1701724P - PRIORITIES OF NURSING RESEARCH: KNOWLEDGE GENERATION OR PRACTICE IMPROVEMENT

Presentation at the International Nursing Symposium with the theme: “Cultivating Nursing Talent & Capacity in Saudi Arabia, Road Map to Excellence” held on 23-24 October 2017 at King Fahad Medical City by Professor Omar Hasan Kasule Sr. MB ChB (MUK). MPH (Harvard), DrPH (Harvard) Chairman of the Institutional Review Board / Research Ethics Committee at King Fahad Medical City, Riyadh.

WHERE IS THE HEART OF MODERN NURSING: THE MAJOR DICHOTOMIES:

• Information vs knowledge (the accomplished lawyer who knows a bit about everything including law)
• Healer vs technician; Physician vs barber.
• Academic nurse vs professional nurse (concept of phronosis)
• Independent profession vs supportive profession 
• Research for knowledge vs research for promotion
• Priorities based on needs vs priorities based on funding sources

WHAT IS THE PURPOSE OF NURSING RESEARCH?:

• Basic research vs applied research?[*]
• Scientific knowledge vs practice improvement?
• The micro vs. the macro?
• Holistic vs specific?
• My conviction is that nursing research should prioritize applied operational research to improve services.

SETTING NURSING RESEARCH PRIORITIES - 1:

• Peltonen LM et al. Nursing Informatics Research Priorities for the Future: Recommendations from an International Survey. Stud Health Technol Inform. 2016;225:222-6.
• Scott ES et al. Nursing Administration Research Priorities: Findings From a Delphi Study. J Nurs Adm. 2016 May;46(5):238-44.
• Garcia AB et al. A systematic review of nursing research priorities on health system and services in the Americas. Rev Panam Salud Publica. 2015 Mar;37(3):162-71.
• Spies LA et al. Uganda nursing research agenda: a Delphi study. Int Nurs Rev. 2015 Jun;62(2):180-6.
• Wynaden D et al. Identifying mental health nursing research priorities: A Delphi study. Contemp Nurse. 2014;47(1-2):16-26.

SETTING NURSING RESEARCH PRIORITIES - 2:

• Di Massimo DS et al. Nursing research priorities in internal medicine nursing practice: an Italian Delphi study. Ann Ig. 2015 Sep-Oct;27(5):760-8.
• Monterosso L et al. Developing a research agenda for nursing and midwifery: a modified Delphi study. Contemp Nurse. 2015;51(1):83-95.
• Sun C et al. Clinical Nursing and Midwifery Research Priorities in Eastern and Southern African Countries: Results From a Delphi Survey. Nurs Res. 2015 Nov-Dec;64(6):466-75.
• Cooley ME et al. The 2014-2018 Oncology Nursing Society Research Agenda. Oncol Nurs Forum. 2015 Sep;42(5):450-65.
• Mayer DK.  Improving Cancer Care Through Nursing Research. Oncol Nurs Forum. 2015 Sep;42(5):439.

SETTING NURSING RESEARCH PRIORITIES - 3:

• Wielenga JM et al. European neonatal intensive care nursing research priorities: an e-Delphi study. Arch Dis Child Fetal Neonatal Ed. 2015 Jan;100(1):F66-71.
• Tume LN et al. An electronic delphi study to establish pediatric intensive care nursing research priorities in twenty European countries*. Pediatr Crit Care Med. 2014 Jun;15(5):e206-13.
• Tume LN. Pediatric Critical Care Nursing Research Priorities-Initiating International Dialogue. Pediatr Crit Care Med. 2015 Jul;16(6):e174-82.
• Schoenly L. Research Priorities in Correctional Nursing Practice: Results of a Three-Round Delphi Study. J Correct Health Care. 2015 Oct;21(4):400-7. 
• Sun C et al. Clinical nursing and midwifery research in African countries: a scoping review. Int J Nurs Stud. 2015 May;52(5):1011-6.

LISTING OF NURSING RESEARCH PRIORITIES (NON-CLINICAL):

• EDUCATION-centered: Education and training, experience, learning, development, knowledge, skills, recruitment and retention, critical thinking, critical judgment, workforce issues, clinical competencies
• SYSTEM-centered: Policy, economics, Healthcare delivery systems organization and management, health care systems, Implementation, evaluation, data management, bioinformatics,
• RESEARCH-centered: biomarkers, large data, dissemination of knowledge.

LISTING OF NURSING RESEARCH PRIORITIES (CLINICAL):

• PRACTICE-centered: nursing practice, Clinical decisions, clinical quality, ethics, nursing protocols, new practice guidelines, professionalism.
• PATIENT-centered: Patient safety, patient engagement, risk reduction, pain, stress, patient outcomes.
• DISEASE/PROCESS-centered: Infectious disease control, cancer, resuscitation, ventilation, intensive care unit, neonatal care, trauma.
• FAMILY-centered: family caregivers, maternal health, impact of illness on the family, intensive care.
• LIFE CYCLE: Palliative and end of life care, aging, children.

