Methodology for the calculation of health expectancies

Transcription

1 Methoology for the calculation of health epectancies 31

2 Metoología para el cálculo e esperanzas e salu 32

3 Methoology for the calculation of health epectancies n theory, the probabilities by age come from the entrance an eit incience rates of each state, in the same way as the calculation of life epectancy probabilities are obtaine from the registere number of eaths. Therefore, probabilities are calculate from the movement observe in a efine perio an provie information on the number of transitions insie an outsie each state. n practice, the construction of this table is inviable as the transition ata between health states is not collecte systematically but rather specific surveys are available (health, isability) which are more or less perioic an which reflect a stock, not a movement. Therefore, the incience of the perio, necessary for calculations has to be estimate base on the available information. 1. Methos of calculation There are three types of metho to carry out this estimate: life table methos base on observe prevalence, life table methos with multiple ecreases, life table methos with ecreases-increases. LFE TABLE METHO BASE ON OBSERVE PREVALENCE. SULLVAN METHO. With respect to health, Saners propose a combine moel of mortality an morbiity in the same life table (Saners, 1964). The ieas was carrie out finally by Sullivan, who built a life table base on the observe prevalence an calculate the first life epectancy free of isability. The Sullivan metho (1971) is toay the one use most often to obtain temporary series an international comparisons on health issues. Construction of the life table The elaboration process of the life table base on observe prevalence is simple. t consists of moifying in the classical life table the function L (the number of years live in the age interval), multiplying it by 1 - the specific prevalence rate by age (t). Now you have (1-t) L which is the number of years live without isability in the age interval. To apply this metho, ata is use from the classical mortality tables an the ata relative to the state j observe in a specific population survey. The specific rates by age t referring to the state of health, isability, etc. are stock ata, in other wors, observe prevalence. Avantages an isavantages of the metho. The main isavantage of the Sullivan metho is the non-observation of transitions between states, but also that they are estimate from observe prevalence. t has been emonstrate that the metho prouce goo estimates when the transition uner stuy is stable over time. With respect to health states, such as isability, changes are sufficiently graual to ensure that observe prevalence is a goo estimator of the changes in a stuie perio. However, the ata must be interprete with caution, especially in estimating, for eample the number of years live with isability, given that we are ealing with a situation of short uration in a state of relatively low incience an therefore the imprecisions of the moel may have greater impact. Up to now the Sullivan metho has been use to calculate health epectancy an other health inicators in at least 49 countries. LFE TABLE METHO WTH MULTPLE ECREASES This moel was evelope initially by actuaries an insurance companies to calculate pensions an inemnities base on average life epectancy before reaching the state of "insurance beneficiary", in other wors, isable, wiow, etc. The moel of multiple ecreases oes not consier eath as the final state on its own but that other life states that may be taken as efinitive are also taken into account. n other 33

4 Methoology for the calculation of health epectancies wors, events that constitute an eit from the life table. n this sense eath an isability are consiere absorbent states. Using a continuous stuy that collects ata in various waves, the number of transitions from the initial state to the absorbent state can be evaluate an in this way the specific survival probabilities in the initial state can be estimate. Applying these probabilities to a hypothetical cohort, the life table relate to these absorbent events can be obtaine. This metho supplies an inicator of the perio that correctly reflects the current health conitions of the population. However, the information that the metho requires is impossible to achieve on a large scale, since it is necessary to carry out population surveys more than once on the same sample. Therefore, its use to calculate internationally comparable health epectancy is inviable. On the other han, the assumption that the isability or health state consiere is absorbent, without possible recovery is not appropriate for health stuies since there eist recoverable isabilities an transitory states of health. inicator will be useful for obtaining the health of the population an for making comparisons between countries, within a country over time or within population subgroups espite the ifferences that may eist in the composition by age. Comparability is even greater if calculations are mae separately for men an women. As this metho has general use, various points shoul be borne in min in orer to facilitate comparison. The same efinitions of state of health must be use The general esign of surveys from which prevalence is obtaine is require to be the same as the estimates of prevalence of states of illness an health are very sensitive to the metho of collection: personal interview, telephone, mail Heath epectancies shoul be calculate on the total population, incluing resients in group establishments, as omission leas to significant bias in elerly populations. t is essential to specify the last age group, the open group, that iffers between surveys an may also affect the comparability of the results. LFE TABLE METHO WTH ECREASES- NCREASES n the 1970s other authors starte to eplore moels capable of maintaining not only eit transitions from the initial state, but also the return to that state. The iea that these moels evelope was to stuy the effect of a specific health program on the population group that the program was irecte at. A life table of increases-ecreases base on a Markov chain has been evelope. 2. Health epectancy with the Sullivan metho. The main reason for calculating a health epectancy is to combine information on mortality an morbiity in one inicator. This Calculation of Life Epectancy Free of isability (LEF) with the Sullivan metho As was commente at the beginning the metho consists of moifying the life table multiplying by 1 - the specific prevalence rate by age (t), by L, the number of years live in the age group. For each age group the life table provies the columns that are require for the calculation of the LEF: l Survivors at age L Number of years live in the age group, +n an the Survey on isabilities, mpairments an State of Health 1999 supplies isability rates by age, t. 34

