Use of appropriate statistical tools in biomedical research: Current trend & status

Background & objectives: Due to lack of appropriate statistical knowledge, published research articles contain various errors related to the design, analysis and interpretation of results in the area of biomedical research. If research contains statistical error, however, costly, it may be of no use and the purpose of the investigation gets defeated. Many biomedical research articles published in different peer reviewed journals may retain several statistical errors and flaws in them. This study aimed to examine the trend and status of application of statistics in biomedical research articles. Study design, sample size estimation and statistical measures are crucial components of a study. These points were evaluated in published original research articles to understand the use or misuse of statistical tools. Methods: Three hundred original research articles from the latest issues of selected 37 journals were reviewed. These journals were from the five internationally recognized publication groups (CLINICAL KEY, BMJ Group, WILEY, CAMBRIDGE and OXFORD) accessible through the online library of SGPGI, Lucknow, India. Results: Among articles assessed under present investigation, 85.3 per cent (n=256) were observational, and 14.7 per cent (n=44) were interventional studies. In 93 per cent (n=279) of research articles, sample size estimation was not reproducible. The simple random sampling was encountered rarely in biomedical studies even though none of the articles was adjusted by design effect and, only five articles had used randomized test. The testing of assumption of normality was mentioned in only four studies before applying parametric tests. Interpretation & conclusions: In order to present biomedical research results with reliable and precise estimates based on data, the role of engaging statistical experts need to be appreciated. Journals must have standard rules for reporting study design, sample size and data analysis tools. Careful attention is needed while applying any statistical procedure as, it will not only help readers to trust in the published articles, but also rely on the inferences the published articles draw.

Designing, molecular docking, and dynamics simulations studies of 1,2,3-triazole clamped Uracil-Coumarin hybrids against EGFR tyrosine kinase

Since the last decade, hybrid drug strategies have attracted many researchers for their improved anti-cancer potential incomparison to single drug components. Complying to this approach, 28 novel Uracil–Coumarin hybrids with differentsized linkers (2–5 carbon atoms) and substituents were designed to occupy the active site of protein epidermal growthfactor receptor (EGFR) tyrosine kinase (Protein Data Bank ID: 1M17). Molecular docking studies were performedfor all ligands (A1-D7) to identify the potential candidate using Schrödinger software. The relative binding affinity ofhybrids toward EGFR was compared with standard Erlotinib on the basis of gScore and Emodel score. Positively, allthe hybrids docked inside the cavity and showed significant interactions, compounds A6, A2, and A7 with short-chainlinker (two carbon atoms) and halogen substituents were found to have more interactions and better docking score thanstandard Erlotinib. The visualization results depicted that compound A6 showed the highest affinity and formed thebest binding pose to the target EGFR with gScore = −8.891 kcal/mol and Emodel score = −100.744 in comparison tostandard Erlotinib (gScore of −8.538 kcal/mol and Emodel score = −80.588). Moreover, a molecular dynamics studyalso reveals that ligand A6 forms a stable complex with root mean square deviation (RMSD) of 0.3 nm and the plateauphase started just after 10 ns (time). Hence, the present research provides computational insights of Uracil–Coumarinhybrids as potential ligands against EGFR tyrosine kinase and in future in vitro investigations of these hybrids mayprove their therapeutic potential against cancer.

Application of student's t-test, analysis of variance, and covariance

Student's t test (t test), analysis of variance (ANOVA), and analysis of covariance (ANCOVA) are statistical methods used in the testing of hypothesis for comparison of means between the groups. The Student's t test is used to compare the means between two groups, whereas ANOVA is used to compare the means among three or more groups. In ANOVA, first gets a common P value. A significant P value of the ANOVA test indicates for at least one pair, between which the mean difference was statistically significant. To identify that significant pair(s), we use multiple comparisons. In ANOVA, when using one categorical independent variable, it is called one-way ANOVA, whereas for two categorical independent variables, it is called two-way ANOVA. When using at least one covariate to adjust with dependent variable, ANOVA becomes ANCOVA. When the size of the sample is small, mean is very much affected by the outliers, so it is necessary to keep sufficient sample size while using these methods.

