Active life expectancy of older people in Mexico

Disability and Rehabilitation, 2005; 27(5): 213 – 219 Active life expectancy of older people in Mexico SANDRA REYES-BEAMAN1, CAROL JAGGER2, CARMEN GARCIA-PEN˜ A3, 3 4 5 ˜ OZ , PETER E. BEAMAN , BRUCE STAFFORD , & ONOFRE MUN NATIONAL GROUP OF RESEARCH ON AGEING3 1 Bedford Primary Care Trust, NHS, UK, 2Department of Health Sciences, University of Leicester, UK, 3Mexican Institute of Social Security, Mexico, 4Social Sciences Department, Loughborough University, UK, 5Centre for Research in Social Policy, Loughborough University, UK (Accepted June 2004) Abstract Background: Social and economic development together with demographic changes and health interventions have resulted in an increase in life expectancy and a rapidly ageing population in Mexico. Whether people will live longer active and independent lives is still, however, unknown. We will address this question, providing the first estimates of active life expectancy by age, sex and local regional area in Mexico. Methods: Active life expectancy was calculated using the Sullivan method with abridged life tables. Information on the older Mexican population covered by the Mexican Institute of Social Security (IMSS) and the number of deaths for the same group in the year 2000 was obtained from the Office for Health Statistics and Information at IMSS in Mexico. Information on ability to perform basic activities of daily living was obtained from the National Survey on Ageing carried out in IMSS during 1998 – 99. Results: For males and females combined, active life expectancy decreased from 26.9 years at 60 years to 5.7 years at 85 years. Women’s life expectancy exceeded that of men but women lived more years dependent. Similarly, older people in geographical areas with longer life expectancy spent a lower proportion of remaining life active. Conclusion: The success in increasing life expectancy above average in some groups of older people covered by IMSS has been accompanied by increments in the proportion of remaining years dependent upon others for help in basic self-care activities. How risk factors and protective factors interact is Introduction still unknown to a large extent. However, for a Population ageing has become a recent cause for number of the most common problems such as concern in developing countries. Although it is diabetes mellitus and cardiovascular diseases, redu- expected they will have time (the so-called ‘demo- cing risk factors (smoking, inactivity, obesity, etc.) graphic bonus’) [1] to plan and prepare for future and promoting healthy life styles have contributed to needs, developing countries are ageing more rapidly delaying the onset of disease and avoiding premature than developed countries did and the resources to death [4]. face present and future challenges are scarce [2]. Delaying the onset and reducing the incidence, As people age the likelihood of becoming chroni- duration and mortality of health problems has cally ill and disabled increases, not only as a result of contributed, to increase population life expectancy. exposure to current risk factors, but also due to However, there is ongoing debate as to genetic factors and the cumulative effects of exposure whether there is a natural limit to the life span to environmental factors earlier in life [3]. The [5, 6] or whether longer life will be accompanied ageing process begins soon after birth but the effect by extra years in good (compression of morbidity) of cumulative factors is mostly observed as people [7 – 9] or poor health (expansion of morbidity) grow older; in consequence older people suffer more [10, 11] with conflicting evidence from different acute and chronic health problems [2]. countries [12]. Correspondence: Sandra Reyes-Beaman, Bedford Primary Care Trust, NHS, UK. Email: [email protected] ISSN 0963-8288 print/ISSN 1464-5165 online # 2005 Taylor & Francis Group Ltd DOI: 10.1080/09638280400006424 214 S. Reyes et al. The concept of health expectancy has been Active life expectancy (ALE) and 95% confidence developed to help to address these issues through a intervals were calculated using Sullivan’s meth- single indicator that combines information on a od[24] with abridged life tables. Basic activities of population’s mortality and morbidity [13 – 15]. daily living (Katz) [25] were used to define indepen- Active life expectancy is one health expectancy dence; those that required help for at least one of the indicator, combining information on mortality and basic activities were classified as inactive or depen- independence