Public Health Agency of Canada
www.publichealth.gc.ca
A Guide to HIV/AIDS Epidemiological and Surveillance Terms
[Previous] [Table of Contents] [Next]
Frequently Used Terms in HIV/AIDS Monitoring and Epidemiological Research in Canada
Person Years
(See also incidence)
Person years describes the length of time of experience or exposure of a group of people who have been observed for varying periods of time. It is the sum total of the length of time each person has been exposed, observed, or at risk.
You will sometimes see person years reported as PY or py.
For example, 100 people studied for one year, or 50 people studied for two years, are both equivalent to 100 person years of observation:
- 100 people × 1 years = 100 person years
- 50 people × 2 years = 100 person years
Person years is often used as the denominator in expressing incidence rate. An example of how the term person years has been used this way can be found under that term.
Person years is also often used as the denominator in expressing incidence density. If some participants in a research study do not stay in the study for the entire length of the study, person years may be used in order to calculate incidence.
For example, let us assume that there were 100 people in a study that continued for five years. Of these, 85 remained until the end of the study, but the other 15 dropped out after only 2 years. Person years would be calculated by:
| + (85 people × 5 years) | = | 425 person years |
| + (15 people × 2 years) | = | 30 person years |
| = 455 person years |
Point Prevalence
(See also prevalence)
The number of people living with HIV divided by a defined population at a specified time (prevalence rate) can be reported as point prevalence. Reporting a prevalence rate as a point prevalence emphazises the "point" in time at which the prevalence rate was calculated.
The "point" can refer to a specific point in calendar time at which the prevalence rate was calculated. This point will be the same for all participants in a research study.
The "point" can also refer to a specific point in a research study that will not necessarily be the same in calendar time for all subjects. For example, the number of people with HIV may be measured at the time (or point) at which each participant first joins a study. This date may be different in calendar time for each participant. However, it will relate to the same specific point in the research study - the point at which each participant joined the study.
As a hypothetical example, let us assume that as inmates enter the Carlson Road Detention Centre, they are interviewed and asked if they have tested positive for HIV. This is the collection of baseline data. There are currently 1,800 inmates at the Detention Centre, and it is recorded that 266 of them said they had previously tested positive for HIV when they were asked the question on the entry interview.
Therefore it can be said that the HIV point prevalence at study entry for 1,800 inmates at the Carlson Road Detention Centre was 14.8%.
Inmates answering yes |
= |
266 |
= |
14.8%(0.1477×100) |
Current number of inmates |
1,800 |
This can also be expressed as,
HIV point prevalence at the time of baseline interview for 1,800 inmates at the Carlson Road Detention Centre was 14.8%.
Population at Risk
The population at risk represents those persons at risk of contracting a disease.
It is often the denominator in a rate, such as incidence or prevalence rates. The size of the population at risk is usually estimated when national or regional rates are calculated. In some research studies of small populations, it may be a number that is exactly known as opposed to one that is estimated.
Positive HIV Test Reports
(See also reported AIDS case and FAQs 7, 8, 10, 11, 13, 14, 18)
A positive HIV test report is documentation of a person's confirmed and reported HIV infection.
These reports include patient information, laboratory data and the activities that put the person at risk for transmission of HIV. The information recorded forms the basis of the surveillance data reported at the provincial and territorial level as well as to the Centre for Infectious Disease Prevention and Control (CIDPC) at the federal level. Information reported to CIDPC does not include names nor does it identify anyone.
The term does not include individuals who may be positive and have not been tested or individuals who have received a positive HIV test result, but the result has not been forwarded to CIDPC.
A positive HIV test report applies to the time of diagnosis of HIV infection, not to the time of infection with HIV. HIV infection may have been much earlier. It is therefore important to refer to positive HIV test reports as new diagnoses of HIV infection that are reported, not as new infections.
Prevalence
(See also point prevelance, prevalence rate and FAQs 18-22)
Prevalence is the total number of people with a specific disease or health condition living in a defined population at a particular time.