Selection of appropriate statistical methods for data analysis

In biostatistics, for each of the specific situation, statistical methods are available for analysis and interpretation of the data. To select the appropriate statistical method, one need to know the assumption and conditions of the statistical methods, so that proper statistical method can be selected for data analysis. Two main statistical methods are used in data analysis: descriptive statistics, which summarizes data using indexes such as mean and median and another is inferential statistics, which draw conclusions from data using statistical tests such as student's t-test. Selection of appropriate statistical method depends on the following three things: Aim and objective of the study, Type and distribution of the data used, and Nature of the observations (paired/unpaired). All type of statistical methods that are used to compare the means are called parametric while statistical methods used to compare other than means (ex-median/mean ranks/proportions) are called nonparametric methods. In the present article, we have discussed the parametric and non-parametric methods, their assumptions, and how to select appropriate statistical methods for analysis and interpretation of the biomedical data.

Descriptive statistics and normality tests for statistical data

Descriptive statistics are an important part of biomedical research which is used to describe the basic features of the data in the study. They provide simple summaries about the sample and the measures. Measures of the central tendency and dispersion are used to describe the quantitative data. For the continuous data, test of the normality is an important step for deciding the measures of central tendency and statistical methods for data analysis. When our data follow normal distribution, parametric tests otherwise nonparametric methods are used to compare the groups. There are different methods used to test the normality of data, including numerical and visual methods, and each method has its own advantages and disadvantages. In the present study, we have discussed the summary measures and methods used to test the normality of the data.

Scales of measurement and presentation of statistical data

Measurement scale is an important part of data collection, analysis, and presentation. In the data collection and data analysis, statistical tools differ from one data type to another. There are four types of variables, namely nominal, ordinal, discrete, and continuous, and their nature and application are different. Graphs are a common method to visually present and illustrate relationships in the data. There are several statistical diagrams available to present data sets. However, their use depends on our objectives and data types. We should use the appropriate diagram for the data set, which is very useful for easily and quickly communicating summaries and findings to the audience. In the present study, statistical data type and its presentation, which are used in the field of biomedical research, have been discussed.

Validation of WHO QOL-BREF instrument in indian adolescents.

Objective. To test the psychometric properties of World Health Organization Quality of Life (WHO QOL-BREF) instrument in Indian adolescents. Methods. Of 1900 schools in Lucknow city, 20 schools were invited to participate. To make WHO QOL-BREF instrument culturally appropriate for Indian adolescents, a minor modification was done by replacing one item in Social domain “Are you satisfied with your sex life?” with “Are you satisfied with the respect you receive from others?”. The revised WHO QOL-BREF was administered to subjects in school after obtaining written parental consent. Results. From August 2007 – January 2008, 525 adolescents were recruited (mean age 14.04±2.09 yr; 52.38 % males). Adolescents reported highest HRQoL in social relations and lowest in environment domain. The instrument showed good internal consistency (Cronbach’s a=0.87; p-value<0.01) as well as good content, construct and predictive validity (pvalues< 0.05). Psychological domain had best predictive validity, whereas, social relations domain had best content validity. Conclusion. The study provides evidence that revised WHO QOL-BREF is a reliable and valid instrument and can be used in Indian adolescents.

Steroid resistant nephrotic syndrome: role of histopathology.

This study was conducted to (1) see the histopathological distribution of different subtypes in steroid resistant nephrotic syndrome (SRNS) and (2) compare the clinical, biochemical parameters and outcome between Minimal Change Disease (MCD) with non-MCD subtypes in response to immunosuppressive therapy. A retrospective analysis was done of data on all biopsy proven children with idiopathic SRNS (no response to 4 weeks of standard prednisone therapy (60 mg/m(2)/day)) referred to our institute over last 12 years. They were treated with one of the following medications: oral or intravenous cyclophosphamide, cyclosporine or combination of dexamethasone and azathioprine. A comparison was done of the demographic clinical and biochemical features different histopathologies. We studied 136 children with SRNS (100 M, 36 F). They accounted for 15.1%(136/900) of all children with idiopathic nephrotic syndrome. Focal segmental glomerulosclerosis (FSGS) was the commonest 80/136 (59%), followed by MCD (17.6%). Children with non-MCD had a significantly greater prevalence of microhematuria as compared to MCD. The other baseline clinical and biochemical features including the glomerular filtration rate (GFR) were similar. After a mean follow up of 46 (8-148) months, a significantly greater children with non-MCD 65/112) continued to be proteinuric as compared to the MCD (3/24) (p=0.0001). FSGS was the commonest cause of SRNS in our patient population. Children with SRNS secondary to MCD are more likely to achieve remission as compared to non-MCD subtypes and have a better long-term prognosis. Hence kidney biopsy is of significant prognostic value in SRNS.