in basic activities of daily living in dent [26]. order to evaluate the expected number of years lived The age and sex-specific prevalence of depen- active or independently. dency in activities of daily living was obtained from The majority of health expectancy estimates have the National Survey on Ageing carried out in been for developed countries, firstly using prevalence 1998 – 99 by IMSS under its strategic projects data and the Sullivan method [14] and recently by programme. The original sample included 21 102 more complex methods with longitudinal data [16 – older subjects distributed proportionally in each 19]. Health expectancy calculations for developing ‘delegacio´n’, with a minimum sample of 500 countries [20, 21] are still relatively uncommon due subjects in the smallest. However, only 17 758 cases to the lack of reliable information on nationally with complete information on basic activities of representative samples of population, life tables or daily living were included in this analysis. Samples demographic data (population and number of deaths in Tabasco and Southern Veracruz were not by age group). When information is available, older collected, and the minimum sample size was not people are not always sufficiently represented in reached in Nuevo Leon, so estimates are not national surveys and thus estimates may be unrepre- available for these ‘delegaciones’. sentative or lack precision [22]. This study aims to In the survey subjects were selected randomly provide evidence on whether the additional years from those insured (not only users) at IMSS using a gained by the increase in life expectancy in certain multi-stage sampling procedure. At the first stage groups of older people has meant further years being two family medicine units were selected randomly in dependent upon others. Additionally it will provide each ‘delegacio´n’. Two consulting rooms were then the first estimates of active life expectancy at a selected randomly in each family medicine unit. regional local level in Mexico using a representative Finally, households located in the geographical area national sample of older people from the Mexican covered by those consulting rooms were visited door Institute of Social Security (IMSS). by door to find people aged 60 years or over covered by IMSS. Once a subject was identified the objectives of the Methodology study were explained as well as the need for a mental To estimate life expectancy by age, sex and ‘delega- status evaluation prior to the formal consent to be cio´n’,{ the number of deaths among IMSS insurers interviewed. The initial evaluation of the mental and the Mexican older population (60 years or over) status was made using a revised validated version of covered by IMSS were obtained from the Office for the MMSE [27]. A subject was considered capable of Health Statistics and Information at IMSS [23]. giving consent and undertaking the interview if they Delegaciones were ranked by life expectancy into obtained a score of 24 or more (out of 30). Below three groups for men and women separately, with this a proxy respondent was approached for consent Group 1 (longest life expectancy), Group 3 (shortest and was then interviewed about the subject. life expectancy), and Group 2 (the remainder) Older people (or proxies) were interviewed face to (table 1). face by trained older retirees (doctors, nurses, social workers) using a structured questionnaire. The quality of the information was guaranteed at different Table I. Cut-off points of life expectancy at 60 years old for men levels: by training the interviewers, through direct and women to define Groups supervision during the fieldwork, repeated interviews Group Men Women to evaluate consistency, and the critical revision of each questionnaire by trained reviewers. Data were 1 (Longest LE) 527.5 527 automatically captured through optical character 2 (Middle LE) 526.5 and 527.5 526 and 527 reading and were then analysed using SPSS. 3 (Shortest LE) Under 26.5 Under 26 Frequencies of dependency in basic activities of daily living by age, sex, delegacio´n and group were {‘Delegacio´n’ refers to the geographical area of Mexico in which the country transferred to the life tables in Excel for the final is divided to organize and deliver local services for those insured at IMSS. calculations. In most cases this is equivalent to states though exceptions are: Mexico State represented by two delegaciones; the Federal District by four Life expectancy, active life expectancy and the delegaciones, and Veracruz by two delegaciones. proportion