TABLE 1: KFMC NURSING RESEARCH 2012-2014

TABLE 2: KFMC NURSING RESEARCH 2015

TABLE 3: KFMC NURSING RESEARCH 2016

TABLE 4: SUMMARY OF KFMC NURSING RESEARCH (NON- CLINICAL) 2012-2016

TABLE 5: SUMMARY OF KFMC NURSING RESEARCH (CLINICAL) 2012- 2016

TABLE 6: COMPARISON OF NURSING RESEARCH: KFMC vs LITERATURE (NON-CLINICAL)

TABLE 7: COMPARISON OF NURSING RESEARCH: KFMC vs LITERATURE (CLINICAL)

GENERAL CONCLUSIONS

• Most nursing research is applied and operational.
• Practice-centered research has higher priority.

NOTE:

* Parmar J et al. Rev Panam Salud Publica. 2015 Jun;37(6):409-14. Almost all of the studies (98%) were applied research and had a descriptive (55%) or qualitative (30%) design. The most prevalent topic was nursing care (23.4%). Health systems and services were the least studied topics. /About 25% of the studies contained some reference to United Nations Millennium Development Goals.

170717P - PRINCIPLES OF EPIDEMIOLOGY HEALTH RESEARCH COURSE: THE PROBABILITY THEORY AS A BASIS FOR DISCRETE RANDOM VARIABLES (RVS)

Presentation at a Course on Principles of Epidemiology, Health Research Faculty of Medicine, King Fahad Medical City, October 11-12, 2017 by Professor Omar Hasan Kasule Sr. MB ChB (MUK). MPH (Harvard), DrPH (Harvard) Chairman of the Institutional Review Board / Research Ethics Committee at King Fahad Medical City, Riyadh.

LECTURE 2: THE PROBABILITY THEORY AS A BASIS FOR DISCRETE RANDOM VARIABLES (RVS)

PROBABILITY AS A CONCEPT:

• The bulk of statistical theory is probability theory since modern inferential statistics depends on probability theory.

• Probability is the modeling of chance random events and a measure of the likelihood of their occurrence.

PROBABILITY AS a CONCEPT, Con’t...:

• Probability is commonly defined as the relative frequency of an event on repeated trials under the same conditions.

• Special mathematical techniques called arrangements, permutations, and combinations, can enable us to calculate the probability space theoretically without having to carry out the trials.

CLASSIFICATION OF PROBABILITY: 

• Probability can be subjective (based on personal feelings or intuition) or objective (based on real data or experience). Objective probability can be measured or computed.

• Prior probability is knowable or calculable without experimentation. The posterior probability is calculable from the results of experimentation.

• Bayesian probability combines prior probability (objective, subjective, or a belief) with new data (from experimentation) to reach a conclusion called posterior probability.

TYPES OF PROBABILITY EVENTS:

• On the scale of exclusion, events are classified as mutually exclusive or non-mutually exclusive. Mutually exclusive events are those that cannot occur together like being dead and being alive.

• On the scale of independence, events are classified as independent or dependent. Under independence, the occurrence of one event is not affected by the occurrence or non-occurrence of another. Independent events can occur at the same instant or subsequently. Some independent events are equally likely while others are not.

• On the scale of exhaustion, two events A and B are said to be exhaustive if between them they occupy all the probability space.

SET THEORY:

• Intersection A n B

 

• Union A u B

 

QUALITATIVE RANDOM VARIABLES:

• Qualitative variables (nominal, ordinal, and ranked) are attribute or categorical with no intrinsic numerical value.
• The nominal has no order, the ordinal has ordered, and the ranked has observations arrayed in ascending or descending orders of magnitude.

QUANTITATIVE (NUMERICAL) DISCRETE RANDOM VARIABLES - 1:

• The discrete random variables are the Bernoulli, the binomial, the multinomial, the negative binomial, the Poisson, the geometric, the hypergeometric, and the uniform.
• The Bernoulli is the number of successes in a single unrepeated trial with only 2 outcomes.

QUANTITATIVE (NUMERICAL) DISCRETE RANDOM VARIABLES - 1, Con’t...:

• The binomial is the number of successes in more than 2 consecutive trials each with a dichotomous outcome.
• The multinomial is the number of successes in several independent trials with each trial having more than 2 outcomes.

QUANTITATIVE (NUMERICAL) DISCRETE RANDOM VARIABLES - 2:

• The negative binomial is the total number of repeated trials until a given number of successes is achieved.
• The Poisson is the number of events for which no upper limit can be assigned a priori.
• The geometric is the number of trials until the first success is achieved.
• The hypergeometric is the number selected from a sub-sample of a larger sample for example selecting males from a sample of n persons from a population N. The uniform has the same value at repeated trials.

