Gac Sanit. 2021;35(2):193–198 Special article Are partner’s features important to understand health at older ages? The Spanish case Jordi Gumàa,∗ , Jeroen Spijkerb a Department of Political and Social Sciences and Expertise Centre for Survey Methodology, Universitat Pompeu Fabra, Barcelona, Spain b Centre d’Estudis Demogràfics, Bellaterra (Barcelona), Spain a r t i c l e i n f o a b s t r a c t Article history: Objective: To analyse health differences among partnered individuals in Spain aged 65-81 consider- Received 11 January 2019 ing their combined (education of both partners and age gap) and household (economic capacity of the Accepted 1 July 2019 household) characteristics. Available online 15 September 2019 Method: A cross-sectional study of the 2015 Spanish sample of the European Union Statistics on Income and Living Conditions survey (EU-SILC) on partnered individuals aged 65-81 years (N = 1787). Using logis- Keywords: tic regression models separately for women and men we obtained odds and predicted probabilities of Spouse/partner having less than good health (95% confidence intervals) according to combined information from both Self-reported health Socio-economic status partners on education and age, the household’s economic capacity and partner’s health status. Education Results: Probabilities of not having good health are significantly less among lower educated women Elderly whose partners are more highly educated (compared to both partners being lower educated) and among Sex differences women whose partner is younger or has good health status. The latter also applies to men. Living in a EU-SILC household without economic difficulties also favours health (both sexes). Spain Conclusions: For both sexes a partner’s health status is the variable that shows the largest effect on elderly partnered Spanish people’s health but women’s health appears to be more sensitive to their partner’s educational attainment and the household’s economic situation. © 2019 SESPAS. Published by Elsevier España, S.L.U. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). ¿Son importantes las características de la pareja para entender la salud en edades avanzadas? El caso español r e s u m e n Palabras clave: Objetivo: Analizar las diferencias de salud entre la población española de 65 a 81 años de edad que vive Pareja en pareja, considerando características individuales, combinadas de ambos miembros y generales de la Salud autopercibida pareja. Estatus socioeconómico Método: Estudio transversal de la muestra española del año 2015 de la Encuesta Europea de Condiciones Educación de Vida de personas mayores (edades 65-81 años) que conviven en pareja (N = 1787). Mediante modelos Personas mayores Diferencias por sexo de regresión logística independientes para mujeres y hombres se obtuvo la estimación de las probabili- EU-SILC dades de no tener buena salud autopercibida (intervalos de confianza del 95%) según la combinación del España nivel de estudios y la edad de ambos miembros de la pareja, así como la capacidad económica del hogar y el estado de salud del cónyuge. Resultados: Las probabilidades de no tener buena salud son significativamente menores entre las mujeres con menor nivel de educación cuyas parejas son más educadas (en comparación con las que tienen menos educación) y entre las mujeres cuya pareja es más joven o tiene un buen estado de salud. Esto último también se aplica a los hombres. Vivir en un hogar sin dificultades económicas también favorece la salud en ambos sexos. Conclusiones: Para ambos sexos, el estado de salud de la pareja es la variable que muestra el mayor efecto sobre la salud de la población mayor española que vive en pareja, aunque la salud de las mujeres parece ser más sensible al nivel educativo de su pareja y a la situación económica general del hogar. © 2019 SESPAS. Publicado por Elsevier España, S.L.U. Este es un artı́culo Open Access bajo la licencia CC BY-NC-ND (http://creativecommons.org/licenses/by-nc-nd/4.0/). ∗ Corresponding author. E-mail address: jordi.guma@upf.edu (J. Gumà). https://doi.org/10.1016/j.gaceta.2019.07.005 0213-9111/© 2019 SESPAS. Published by Elsevier España, S.L.U. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc- nd/4.0/). Descargado para Anonymous User (n/a) en Autonomous University of Barcelona de ClinicalKey.es por Elsevier en marzo 24, 2021. Para uso personal exclusivamente. No se permiten otros usos sin autorización. Copyright ©2021. Elsevier Inc. Todos los derechos reservados. 194 J. Gumà, J. Spijker / Gac Sanit. 