11 Attributable risk percent I-I AF where I is the incidence rate(proportion) in the total population and I is the incidence rate(proportion)among the unexposed. The formula AF 1+P(RR-1) (where RR is the rate ratio, I /I and P is the proportion exposed in the entire population) is often cited, but it is biased if the rr is adjusted for confounders, as is normally the case. A formula that does not suffer from this problem P(RR-1) AF RR (where P is the exposure prevalence among cases) Attributable fractions may also be calculated for other measures of disease frequency; e.g the attributable fraction for the caseload over a defined time period is (4-A)yA2 where A, is the caseload in the population and a is what the caseload would be if everyone were not exposed. This is not the same as the quantity computed by rates because the rate fraction does not account for the effect of exposure on person time ATTRIBUTABLE NUMBER The excess caseload of a specific outcome attributable to an exposure over a defined time period. If there is no bias or confounding and the exposure has negligible effect on the person-time at risk, it may be estimated using the formula AN=T(I-I where I is the incidence rate among the exposed. I is the incidence rate among the unexposed and T is the person-time in the exposed population during the period in question. ATTRIBUTABLE PROPORTION See ATTRIBUTABLE FRACTION ATTRIBUTABLE RATE, ATTRIBUTABLE RISK (Syn: causal rate difference, causal risk difference) The proportion of the rate(risk)of a disease or other outcome in exposed individuals that can be attributed to the exposure. This measure is estimated by sub- tracting the rate(risk)of the outcome(usually, incidence or mortality) among the unex- posed from the rate(risk)among the exposed individuals; this estimate assumes that causes other than the one under investigation have had equal effects on the exposed and unexposed groups. Unfortunately, this term has been used to denote a number of different concepts, including the ATTRIBUTABLE FRACTION in the population, the attribut able fraction among the exposed, the POPULATION EXCESS RATE, and the RATE DIFFERENCE. See alSo ABSOlute RISK REDUCTION: IMPACT NUMBERS ATTRIBUTABLE RISK(EXPOSED) This term has been used with different connota ons to denote the attributable fra and the excess risk A/ exposed. See also ATTRIBUTABLE FRACTION(EXPOSED): RATE DIFFERENCE. RIBUTABLE RISK PERCENT Attributable fraction expressed as a percentage of the total rate or risk rather than as a proportion
AF I I I p p u p = − where Ip is the incidence rate (proportion) in the total population and Iu is the incidence rate (proportion) among the unexposed. The formula AF P P RR e e p = − + − ( ) ( ) RR 1 1 1 (where RR is the rate ratio, Ie /Iu and Pe is the proportion exposed in the entire population) is often cited, but it is biased if the RR is adjusted for confounders, as is normally the case. A formula that does not suffer from this problem is AF P RR RR c p = ( ) −1 (where Pc is the exposure prevalence among cases). Attributable fractions may also be calculated for other measures of disease frequency; e.g., the attributable fraction for the caseload over a defi ned time period is (Ap – Au )/Ap where Ap is the caseload in the population and Au is what the caseload would be if everyone were not exposed. This is not the same as the quantity computed by rates because the rate fraction does not account for the effect of exposure on person time. ATTRIBUTABLE NUMBER The excess caseload of a specifi c outcome attributable to an exposure over a defi ned time period. If there is no bias or confounding and the exposure has negligible effect on the person-time at risk, it may be estimated using the formula AN = Te (Ie – Iu ) where Ie is the incidence rate among the exposed, Iu is the incidence rate among the unexposed, and Te is the person-time in the exposed population during the period in question. ATTRIBUTABLE PROPORTION See attributable fraction ATTRIBUTABLE RATE, ATTRIBUTABLE RISK (Syn: causal rate difference, causal risk difference) The proportion of the rate (risk) of a disease or other outcome in exposed individuals that can be attributed to the exposure. This measure is estimated by subtracting the rate (risk) of the outcome (usually, incidence or mortality) among the unexposed from the rate (risk) among the exposed individuals; this estimate