HIV prevalence among Canadians is the total number of people living with HIV infection (including those with AIDS) in Canada at a particular time.
Prevalence may be expressed as a number or a rate (see below).
Prevalence is different from incidence; however, the two terms are frequently confused.
Prevalence Rate
(See also point prevelance)
The prevalence rate is the number of existing cases of a disease at a specified time divided by a defined population that is "at risk" of experiencing the condition.
| Prevalence rate = | Number of existing events in a specified period |
| Number of people exposed to risk in this period |
It is very important to know the number of people living with a disease in relation to the total population at risk of contracting that disease. For example, 25 people with HIV infection in a population at risk of 100 people has a different meaning from 25 people with HIV infection in a population at risk of 500 people. To take account of this, prevalence is expressed as a rate.
Using this example, we would calculate prevalence rate as either:
Prevalence rate = |
25 people who currently have HIV |
= 0.25 |
Population of 100 people |
or
Prevalence rate = |
25 people who currently have HIV |
= 0.05 |
Population of 500 people |
In the first calculation, the prevalence rate is 0.25 (× 100 = 25%). This means that in a group of 100 people, we would expect 25% of them or, 25 of 100, to have HIV at the specified time period, such as the year 2000. In the second calculation, we would expect 5% or 5 of 100 people.
As another example, let us calculate the prevalence rate of HIV in a general practice. A physician has 1,000 patients registered with him. Looking at his records at the end of 2001 for an annual audit, he notes that 200 of his patients were living with HIV infection (prevalence) at the time of the audit.
The HIV prevalence rate among the patients registered in his practice at the end of 2001 is the existing number of people with HIV infection at the end of 2001 (the specific point in time) divided by the total number of registered patients (the population at risk).
The HIV prevalence rate among the physician's patients at the end of 2001:
existing number of patients with HIV |
= |
200 |
= |
0.20 (20%) |
population at risk |
1,000 |
Proportion
A proportion is a type of ratio in which the numerator is included in the denominator. It is an expression of a comparative part or share of the total at a specified period of time. In HIV/AIDS research, the proportion can be the number of people with a common characteristic, such as gender, as compared with the total population that shares the same "event", such as HIV infection, at the same specified period of time.
A proportion is calculated by dividing the number of people with a common characteristic at a given time period by the total population that shares the same event in the same time period.
For example, HIV and AIDS Among Women in Canada in the May 2001 HIV/AIDS Epi Updates states that "Women represent an increasing proportion of reported HIV cases in Canada, and accounted for 24% of positive HIV test reports in 2000". (Bureau of HIV/AIDS, STD and TB, Centre for Infectious Disease Prevention and Control, Health Canada, 2001b)
In other words, the proportion of women among positive HIV test reports was 24% in the year 2000.
This was calculated by:
| Positive HIV reports among women in 2000 | = | 484 | = | 0.237 |
| Positive HIV reports among adults in 2000 | 2,039 |
This can be rounded up to 24% (0.237 × 100).
P Value
The symbol p (or P) means probability. It appears frequently in the "Results" section of published research papers or reports.
The symbol p followed by the mathematical symbol < (less than) 0.05 is used to indicate that the result could be expected to occur by chance less often than five times in 100, or once out of 20. Another way of expressing this is to say that the result is "significant at the 5% level". The p value is closely linked to confidence intervals.
For many routine research studies, a 5% level of statistical significance is considered to be good enough. However, if the findings are likely to have very important consequences for medical interventions or public policy, for example, a higher level of statistical significance is demanded, such as p < 0.01. A statistical significance level of less than 0.01 means the result might occur by chance less than one time in a 100. In other words, the smaller the p value, the less likely the results happened by chance.
If there is an asterisk (*) by a p value, it indicates that the result had statistical significance.
[Previous] [Table of Contents] [Next]
To share this page just click on the social network icon of your choice.