5 Methoology for the calculation of health epectancies This way, life epectancy an the LEF are obtaine from these functions: n ( L ) l n n ((1 t ) L ) The calculation of LEF is illustrate with a numerical eample. Life Epectancy Free of isability by age calculate with the Sullivan metho. National total, men. L Number of years live in the age group, +n l Survivors at age T Future life years from age L isability rate in the age group, +n (1- t ) L Number of years live without isability in the age group, +n T Future life years without isability from age e Life epectancy at age LEF Life Epectancy Free of isability at age l n There is a singularity in the first age group in the table, 0 to 5 years. This is ue to the isability stuy that has been carrie out in ifferent questionnaires an efinitions for the population of 0 to 5 years an those over 6 years ol. Therefore this age istribution was use when obtaining these rates. Health epectancy provie are calculate inepenently for men an women. The rates use in the calculation of life epectancy for this stuy, obtaine from the information provie by the isabilities survey, are the following: LEFC. Life Epectancy Free of Chronic isease: rate of people with at least one chronic isease. This rate has been obtaine from the survey health moule. LEPH. Life Epectancy in Goo Perceive Health: the state of perceive health has been aske for following WHO recommenations by means of the question'what is your general state of health in your opinion?' with 5 possible answers: very goo, goo, average, ba, an very ba. The rate of people who have an average, ba or very ba state of health has been use, this also following international recommenations. This rate has been obtaine from the survey health moule. 35