assumes that causes other than the one under investigation have had equal effects on the exposed and unexposed groups. Unfortunately, this term has been used to denote a number of different concepts, including the attributable fraction in the population, the attributable fraction among the exposed, the population excess rate, and the rate difference. See also absolute risk reduction; impact numbers. ATTRIBUTABLE RISK (EXPOSED) This term has been used with different connotations to denote the attributable fraction among the exposed and the excess risk among the exposed. See also attributable fraction (exposed); rate difference. ATTRIBUTABLE RISK PERCENT Attributable fraction expressed as a percentage of the total rate or risk rather than as a proportion. 11 Attributable risk percent
Attributable risk percent(exposed ATTRIBUTABLE RISK PERCENT(EXPOSED) The attributable fraction among the exposed, expressed as a percentage of the total rate or risk among the exposed. See also ATTRIBUTABLE FRACTION(EXPOSED) ATTRIBUTABLE RISK PERCENT (POPULATION)The attributable fraction in the population, expressed as a percentage of the total rate or risk in the population. See also ATTRIBUTABLE FRACTION(POPULATION) ATTRIBUTABLE RISK(POPULATION) This term has been used with different conno- tations to denote the attri fraction in the population and the population excess risk. See also ABSOLUTE RISK REDUCTION; ATTRIBUTABLE FRACTION(POPULATION ) POPULA TION EXCESS RATE ATTRIBUTE A qualitative characteristic of an individual or an item ATTRITION Reduction in the number of participants in a study as it progresses (i.e, dur- ing FOLLow-UP of a CoHORT). Losses may be due to withdrawals, DROPOUTS, or protocol deviations. 1 See alSo ceNSORING ATTRITION BIAS A type of SELECTION BIAS due to systematic differences between the study groups in the quantitative and qualitative characteristics of the pro- cesses of loss of their members during study conduct, i.e., due to ATTRITION among subjects in the study. Different rates of losses to follow-up in the exposure groups may change the characteristics of these groups irrespective of the studied AUDIT 1. An examination or review that establishes the extent to which a condition, process, or performance conforms to predetermined standards or criteria. Assessment or review of any aspect of HEALTH CARE to determine its quality, audits may be carried out on the provision of care, compliance with regulations, community response. completeness of records, etc 2. An evaluation of the quality of health care, the use of resources, and outcomes. See alSo HEALTH SERVICES RESEARCH 3. The process of checking whether the accounts of an institution, company, or association are complete, accurate, and consistent; whether they agree with other records of activity; and whether they comply with legal requirements and professional standards. AUSTRALIA ANTIGEN Hepatitis B surface antigen(HBsAg). So called because it wa first identified in an Australian aborigine HBsAg is a BIOMARKER for the prevalence of infection with the virus of hepatitis B AUTONOMY. RESPECT FOR 1. In ETHICS, the principle of respect for human dignity and the right of individuals to decide things for themselves. arch, this principle of INFORMED CONSENT. It can confict with the need to protect the population from identified risks(e. g, risks related to contagious disease)and with the need or access to personally identifiable health-related data and information. See also CONFIDENTIALITY, CONSENT BLAS. PRIVACY AUTOPSY DATA Data derived from autopsied deaths; used, for instance, to study aspect of the natural history of disease or trends in frequency of disease. Autopsies are done on nonrandomly selected persons; findings should therefore be generalized only with great caution See also BIAS IN AUTOPSY SERIES