PLOT OF THE BINOMIAL DISTRIBUTION:

PLOT OF THE NEGATIVE BINOMIAL DISTRIBUTION:

PLOT OF THE POISSON DISTRIBUTION:

PLOT OF THE GEOMETRIC DISTRIBUTION:

 

170717P - PRINCIPLES OF EPIDEMIOLOGY HEALTH RESEARCH COURSE: THE PROBABILITY THEORY AS A BASIS FOR CONTINUOUS RANDOM VARIABLES

Presentation at a Course on Principles of Epidemiology Health Research Faculty of Medicine, King Fahad Medical City October 11-12, 2017 by Professor Omar Hasan Kasule Sr. MB ChB (MUK). MPH (Harvard), DrPH (Harvard) Chairman of the Institutional Review Board / Research Ethics Committee at King Fahad Medical City, Riyadh.

LECTURE 3: THE PROBABILITY THEORY AS A BASIS FOR 

CONTINUOUS RANDOM VARIABLES

QUANTITATIVE (NUMERICAL) CONTINUOUS RANDOM VARIABLES:

• The continuous random variables can be natural such as the normal, the exponential, and the uniform, or artificial such as chi-square, t, and F variables.
• The normal represents the result of a measurement on the continuous numerical scale such as height and weight.
• The exponential is the time until the first occurrence of the event of interest.
• The uniform represents the results of a measurement and takes on the same value at repeated trials. 

THE NORMAL DISTRIBUTION

THE EXPONENTIAL DISTRIBUTION

 

THE UNIFORM DISTRIBUTION

THE CHI-SQUARE DISTRIBUTION

THE STUDENT t DISTRIBUTION

THE F DISTRIBUTION

THE CONTINUOUS R.V: THE INTERVAL OR THE RATIO SCALES:

• Only 2 measurements are made on the interval scale, the calendar, and the thermometer. The rest of the measurements are on the ratio scale.
• The interval scale has the following properties: the difference between 2 readings has a meaning, the magnitude of the difference between 2 readings is the same at all parts of the scale, the ratio of 2 readings has no meaning, zero is arbitrary with no biological meaning, and both negative and positive values are allowed.
• The ratio scale zero has the following properties: zero has a biological significance, values can only be positive; the difference between 2 readings has a meaning, the ratio of 2 readings has a meaning and can be interpreted, and intervals between 2 readings have the same meaning at different parts of the scale.

170717P - PRINCIPLES OF EPIDEMIOLOGY HEALTH RESEARCH COURSE: THE NORMAL CURVE AND PROBABILITY

Presentation at a Course on Principles of Epidemiology Health Research Faculty of Medicine, King Fahad Medical City October 11-12, 2017 by Professor Omar Hasan Kasule Sr. MB ChB (MUK). MPH (Harvard), DrPH (Harvard) Chairman of the Institutional Review Board / Research Ethics Committee at King Fahad Medical City, Riyadh.

LECTURE 5: THE NORMAL CURVE AND PROBABILITY

INTRODUCTION:

• Abraham de Moivre first described the formula for the normal curve in 1744. In the 19th century, Pierre Simon Laplace and Carl Friedrich Gauss rediscovered the normal curve, each working independently.

• Around 1835, Adolph Quetelet first used the normal curve as an approximation to the histogram.

• The normal curve may be one of the unifying principles of nature reflecting sunan al llaah. The normal curve fits so many natural data distributions, making it very useful in statistics.

• The normal curve can be used for data that is initially not normally distributed. Such data can be made normally distributed by suitable mathematical transformations.

• The binomial, the Poisson, the t, and the chi square distributions become normal curves if the sample size is large enough.

PROPERTIES & CHARACTERISTICS OF THE NORMAL CURVE:

• The normal curve is described fully by its mean and its standard deviation. A standardized normal curve has mean = 0 and standard deviation = 1. 

• Two curves may have the same mean but different standard deviations. Two curves may have different means but the same standard deviation. 

• The normal curve is perfectly symmetrical about the mean. 

• Although continuous, it models discrete data well for large sample sizes. It is asymptotic i.e. approaches the x-axis but never touches it. data

NORMAL CURVE SHOWING MEAN AND STANDARD DEVIATION: 

USE OF THE NORMAL CURVE FOR NON-NORMAL DATA:

• Before using the normal curve to model a data set, tests have to be carried out to test the normality of the data.

• These tests include: a bell-shaped histogram, a straight line on probability paper, and use of special computer programs. 

• If the data is not normal it can be normalized by logarithmic, power, reciprocal, or Z-score transformation.

THE Z-SCORE and the AREA UNDER THE CURVE:

• The Z score is the deviation of a measurement from the mean measured in SD units. The standard normal variable, z, has mean 0 and variance 1, written as z ~ N (0,1).

• Z scores are used to compare different data sets, to determine a cut-off or critical value, and to replace the original variable in analysis.

• The area under the curve is relative frequency or probability. Mean +/- 1SD covers 68% of observations. Mean +/- 2 SD covers 95% of observations. Mean +/- 3 SD covers 99% of observations.

• The area under the curve between mean – 2 SD and mean + 2 SD is the probability of 95% confidence interval (CI).

NORMAL CURVE SHOWING z SCORES:

ESTIMATION:

• There are 3 types of estimates: the point estimate, the pooled estimate, and the interval estimate.

• Point estimation being just one point may be in error.

• Pooled estimation is a weighted combination of parameters from more than one population or sample.

