Adverse events case-control studies

Figueras, A., Tato, F., Fontainas, J. and Gestal-Otero, J.J. (1999). Influence of physician s attitudes on reporting adverse drug events A case-control study. Medical Care 37 809-814. 

Case-control studies start with patients that had the event of interest, often an adverse event (such as phocomelia), and compare the previous events (such as medications used) in the patients lives to those in a group of control patients who did not have the event of interest. These studies are especially useful to generate ideas about causes of uncommon events. The example of thalidomide-induced phocomelia is a classic example of the use of this epidemiological approach. 

Adverse events listed in clinical trials, case reports, and for rare events epidemiological studies, such as case control studies. Data from national side-effects registers or from the WHO register can be of value, but should be interpreted with caution. 

Retrospective case-control studies reported that immediate-acting nifedipine increased the risk of myocardial infarction in patients with hypertension. Slow-release and long-acting vasoselective calcium channel blockers are usually well tolerated. Flowever, dihydropyridines, compared with angiotensin-converting enzyme (ACE) inhibitors, have been reported to increase the risk of adverse cardiac events in... 

Case report forms and narratives should be provided for all patients who died, discontinued from a study due to an adverse event, or experience a serious adverse event. Case report forms for all patients involved in pivotal well-controlled studies should be available upon request. 

Case-control studies have been used extensively to assess the safety of pharmaceuticals. There are many examples of case-control studies that have identified important associations between drugs and adverse health events vaginal cancer and diethylstilbestrol, Reye s syndrome and aspirin, peptic ulcer disease and nonsteroidal antiinflammatory drugs, and venous thromboembolism and oral contraceptives. Data from case-control studies are used to calculate an odds ratio, which is the ratio of the odds of developing the disease for exposed patients to the odds of developing the disease for unexposed patients. 

On the other hand, certain drugs have gone through the fast-track process and, at the time of authorisation, information on the safety of these medications has been limited, especially for children. In such cases non-interventional observations such as cohort or case-control studies could be a valuable tool to evaluate adverse events. 

A case-control study is a retrospective analysis it is generally easier to administer than a cohort study. Cases of diseases or events are identified. Controls and patients exposed to the treatment are selected from the source population. The exposure status of the two groups is compared using the odds ratio, an estimate of relative risk of exposure and non-exposure. Case-control studies are less expensive than cohort studies, but provide weaker empirical evidence than well-executed cohort studies. These studies are useful for identifying the relationship between drug treatments with one specific rare adverse event, or for identifying risk factors for adverse events. Risk factors can include renal and hepatic insufficiency that might modify the risk profile. 

Case-control studies are best for studying rare adverse events that take a long time to develop. A disadvantage of this type of study is that it is based on memory and recall, which can be biased, as well as on medical records, which can be incomplete. 

White JR, Walczak TS, Marino SE, Beniak TE, Leppik IE, Bimbaum AK. Zonisamide discontinuation due to psychiatric and cognitive adverse events a case-control study. Neurology 2010 75(6) 513-8. 

Campbell RJ, GiU SS, BronskiU SE, Paterson JM, Whitehead M, Bell CM. Adverse events with intravitreal injection of vascular endothelial growth factor inhibitors nested case-control study. BMJ July 4, 2012 345 e4203. http //dx.doi.org/10.1136/bmj.e4203. PubMed PMID 22763393 PubMed Central PMOD PMC3389519. 

Study and control groups. Investigators need to consider whether the adverse events that occur are due to abnormalities in the distribution, metabolism, and excretion of drugs as a result of underlying disease. These analyses could be systematically facilitated by having standardized ways of measuring blood (and in some cases, tissue) levels of drugs and their metabolites. 

Despite the history of robust BZ anxiolytic impact in adults, controlled studies, open studies, and case reports of BZs for pediatric anxiety have not been impressive. Concerns about BZ-related adverse events, such as behavioral disihnhibition, have since slowed interest in controlled studies of BZs for treatment of pediatric anxiety disorders. Three small placebo-controlled studies of BZs for pediatric anxiety disorders have been published and none demonstrated... 

In a placebo-controlled study of 1142 hypercholestero-lemic patients treated with pravastatin for 8-16 weeks, the numbers of adverse drug experiences were similar in the treated and untreated individuals (1). Rash was the only adverse clinical event that was different (4.0 versus 1.1%). However, in the same patients withdrawal of therapy during follow-up was thought to be necessary in 3.2% of those given pravastatin alone. Myopathy was observed in one instance only, and increases in creatine kinase activity in those taking pravastatin did not differ significantly from controls. There were marked persistent increases in transaminases in 1.1%, with no cases of symptomatic hepatitis. Pravastatin is believed to have a particularly low potential for nervous system-related adverse effects, as it has not been shown to enter the cerebrospinal fluid, and clinical experience suggests that muscle toxicity occurs less often with pravastatin than with lovastatin (2). 

In a placebo-controlled study in 48 patients with active rheumatoid arthritis, CD4 blockade produced clinical benefit (5). Adverse events were reported in 97% of the patients, compared with 73% of those given placebo. In both groups most of the events were mild to moderate. Serious adverse events were reported in five patients who received anti-CD4 syncope/vasovagal attacks (n = 3), back pain (n = 1), abdominal pain/rectal bleeding (n = 1). Skin rashes occurred in 62% of the patients who received the antibody. In five cases a skin biopsy was performed, and showed a cellular infiltration centered on the blood vessels, suggesting a drug-induced vasculitis. 

Intravenous lidocaine has been used to treat severe chronic daily headache in 19 patients (three men, median age 37 years) (9). There were adverse effects during four infusions of lidocaine hyperkalemia (6.4 mmol/1), which did not resolve after withdrawal of lidocaine transient hypotension (75/50 mmHg), which was attributed to concomitant droperidol an unspecified abnormality of cardiac rhythm and on another occasion a transient bradycardia and chest pain with a normal electrocardiogram, fever, and intractable nausea. The study was neither randomized nor placebo-controlled, and in no case was the adverse event strongly associated with the administration of lidocaine. 

In a retrospective view of 63 patients who received intravenous milrinone for more than 24 hours for advanced cardiac failure, the mean dose was 0.43 micro-gram/kg/minute and the mean duration of therapy 12 (range 1-70) days (14). After 24 hours of therapy there was significant improvement in pulmonary artery pressure, pulmonary capillary wedge pressure, and cardiac index. Because of the nature of the study, which was not placebo-controlled, it is impossible to be sure what events could have been attributed to the milrinone. However, the authors reported five cases of asymptomatic, non-sustained ventricular tachycardia, six of symptomatic ventricular tachycardia, and three deaths, one in ventricular tachycardia and two in heart failure. There was no difference in the incidence of these adverse events in patients who received milrinone for more than 7 days compared with the others. 

---