ATTRIBUTABLE RISK PERCENT (EXPOSED) The attributable fraction among the exposed, expressed as a percentage of the total rate or risk among the exposed. See also attributable fraction (exposed). ATTRIBUTABLE RISK PERCENT (POPULATION) The attributable fraction in the population, expressed as a percentage of the total rate or risk in the population. See also attributable fraction (population). ATTRIBUTABLE RISK (POPULATION) This term has been used with different connotations to denote the attributable fraction in the population and the population excess risk. See also absolute risk reduction; attributable fraction (population); population excess rate. ATTRIBUTE A qualitative characteristic of an individual or an item. ATTRITION Reduction in the number of participants in a study as it progresses (i.e., during follow-up of a cohort). Losses may be due to withdrawals, dropouts, or protocol deviations.31 See also censoring. ATTRITION BIAS A type of selection bias due to systematic differences between the study groups in the quantitative and qualitative characteristics of the processes of loss of their members during study conduct, i.e., due to attrition among subjects in the study. Different rates of losses to follow-up in the exposure groups may change the characteristics of these groups irrespective of the studied intervention.32,33 AUDIT 1. An examination or review that establishes the extent to which a condition, process, or performance conforms to predetermined standards or criteria. Assessment or review of any aspect of health care to determine its quality; audits may be carried out on the provision of care, compliance with regulations, community response, completeness of records, etc. 2. An evaluation of the quality of health care, the use of resources, and outcomes. See also health services research. 3. The process of checking whether the accounts of an institution, company, or association are complete, accurate, and consistent; whether they agree with other records of activity; and whether they comply with legal requirements and professional standards. AUSTRALIA ANTIGEN Hepatitis B surface antigen (HBsAg). So called because it was fi rst identifi ed in an Australian aborigine. HBsAg is a biomarker for the prevalence of infection with the virus of hepatitis B. AUTONOMY, RESPECT FOR 1. In ethics, the principle of respect for human dignity and the right of individuals to decide things for themselves. 2. In epidemiological practice and research, this principle is central to the concept of informed consent. It can confl ict with the need to protect the population from identifi ed risks (e.g., risks related to contagious disease) and with the need for access to personally identifi able health-related data and information. See also confi dentiality; consent bias; privacy. AUTOPSY DATA Data derived from autopsied deaths; used, for instance, to study aspects of the natural history of disease or trends in frequency of disease. Autopsies are done on nonrandomly selected persons; fi ndings should therefore be generalized only with great caution. See also bias in autopsy series. Attributable risk percent (exposed) 12
AUXILIARY HYPOTHESIS BIAS A form of RESCUE BIAS and thus of INTERPRETIVE BIAS which occurs in introducing ad hoc modifications to imply that an unanticipated finding would have occurred otherwise had the experimental conditions been different. Because experimental conditions can easily be altered in many ways, adjusting a hypothesis is a versatile tool for saving a cherished theory. 4 AVERAGE 1. In science, loosely, the ARITHMETIC MEAN. The arithmetic average of a set of n numbers is the sum of the numbers divided by n 2. A measure of location either the mode or in the case of numerical data the median 3. Distribution of aggregate inequalities in a series among all the members of the series so as to equalize them. See also MEASURE OF CENTRAL TENDENCY. 4. In everyday speech, ordinary, usual, or NORMAL; the normal or typical amount. AVERAGE LIFE EXPECTANCY See EXPECTATION OF LIFE AXIS 1. One of the dimensions of a graph. A two-dimensional graph has two axes, the horizontal or x axis and the vertical or y axis. Mathematically, there may be more than two axes, and graphs are sometimes drawn with a third dimension. See also ABSCISSA, ORDINATE. 2. In NOSOLOGY, an axis of classification is the conceptual framework(e. g, etiological, pographical, psychological, sociological). The INTERNATIONAL CLASSIFICATION OF SEASES, for example, is multiaxial: the primary axis is topographical (ie, body systems), while secondary axes relate to etiology, manifestations of disease, detail of sites affected, severity, etc