• In Interval estimation the confidence interval is stated as the lower confidence level and the upper confidence level using a usual or customary confidence of 95%.

• In a common sense way the 95% confidence interval (CI) means that we are 95% sure that the true value of the parameter is within the interval.

VALIDITY vs. PRECISION:

• Validity tells us how well an instrument measures what it is supposed to measure.

• The mean is a measure of validity (parameter of location).

• The standard deviation is a measure of precision (spread).

• Validity and precision are both desirable but may not always be achieved simultaneously. A valid measurement may not be precise. A precise measurement may not be valid.

170717P - PRINCIPLES OF EPIDEMIOLOGY HEALTH RESEARCH COURSE: THE SCIENTIFIC METHOD AND DEDUCTIVE-INDUCTIVE HYPOTHESES IN HEALTH SCIENCE RESEARCH

Presentation at a Course on Principles of Epidemiology Health Research Faculty of Medicine, King Fahad Medical City October 11-12, 2017 by Professor Omar Hasan Kasule Sr. MB ChB (MUK). MPH (Harvard), DrPH (Harvard) Chairman of the Institutional Review Board / Research Ethics Committee at King Fahad Medical City, Riyadh.

LECTURE 1: THE SCIENTIFIC METHOD AND DEDUCTIVE-INDUCTIVE HYPOTHESES IN HEALTH SCIENCE RESEARCH 

EMPIRICISM:

• Epidemiological methodology, following the scientific method, is empirical

• Epidemiology relies on and respects only empirical findings.

• Empiricism refers to reliance on physical proof.

INDUCTIVE VS DEDUCTIVE INFERENCE:

• Epidemiological methodology, following the scientific method, is inductive

• Inductive inference = from the specific to the general

• Induction is building a theory on several individual observations

• Deductive inference is from the general to the specific

RELATIVITY VS. ABSOLUTISM:

• Nothing is absolute, everything is relative

• Science is not deterministic or absolute

• Some sciences are deterministic than others for example laboratory data vs epidemiological data

CLASSICAL VS BAYESIAN INFERENCES:

• Classical inference depends only on the data collected at the moment. It assumes starting the experiment with a clean slate

• Bayesian inference combines prior information (objective, subjective, or a belief) with new information (from experimentation) to reach a conclusion

• Bayesian inference is a good representation of how conclusions are made from empirical observation in real life

STATISTICAL VS SUBSTANTIVE QUESTIONS AND CONCLUSIONS:

• An investigator starts with a substantive question that is formulated as a statistical question.

• Data is then collected and is analyzed to reach a statistical conclusion. 

• The statistical conclusion is used with other knowledge to reach a substantive conclusion. 

• Statistics has a limitations: it gives statistical and not substantive answers.

• The statistical conclusion refers to groups and not individuals. 

• The statistical conclusion summarizes but does not interpret data. 

170717P - PRINCIPLES OF EPIDEMIOLOGY HEALTH RESEARCH COURSE: STUDY ANALYSIS AND INTERPRETATION: MEASURES OF ASSOCIATION and EFFECT

Presentation at a Course on Principles of Epidemiology, Health Research Faculty of Medicine, King Fahad Medical City, October 11-12, 2017 by Professor Omar Hasan Kasule Sr. MB ChB (MUK). MPH (Harvard), DrPH (Harvard) Chairman of the Institutional Review Board / Research Ethics Committee at King Fahad Medical City, Riyadh.

LECTURE 12-A: STUDY ANALYSIS AND INTERPRETATION: MEASURES OF ASSOCIATION and EFFECT:  

GENERAL CONCEPTS:

• Data analysis involves construction of hypotheses and testing them.

• Simple manual inspection of the data can help identify outliers, assess the normality of data, identify commonsense relationships, and alert the investigator to errors in computer analysis.

• Two procedures are employed in analytic epidemiology: tests for association and measures of effect. The association test is done first. The assessment of the effect measures is done after finding an association. Measures of effect are applied to discrete data.

• Measures of trend can discover relationships that are too small to be picked up by association and effect measures.

TESTS OF ASSOCIATION FOR CONTINUOUS DATA:

• The t-test is used for two sample means.

• Analysis of variance, ANOVA (F test) is used for more than 2 sample means.

• 1-way ANOVA involves one factor (explanatory variable).

• 2-way ANOVA involves 2 factors.

• Multiple analysis of variance, MANOVA, is used to test for more than 2 factors.

• Linear regression is used in conjunction with the t test for data that requires modeling.

TESTS OF ASSOCIATION FOR DISCRETE DATA:

• The Spearman chi-square test is used to test association of 2 or more proportions in contingency tables.

• The exact test is used to test proportions for small sample sizes.

• The Mantel-Haenszel chi-square statistic is used to test for association in stratified 2 x 2 tables.

• The chi-square works best for approximately Gaussian distributions. J

MEASURES OF EFFECT:

• The Mantel-Haenszel Odds Ratio is used for 2 proportions in a single or stratified 2x2 contingency table.

• Logistic regression can be used as an alternative to the MH procedure.

• For paired proportions, a special form of the Mantel-Haenszel OR and a special form of logistic regression called conditional logistic regression are used.