6 Methoology for the calculation of health epectancies LEF. Life Epectancy Free of isability: rate of general isability (rate of limitations in the case of chilren 0 to 5 years ol). Life Epectancy Free of Severe isability: rate of persons with some severe or total isability limitation (categories 3 with serious severity an 4 cannot carry out the severity activity variable. Life epectancy free of isabilities that nee assistance: rate of persons who receive some assistance or who o not receive assistance but who nee it Life Epectancy Free of isabilities for Activities of aily Living: rate of persons with some isability in ALs: carrying out changes in boy position; getting up, going to be, moving within the househol, walking without transport, going to the toilet, controlling nees, ressing, eating an rinking, taking care of shopping, meals, cleaning an ironing of clothes an maintenance of the house an well being of members of the family. Chilren from 0 to 5 have not been inclue in the calculation of this inicator. Life Epectancy Free of isabilities of Mobility: Persons 25 years ol an over: rate of persons with a group 5 isability. Getting aroun (5.1. Changes an maintenance of various boy positions, 5.2. Getting up, going to be, remaining staning or seate an 5.3. moving aroun within the house) an Moving aroun outsie the house (7.1. Walking without transport, 7.2. Moving aroun in public transport an 7.3. riving own vehicle). Chilren 0 to 5: rate of chilren with limitation 1. Significant elays to remain seate, staning or starting to walk or ifficulties walking or weakness or rigiity in the legs. Life Epectancy Free of isability in Taking Care of Oneself: rate of people with a group 8 isability. Taking care of oneself (8.1 Go to the toilet alone: wash an take care of appearance, 8.2 Control of nees an use the toilet alone, 8.3 ress, unress, ress up an eat an rink). Chilren 9 to 5 have not been inclue in the calculation. Life Epectancy Free of isabilities to carry out omestic Tasks: rate of people with a group 9 isability. Carrying out househol tasks. (9.1 Taking care of shopping an control of supplies an services, 9.2. Taking care of meals, 9.3. Taking care of cleaning an ironing of clothes, 9.4. Taking care of cleaning an maintenance of the house an 9.5. Taking care of the well being of other members of the family). Chilren 0 to 5 have not been inclue in the calculation of this inicator since group 9 isabilities are only consiere from the age of 10. Life Epectancy Free of isability for Seeing: Persons 25 years ol an over: rate of persons with a group 1 isability. Seeing (1.1. isability for receiving images, 1.2. isability for visual tasks as a whole an 1.3. isability for etaile visual tasks). Chilren 0 to 5: rate of chilren with limitation 2. Significant ifficulties seeing, or 3. Total blinness. Life Epectancy Free of isabilities Hearing: Persons 25 years ol an over: rate of persons with a group 2 isability. Hearing (2.1. isability to receive any soun, 2.2. isability to hear strong souns an 2.3. isability for listening to speech). Chilren 0 to 5: rate of chilren with limitation 4. Significant ifficulties hearing, or 5. Total eafness. Life Epectancy Free of Osteoarticular eficiencies: rate of persons with a group 5 impairment. Osteoarticular impairments (5.1. Hea, 5.2. Vertebral column, 5.3. Superior etremities an 5.4 nferior etremities). The information that is offere on health epectancy calculation methos as well as the methoology use to obtain them have been etracte from «Selection of a Coherent Set of Health nicators». Final raft. A First Step Towars A User's Guie to Health Epectancies for the European Union', J-M Robine, C. Jagger an V. Egii. Montpellier (France), Euro-REVES, June

7 Methoology for the calculation of health epectancies The inicators that are presente in the first five sections provie information on the group of people with isability an on people without isability. The ratio between the inicators of both population groups is also offere, resulting in a more specific iea on the relation eisting between them. The inicators from point si make a reference to the househols in which isable people live an those where they o not live. The formulas use for the calculation of each one of the inicators are etaile below. 1. Percentage of people marrie or living with a partner C C 1 is the number of isable people who are marrie or cohabit with their partner is the total number of isable persons t is similarly efine: 2. Percentage of people who live alone S S 2 is the number of isable persons who live alone is the total number of isable persons t is similarly efine: N NS 2 N 2, RATO 2 N 2 accoring to age group (20-39, 40-59, 60-79, 80 an + an total) an se (male, female an both sees). N NC 1 N an the ratio between the inicator for the 1, RATO 1 N 1 accoring to age group (20-39, 40-59, 60-79, 80 an + an total) an se (male, female an both sees). 3. Stanarise literacy an illiteracy percentages 3.1. Stanarise illiterates percentage by physical an psychological reasons AnF, 3,1 w AnF, group (10-17, 18-24, 25-34, 35-44, 45-54, 55-64, 65-74, 75 an +) who are illiterate for physical or psychological reasons. 37

8 Methoology for the calculation of health epectancies w group is the weight of age group in the Spanish general population 10 years an over t is similarly efine: n N AnF, 3,1 w n 3,1, RATO 3,1 N 3,1 accoring to age group (10-44, 45-64, 65-79, 80 an + an total) an se (male, female an both sees) Stanarise percentage of illiterates by other reasons AnO, 3,2 w AnO, is the number of isable people in age group (10-17, 18-24, 25-34, 35-44, 45-54, 55-64, 65-74, 75 an +) who are illiterate for other reasons. w group is the weight of age group in the Spanish general population 10 years an over t is similarly efine: n N AnO, 3,2 w n 3,2, RATO 3,2 N 3,2 accoring to age group (10-44, 45-64, 65-79, 80 an + an total) an se (male, female an both sees) Stanarise literacy percentage Alf, 3,3 w Alf, group (10-17, 18-24, 25-34, 35-44, 45-54, 55-64, 65-74, 75 an +) who know how to rea an write w group is the weight of age group in the Spanish general population 10 years an over t is similarly efine: n N Alf, 3,3 w n 3,3, RATO 3,3 N 3,3 accoring to age group (10-44, 45-64, 65-79, 80 an + an total) an se (male, female an both sees). 38