2021;35(2):193–198 Introduction health. Reasons that have been linked to this include shared living conditions,19 similar health-related behaviour20 and the already The protective health effect of living with a partner is well- mentioned health risks when caring for an ill partner.21 known,1–3 but we know less about health differences among Why study Spain? While studies on the relationship between partnered individuals. Although the link between health and indi- joint (partner and own) education profiles and health status have vidual socio-economic information from each partner member has focused on the United States or northern, western and central already been studied,4,5 such research tested the effect of the European countries, no country from southern Europe has yet partner’s socio-economic characteristics in addition to the ego’s been analysed. Moreover, Spain is an especially interesting case characteristics rather than combining the socio-economic charac- study because during the last half-a-century it experienced a pro- teristics of both partners. The latter is the aim of the present study. cess of rapid political and social change, transitioning from an The health advantage of people co-residing with a partner has ultra-conservative dictatorship (1939-1975) to a democracy-based been extensively explained by a set of benefits related to living market economy. The educational expansion and the massive inser- with a partner: first, being subjected to social control, particu- tion of women into the labour market started later in Spain than in larly from partners. Especially partnered men may be more often other Western European countries.22 Accordingly, the productive encouraged to abandon risky behaviours than single individuals.3 and reproductive roles are still very much gendered for the older Second, the combination of both partners’ social nets creates a cohorts studied here, whereby the female socio-economic status is wider source of possible resources to reduce the impact of events almost totally determined by their partners’ status.23 that may negatively impact health, such as becoming unemployed Within these socioeconomic, gender and cohort specificities of or widowed.6 And third, the optimization of common resources Spain, the study’s aim is to determine whether, besides own edu- through economies of scale and specialization of tasks within the cational attainment also characteristics of the partner (educational household enhances the socioeconomic status of both partners.6 attainment and health status) and union (household capacity to The latter two health benefits also have a socio-economic gradient.7 make ends meet and the spousal age gap) affect individual health When measuring socioeconomic status of individuals, educa- status among co-residing individuals aged 65 years and older. We tion is a frequently used indicator of adult socioeconomic status contextualize our results within the high level of gender inequality in studies on health differences because of its stability over time typical of older Spanish cohorts. Our research questions are: 1) do (unlike income or employment status) given that it is typically partner and couple features complement what we currently know completed early in adulthood.8 Using education also avoids prob- about health status variability of the Spanish older population? and lems of reverse causation when linking it with health outcomes 2) how does this vary by sex? in older ages.9 It also has a well-established negative association with long-term risky health behaviours like smoking or drinking10 and facilitates the effective use of health services.11 According to the social causation hypothesis, an individual’s educational level Method is also assumed to affect material and psychosocial factors that in turn have effects on health. In this study, we extent this assertion Data come from the Spanish cross-sectional sample of the Euro- by purporting that partner’s education has an additional effect on pean Union Statistics on Income and Living Conditions (EU-SILC) health. survey (2015 wave). Our target population are partnered Spaniards Indeed, previous studies have found positive associations where both members of the couple are aged 65-81 (N = 6569). The between partner’s educational attainment and individual’s health, lower age boundary was fixed at 65 because this was the legal age including self-perceived health,5 risky behaviour12 as well as at retirement at the time of the interview; 81 is the age of the oldest mortality.4 All studies stated that the inclusion of partner’s edu- respondents. EU-SILC has information on socioeconomic status and cation adds meaningful information to better understand health health from all members living in a private household, although inequalities, even after controlling for own educational level. A 32 couples which accomplished the age criteria were discarded common explanation was that couples usually pool their resources because of incomplete data. Only native Spaniards who live with a (material and non-material). partner were analysed (56% of the age-specific sample), irrespective In line with the latter, previous studies on health inequalities of their marital status (due to the low (<1%) prevalence of marital have indicated household capacity to make ends meet to be an statuses other than married). Given our interest in possible sex dif- accurate indicator of poverty as it summarises the economic con- ferences in the effect of partner characteristics on individual health, text of a household.13 This is of special interest to our study as same sex couples were also removed from the analysis. This left us relatively few elderly Spanish women participated in the labour with 1787 couples with complete information on all variables. market, meaning that their socioeconomic status is dependent on Our dependent variable is individual self-reported health status, their partner’s labour trajectory. Moreover, household income is measured by the survey question ‘How is your health in general?’