AUXILIARY HYPOTHESIS BIAS A form of rescue bias and thus of interpretive bias, which occurs in introducing ad hoc modifi cations to imply that an unanticipated fi nding would have occurred otherwise had the experimental conditions been different. Because experimental conditions can easily be altered in many ways, adjusting a hypothesis is a versatile tool for saving a cherished theory.34 AVERAGE 1. In science, loosely, the arithmetic mean. The arithmetic average of a set of n numbers is the sum of the numbers divided by n. 2. A measure of location, either the mode or, in the case of numerical data, the median or the mean. 3. Distribution of aggregate inequalities in a series among all the members of the series, so as to equalize them. See also measure of central tendency. 4. In everyday speech, ordinary, usual, or normal; the normal or typical amount. AVERAGE LIFE EXPECTANCY See expectation of life. AXIS 1. One of the dimensions of a graph. A two-dimensional graph has two axes, the horizontal or x axis and the vertical or y axis. Mathematically, there may be more than two axes, and graphs are sometimes drawn with a third dimension. See also abscissa; ordinate. 2. In nosology, an axis of classifi cation is the conceptual framework (e.g., etiological, topographical, psychological, sociological). The International Classifi cation of Diseases, for example, is multiaxial: the primary axis is topographical (i.e., body systems), while secondary axes relate to etiology, manifestations of disease, detail of sites affected, severity, etc. 13 Axis
B BACKGROUND LEVEL. RATE The concentration. often low, at which some substance agent, or event is present or occurs at a particular time and place in the absence of a spe fic hazard or set of hazards under investigation. An example is the background level of the naturally occurring forms of ionizing radiation to which we are all exposed. ACTERIA(singular: bacterium) Single-celled organisms found throughout nature, which can be beneficial or cause disease BAR CHART(Syn: bar diagram) A graphic technique for presenting DISCRETE DATA organ ized in such a way that each observation can fall into one and only one category of the variable. Frequencies are listed along one axis and categories of the variable along the other axis. The frequencies of each group of observations are represented by the lengths of the corresponding bars. See also HISTOGRAM. BAR DIAGRAM See BAR Chart BARKER HYPOTHESIS See DEVELOPMENTAL ORIGINS HYPOTHESIS BARRIER METHOD Contraceptive method that interposes a physical barrier between erm and ovum(e. g, condom, cervical cap, diaphragm) BARRIER NURSING(Syn: bedside isolation) Nursing care of hospital patients that min mizes the risks of cross-infection by use of antisepsis, gowns, gloves, masks for nursing staff, and isolation of the patient, preferably alone in a single room. See also uNIVERSA BASELINE DATA A set of data collected at the beginning of a study BASE POPULATIO e POPUlATIoN, SoURCE. BASE STUDY See STUDY BASE. BASIC REPRODUCTIVE RATE (R)A measure of the number of infections duced, on average, by an infected individual in the early stages of an epidemic, virtually all contacts are susceptible. Some authors use the symbol z, for BAYESIAN STATISTICS A method of statistical inference that begins with formulation of probabilities of hypotheses(called prior probabilities) before the data under analysis are taken into account. It then uses the data and a model for the data probability(usu ally the same model used by other methods, such as a LOGISTIC MODEL) to update the probabilities of the hypotheses. The resulting updated probabilities are called posterior probabilities. Central to this updating is BAYES'THEOREM, although not all Bayesian methods require explicit use of the theorem and not all uses of the theorem are Baye sian methods. Bayesian statistics can be used alongside or in place of other methods for
B BACKGROUND LEVEL, RATE The concentration, often low, at which some substance, agent, or event is present or occurs at a particular time and place in the absence of a specifi c hazard or set of hazards under investigation. An example is the background level of the naturally occurring forms of ionizing radiation to which we are all exposed. BACTERIA (singular: bacterium) Single-celled organisms found throughout nature, which can be benefi cial or cause disease. BAR CHART (Syn: bar diagram) A graphic technique for presenting discrete data organized in such a way that each observation can fall into one and only one category of the variable. Frequencies are listed along one axis and categories of the variable along the other axis. The frequencies of each group of observations