Premature coronary artery disease in North India: an angiography study of 1971 patients.

BACKGROUND: South Asians, specially Indians, show increased risk for atherosclerosis and have the highest mortality rates due to coronary artery disease amongst all ethnic groups studied so far. We aimed to find out the differences in clinical-biochemical and angiographic profile of young patients versus older patients with angiographically proven atherosclerotic coronary artery disease. METHODS AND RESULTS: Group I (n=828) consisted of patients with age above 55 years (mean age: 63.15 +/- 5.76 years), group II (n=924, mean age: 49.13 +/- 4.25 years) consisted of patients between age 41-55 years and group III (n=219) consisted of patients with age < or = 40 years (mean age: 37.37 +/- 2.92 years). Among the conventional risk factors, smoking was significantly more frequent in group III, while diabetes mellitus and systemic hypertension were more prevalent in groups II and I. Q wave myocardial infarction was more frequently present in groups II and III. Only about one-third of the entire patient population in the myocardial infarction group received thrombolytic therapy. Total cholesterol, low-density lipoprotein cholesterol, and triglyceride levels were significantly higher in younger patients (groups II and III), while high-density lipoprotein cholesterol was significantly low in whole cohort but more so in older patients. Single vessel involvement was more common in group III, while multi-vessel involvement, diffuse disease and fluoroscopic calcium were more common in groups I and II. CONCLUSIONS: Significant differences were observed in the clinical, biochemical and angiographic profile of young patients with coronary artery disease as compared to elderly patients. The younger cohort had more atherogenic lipid profile, higher prevalence of smoking and more frequent single vessel disease. We observed that total cholesterol/high-density lipoprotein cholesterol ratio was a better predictor of coronary artery disease as compared to individual lipid levels.

Association of common carotid intima-media thickness and lipoprotein(a) with coronary artery disease.

BACKGROUND: Carotid artery intimal medial thickness is a simple, non-invasive and reproducible clinical tool to evaluate atherosclerosis and predict coronary artery disease. Lipoprotein(a) levels are related to both atherogenesis and thrombogenesis and may be a key link between lipid and coronary artery disease. This study evaluated the association of carotid intimal medial thickness and lipoprotein(a) with coronary artery disease. METHODS AND RESULTS: We studied 185 randomly selected patients hospitalized for coronary angiogram in our institute. There were 110 angiographically proven patients of coronary artery disease with mean age of 55.8 +/- 9 years (range 34-72 years) and 75 subjects with normal coronary artery anatomy with mean age of 54.8 +/- 8 years (range 34-68 years). The mean carotid intimal medial thickness of subjects with coronary artery disease was significantly higher than in subjects without coronary artery disease (0.84 +/- 0.16 mm v. 0.65 +/- 0.15 mm, p<0.001). The mean carotid intimal medial thicknesses in patients with triple vessel, double vessel and single vessel disease were 0.96 +/- 0.12 mm, 0.84 +/- 0.11 mm and 0.78 +/- 0.13 mm, respectively (p=0.05). The mean lipoprotein(a) of subjects with coronary artery disease was significantly higher than in subjects without coronary artery disease (35.9 +/- 22.3 mg/dl v. 19.1 +/- 21.2 mg/dl, p<0.001). Mean lipoprotein(a) levels in subjects with carotid intimal medial thickness <0.80 was 26.4 +/- 24.2 mg/dl and in subjects with carotid intimal medial thickness > or = 0.80 was 32.1 +/- 22.1 mg/dl (p=0.05). CONCLUSIONS: There is a strong correlation between carotid and coronary atherosclerosis and carotid intimal medial thickness is a good predictor of presence and extent of coronary artery disease. Lipoprotein(a) level is a powerful independent risk factor for atherosclerosis. Carotid intimal medial thickness and lipoprotein(a) in conjoint can predict coronary artery disease reliably.