of remaining years spent active were  Active life expectancy of older people in Mexico 215 compared by age, sex, and Group (defined by life When areas were grouped according to the level of expectancy). life expectancy, for men and women separately, those under 80 years of age in Group 2 (middle life expectancy) and Group 3 (shortest life expectancy) Results spent a similar between them, but a greater propor- tion of remaining life active than men in Group 1 General characteristics of the sampled population (longest life expectancy) (figures 3 and 4 and table Just over half of the sample (56.4%) were women, 2). Differences by sex reported earlier were also around half were aged 60 – 69 years old (57.9%) and observed in each Group. 9.7% were aged 80 or over. Almost a quarter (23.8%) of subjects did not have any formal Discussion education. The need for help in at least one basic activity of daily living was reported by 5.4% of older Active life expectancy, the remaining number of insurers, these then being classified as inactive or years at a particular age that an individual lives dependent. without dependency in activities of daily living, is a comprehensive indicator that summarises the mortality and disability experience of a real Total and active life expectancy population [15, 24]. This indicator is useful For the total population sampled, life expectancy because it can improve the understanding of a decreased from 26.9 years at age 60 to 5.7 years at population’s needs and highlight health inequal- age 85. Life expectancy for women followed the ities [28, 29]. However, as many other indicators, same trend as that for men, but was always higher at it depends crucially on the quality of mortality all ages (figure 1 and table 2). and morbidity data, being affected by under- Active life expectancy for older people insured by estimation of the number of deaths, lack of IMSS decreased from 24.5 years at age 60 years to consensus about definitions of morbidity, and the 4.3 years at age 85. Hence 90% of remaining years at non-representative information on activities of 60 years will be active or independent, compared to daily living [30]. 75% of remaining years at age 85. There were no In Mexico, the quality of such information is still significant differences in years of life spent active improving. It is well known that the most deprived between men and women, implying that the excess areas need to make further progress in guarantee- total years women live over men are spent inactive. ing the quality of their data and this factor could In addition, the proportion of remaining life spent have affected our results. However, integrating active decreased faster with age in women, so estates within groups defined by life expectancy widening the gap between men and women of the has showed that the findings were consistent same age (a gap of 2% at age 60 compared to 9% at regardless of the level of deprivation of individual age 85) (figure 2). ‘delegaciones’. Figure 1. Life expectancy by age and sex, IMSS population. 216 S. Reyes et al. Table II. Life expectancy (LE), active life expectancy (ALE) and 95% confidence interval for ALE by group and sex Men LE ALE 95% CI UL 95% CI LL %ALE/LE Group 1 (Longest LE) 60 years 28.20 25.57 26.06 25.08 90.7 65 years 23.72 21.20 21.70 20.71 89.4 70 years 19.26 16.94 17.44 16.44 88.0 75 years 14.81 12.60 13.11 12.09 85.1 80 years 10.27 8.42 8.93 7.91 82.0 85 years 5.65 4.41 4.84 3.98 78.1 Group 2 (Middle LE) 60 years 26.65 24.58 24.98 24.18 92.2 65 years 22.44 20.45 20.87 20.04 91.2 70 years 18.29 16.40 16.83 15.98 89.7 75 years 14.17 12.42 12.86 11.98 87.7 80 years 9.98 8.47 8.92 8.01 84.8 85 years 5.65 4.49 4.91 4.08 79.5 Group 3 (Shortest LE) 60 years 25.73 23.74 24.14 23.35 92.3 65 years 21.67 19.70 20.11 19.29 90.9 70 years 17.69 15.86 16.28 15.44 89.6 75 years 13.75 12.05 12.50 11.61 87.7 80 years 9.81 8.29 8.76 7.83 84.6 85 years 5.65 4.65 5.06 4.24 82.2 National IMSS 60 years 26.47 24.44 24.65 24.22 92.3 65 years 22.24 20.28 20.50 20.06 91.2 70 years 18.06 16.24 16.47 16.02 89.9 75 years 13.90 12.21 12.45 11.98 87.9 80 years 9.68 8.21 8.45 7.98 84.9 85 years 5.28 4.23 4.45 4.01 80.0 Women LE ALE 95% CI UL 95% CI LL % Group 1 (Longest LE) 60 years 27.76 24.50 24.94 24.06 88.2 65 years 23.23 20.04 20.48 19.59 86.2 70 years 18.76 15.62 16.07 15.16 83.3 75 years 14.31 11.34 11.80 10.88 79.2 80 years 9.87 7.17 7.64 6.70 72.7 85 years 5.28 3.48 3.86 3.11 65.9 Group 2 (Middle LE) 60 years 26.54 23.98 24.46 23.50 90.4 65 years 22.21 19.66 20.15 19.18 88.6 70 years 17.97 