Lesson 7: ART Critique a case-control study

Assessing the risk of AMD with statin use

In this step, we ask to read the following summary of a case-control study that aimed to assess the risk of age-related macular degeneration (AMD) associated with statin use. As you read, critically appraise the study’s methodology and findings.

A case-control study of age-related macular degeneration and use of statins. Smeeth et al. (BJO 2005).

Summary: Age-related macular degeneration (AMD) is the leading cause of blindness in industrialised countries. Studies have suggested that statins may have a protective effect against the disease but these have produced conflicting results. This case-control study used data from the United Kingdom General Practice Research Database to assess the risk of AMD associated with statin use. This was a very large study with 18,007 cases of AMD compared with 86,169 matched controls. The primary outcome was the association between exposure to statins and AMD.

Methods: The General Practice Research Database (GPRD) was set up in 1987 and contains complete prescribing and diagnostic information for over three million people in the UK. Geographically, and in size, the participating practices are representative of practices in England and Wales. The age and sex distribution of the population is similar to the UK population.

The quality and completeness of the information in the database is high and has been validated in a number of independent studies. Anonymised data available include all drug prescriptions, consultations and diagnoses.

For the study, a case was defined as any person aged 50 years or over who had the first diagnosis of AMD (by eye specialists) while registered with a practice participating in the GPRD. The diagnosis was validated for 50 cases and was confirmed for 43 (93.5%) of them. From the information recorded,
the researchers were unable to identify the stage of AMD at diagnosis.

Cases were matched on age within five years, sex and general practice to five controls. Controls had to be registered with the practice on the diagnosis date of the case and were excluded if there was no clinical data
recorded during the observation period.

Statin exposure was coded as ever or never if there was a prescription for any statin recorded in the database. The mean daily dose of statin was coded as low (10 mg or less), moderate (20 mg), or high (30 mg or more), and the type of statin was recorded. Data on potential confounding factors and some other drugs (aspirin, fibrates, HRT) were collected. In addition to the primary outcome, the researchers assessed the daily dose effect, the effect by type of statin, and the total number of prescriptions for statins.

Results and conclusion: 2.1% of cases versus 1.6% of controls had at least one prescription for a statin. Of these, the median period of exposure to statins before the date of diagnosis of the case was 1.2 years versus 1.3 years for cases and controls, respectively. Of all those exposed to statins,
92% were exposed for less than five years. The majority of both the cases and controls were exposed to only one statin, simvastatin.

There was no evidence that the risk varied by dose or type of statin, or duration of use. The authors concluded that “The results of the present study suggest that at the usual doses used in clinical practice, short to medium term exposure to statins is not associated with a decreased risk of AMD.

* Includes coronary artery disease, cerebrovascular disease, peripheral vascular disease and any other diagnosed atherosclerosis.** Excluding statins or aspirin.

*Adjusted for all potential confounders significantly associated with both AMD and statin use (that is, consultation rate (quintiles), smoking, alcohol intake, BMI, atherosclerosis, hyperlipidaemia, heart failure, diabetes, hypertension, cardiovascular drug use (excluding aspirin or statin), fibrate use). People with missing data are included as separate strata.

*Adjusted for all potential confounders significantly associated with both AMD and statin use (that is, consultation rate (quintiles), smoking, alcohol intake, BMI, atherosclerosis, hyperlipidaemia, heart failure, diabetes, hypertension, cardiovascular drug use (excluding aspirin or statin), fibrate use). People with missing data are included as separate strata.
** People who received a prescription for more than one type of statin
.

Discussion: Think about the following questions in relation to this study:

  • Why was this study superior to previous studies?
  • What were the possible biases that could have occurred in the selection of the cases?
  • What advice about statins would you give to a patient with a family history of AMD?