. known to be associated with individual mental and self-perceived The possible answers were dichotomised into good health (“good” health.14,15 and “very good” health) and less than good health (“fair”, “bad” and There are also other partner’s features that can modify indi- “very bad” health). vidual health and mortality risks. The spousal age gap has been The co-variables are: associated with survival probabilities, although differently by gen- der: having a younger spouse is beneficial for men but detrimental for women, whereas having an older spouse is detrimental for • Educational attainment of both partners: low or no education both.16 Positive effects include psychological improvements of (ISCED 0); medium education (ISCED 1-2); and high education having a younger partner as well as selection effects (healthier (ISCED 3-6). Information from both partners were then com- individuals have higher chances of attracting younger partners).16 bined into seven categories: both partners low education; ego Conversely, being the caregiver of one’s partner is a known risk low education-partner medium or high education; ego medium factor for physical and mental health,17 a task that women ful- education-partner low education; both partners medium educa- fill more often than men due to their higher life expectancy and tion; ego medium education-partner high education; ego high the traditional gender role norms roles.18 Partner’s health sta- education-partner low or medium education; and both partners tus is another characteristic of the partner affecting individual high education (see Table I in online Appendix for further detail). Descargado para Anonymous User (n/a) en Autonomous University of Barcelona de ClinicalKey.es por Elsevier en marzo 24, 2021. Para uso personal exclusivamente. No se permiten otros usos sin autorización. Copyright ©2021. Elsevier Inc. Todos los derechos reservados. J. Gumà, J. Spijker / Gac Sanit. 2021;35(2):193–198 195 • Household ability to make ends meet: “easily or very easily”, make ends meet) or in reverse order (e.g., the combined educational “fairly easily”, “with some difficulty” and “with difficulty or great level of each partner and the age gap between partners). Regarding difficulty”. health status, the values for the other sex in the table correspond • Age gap between partners: wife older than husband; no age gap to the results for the variable “health status of partner”. (0 to 2-years older husband); husband moderately older (3-5 As expected, women report worse health than men (58% of years); and husband much older (6 + years). women reported to have less than good health vs. 55% of men), • Partners’ health status: this information is obtained from the even though they were on average 2.1 years younger. 39% of women interviewee’s partner’s answer to the same question as the had at least a three-year older partner, while 15% were older them- dependent variable and contains the same two categories. selves. Regarding educational attainment, homogamy was high among the studied cohort (70%). The most frequent situation was The possible effects of the variables of interest on individuals’ both partners with medium education (40%), followed by both with health status were assessed by separate logistic regression mod- low education (18%) and couples where the male partner is high els for men and women to assess possible gender differences. We educated and the female partner medium educated (12%). The least present predicted probabilities of poor health with 95% confidence frequent situation was the opposite (3%). With regard to the house- intervals that include all variables. Predicted probabilities allow the hold’s ability to make ends meet, 53% declared to have financial direct comparison of results from different models24 (odds ratios difficulties. are reported in Table II in online Appendix). We additionally con- ducted a dominance analysis to determine the relative contribution from each variable in the total explained variance.25 Besides edu- cation and household ability to make ends meet, we also tested Multivariate analysis the effect of overall household income and the individual rela- tive contribution to the overall household income. However, they Table 2 displays the predicted probabilities of having less than were excluded from the final model because they did not pro- good health. Only the results for the final model which includes all duce a significant contribution (see Table III in online Appendix). variables of interest