Pharmacoepidemiological studies cohort study

Mackay, F., Dunn, N., Wilton, L., Pearce, G., Freemantle, S., Mann, R. (1997). A comparison of fluvoxamine, fluoxetine, sertraline, and paroxetine examined by observational cohort studies. Pharmacoepidemiology and Drug Safety, 6, 235-246. 

Observational studies, where the drug is observed epidemiologically under conditions of normal use in the community, i.e. pharmacoepidemiology. Techniques used for postmarketing (Phase 4) studies include the observational cohort study and the case-control study. The systems are described on page 69. 

A case-control study may be nested within a cohort smdy, i.e. cases and controls are all drawn from a clearly defined cohort. This is an efficient design which is now commonly used in pharmacoepidemiology. Attempts are made in the design and analysis to nimirnise possible biases, and to identify and adjust for confounding factors. Typically a cohort study will measure both absolute and relative risks whereas a case-control study will usually only measure odds ratios which generally approximate to relative risks, hi both cases the data may be summarised in two-by-two tables, as shown in the following examples ... 

Pancreas Case reports and pharmacoepidemiological studies have provided evidence that statins may cause pancreatitis. However, data from the SHARP study, a placebo-controlled study of the effects of a combination of simvastatin and ezetimibe on cardiovascular events in patients with chronic kidney disease, showed a reduction in the number of cases of pancreatitis [75] [40 -]. Recent prospective cohort study with 1062 subjects of whom 92 were taking statins found severe pancreatitis was more common in the statin nonuser than statin user. Pancreatitis-related mortality was higher in the statin nonuser, and among patients who developed severe acute pancreatitis, statin users showed lower Ranson s and APACHE II scores and lower maximal CRP, suggesting that prior statin treatment reduced morbidity and mortality in acute pancreatitis [76]. 

---