Glossary

The number of new cancer cases diagnosed during a specified period of time in a given population.

Number of deaths from cancer during a specific period of time in a given population.

Number of (incident cases / deaths) during a given period of time divided by the number of years in this period.

NNumber of (incident cases / deaths) of a type of cancer during the study period divided by the total cancers / total deaths from cancer in the same period of time in the study population. It is expressed as a percentage.

It is calculated by dividing the total number of (new cancers / deaths) occurring in a certain specific population during a specific time period the total number of person-years of observation and multiplying the result, usually, 100.000. This type of rate is affected by the age structure of populations and is not appropriate for comparison, although it does reflect the true value of the rate of cancer.

$$\text{TB} = \Bigg(\frac{\text{Number of} \enspace {(\frac{\text{incident cases}}{\text{deaths}}) \enspace \text{in the period}}}{\text{number of people} - \text {year at risk}} \Bigg) * 100000$$

It is the rate of (incidence / mortality) for a specific age group. It is calculated by dividing the number of (cases / deaths) that occur in an age group by the corresponding person- years of observation and multiplying, generally, by 10.000.

Age-adjusted rates are used to compare rates of (incidence / mortality) between different populations that differ from each other in their demographic structure or from the same population in different periods of time (temporal evolution) due to the close relationship between the appearance of the cancer and age and different levels of aging of different populations. To avoid this influence of age, the rates adjusted by the direct method are calculated, normally taking the world standard population or the European standard population as a reference.
The adjusted rate is a summary measure of the rate that a given population would have if it had the same age structure as the population considered standard. It is generally expressed as the number of (cancers / deaths) per 100,000 people per year.

$$\text{AT} = \frac{\textstyle\sum_{i=1}^{18}{wi}*{\text{TEE}}_i}{\textstyle\sum_{i=1}^{18}{wi}_i}*\frac{N}{N-SE}$$ Where:

• w are the weights of each age group in the standard population (world or European).
• N is the number of (incident cases / deaths) in the period.
• SE is the number of (incident cases / deaths) with no known age in the period

Depending on the populations that we want to compare, we can use the following types of standard populations:

• World standard population: When we want to compare ourselves with countries around the world.
• European standard population: We use it when we compare ourselves with European countries, since our age structure is more similar. In Europe, that of Waterhouse et al., From 1976, continues in force. Eurostat has been proposing a new population for the calculation of standardization since 2013 to more accurately reproduce demographic change in Europe
• Others: When you want to compare between populations of territorial areas is smaller, sometimes other pyramids are used. For example, if data only from Spanish cancer registries are used, a pyramid of the entire Spanish population can be used. The disadvantage of using these pyramids is that the results cannot be compared with results from registries in other countries.

Standard populations
Age	World P. 1960	European 1976	European 2013	European 2013 *	Age
<1	2400	1600	1000	1000	<1
1-4	9600	6400	4000	4000	1-4
5-9	10000	7000	5500	5500	5-9
10-14	9000	7000	5500	5500	10-14
15-19	9000	7000	5500	5500	15-19
20-24	8000	7000	6000	6000	20-24
25-29	8000	7000	6000	6000	25-29
30-34	6000	7000	6500	6500	30-34
35-39	6000	7000	7000	7000	35-39
40-44	6000	7000	7000	7000	40-44
45-49	6000	7000	7000	7000	45-49
50-54	5000	7000	7000	7000	50-54
55-59	4000	6000	6500	6500	55-59
60-64	4000	5000	6000	6000	60-64
65-69	3000	4000	5500	5500	65-69
70-74	2000	3000	5000	5000	70-74
75-79	1000	2000	4000	4000	75-79
80-84	500	1000	2500	2500	80-84
>=85	500	1000	2500	1500	85-89
				800	90-94
				200	>=95

It is a type of adjusted rate that only takes into account the age groups between 35 and 64 years. Its interest lies in the fact that it is considered that the cases of the higher age groups are more difficult to register completely and that in the lower age groups, the probability of the appearance of cancer is very low and take them into account in the calculation of global rates distorts the results to some extent. It is generally expressed as the number of cancers per 100,000 person-years.

It is the sum of the age-specific rates, up to a limit, usually 75 years.

The risk or cumulative probability of (developing / dying from) cancer is calculated by adding the number of people who (develop / die from) cancer up to a given age group and dividing by the sum of people who were at risk. It indicates the probability that an individual has to develop / die from cancer over a period of his life, in the absence of another cause of death. It is usually calculated for two ranges: from 0 to 64 years and from 0 to 75 years.

$$\text{Risk} = \Bigg(1-exp\bigg(\frac{\text{TAC}}{100}\bigg)\Bigg) * 100$$

Survival analysis consists of estimating the probability that a patient diagnosed with cancer will survive more than a certain time. When applied to a series of patients, the proportion of these who survive more than a given time is estimated. Survival rates are the most direct indicators of the severity of cancer and the impact of treatment. Population survival is measured in cancer registries, which is generally lower than in case series or clinical trials, since these do not include patients who do not have adequate treatment or are not eligible for clinical trials.

It is the probability that a patient diagnosed with cancer will survive all causes of death during a specific time interval. Observed survival does not consider cause of death, just look at who is alive and who is not. It is sometimes known as general survival.

This survival is calculated from the cohort of cancer cases under study using the actuarial method or the Kaplan- Meier method.

Two types of mortality intervene in the group of cancer patients, a mortality derived from the cancer that we study and another unrelated to this cancer.

Net survival is survival in the absence of causes of death other than cancer. Represents a hypothetical situation where cancer is the only attributable cause of death. We can estimate net survival using cause-specific survival, relative survival, or the Pohar-Perme estimator.

It is an estimate of net survival that is calculated using the cause of death indicated on death certificates to estimate the proportion of deaths from cancer.

Specific survival consists of calculating observed survival excluding deaths from causes other than cancer. The problem of calculating the survival and s that often cannot determine the cause of death for either the cause of death is unknown, or the quality of death certificates is low or cannot know whether the tumour has contributed or not to the death of the patient.

It is an estimate of net survival that is defined as the ratio between observed survival that ignores the cause of death and expected survival in a group of people with the same initial age and sex distribution, but without the specific disease we are studying. . Depending on the method for calculating the estimate of the expected survival rate in the group of patients, we will have the relative survival by the Ederer I method, the Ederer II method or the Hakulinen method. The relative survival rate is interpreted as the proportion of patients who survive after a certain follow-up time, in the hypothetical situation where the cancer in question is the only possible cause of death and measures the excess mortality to which a diagnosis is associated of cancer.

The Pohar-Perme estimator is a weighted version of the Ederer II estimator that uses the relative survival framework, but does not estimate relative survival.

The Pohar-Perme estimator , unlike relative survival, presents an unbiased estimator of net survival. For five-year survival, the Pohar-Perme estimates are similar to the estimates by the Ederer I and II and Hakulinen methods, with differences at higher times.