and controls for age are shown here. In gen- To check the robustness of our results, we replicated our mod- eral, the female model that includes common features of the couple els for two other health outcomes available in EU-SILC (suffer explains variability in self-perceived health in Spain better than the from any chronic [long-standing] illness or condition; limitation in male model (r2 = 0.18 vs. 0.15). This is therefore consistent with activities because of health problems). Both outcomes were opera- two earlier studies by the authors that considered the middle-aged tionalized as a dichotomous variable (1 = yes; 0 = no) and results are adult population.26,27 In addition, the obtained predicted probabil- provided in Tables IV and V in online Appendix, and are briefly ities are generally higher for women than for men for all variable commented on in the discussion. categories, although not statistically significant (p >0.05). Regarding the observed difference in each sex-specific model, Results the variable that combines the educational attainment of both partners showed significantly higher predicted probabilities of Descriptive analysis having less than good health among low-educated women with low-educated partners (the reference category) compared to Table 1 displays the characteristics of our working sample. As medium-educated women with equal or higher educated part- common information is analysed for both partners, the sex-specific ners as well as higher-educated women irrespective of their values are either the same for both sexes (e.g., household ability to partner’s educational level. Regarding men, coresiding with a higher-educated partner did not influence their health. Only having the same educational level as their partner (medium Table 1 Sample characteristics of partnered individuals aged 65-81 years by sex. Spain, 2015. or high) was associated with better health than the reference category. Women Men Turning to the age gap between both partners, the only observed Health status 41.8% 44.5% difference (albeit at the 0.1 level) was for women whose part- Good and very good health 58.2% 54.5% ner was at least 6 years older. They had a higher probability of Fair, poor or very poor health having less than good health than the reference category (no age Education of both partners gap). Both partners low education (ref.) 18.2% 18.2% Partner’s health status is the variable that showed the largest Ego low education and partner medium or high education 8.5% 7.0% Ego medium education and partner low education 6.8% 7.3% effect on one’s health: the proportion of women with less than good Both partners medium education 40.3% 40.3% health equals 0.72 when the partner has the same health status, but Ego medium education and partner high education 12.2% 3.2% just 0.43 if his health is good. Similar results are observed for men Ego high education and partner low or medium education 3.4% 13.4% (respectively, 0.68 and 0.38). Both partners AGE surveyhigh education 10.6% 10.6% The last tested variable, household capacity to make ends meet, Age gap between partners was also significant for both sexes, although the only difference Female partner older 15.3% with the reference category (with (great) difficulty) was for those No age gap or male partner 1-2 years older 45.9% Male partner 3-5 years older 25.0% who have it easy to make ends meet. Male partner 6 or over years older 13.8% In terms of the relative explanatory capacity of the independent variables (Table 3), results are very similar for both sexes. Part- Household ability to make ends meet With (great) difficulty 22.3% ner’s health status explains the most, contributing to the 47% of With some difficulty 30.7% the explanatory capacity of the female and 52% of the male model, Fairly easily 33.6% followed by ego’s age (18% for both sexes) and education of both Easily 13.4% partners (14% for women and 11% for men). Economic capacity of Total (n) 1787 1787 the household and former occupation of the ego and partner explain Note: Data weighted by the survey’s personal cross-sectional weight. between 4% and 8% of the variance in both models. The couple’s age Source: Spanish sample EU-SILC 2015. gap has the least explanatory power (1% in each case). Descargado para Anonymous User (n/a) en Autonomous University of Barcelona de ClinicalKey.es por Elsevier en marzo 24, 2021. Para uso personal exclusivamente. No se permiten otros usos sin autorización. Copyright ©2021. Elsevier Inc. Todos los derechos reservados. 196 J. Gumà, J. Spijker / Gac Sanit. 