Introduction | Part 1: Hierarchy of Evidence | Part 2: Case Reports | Part 3: Cross-Sectional Studies | Part 4: Case Control Studies | Part 5: Cohort Studies | Part 6: Randomized Controlled Studies | Part 7: Meta-Analysis | Part 8: Systematic Reviews

Related: The Curbside Consultation | The Focused Clinical Question

Status: Draft (in development)

Case-control studies (or surveys) involve choosing a population with a defining problem or characteristic, such as:

  1. Cats with gastrointestinal lymphoma
  2. Doberman pinschers who are not screened for dilated cardiomyopathy

and compares this group with a control group with similar features

  1. Cats without gastrointestinal lymphoma
  2. Doberman pinschers who are screened for cardiomyopathy

These studies measure the frequency of the target disease or characteristics in both groups and attempts to reveal a potential association between them. Their distinctive feature is they begin after the onset of disease (or failure to develop the disease) in the individual patient. Therefore, they are inherently retrospective and seek to learn about patients before their disease began. In other words, patients with the target outcome (e.g. cats with gastrointestinal lymphoma) are identified first. Once identified, their medical history is subsequently evaluated during the period prior to the onset of their illness to try to identify any predisposing factors.

The term "case" refers to the cases of patients who have developed the target disease. Cases may be newly diagnosed patients (i.e. incidence cases) or they may be patients with with a pre-existing or chronic disease (i.e. prevalence cases). Control refers to the group of control patients who have not developed the disease.

Example

Incidence and prognostic value of low plasma ionized calcium concentration in cats with acute pancreatitis:

46 cases (1996-1998), Kimmel SE, Washabau RJ, Drobatz KJ, J Am Vet Med Assoc 2001 Oct 15;219(8):1105-9

The objective of this study was to determine the incidence and prognostic significance of low plasma ionized calcium in cats with clinical signs of acute pancreatitis (AP). A retrospective, case-control study, 46 cats with acute pancreatitis (the cases) were compared with 92 cats without this disease (the controls). A retrospective review of medical records gleaned results of ionized calcium and other clinical pathological parameters. Cats with the illness were grouped by those who survived, died or were euthanized. Results indicate that cats with AP had significantly lower media plasma ionized calcium (1.06 mmol/l vs. 1.12 mmol/l). Conclusions expressed by the authors were that low ionized calcium concentrations was not only common, but carries a poorer prognosis in cats with AP and a cutoff of ionized calcium less than or equal to 1.00 mmol/l carries a guarded prognosis and justification for aggressive intervention.

 

Another Example

Incidence and prognostic value of low plasma ionized calcium concentration in cats with acute pancreatitis:

46 cases (1996-1998), Kimmel SE, Washabau RJ, Drobatz KJ, J Am Vet Med Assoc 2001 Oct 15;219(8):1105-9

The objective of this study was to determine the incidence and prognostic significance of low plasma ionized calcium in cats with clinical signs of acute pancreatitis (AP). A retrospective, case-control study, 46 cats with acute pancreatitis (the cases) were compared with 92 cats without this disease (the controls). A retrospective review of medical records gleaned results of ionized calcium and other clinical pathological parameters. Cats with the illness were grouped by those who survived, died or were euthanized. Results indicate that cats with AP had significantly lower media plasma ionized calcium (1.06 mmol/l vs. 1.12 mmol/l). Conclusions expressed by the authors were that low ionized calcium concentrations was not only common, but carries a poorer prognosis in cats with AP and a cutoff of ionized calcium less than or equal to 1.00 mmol/l carries a guarded prognosis and justification for aggressive intervention.

 

Representation: Journals
From 2000-2010, case-control studies were reported most often in JAVMA (1403), Journal of Veterinary Internal Medicine (468), Veterinary Surgery (352), Veterinary Record (287), Equine Veterinary Journal (280), Journal of the American Animal Hospital Association (252), Journal of Small Animal Practice (220), Veterinary Pathology (169), Australian Veterinary Journal (152) and Veterinary Radiology and Ultrasound (148), respectively.

 

Representation: Concepts
Top concept representation (MeSH Terms), excluding case-control studies, were retrospective studies, dogs and dog diseases, horses and horse diseases, cats and cat diseases, treatment outcome and risk factors.

Advantages

These retrospective studies are relatively easy to perform, as long as medical record retrieval is not difficult, because they consist largely of a review of medical records (often at a single institution). Collection of patient data is generally easier for incidence cases because there is a greater likelihood that access to medical records are more easily obtained (e.g. electronic medical record systems) as well as better recollection of clinical procedures, interventions, etc. Patients are not enrolled as in a prospective study, nor is a target population surveyed as in a cross-sectional study. Because of this, they can be relatively of short duration and may be performed at a lower cost. They are also better for rare diseases because these patients may be more thoroughly selected from a larger pool of medical record data.

Disadvantages:

As previously mentioned, the retrospective nature of these studies may prohibit the collection of relevant data, especially with prevalence cases. Assigning patients as well as gathering information about both groups (cases and controls) faces the same obstacles as in prospective studies (e.g. cohort and random-controlled trials). There must be a high degree of certainty or a guarantee that the methods used between both groups are equal and that errors in assignment (e.g. avoiding placing patients with occult disease into the control group) for example are avoided.


This series has been loosely organized from a set of lectures given by the author within graduate courses in Biomedical Informatics beginning in 2002. Content is being edited to improve organization, depth, correct inaccuracies as well as updates with new information during Winter/Spring 2012. Feedback is greatly appreciated. © 2011, Stuart Turner.