Changing cluster composition in cluster randomised controlled trials: design and analysis considerations.

BACKGROUND: There are many methodological challenges in the conduct and analysis of cluster randomised controlled trials, but one that has received little attention is that of post-randomisation changes to cluster composition. To illustrate this, we focus on the issue of cluster merging, considering the impact on the design, analysis and interpretation of trial outcomes. METHODS: We explored the effects of merging clusters on study power using standard methods of power calculation. We assessed the potential impacts on study findings of both homogeneous cluster merges (involving clusters randomised to the same arm of a trial) and heterogeneous merges (involving clusters randomised to different arms of a trial) by simulation. To determine the impact on bias and precision of treatment effect estimates, we applied standard methods of analysis to different populations under analysis. RESULTS: Cluster merging produced a systematic reduction in study power. This effect depended on the number of merges and was most pronounced when variability in cluster size was at its greatest. Simulations demonstrate that the impact on analysis was minimal when cluster merges were homogeneous, with impact on study power being balanced by a change in observed intracluster correlation coefficient (ICC). We found a decrease in study power when cluster merges were heterogeneous, and the estimate of treatment effect was attenuated. CONCLUSIONS: Examples of cluster merges found in previously published reports of cluster randomised trials were typically homogeneous rather than heterogeneous. Simulations demonstrated that trial findings in such cases would be unbiased. However, simulations also showed that any heterogeneous cluster merges would introduce bias that would be hard to quantify, as well as having negative impacts on the precision of estimates obtained. Further methodological development is warranted to better determine how to analyse such trials appropriately. Interim recommendations include avoidance of cluster merges where possible, discontinuation of clusters following heterogeneous merges, allowance for potential loss of clusters and additional variability in cluster size in the original sample size calculation, and use of appropriate ICC estimates that reflect cluster size.

Role of whole-body staging computed tomographic scans for detecting distant metastases in patients with bladder cancer.

OBJECTIVE: This study aimed to establish the incidence of distant metastases on whole-body computed tomographic (CT) scans in patients with newly diagnosed bladder cancer and to determine whether there is a significant difference in the incidence of metastases in patients with superficial and muscle invasive cancers. MATERIALS AND METHODS: A total of 201 patients who had a proven histological diagnosis of transitional cell carcinoma of the bladder and a whole-body staging CT scan at diagnosis were identified from our MDT database during a 36-month period. Imaging was retrospectively reviewed with view to recording site, if any, of distant metastases. RESULTS: Of 201 patients, 11 (5.5%) were found to have distant metastases on CT. In univariable models, staging was not associated with either age (odds ratio, 0.98; 95% confidence interval, 0.92-1.04; P = 0.4) or sex (Fisher exact test, P = 0.07). Mean (SD) age was 74.1 (10.5) years. There was a significant association between staging and metastasis (odds ratio, 19.9; 95% confidence interval, 3.2-infinity; P = 0.0003). Of the patients, 7% of males had metastases versus 0% of the females. CONCLUSIONS: Staging CT scans for assessment of distant metastatic disease in patients with newly diagnosed bladder cancer can be restricted to patients with muscle invasive disease.

Screening for osteoporosis in anorexia nervosa: prevalence and predictors of reduced bone mineral density.

OBJECTIVE: Decreased bone mineral density (BMD) in anorexia nervosa (AN) can be detected easily by dual-energy X-ray absorptiometry (DXA). This study was designed to assess the prevalence of osteoporosis and osteopenia in AN, identify predictors, and determine the diagnostic yield of screening with DXA. METHOD: DXA was used to screen 59 unselected adult patients with a history of AN. RESULTS: Osteoporosis was identified in 18 patients (31%) and osteopenia in 30 (51%). The spine had a lower mean T-score than either the hip or femur. BMI significantly predicted T-score (p = 0.0006) and the odds of having osteoporosis (p = 0.0188). There was a significant association between use of oestrogens and the presence of osteoporosis or osteopenia (p = 0.0491). There was no significant association between duration of AN and T-score. A duration of AN of less than 1 year was found in 12% of those with osteoporosis. CONCLUSION: BMI is a strong predictor of BMD in AN. DXA is an effective screening tool and should probably be offered routinely.

Rectal cancer: incidence of pulmonary metastases on thoracic CT and correlation with T staging.

PURPOSE: The aim of the study was to evaluate the incidence of pulmonary metastases detected on thoracic computed tomography in patients with rectal cancer and assess the association between the incidence of pulmonary metastases and the stage of the rectal tumor. MATERIALS AND METHODS: Fifty-six consecutive patients who were diagnosed with rectal cancer over a 22-month period were included in the study. These patients had local tumor staging with a pelvic magnetic resonance imaging and staging computed tomographic scan of the chest and upper abdomen immediately after the magnetic resonance imaging. Two radiologists retrospectively reviewed all the thoracic imaging performed on these patients for the presence of metastases. The presence of a parenchymal lung nodule (greater than or equal to 1 cm if single and 0.5 cm if multiple) with a soft tissue component without calcification on lung and mediastinal window settings was considered positive for the presence of metastasis. All other patients were considered as not having any lung metastases. RESULTS: Of the 56 patients, 10 (17.9%) had evidence of pulmonary metastases on computed tomography. Of the 56 patients, there were 3 patients with stage T1, 24 with T2, 26 with T3, and 3 with stage T4 tumors. Of these 10 patients, 1 had a stage T2 tumor, 7 had T3, and 2 had stage T4 tumors. Statistical analysis using exact logistic regression showed the odds of getting lung metastases is an increasing function of tumor grade. CONCLUSIONS: There is a high incidence of lung metastases in patients with rectal cancer, and thoracic computed tomographic scanning should be performed as part of a staging protocol in all patients before any form of treatment is planned. There is a higher incidence of lung metastases with higher T stage.