9 Methoology for the calculation of health epectancies 4. Stanarise percentage of people with seconary or higher stuies For the calculation the following levels have been consiere as seconary or higher stuies: secon year general seconary eucation, higher professional eucation an university stuies or equivalent. Sec, 4 w Sec, group (18-24, 25-34, 35-44, 45-54, 55-64, 65-74, 75-74, 75 an +) who have finishe seconary or higher stuies. w group is the weight of age group in the Spanish general population 18 years an over t is similarly efine: n N Sec, 4 w n 4, RATO 4 N 4 accoring to age group (18-44, 45-64, 65-79, 80 an + an total) an se (male, female an both sees). 5. Stanar employment an unemployment rates of the population from 16 to 64 years ol 5.1. Stanarise employment rate E, w E, 5,1 w is the number of people with isability in age group (16-24, 25-34, 35-44, 45-54, 55-64) who have a job (working or with job but temporarily absent) group is the weight of age group in the Spanish general population 16 to 64 years ol t is similarly efine: n N E, 5,1 w n 5,1, RATO 5,1 N 5,1 accoring to age group (16-24, 25-44, 45-64, 80 an + an total) an se (male, female an both sees) Stanarise unemployment rate. P, P, 5,2 wa, A, is the number of people with isability in age group (16-24, 25-34, 35-44, 45-39

10 Methoology for the calculation of health epectancies A, w 54, 55-64) who are unemploye (unemploye who are looking for their first job or unemploye who have worke before). is the number of people with isability in age group who are active (with job or unemploye) is the weight of age group in the active Spanish general population (with job or unemploye) 16 to 64 years ol t is similarly efine: n N P, 5,2 wa, n A, 5,2, RATO 5,2 N 5,2 accoring to age group (16-24, 25-44, 45-64, 80 an + an total) an se (male, female an both sees). NSn is the number of househols with number of isable persons equal to n. is the total number of househols accoring to the size of the househol (1, 2, 3, 4, 5, 6 or more members an Total) istribution of househols accoring to the size of the househol 6 inicators are obtaine, one for each of the variables Siz, which is the size of the househol an takes the values 1, 2, 3, 4, 5 an 6 or more members: Siz1, Siz2,, Siz6+. Sizt Siz t Siz t is the number of househols with size equal to t. is the total number of househols accoring to the size of the househol (1, 2, 3, 4, 5, 6 or more members an Total). 6. istribution of househols 6.1. istribution of househols accoring to number of isable persons who live in them 4 inicators are obtaine, one for each variable value NS0, which is the number of isable people who live in the househol an takes values 2, 3, 0 an 1 or more: NS0, NS1, NS2, NS3+. NS n NS n 6.3. istribution of househols with people 65 an over accoring to number of isable people 65 an over who live in them 4 inicators are obtaine, one for each variable value NS65, which is the number of isable people 65 an over who live in the househol an takes values 0, 1, 2 an 3 or more: H65 NS0, H65 NS1, H65 NS2, H65 NS3+. H H NS 65 n NS 65 n H 65 H65 NS65n is the number of househols with persons 65 an over where there live a number of isable people 65 an over equal ton. 40

11 Methoology for the calculation of health epectancies H65 is the total number of househols with persons 65 an over accoring to the size of the househol (1, 2, 3, 4, 5, 6 or more members an Total) istribution of househols with people 65 an over accoring to size of househol 6 inicators are obtaine, one for each of the variables Siz, which is the size of the househol an takes the values 1, 2, 3, 4, 5 an 6 or more members: H65 Siz1, H65 Siz2,., H65 Siz6+. H65 H65 Siz t Siz t H65 H65 Sizt is the number of househols with persons 65 an over with size equal to t. H65 is the total number of househols with people 65 an over accoring to the number of isable persons 65 an over who are in the househol (0, 1, 2, 3 or more isable an Total). 41