MEASURES OF EFFECT, Con’t.:

• Excessive disease risk is measured by Attributable Risk, Attributable Risk Proportion, and Population Attributable Risk. 

• Variation of an effect measure by levels of a third variable is called effect modification by epidemiologists and interaction by statisticians. 

VALIDITY and PRECISION:

• Validity is a measure of accuracy generally using measures of central tendency

• Precision measures variation in the estimate using variance or confidence interval

• Reliability is reproducibility.

• Internal validity is concerned with the results of each individual study.

• External validity is generalizability of results from a single large sample study or several small sample studies.

META ANALYSIS:

• Meta-analysis refers to methods used to combine data from more than one study to produce a quantitative summary statistic.

• Meta-analysis enables computation of an effect estimate for a larger number of study subjects thus enabling picking up statistical significance that would be missed if analysis were based on small individual studies.

• The summary effect measure, OR or b, is computed from the effect measures of individual studies using weighted logistic regression or computing a MH weighted average in which the weight of each measure is the inverse of its precision i.e. 1/(se)2.

LECTURE 12-B: STUDY ANALYSIS AND INTERPRETATION: SOURCES AND TREATMENT OF BIAS:

MISCLASSIFICATION BIAS:

• Misclassification is inaccurate assignment of exposure or disease status. It may be random or non-random

• Misclassification bias is classified as information bias, detection bias, and proto-pathic bias.

• Information bias is systematic incorrect measurement on response due to questionnaire defects, observer errors, respondent errors, instrument errors, diagnostic errors, and exposure mis-specification.

MISCLASSIFICATION BIAS, Con’t.:

• Detection bias arises when disease or exposure are sought more vigorously in one comparison more than the other group.

• Protopathic bias arises when early signs of disease cause a change in behaviour with regard to the risk factor. Misclassification bias can be prevented by using double-blind techniques to decrease observer and respondent bias.

• Treatment of misclassification bias is by the probabilistic approach or measurement of inter-rater variation.

SELECTION BIAS:

• Selection bias arises when subjects included in the study differ in a systematic way from those not included.

• Selection bias due to disease ascertainment procedures includes publicity, exposure, diagnostic, detection, referral, self-selection, and Berkson biases.

• The Hawthorne self-selection bias is also called the healthy worker effect since sick people are not employed or are dismissed.

SELECTION BIAS, Con’t.:

• The Berkson fallacy arises due to differential admission of some cases to the hospital in proportions such that studies based on the hospital give a wrong picture of disease-exposure relations in the community.

• Selection bias during data collection is represented by non- response bias and follow-up bias.

• Prevention of selection bias is by avoiding its causes that were mentioned above. There is no treatment for selection bias once it has occurred.

CONFOUNDING BIAS:

• Confounding is the mixing up of effects. Confounding bias arises when the disease-exposure relationship is disturbed by an extraneous factor called the confounding variable, related to both disease and exposure but unequally distributed.

• Prevention of confounding at the design stage by eliminating the effect of the confounding factor can be achieved using 4 strategies: pair-matching, stratification, randomisation, and restriction.

CONFOUNDING BIAS, Con’t.:

• Confounding can be treated at the analysis stage by various adjustment methods (both non-multivariate and multi-variate).

• Non-multivariate treatment of confounding employs standardization and stratified Mantel-Haenszel analysis.

• Multivariate treatment of confounding employs multivariate adjustment procedures: multiple linear regression, linear discriminant function, and multiple logistic regression.

MIS-SPECIFICATION BIAS:

• This type of bias arises when a wrong statistical model is used.

• An example use of parametric methods for non-parametric data biases the findings.

SURVEY ERROR and SAMPLING BIAS:

• Total survey error is the sum of the sampling error and three non-sampling errors (measurement error, non-response error, and coverage error). 

• Sampling error decreases with increasing sample size. 

• Sampling bias, positive or negative, arises when results from the sample are consistently wrong (biased) away from the true population parameter. 

• The sources of bias are: incomplete or inappropriate sampling frame, use of a wrong sampling unit, non-response bias, measurement bias, coverage bias, and sampling bias.

170717P - PRINCIPLES OF EPIDEMIOLOGY HEALTH RESEARCH COURSE: SIX PROPERTIES OF RVS AS A BASIS FOR DESCRIPTIVE STATISTICS

Presentation at a Course on Principles of Epidemiology, Health Research Faculty of Medicine, King Fahad Medical City, October 11-12, 2017 by Professor Omar Hasan Kasule Sr. MB ChB (MUK). MPH (Harvard), DrPH (Harvard) Chairman of the Institutional Review Board / Research Ethics Committee at King Fahad Medical City, Riyadh.

