Prevention of disease complications through diagnostic models: how to tackle the problem of missing data?

Diagnostic models are frequently used to assess the role of risk factors on disease complications, and therefore to avoid them. Missing data is an issue that challenges the model making. The aim of this study was to develop a diagnostic model to predict death in HIV/ AIDS patients when missing data exist. HIV patients [n=1460] referred to Voluntary Consoling and Testing Center [VCT] of Shiraz southern Iran during 2004-2009 were recruited. Univariate association between variables and death was assessed. Only variables which had univariate P< 0.25 were selected to be offered to the Multifactorial models. First, patients with missing data on candidate variables were deleted [C-C model]. Then, applying Multivariable Imputation via Chained Equations [MICE], missing data were imputed. Logistic regression was fitted to C-C and imputed data sets [MICE model]. Models were compared in terms of number of variables retained in the final model, width of confidence intervals, and discrimination ability. About 22% of data were lost in C-C model. Number of variables retained in the C-C and MICE models was 2 and 6 respectively. Confidence Intervals [C.I.] corresponding to C-C model was wider than that of MICE. The MICE model showed greater discrimination ability than C-C model [70% versus 64%]. The C-C analysis resulted to loss of power and wide CI's. Once missing data were imputed, more variables reached significance level and C.I.'s were narrower. Therefore, we do recommend the application of the imputation method for handling missing data

Low incidence of colorectal cancer in Kerman province, Iran

In Iran colorectal cancer [CRC] incidence varies among various places. But we do not have any validate data that deeply explored the difference of CRC in Kerman province by national average. The aim of this study was to evaluate the incidence of colon cancer within a 12-year period in Kerman province and to find whether the incidence of CRC in Kerman compared with the total incidence found in Iran. Data on colorectal cancer was collected from all histopathology departments around the Kerman Province during 1991 - 2002 retrospectively. The crude and age-standardized incidence rates per 1 million populations were calculated based on the 1996 census data and the population growth rate. During this study total number of 551 new cases of colorectal cancer in Kerman province had been diagnosed. Age Standardized Rate of colon cancer in males and females were 50 [95% CI: 44-56] and 53 [95% CI: 46-59] cases per 1,000,000 population per year, respectively. The risk ratio in females relative to males was not significant in any type of colon cancer. We did not find any difference for each year during the entire study period. The ASR of colorectal cancer in Kerman province was quite lower than the average rate in the whole country. This study showed that, the risk of acquiring colon cancer was constant. We concluded that the risk of colorectal cancer in Kerman province was much lower than the entire country

[How to estimate the sample size in special conditions? [part two]]

In the previous paper, the basic concepts of sample size calculation were presented. This paper explores main post-calculation adjustments of the sample size calculation in special circumstances such as multiple group comparisons, unbalanced studies [with unequal number of subjects in different groups]; sample size correction for missing data, and adjustment for finite population size. In addition, the concept of design effect in multi-stage sampling and its impact on the sample size are presented. We then focused on the sample size estimation when we have to use non-parametric statistical tests for data analysis. The concept of power-efficacy of parametric versus non-parametric methods and its use in the correction of sample size has been explained

Therapeutic effect of bone marrow stromal cells and coenzyme Q10 on rat models of parkinson's disease

The assessment of the ability of combined treatment of bone marrow stromal cells graft [BMSCs] and oral administration of Coenzyme [CoQIO] in rat model of Parkinson disease as a good substitute for common current Parkinson treatments, and the comparison of this combined treatment method with alone application of these treatments. In this experimental study of male Wistar rats were used. They were divided into six groups: control, sham, lesion, treatment groups with oral administration of CoQIO, treatment with graft BMSC and combined treatment with graft BMSC and oral administration of CoQIO. Oral administration of CoQIO with 200 mg/kg/daily dose started a week before the model creation procedure and continued throughout the whole treatment period. The laboratory model of Parkinson disease in rats was performed by injecting 2.5 microlitre saline solution 0.9% containing 8 micrograms 6-hydroxy dopamine [6-OHDA] and 0.2% ascorbic acid in substantia nigra pars compacta. Also in sham group the same volume solution saline-ascorbic was injected. BMS Cells were labeled by 5-Bromo-2'-deoxyuridine [Brdu] before transplantation. Behavioral assessment before creating the model, two weeks after creating the model and eight weeks after cell transplantation was performed. At the end of second month of treatment, Immunohistochemistry and histology Studies were performed. Behavioral assessment of two groups of alone treatments indicated the equal recovery in comparison with lesion group [p<0.01] while combined treatment of BMSC and Co Q10 showed a considerable recovery compared with lesion group [p<0.001]. In addition according to histological studies, no sign of gliosis and graft rejection was seen. Immunohistochemistry studies of Brdu indicate that the cells are alive after two month of application in host tissue. Cell count assessment showed that the number of neural cells in combined treatment of BMSC and Co Q10 was significant difference with others experimental groups [p<0.001]. The combined use of two neuroprotective treatment and replacement therapy can have a more effective role in the treatment of Parkinson's disease in comparison of alone treatment protocols

Early outcome of concurrent mitral valve replacement and coronary artery bypass grafting

Concomitant coronary artery bypass surgery [CABG] in patients undergoing mitral valve replacement [MVR] has been shown to be an important risk factor for hospital mortality. We evaluated preoperative characteristics, postoperative complications, in-hospital mortality rate, and length of stay in hospital for patients undergoing concurrent CABG with MVR. Preoperative and postoperative clinical data from 175 patients undergoing concurrent CABG with MVR operation at Tehran Heart Center from 2002 through 2006 were collected and entered into a database. Information was obtained by clinical and case note review as well as detailed questionnaires to physicians and patients. Mean age of patients was 57.95 +/- 10.54 years and 51.4% were male. Mean New York Heart Association [NYHA] score was 2.46 +/- 0.84. Among studied patients, 18.3% and 2.9% underwent aortic and tricuspid valve replacement, respectively. In-hospital mortality was 6.9% and 96.0% of patients were hospitalized >/= 14 days. History of congestive heart failure [P = 0.027] and postoperative brain stroke [P = 0.004] were independent predictors for in-hospital mortality. Exact considering of congestive heart failure and postoperative brain stroke related to in-hospital mortality in concurrent CABG with MVR operation are necessary