A group of researchers wants to identify factors related to hospital-acquired bacteremia caused by methicillin-resistant Staphylococcus aureus (MRSA). A total of 45 patients were enrolled in the study after MRSA was isolated from a blood sample taken from them on the third or subsequent day after admission. In addition, 90 patients admitted to the hospital over the same period with a length of stay 2 days who did not have bacteremia were randomly selected. The frequency of factors such as insertion of a central line or urinary catheter and surgical site infection were then compared between the 2 groups. Which of the following is the most appropriate null hypothesis for this study?

Question

A group of researchers wants to identify factors related to hospital-acquired bacteremia caused by methicillin-resistant Staphylococcus aureus (MRSA).  A total of 45 patients were enrolled in the study after MRSA was isolated from a blood sample taken from them on the third or subsequent day after admission.  In addition, 90 patients admitted to the hospital over the same period with a length of stay >2 days who did not have bacteremia were randomly selected.  The frequency of factors such as insertion of a central line or urinary catheter and surgical site infection were then compared between the 2 groups.  Which of the following is the most appropriate null hypothesis for this study?

Quizplus · Accepted Answer

A statistical hypothesis is an initial assumption (that may or may not be true) regarding population parameters (eg, mean values in two different groups) or the relationship between variables within a population (eg, association between an exposure and a disease). The null hypothesis (H₀) is a statement of no difference or no association. The alternative hypothesis (H_a) is a statement of difference or association. In this case, patients are initially identified based on their disease status (ie, cases: patients with methicillin-resistant Staphylococcus aureus [MRSA] bacteremia, controls: patients without MRSA bacteremia). The frequency of factors (ie, exposure to insertion of a central line or urinary catheter, surgical site infection) is subsequently compared between cases and controls. Therefore, this is a case-control study; the most appropriate measure of association between an exposure and a disease in case-control studies is the odds ratio (OR), which represents the odds of disease among exposed patients relative to nonexposed patients. The OR may be interpreted as follows: OR = 1.0 (null value) indicates that the odds of exposure among cases are the same as the odds of exposure among controls; therefore, exposure is not associated with the disease. OR ≠ 1.0 indicates that the odds of exposure are lower (ie, OR <1) or higher (ie, OR >1) among cases than among controls; therefore, exposure is associated with the disease. Because H₀ is a statement of no association, the most appropriate H₀ for the study is that the OR = 1 (Choice D). (Choices A, B, E, and F) Measures of association based on risk/incidence (eg, relative risk, hazard ratio) are calculated in cohort studies and experimental designs, in which participants are initially selected based on exposure status (ie, exposed or nonexposed to a risk factor or treatment) and then are followed over time to assess development of disease. However, in case-control studies, cases (by definition) already have the disease. Therefore, case-control studies cannot calculate and compare the risk/incidence of disease between the case and control groups. Instead, case-control studies must compare the odds of exposure (to a risk modifier) between cases and controls. Educational objective: The odds ratio (OR) is a measure of association used in case-control studies. It quantifies the relationship between an exposure and a disease; its null value (ie, null hypothesis value) is always 1 (ie, OR = 1).

A Group of Researchers Wants to Identify Factors Related to Hospital-Acquired