LECTURE # 4: SIX PROPERTIES OF RVS AS A BASIS FOR DESCRIPTIVE STATISTICS

THE 6 PROPERTIES

1. EXPECTATION: The expectation of a random variable is a central value around which it hovers most of the time. 
2. VARIANCE: The variation of the random variable around the expectation is measured by its variance.
3. COVARIANCE: Covariance measures the covariability of the two random variables.
4. CORRELATION: Correlation measures the linear relation between two random variables.
5. SKEWNESS: Skewness measures the bias of the distribution of the random variable from the center. 
6. KURTOSIS: Kurtosis measures the peakedness of the random variable at the point of its expectation 

NORMAL DISTRIBUTION SHOWING EXPECTATION:

 

NORMAL DISTRIBUTION SHOWING VARIANCE:

SCATTER PLOT SHOWING CORRELATION:

DISTRIBUTION SHOWING POSITIVE & NEGATIVE SKEW:

DISTRIBUTION SHOWING KURTOSIS:

170717P - PRINCIPLES OF EPIDEMIOLOGY HEALTH RESEARCH COURSE: MIXED DATA ANALYSIS (DISCRETE AND CONTINUOUS) CORRELATION

Presentation at a Course on Principles of Epidemiology, Health Research Faculty of Medicine, King Fahad Medical City, October 11-12, 2017 by Professor Omar Hasan Kasule Sr. MB ChB (MUK). MPH (Harvard), DrPH (Harvard), Chairman of the Institutional Review Board / Research Ethics Committee at King Fahad Medical City, Riyadh.

LECTURE # 10-A: MIXED DATA ANALYSIS (DISCRETE AND CONTINUOUS) CORRELATION

CORRELATION:

• Correlation analysis is used as preliminary data analysis before applying more sophisticated methods. 

• Correlation indicates only association; the association is not necessarily causative. It measures linear relation and not variability. 

OVERVIEW OF NON-PARAMETRIC ANALYSIS FOR CONTINUOUS DATA:

• Non-parametric methods were first introduced as rough, quick and dirty methods and became popular because of being un- constrained by normality assumptions. 

• They are about 95% as efficient as the more complicated and involved parametric methods. They are simple, easy to understand, and easy to use. 

• Generally non-parametric methods are used where parametric methods are not suitable. 

OVERVIEW OF NON-PARAMETRIC ANALYSIS FOR CONTINUOUS DATA: Con’t…

• The first step in correlation analysis is to inspect a scatter plot of the data to obtain a visual impression of the data layout and identify out-liers.

• Then Pearson’s coefficient of correlation (product moments correlation), r, is the commonest statistic for linear correlation. 

OTHER CORRELATION COEFFICIENTS:

• When the relation between x and y is influenced by a third variable, the coefficient of partial correlation explains the net relationship.

• The correlation ratio, used for curvilinear relations, is interpreted as the variability of y accounted for by x. 

OTHER CORRELATION COEFFICIENTS, Con’t. - 1:

• The biserial or tetrachomic correlation coefficient is used in linear relations when one variable is quantitative and the other is qualitative. 

• The contingency coefficient is used for 2 qualitative nominal (i.e. unordered) variables each of which has 2 or more categories. 

• The coefficient of mean square contingency is used when both variables are qualitative. 

OTHER CORRELATION COEFFICIENTS, Con’t. - 2:

• The multiple correlation coefficient is used to describe the relationship in which a given variable is being correlated with several other variables.

• The partial correlation coefficient denotes the conditional relation between one independent variable and a response variable if all other variables are held constant. 

THE COEFFICIENT OF DETERMINATION, r2:

• The square of the linear correlation coefficient is called the coefficient of determination. 

• It is the proportion of variation in the dependent variable, y, explained by the variation in the independent variable, x.

NON-PARAMETRIC CORRELATION ANALYSIS:

• The Spearman rank correlation coefficient is used for non-normal data for which the Pearson linear correlation coefficient would be invalid.

• The advantage of rank correlation is that comparisons can be carried out even if actual values of the observations are not known. It suffices to know the ranks. 

LECTURE # 10-B: MIXED DATA ANALYSIS (DISCRETE AND CONTINUOUS): REGRESSION 

LINEAR REGRESSION - 1:

• Regression to the mean, first described by Francis Galton (1822- 1911) is one of the basic laws of nature, sunan al llah fi al kawn.

• Parametric regression models are cross sectional (linear, logistic, or log-linear) or longitudinal (linear and proportional hazards).

• Regression relates independent with dependent variables.

LINEAR REGRESSION - 2: 

• The simple linear regression equation is y=a + bx where y is the dependent/response variable, a is the intercept, b is the slope/regression coefficient, and x is the dependent/predictor variable.

• Multiple linear regression, a form of multivariate analysis, is defined by y=a+b1x1 + b2x2 + ...bnxn. 

• Linear regression is used for prediction (intrapolation and extrapolation) and for analysis of variance.

LOGISTIC REGRESSION:

• Logistic regression is non-linear regression with y dichotomous/binary such that logit (y) = a+b1x1 + b2x2 + ...bnxn 

• Logistic regression is used in epidemiology because of a dichotomized outcome variable and direct derivation of the odds ratio from the regression coefficient as shown in the formula OR = eβ.

• Multiple logistic regression is used for matched analysis, stratified analysis to control for confounders, and prediction.

FITTING REGRESSION MODELS:

• Step-up or forwards selection starts with a minimal set of x variables and one x variable is added at a time.

