Clinical Biostatistics 1

Clinical Biostatistics 1

Study Designs • Observational studies • Descriptive or case series • Case-control studies (retrospective) • Cause and incidence of disease • Identification of risk factors • Cross sectional studies, surveys (prevalence) • Disease description • Diagnosis and staging • Disease processes, mechanisms

Study Designs • Observational studies • Cohort studies (prospective) • Causes and incidence of disease • Natural history, prognosis • Identification of risk factors • Historical cohort studies

Study Designs • Experimental studies • Controlled trials • Parallel or concurrent controls • Randomized • Not randomized • Sequential controls • Self controlled • Cross-over • External controls (including historical) • Studies with no controls

Study Designs • Meta analysis

exposed cases unexposed exposed controls unexposed time Onset of study QUESTION: WHAT HAPPENED? Case Control

With outcome Subjects selected for the study Without outcome Onset of study time No direction of inquiry QUESTION: WHAT IS HAPPENING? Cross sectional study

With outcome Subjects (exposed) Without outcome With outcome Cohort selected for study Controls (unexposed) Without outcome time Onset of study Direction of inquiry QUESTION: WHAT WILL HAPPEN? Cohort Study Design

With outcome Subjects (exposed) Without outcome With outcome Cohort selected for study Controls (unexposed) Without outcome time Onset of study Direction of inquiry QUESTION: WHAT WILL HAPPEN? Historical Cohort Study Design

SURVEY today CASE CONTROL COHORT HISTORICAL COHORT Time relationship

With outcome Experimental subjects Without outcome With outcome Subjects meeting entry criteria Controls Without outcome intervention time Onset of study Randomized Clinical Trial Design

With outcome With outcome Experimental subjects Experimental subjects Without outcome Without outcome With outcome With outcome Subjects meeting entry criteria Controls Controls Without outcome Without outcome time Washout period intervention intervention Onset of study RCT with Crossover Design

With outcome Subjects Without outcome With outcome Results from previous historical study Without outcome intervention time Onset of study Trial with external control Design

Exercise 1 • Kremer and coworkers (1987) designed a study to determine the efficacy of fish oil dietary supplements in treating rheumatoid arthritis. They were particularly interested in the effect of the fish oil on the inhibition of neutrophil leukotriene levels. The study involved a group of 40 patients with class 1, 2, 3 rheumatoid arthritis; each patient was given either a dietary supplement or a placebo for 14 weeks, but the treatment assignment was not randomized. From weeks 14-18, all patients took a placebo for this 4-week-period; then they were given the opposite treatment (dietary supplement or placebo) from weeks 1-14 for the next 14 weeks crov

Exercise 2 • A study by O’Malley and Fletcher (1987) looked at the efficacy of the breast self examination (BSE) as a screening test for breast cancer by reviewing studies published on this topic. The authors found the sensitivity of BSE (the percentage of women with breast cancer who have a positive BSE) to be much lower than the sensitivity of a clinical breast examination or mammography. Although training increases the use of BSE and its sensitivity, the number of false positives (women without breast cancer who have a positive BSE) also increases. The authors suggest the need of a controlled trial on BSE before advocating its use as a screening device. rvw

Exercise 3 (part 1) • A study of 252 adult day care providers at 6 day care centers and children from 3 of the centers was undertaken to determine the occupational risk of CMV infection among day care providers (Murphy et al 1991). This virus has been found in the serum of up to 80% of children at some day care centers and the investigators were interested in studying the extent of the disease among day care providers. All day care centers in the study accepted children from ages 2 weeks to 5 years. Each center has between 60 and 231 children at the time of the study. After ff up of 2.5-4.5 years, depending on the center, the investigators reported an overall annual seroconversion rate of 8% among providers.

Exercise 3 (part 2) • Although the rates of conversion varied a great deal among centers, conversion rates appeared to be associated with the level of CMV excretion by children at the centers. ch

Exercise 4 • Kilbourne and colleagues (1983) investigated an epidemic in Spain involving multiple organ systems. Patients presented with cough, dyspnea, pleuritic chest pain, headache, fever, and bilateral pulmonary infiltrates. Although an infectious agent was first suspected, a strong association with cooking oil sold as olive oil but containing a high proportion of rapeseed oil was detected. Epidemiologic studies found that virtually all patients had ingested such oil but that unaffected persons had rarely done so. cc

Exercise 5 • Knutson and associates (1981) treated wound, burn, and ulcer patients using granulated sugar combined with povidone iodine (PI). The study was undertaken from Jan 1976 to Aug 1980 during which time, 759 patients were treated. Of these 154 were treated with standard therapy and the remaining 605 were treated with sugar and PI. Uniformity in treatment and judgment regarding the healing process were enhanced by using 3 physician investigators to oversee the process and by documenting wound healing with 35 mm transparencies. The investigators reported that a much lower percentage of patients treated with sugar-PI required skin grafts. nrct

Exercise 6 (part 1) • Colditz and colleagues (1987) reported on the relationship between menopause and the risk of coronary heart disease in women. Subjects in the study were selected from the Nurses’ Health Study originally completed in 1976. the study included 120,000 married female registered nurses aged 30-55. Colditz et al identified 116,000 of these women who were premenopausal or who had a known type of menopause and did not have a diagnosis of coronary heart disease at the beginning of the study. The investigators were interested in determining whether the occurrence of menopause alters the risk of coronary heart disease -

Exercise (Part 2) • specifically whether the influence of menopausal status is altered by the use of postmenopausal estrogen. The original survey provided information on the subjects’ age, parental history of myocardial infarction, smoking status, height, weight, use of oral contraceptives or post menopausal hormones, and history of MI or angina pectoris, diabetes, HPN, or high serum cholesterol levels. Follow up surveys were done in 1978, 1980, and 1982 and the data were 94.2% complete chpr

Exercise 7 • Bartle and co-workers (1986) designed a study to examine the association between NSAID use and acute non-variceal bleeding in the upper GIT. The association between consumption of ASA and UBI bleeding is well established. However there is no information available on non-ASA NSAID’s. The medical records were reviewed to obtain medication histories of 57 consecutive patients with non-variceal acute UGI hemorrhage presenting at a medical center to compare them with 123 sex-matched and age-matched controls. The investigators found that a larger proportion of patients than of controls had taken NSAID’s. cc

Scales of measurement • Nominal scales • Dichotomous or binary • Also called qualitative observations • Also called categorical observations • Presented as percentage or proportions • Ordinal scales • Ex: Ca stage, Arthritis grade, Quality of life grade

Scales of measurement • Numerical scales • Quantitative observations • Types • Interval or continuous • Discrete

Clinical Biostatistics 1