are represented by the lengths of the corresponding bars. See also histogram. BAR DIAGRAM See bar chart. BARKER HYPOTHESIS See developmental origins hypothesis. BARRIER METHOD Contraceptive method that interposes a physical barrier between sperm and ovum (e.g., condom, cervical cap, diaphragm). BARRIER NURSING (Syn: bedside isolation) Nursing care of hospital patients that minimizes the risks of cross-infection by use of antisepsis, gowns, gloves, masks for nursing staff, and isolation of the patient, preferably alone in a single room. See also universal precautions. BASELINE DATA A set of data collected at the beginning of a study. BASE POPULATION See population, source. BASE, STUDY See study base. BASIC REPRODUCTIVE RATE (R0 ) A measure of the number of infections produced, on average, by an infected individual in the early stages of an epidemic, when virtually all contacts are susceptible. (Some authors use the symbol Z0 for basic reproductive rate.) BAYESIAN STATISTICS A method of statistical inference that begins with formulation of probabilities of hypotheses (called prior probabilities) before the data under analysis are taken into account. It then uses the data and a model for the data probability (usually the same model used by other methods, such as a logistic model) to update the probabilities of the hypotheses. The resulting updated probabilities are called posterior probabilities. Central to this updating is Bayes’ theorem, 35 although not all Bayesian methods require explicit use of the theorem and not all uses of the theorem are Bayesian methods. Bayesian statistics can be used alongside or in place of other methods for 15
Bayestheorem Female 16 Bar chart. HIV seroprevalence among patients with sexually transmitted disease by gender, selected Af countries, 1990-1993. From Mann J M, Tarantola D J M, eds. AIDS in the World II. New York: Ox University Press, 1996, P. 47. many purposes(e.g, evaluation of diagnostic tests, studies of disease progression, and analyses of geographic studies, clinical trials, cohort studies, and case-control studies BAYES'THEOREM A theorem of probability named for Thomas Bayes(1702-1761). an English clergyman and mathematician; his Essay Towards Solving a Problem in the Doctrine of Chances(1763, published posthumously) contained this theorem In epide- nology, it is often used to obtain the probability of disease in a group of people with some characteristic on the basis of the overall rate of that disease(the prior probability of disease) and of the likelihoods of that characteristic in healthy and diseased indi- viduals. The most familiar application is in CLINICAL DECISION ANALYSIS, where it is used for estimating the probability of a particular diagnosis given the appearance of some symptoms or test result. A simplified version of the theorem is (DS)= P(SID)P(D) P(SIDPOD)
many purposes (e.g., evaluation of diagnostic tests, studies of disease progression, and analyses of geographic studies, clinical trials, cohort studies, and case-control studies). BAYES’ THEOREM A theorem of probability named for Thomas Bayes (1702–1761), an English clergyman and mathematician; his Essay Towards Solving a Problem in the Doctrine of Chances (1763, published posthumously) contained this theorem. In epidemiology, it is often used to obtain the probability of disease in a group of people with some characteristic on the basis of the overall rate of that disease (the prior probability of disease) and of the likelihoods of that characteristic in healthy and diseased individuals. The most familiar application is in clinical decision analysis, where it is used for estimating the probability of a particular diagnosis given the appearance of some symptoms or test result. A simplifi ed version of the theorem is P DS P SD P D P SD P D P S P ( ) | (| ) ( ) ( | ) ( ) + ( |D) (D) = Bayes’ theorem 16 HIV seroprevalence (percent) Nairobi, Kenya 1992 Kigali, Rwanda 1991 Johannesburg, South Africa (black population) 1994 Kampala, Uganda 1990 Lusaka, Zambia 1991 Brazzaville, Congo 1990 Dar es Salaam, Tanzania 1991 0 10 20 19 15 69 Female 48 26 19 50 43 21 19 69 60 20 16 30 40 50 60 70 80 Male Bar chart. HIV seroprevalence among patients with sexually transmitted disease by gender, selected African countries, 1990–1993. From Mann J M, Tarantola D J M, eds. AIDS in the World II. New York: Oxford University Press, 1996, p. 47