15.44 15.94 14.93 85.9 75 years 13.78 11.27 11.80 10.74 81.8 80 years 9.57 7.35 7.89 6.80 76.8 85 years 5.28 3.82 4.26 3.38 72.3 Group 3 (Shortest LE) 60 years 25.34 22.89 23.33 22.46 90.3 65 years 21.21 18.81 19.26 18.36 88.7 70 years 17.20 14.85 15.32 14.38 86.4 75 years 13.26 10.95 11.45 10.46 82.6 80 years 9.30 7.34 7.85 6.82 78.9 85 years 5.28 3.84 4.31 3.38 72.7 National IMSS 60 years 27.32 24.49 24.75 24.23 89.7 65 years 22.98 20.18 20.44 19.92 87.8 70 years 18.74 15.98 16.25 15.71 85.3 75 years 14.54 11.87 12.15 11.59 81.7 80 years 10.34 7.93 8.22 7.64 76.7 85 years 6.02 4.28 4.55 4.02 71.1  Active life expectancy of older people in Mexico 217 Figure 2. Proportion of remaining life spent active by age and sex. Figure 3. Proportion of remaining life spent active for men by age and group. Figure 4. Proportion of remaining life spent active for women by age and group. 218 S. Reyes et al. This is the first time that indicators have been Acknowledgements published for Mexico at a local regional level, and in particular it is the first time that calculations for older This project was made possible by the long term people covered by the IMSS are presented. Estimat- funding provided by the National Council of Re- ing the number of years that older people would be search and Technology in Mexico (L0039-M9608) inactive or would require help to satisfy their basic and the Mexican Institute of Social Security. self-care needs, will be useful in predicting future care needs in the IMSS population where the ageing process is more accelerated than the general Mexican References population [2]. 1. Guimaraes R, Acun˜a G, Caravias J, Tudela F, Provencio E, The greater survival of women over men has often Samaniego JL. The sustainability of development in Latin been reported, and the IMSS population in Mexico America and the Caribbean: Challenges and opportunities. is no different. However, it is important to highlight Regional preparatory conference of Latin America and the here that although older women covered by IMSS Caribbean for the World Conference on Sustainable Devel- opment. LC/G.2145 (CONF. 90/3). October 2001. http:// live longer, the extra years gained are spent www.rolac.unep.mx/foroalc/brasil2001/finales/CRPSusti.pdf. dependent in basic activities of daily living. Years Accessed on 2nd September 2002. of active life were not significantly different between 2. Reyes S. Population Ageing in the Mexican Institute of Social men and women at any age but the proportion of Security: Health Policy and Economic Implications. IMSS- remaining years spent active decreased with age FunSalud Publication. Grupo Edicio´n. Me´xico, 2001. pp. 1 – 298. ISBN. 968-5018-53-7. more rapidly among women, thus increasing the 3. WHO. A Life Course Perspective of Maintaining Indepen- gender gap. dence in Older Age. WHO/HSC/AHE/99.2. Geneva: WHO. We chose to group the ‘delegaciones’ on the basis 1999. pp. 1 – 19. of life expectancy so as to be able to compare active 4. Yu S, Yarnell JWG, Sweetnam PM, Murray L. What level of physical activity protects against premature cardiovascular life expectancy controlled by level of the life death? The Caerphilly study. Heart 2003;89:502 – 506. expectancy. We found that the longer the life 5. Olshansky SJ. Mortality crossovers and selective survival in expectancy within a group, the greater proportion human and nonhuman populations. New developments in of remaining life dependent in ADLs. This is a major mortality. Gerontologist 1995;35:583 – 587. contribution to the knowledge about the ageing 6. Olshansky SJ, Carnes BA, Cassel C. The future of long life. process in Mexico and in particular in the IMSS. Science 1998;281:1612 – 1613. 7. Fries JF. Aging, natural death, and the compression of The success of social, economic, technological, and morbidity. N Engl J Med 1980;303:130 – 135. medical development appears to be extending the 8. Kalache A, Aboderin I, Hoskins I. Compression of morbidity period of life in which older people live dependent and active ageing: key priorities for public health policy in the upon others for solving their basic needs. Whether 21st century. Bulletin of the World Health Organisation this extension of the period of dependency is a 2002;80:243 – 244. 9. Hubert HB, Bloch DA, Oehlert JW, Fries JF. Lifestyle habits common pattern to developing countries has yet to and compression of morbidity. Journal of Gerontology: be shown. Medical Sciences 2002;57:347 – 351. Although IMSS has been the historical example 10. Gruenberg EM. The failure of success. Milbank Memorial for the Mexican health system in promoting health, Fund Quart 1977;55:3 – 24. 