Glossary terms for 'I'
|Incidence||The proportion of subjects who develop an outcome during the follow-up period; sometimes called incidence proportion or cumulative incidence. For example, the investigators found that pregnant vegetarians had a lower incidence of pre-term delivery than pregnant women who ate meat. |
|Incidence rate||The rate (see below) at which a particular disease or outcome occurs in a group of subjects previously free of that disorder. Usually calculated as the number of new cases of the outcome divided by the person-time (see below) at risk. For example, the incidence rate of myocardial infarction was 35.3 per 1000 person-years in middle-aged men, about twice the rate (17.4 per 1000 person-years) in middle-aged women. See also person-time.|
|Incidence-density sampling||Within a nested case-control study, a technique to sample controls when an important exposure changes with time; thus the exposure needs to be measured at a similar time in both cases and controls. For example, a nested case-control study to determine whether use of antihistamine medications, which varies seasonally, increases the short-term risk of hip fractures (presumably due to an increased risk of falling), used incidence-density sampling of controls, such that a control?s use of an antihistamine was measured during the same month that a hip fracture occurred in a case.|
|Inclusion criteria||A list of attributes required of a potential subjects for a study. For example, the inclusion criteria for the study were ages 18 to 65 years, residence in San Francisco, and no prior history of depression. See also exclusion criteria. |
|Independent variable||See predictor variable.|
|Inference||The process of drawing conclusions about a population based on observations in a sample. For example, because twice as many cases of bladder cancer as controls reported drinking well water (P = 0.02), the investigators made the inference that consumption of well water increases the risk of bladder cancer in the population. |
|Intended sample||The group of subjects the investigator intended to include in a study, as described in the study protocol. For example, the intended sample consisted of women with breast cancer who were seen initially for treatment on a Monday or Thursday at Longview Hospital (the days that the investigator or her research staff were available) and who were within 6 weeks of their original diagnosis, during the period from January 1, 2013 through June 30, 2014. See also accessible population and sample.|
|Intention-to-treat analysis||In a randomized trial, the process of comparing subjects based on the group to which they were randomly assigned, even if this is not the same as the treatment they actually received. This is the most rigorous form of analysis. For example, the investigators performed an intention-to-treat analysis to determine whether random assignment to receive 6 months of psychotherapy improved symptoms of anxiety as compared with random assignment to a control group that received a pamphlet about stress reduction. See also per-protocol analysis. |
|Intervention||In a randomized trial, the active treatment that subjects receive. Often used as an adjective (intervention group). For example, in a randomized trial of psychotherapy for the treatment of anxiety, the intervention consisted of 6 months of weekly one-hour sessions with a licensed psychologist that emphasized cognitive-behavioral approaches. See also the second definition of control. |
Glossary material from Hulley SB et al. Designing Clinical Research, 4th ed. Philadelphia, Lippincott Williams & Wilkins, 2013.