• Step-down or backward elimination starts with a full model and one variable is eliminated at a time.

• Step-wise selection is a combination of step up and step down selection. 

• Variables are retained or eliminated on the basis of their p-value.

ASSESSING REGRESSION MODELS:

• The best model is one with the highest coefficient of determination.

• The coefficient of determination defined as r2 varies 0-1.0 and is a measure of goodness of fit. 

LECTURE 10-C: MIXED DATA ANALYSIS (DISCRETE AND CONTINUOUS): TIME SERIES ANALYSIS and SURVIVAL ANALYSIS

TIME SERIES ANALYSIS:

• Longitudinal data is summarized in the following ways: graphical presentation, longitudinal regression, auto-regression, autocorrelation, repeated measures ANOVA, and tests for trend.

• A time series plot of y against time shows time trends, seasonal patterns, random / irregular patterns, or mixtures of the above.

• Forecasts can be made using time series.

TIME SERIES ANALYSIS, Con’t.:

• Longitudinal regression models, additive or multiplicative, can be used to model time-varying data. A

• Auto-regression is a regression model relating a variable to its immediate predecessor.

• Autocorrelation is correlation between a variable and its lagged version (immediate predecessor). 

• A chi-square test for trend can be constructed for 2 x k contingency tables where k represents time periods.

INTRODUCTION TO SURVIVAL ANALYSIS:

• Survival analysis is used to study survival duration and the effects of covariates on survival. It uses parametric methods (Weibull, lognormal, or gamma) or non-parametric methods (life- table, Kaplan-Maier, and the Proportional hazards). 

• Time is measured as time to relapse, length of remission, remission duration, survival after relapse, time to death, or time to a complication. 

INTRODUCTION TO SURVIVAL ANALYSIS, Con’t.: 

• The best zero time is point of randomization. Other zero times are: enrolment, the first visit, first symptoms, diagnosis, and start of treatment.

• Problems of survival analysis are censoring, truncation, and competing causes of death. Censoring is loss of information due to withdrawal from the study, study termination, loss to follow-up, or death due to a competing risk. 

NON-REGRESSION SURVIVAL ANALYSIS:

• Two non-regression methods are used in survival analysis: The life-table and the Kaplan-Maier methods.

• The life-table methods better with large data sets and when the time of occurrence of an event cannot be measured precisely. It leads to bias by assuming that withdrawals occur at the start of the interval when in reality they occur throughout the interval.

NON-REGRESSION SURVIVAL ANALYSIS, Con’t.:

• The Kaplan-Maier method is best used for small data sets in which the time of event occurrence is measured precisely.

REGRESSION METHODS FOR SURVIVAL ANALYSIS:

• The Proportional hazards, a semi-parametric method proposed by Sir David Cox in 1972, is the most popular regression method for survival analysis.

• It is used on data whose distribution is unknown.

COMPARING SURVIVAL CURVES:

• Proportional hazards, a semi-parametric method proposed by Sir David Cox in 1972, is the most popular regression method for survival analysis. 

• It is used on data whose distribution is unknown. 

COMPARING SURVIVAL CURVES:

• The non-parametric methods for comparing 2 survival distributions are: Gehan’s generalized Wilcoxon test, the Cox- Mantel test, the log-rank test, Peto’s generalized Wilcoxon test, the Mantel-Haenszel test, and Cox’s F test.

• The parametric tests are the likelihood ratio test and Cox’s F test. The log-rank test is more sensitive if the assumptions of proportional hazards hold. 

170717P - INSTITUTIONAL REVIEW BOARD: PROCEDURES AND PROBLEMS

Presentation at a Course on IRB held on July 17-18, 2017, at MOH Jeddah by Professor Omar Hasan Kasule Sr. MB ChB (MUK). MPH (Harvard), DrPH (Harvard), Chairman of the Institutional Review Board / Research Ethics Committee at King Fahad Medical City, Riyadh.

FUNCTIONS OF REC:

• Initial evaluation and approval of research proposals to make sure they fulfill the requirements of the Saudi regulations on human research and the international consensus Good Clinical Practice Guidelines. Above these regulations, the REC must ensure the highest ethical standards in any research.
• Follow-up evaluation and approval of matters arising in the course of the research: protocol amendments, study termination/completion.
• Monitoring of study execution by checking on vital issues such as proper consenting procedures, confidentiality of the data, and complete and up-to-date documentation, completeness and quality of AE reporting[1], and any ethical violations/protocol deviations.

MEMBERSHIP OF IRB - 1:

• Membership of the REC must have a diversity of medical professional competencies (clinical and non-clinical) to make sure that for every project reviewed there is a member from the relevant discipline and not necessarily the sub-discipline. Physicians tend to dominate RECs[2]. 
• Membership should comprise representatives of the major medical and surgical specializations practiced in the hospital or region, hospital physicians, hospital nursing staff, general practitioners, pharmacists, statisticians, ethicists, and lay persons from the community. 
• At least one of the members must be a normal community representative with no affiliation to the institution. This member should not have any connection with medical work. 