11. Verbrugge LM. Longer life but worsening health? Trends in preventing disease and reducing damage, and its health and mortality of middle-aged and older persons. population has enjoyed a life expectancy above the Milbank Q 1984;62:475 – 519. Mexican average, our findings suggest that much 12. Robine JM, Jagger J, Mathers CD, Crimmins E, Suzman R. needs to be done to promote healthy and active Determining Health Expectancies (pp. 1 – 444). Chichester, ageing. The infrastructure to promote healthier life UK: Wiley; 2002. styles and the medical facilities and trained staff to 13. Sanders BS. Measuring community health levels. Am J Public Health 1964;54:1063 – 1070. meet other health needs exist. The life course 14. Sullivan DF. A single index of mortality and morbidity. perspective [31] could be a useful approach in HSMHA Health Reports 1971;86:347 – 354. promoting a healthier life-style in early stages in 15. Molla MT, Wagener DK, Madans JH. Summary measures order to have an impact in later life. In the meantime, of population health: methods for calculating healthy measures to prevent or delay the onset of chronic life expectancy. Healthy People 2010 Stat Notes 2001;21:1 – 11. diseases and disability among older people in Mexico 16. Crimmins E, Saito Y, Hayward MD. Sullivan and multistate are urgently needed [2]. methods of estimating active life expectancy: two methods, Calculations using other health indicators and for two answers. In: JM Robine, CD Mathers, MR Bone, I the entire Mexican population as well as using Romieu (eds) Calculation of Health Expectancies; Harmoni- longitudinal data are needed to confirm in the zation, Consensus Achieved and Future Perspectives. Montrouge: John Libbey Eurotext, 1993;155 – 160. future these initial findings for the IMSS older 17. Crimmins EM, Hayward MD, Saito Y. Changing mortality population, and to monitor the effect of ongoing and morbidity rates and the health status and life expectancy interventions. of the older population. Demography 1994;31:159 – 175.  Active life expectancy of older people in Mexico 219 18. Crimmins EM, Saito Y, Ingegneri D. Trends in disability-free 26. Singer BH, Manton KG. The effect of changes on projections life expectancy in the United States, 1970 – 90. Popul Dev Rev of health services needs for the old population in the United 1997;23:555 – 572. States. Proceedings of the National Academy of Sciences of 19. Laditka SB, Wolf DA. New methods for analysing active life the United States of America 1998;95:15618 – 15622. expectancy. J Aging Health 1998;10:214 – 241. 27. Reyes-Frausto S, Beaman P, Jagger C, Garcia-Pen˜a C, Villa 20. Saito Y, Qiao X, Jitapunkul S. Health Expectancy in Asian MA, Heres J, Co´rdova A. Validation of a modified version Countries. In: JM Robine, J Jagger, CD Mathers, E of the Mini-Mental State Examination (MMSE) in Crimmins, R Suzman (eds) Determining Health Expectan- Spanish. Aging, Neuropsychology, and Cognition cies. (pp. 1 – 444). Chichester, UK: Wiley; 2002. 2004;11:1 – 11. 21. Mathers CD, Murray C, Lopez AD, Salomon JA, Sadana R. 28. Hubert HB, Bloch DA, Oehlert JW, Fries JF. Lifestyle habits Global patterns of health expectancy in the year 2000. In: JM and compression of morbidity. J. Gerontol. A Biol. Sci. Med. Robine, J Jagger, CD Mathers, E Crimmins, R Suzman (eds) Sci. 2002;57:M347 – M351. Determining Health Expectancies (pp. 1 – 444). Chichester, 29. Bronnum-Hansen H, Juel K. Abstention from smoking UK: Wiley; 2002. extends life and comes morbidity: a population base study of 22. Ham R. Esperanzas de vida y expectativas de salud en las health expectancy among smokers and never smokers in edades avanzadas. Estudios Demogra´ficos y Urbanos Denmark. Tobbaco Control 2001;10:273 – 278. 2001;48:545 – 560. 30. Perenboom R, Van Oyen H, Mutafova M. Health Expectan- 23. Divisio´n Te´cnica de Informacio´n Estadı´stica en Salud. http:// cies in European Countries. In: JM Robine, J Jagger, www.imss.gob.mx/nr/imss/dpm/dties/default.htm CD Mathers, E Crimmins, R Suzman (eds) Determining 24. Jagger C. Health Expectancy by the Sullivan Method: a Health Expectancies (pp. 1 – 444). Chichester, UK: Wiley; Practical Guide. Tokyo: Nihon University, 1999. (NUPRI 2002. Research Paper Series No. 68). 31. WHO. A Life Course Perspective of Maintaining Indepen- 25. McDowell I, Newell C. Measuring Health. A Guide to Rating dence in Older Age. WHO/HSC/AHE/99.2. Geneva, Scales and Questionnaires. New York: Oxford University 1999;pp. 1 – 19. Press, 1996;1 – 523.