A tertiary care hospital-based study of conventional risk factors including lipid profile in proven coronary artery disease.

BACKGROUND: The prevalence and mortality rates of coronary artery disease have been known to be higher in the Indian than the Western population. Most data on lipid levels in Indians have been obtained from studies on migrant Asian Indians. There are insufficient data on lipid profile and other conventional risk factors in Indian patients living within India. METHODS AND RESULTS: The study included 2656 consecutive patients who underwent coronary angiography between March 1998 and February 2002. Of these, 2399 subjects had angiographically proven coronary artery disease (group 1) while 257 had normal coronary arteries (group 2). Lipid values were measured in the fasting state on the morning the coronary angiography was done. Patients receiving lipid-lowering agents, those having renal, hepatic or thyroid disorders, patients presenting within 8 weeks after acute myocardial infarction, and patients who were taking noncardiac drugs that affect the lipid profile were excluded from the study. Other conventional risk factors were also recorded. In subjects with coronary artery disease and normal coronary arteries, the levels of mean total cholesterol recorded were 178.5+/-42.1 mg/dl v. 154.1+/-40.2 mg/dl (p<0.001), high-density lipoprotein cholesterol 30.6+/-9 mg/dl v. 27.3+/-6.8 mg/dl (p<0.001), low-density lipoprotein cholesterol 109.8+/-35.4 mg/dl v 93.6+/-33.9 mg/dl (p<0.001), and triglyceride 190.7+/-95.4 mg/dl v. 157.6+/-73.5 mg/dl (p<0.001), respectively. In subgroup analysis by age, the younger coronary artery disease group (< or = 40 years) had significantly higher total and low-density lipoprotein cholesterol levels than the older group (> 40 years), viz. 194.6+/-51.4 mg/dl v. 176.3+/-40.2 mg/dl (p<0.001), and 118.3+/-39.6 mg/dl v. 108.7+/-36.1 mg/dl (p=0.001). Triglyceride levels were not significantly different [211.7+/-105.1 mg/dl v. 187.8+/-93.6 mg/dl (p=ns)], being equally high in both subgroups and, although high-density lipoprotein cholesterol levels were different, this difference was minimal, being equally low in both [32.7+/-9.5 mg/dl v. 30.3+/-9.0 mg/dl (p=ns)]. In the subgroup analysis of those with coronary artery disease, diabetics had significantly lower total cholesterol [174+/-41.1 mg/dl v. 180.4+/-42.4 mg/dl (p<0.001)] and low-density lipoprotein cholesterol levels [105.8+/-34 mg/dl v. 111.5+/-35.8 mg/dl (p<0.001)] than non-diabetics, whereas triglyceride and high-density lipoprotein cholesterol levels were not significantly different, triglycerides being equally high in both [186.2+/-95.5 mg/dl v. 192.5+/-95.2 mg/dl (p=ns)], and high-density lipoprotein equally low in both [30.9+/-9.3 mg/dl v. 30.5+/-9 mg/dl (p=ns)]. The commonest associated conventional risk factor in diabetics was hypertension and, in the younger age group (< or = 40 years), it was smoking and a positive family history of premature coronary artery disease. CONCLUSIONS: We conclude that in north Indians, coronary artery disease occurs at much lower levels of total cholesterol and low-density lipoprotein cholesterol than other populations, and high triglyceride and low high-density lipoprotein levels are more of a universal phenomenon in this population. Younger patients have a more atherogenic lipid profile than the older subgroup with coronary artery disease, and smoking and a family history of premature coronary artery disease are the most common associated risk factors. Total cholesterol levels seem to play a lesser role in the occurrence of coronary artery disease in diabetics, the presence of which is in itself overwhelming for the occurrence of coronary artery disease.