MEMBERSHIP OF IRB - 2:

• In selecting members, attempts should be made to make sure that all genders and age groups are well represented.
• Members should be selected on their own personal merit as people with knowledge, skills, and sound judgment. They should discuss the proposals as individuals and not representatives of any department, unit, organization, or profession.
• Members are appointed by the hospital, the health authority in the region, or the government.
• The period of service on the committee is usually three years. Membership may be renewed. Staggered renewal of membership. A member should serve 2-3 terms.

SCOPE OF RESEARCH REVIEW BY REC:

• The committee assesses research proposals and protocols that have ethical implications: (a) research on patients, volunteers, the recently dead, fetal or embryological tissues (b) research with potential to breach confidentiality.
• All research, however, must be submitted to the REC Chairman because the researcher cannot be trusted to determine the classification.
• The Chairman will determine which proposals are exempt (no risk, no ethical issues) and which are expedited (minimal risk) and approve them immediately.
• The Chairman will determine and approve expedited proposals (minimal patient risk) and, if need be, can consult one or more members of the committee. 
• Full review is done for research with human intervention.

DOCUMENTS SUBMITTED FOR REC REVIEW:

• Exempt proposals submitted on a special form with the proposal attached. The form must be signed by the PI and HOD.
• Full and expedited: Protocols, investigator brochures, and consent documents of proposals with potentially significant patient risk are sent to all REC members at least 2 weeks in advance for review, with those from the relevant disciplines being asked to make a more detailed analysis and make a presentation to the REC.
• The CVs of investigators and certification of GCP must be attached. 

REC MEETING:

• The committee must be provided with adequate secretarial and logistical assistance to carry out its functions well. 
• A quorum of at least half of the members will be necessary for holding a meeting. Members who cannot attend can send written comments and may/may not vote.
• The committee meets in private to preserve confidentiality. Others argue that they should meet in public for more transparency.
• If the issues are complicated, the investigator may be invited to the meeting to explain.
• Any member of the committee involved in a project will recuse himself when that project comes up for discussion. 

CRITERIA OF REC APPROVAL - 1: 

• The two main considerations in REC decisions are: informed consent and patient safety. 
• Informed and voluntary consent following full disclosure of objectives, risks and benefits of the research, and the right to abstain or withdraw from the study. Special scrutiny of proxy consent for the vulnerable will be made to ensure there is no abuse[4,5,6]. The investigator must submit reasons in writing in cases in which full disclosure is deemed inappropriate. 
• Patient safety based on careful weighing of benefits and risks.
• Confidentiality (use of certificates of confidentiality[7]) and security of the data.

CRITERIA OF REC APPROVAL - 2:

• Scientific merit: objectives clearly stated and attainable, research design and statistical methods are adequate to produce clinically and scientifically useful results. Outside experts may be consulted. The committee will compare the scientific merit and benefit of the research against risks and costs to patients. Decision-making procedures may improve the risk-benefit discourse[8].
• Qualification of the researchers based on a study of their detailed resumes.
• Adequacy of research facilities.
• No conflicts of interest. 

DECISIONS OF IRB:

• REC decisions are best taken by consensus, but if this is not possible, the decision will be based on a simple majority of the members attending if the quorum is assured. Minority views should be recorded.
• The decision of the committee may be full approval, conditional approval, deferment, or rejection. Reasons should be provided for projects approved conditionally or those that are rejected. If the investigator fulfills the missing information, the chairman may approve a conditional approval without returning to the full committee. 

FOLLOW-UP OF RESEARCH:

• The committee must monitor progress of the research project and must receive reports of all adverse reactions, whether related to the drug tested or not. AE reports from all sites of multi-center trials must be submitted. 
• Members of the committee can make on-site inspections to make sure that the approved protocol is adhered to and to inspect research documents and records.
• Regular monitoring meetings are held to review the following: progress of recruitment of research subjects, changes to the protocol, adverse reactions, the process of informed consent, refusals and withdrawals, and case record forms.
• Warning letters[9]
• Audits 

OVERSIGHT OF IRB

• The committee keeps full records of all its actions. Records are not privileged if a suit arises.
• It submits an annual report listing all proposals considered in the past year, the number approved, and any matters that deserve attention from higher authorities.
• The Chairman of IRB reports to the highest official in the institution. 
• Tool for REC self-assessment[10].
• Researchers have a right of appeal. 

REFERENCES:

1. Clin Cancer Res. 2009 Jun 1;15(11):3850-5.
2. BMC Med Ethics. 2010 Jun 28;11:12.
3. J Med Ethics. 2008 Aug;34(8):631-5.
4. Arch Dis Child. 2010 Nov;95(11):915-7. 
5. J Med Ethics. 2009 Jun;35(6):377-81.
6. Crit Care Med. 2010 Nov;38(11):2146-54.
7. PLoS One. 2012;7(9):e44050.
8. BMC Med Ethics. 2012 Apr 20;13:6.
9. Indian J Med Ethics. 2011 Oct-Dec;8(4):211-4.
10. J Empir Res Hum Res Ethics. 2010 Sep;5(3):85-96; quiz 97-8. 13