarization was the difference between QTcF after administration versus predose. The observed upper bounds of the 95% one-sided confidence intervals were 3.0 ms (5 mg) and 4.5 ms (20 mg). Arithmetic mean curves of QTcF intervals over time of both doses of teduglutide and of placebo were almost superposable. Assay sensitivity for the positive control moxifloxacin was shown. In conclusion, teduglutide at intended therapeutic and supratherapeutic doses had no effect on cardiac repolarization. No safety concerns were identified. Treatment with teduglutide was well tolerated.

Nonparametric confidence intervals for the integral of a function of an unknown density

Given a random sample of size n from an unknown distribution function F on ℝ with finite derivatives and density f, we wish to estimate for a smooth function L. Examples are∈t f 2, the differential entropy and the Kullback–Leibler distance. We estimate f using a kernel estimate [fcirc] based on a kernel of order p, say. We show that {[fcirc](x i ), i=1, …, s} satisfies the Cornish–Fisher assumption with respect to m=nh. It follows that the corresponding estimate θˆ has a bias of magnitude O(h q +m −1), where p≤q≤2p depends on L. We show that the variance of θˆ has magnitude O(n −1) for a suitable bandwidth. For the regular case, we give one-sided and two-sided confidence intervals for θ with errors of magnitude O(M −1/2) and O(M −1), where M=nh 2. We present simulation studies to show the practical values of the results.

Simultaneous testing for the successive differences of exponential location parameters under heteroscedasticity

In this paper, the design-oriented two-stage and data-analysis one-stage multiple comparison procedures for successive comparisons of exponential location parameters under heteroscedasticity are proposed. One-sided and two-sided simultaneous confidence intervals are also given. We also extend these simultaneous confidence intervals for successive differences to a larger class of contrasts of the location parameters. Upper limits of critical values are obtained using the recent techniques given in Lam [Lam, K., 1987. Subset selection of normal populations under heteroscedasticity. In: Proceedings of the Second International Advanced Seminar/Workshop on Inference Procedures Associated with Statistical Ranking and Selection, Sydney, Australia; Lam, K., 1988. An improved two-stage selection procedure. Communications in Statistics Simulation and Computation. 17 (3), 995–1006]. These approximate critical values are shown to have better results than the approximate critical values using the Bonferroni inequality developed in this paper. Finally, the application of the proposed procedures is illustrated with an example.

CNS irradiation in pediatric acute myleoid leukemia: Equal results by 12 or 18 Gy in studies AML‐BFM98 and 2004

The impact of preventive central nervous system irradiation (CNS-RT) in childhood acute myeloid leukemia (AML) is still discussed. As results of study AML-BFM87 revealed an increased risk for relapse when CNS-RT was not performed, studies AML-BFM98 and -2004 randomized CNS-RT of 18 or 12 Gy in order to evaluate the efficacy of the lower dose and to reduce late effects. To achieve a power of 80% for non-inferiority (range 11%) 240 patients per group were required. Out of 722 eligible patients, 486 patients <18 years were randomized to receive 12 Gy (n = 249) or 18 Gy (n = 237). Since this was a non-inferiority study, the analysis was performed for patients as treated (12 Gy: n = 252 and 18 Gy: n = 219). Five-year survival, event-free survival and cumulative incidence of relapse were similar in patients who received 12 or 18 Gy, respectively (82 ± 3% vs. 79 ± 3%, 68 ± 3% vs. 63 ± 3%, and 30 ± 3% vs. 34 ± 3%). The lower limit of the one-sided confidence interval for the -5% difference in 5-years pEFS was 2%. There were six relapses with CNS involvement (one in the 12 Gy, and five in the 18 Gy group). Results demonstrate no disadvantage for patients irradiated with a reduced CNS dose of 12 Gy.