2021;35(2):193–198 Table 2 Predicted probabilities of having less than good health from the logistic regression models for partnered individuals aged 65-81 by sex. Spain, 2015. Women Men Probability CI Probability CI Education of both partners Both partners low education (ref.) 0.66 0.61 0.72 0.63 0.57 0.69 Ego low education and partner medium or high education 0.68 0.58 0.77 0.57 0.48 0.67 Ego medium education and partner low education 0.64 0.55 0.74 0.61 0.51 0.70 a b Both partners medium education 0.59 0.55 0.63 0.53 0.49 0.57 a Ego medium education and partner high education 0.56 0.49 0.63 0.65 0.51 0.78 b c Ego high education and partner low or medium education 0.46 0.32 0.60 0.54 0.47 0.61 d b Both partners high education 0.42 0.34 0.51 0.48 0.40 0.56 Age gap between partners Female partner older 0.56 0.50 0.63 0.55 0.49 0.61 No age gap or male partner 1-2 years older (ref.) 0.58 0.55 0.62 0.56 0.52 0.59 Male partner 3-5 years older 0.57 0.52 0.62 0.55 0.50 0.60 c Male partner 6 + years older 0.65 0.59 0.71 0.56 0.49 0.62 Partner health status Good and very good health (ref.) 0.43 0.39 0.47 0.38 0.34 0.43 d d Fair, poor or very poor health 0.72 0.68 0.75 0.68 0.64 0.71 Household ability to make ends meet With difficulty (ref.) 0.63 0.58 0.68 0.59 0.54 0.64 With some difficulty 0.59 0.55 0.64 0.60 0.56 0.64 a Fairly easily 0.57 0.53 0.62 0.51 0.47 0.56 a a Easily 0.52 0.46 0.59 0.47 0.40 0.55 Pseudo R-Square 0.17 0.15 Total (N) 1787 1787 CI: confidence interval. a p < 0.05. b p < 0.01. c p < 0.1. d p < 0.001. All the models are controlled for age. Data weighted by the survey’s personal cross-sectional weight. Source: Spanish sample EU-SILC 2012. Table 3 Dominance analysis of the logistic regression models of having less than good health from the sex-specific logistic regression models for partnered individuals aged 65-81. Spain, 2015. Women Men Standardized Domin. Stat. Ranking Standardized Domin. Stat. Ranking Education of both partners 0.20 3 0.15 3 Age gap between partners 0.01 5 0.01 5 Health status of partner 0.52 1 0.57 1 Household ability to make ends meet 0.07 4 0.10 4 Age 0.20 2 0.18 2 Total explained variance 1.00 1.00 Data weighted by the survey’s personal cross-sectional weight. Source: Spanish sample EU-SILC 2015. Discussion established) may have a cumulative effect on health as a conse- quence of behavioural factors (smoking, diet, alcohol consumption, Our study contributes to the growing body of research on health etc.). among partnered individuals. Moreover, we did this by also includ- There are also gender implications from having a partner in ing characteristics from both partners combined rather than that poor health. When the husband is in poor health, the wife’s from each partner separately. Results showed for Spanish elderly health is likely to deteriorate due to the increase in their domes- aged 65-81 that women’s health is more sensitive to the charac- tic workload.28 This might explain why, according to our results, teristics of the partner and household as more of the variability living with a much older husband penalised female health. Con- in female health status was explained by the tested variables. Co- versely, the additional workload is lower for husbands when their residing with a partner not in good health is associated with almost wives have poor health as this is usually shared with other, mainly double the probability of not having good health. One possible female, relatives, particularly daughters or daughters-in-law.29 reason for this association is selective mating as two individuals Women’s health was slightly more sensitive to their own and with similar behaviour are more likely to match up, irrespec- partner’s educational level than men. First, their health was signif- tive of socioeconomic status.2 Partners also influence each other’s icantly (p < 0.01) better if their educational level was higher than lifestyle during their relationship (for instance, it is more diffi- their partner’s, a result that was less significant (p < 0.1) in the case cult to quit smoking if one’s partner also smokes) and partner of men. Second, in the context of the known educational gradient resemblance in health increases with relationship duration.21 This in health, there is some suggestion that risk factors leading to implies that assortative mating at younger ages (when couples are health deterioration among women are also mediated through Descargado para Anonymous User (n/a) en Autonomous University of Barcelona de ClinicalKey.es por Elsevier en marzo 24, 2021. Para uso personal exclusivamente. No se permiten otros usos sin autorización. Copyright ©2021. Elsevier Inc. Todos los derechos reservados. J. Gumà, J. Spijker / Gac Sanit. 2021;35(2):193–198 197 the educational status of their partner. However, this is mainly cannot disclose whether the results for Spain are transferable to observed among women with medium education (the probability other countries as only one country was analysed. A comparative of having less than good health declines as the educational level study for this age cohort between European countries characterised of the partner becomes higher). The lower level of association for by different levels of gender inequality is therefore another recom- men in this case is consistent with results for the Netherlands5 and mendation for future study. accords with the gendered social roles that typifies this Spanish generation where few women contributed financially to household income30 and were therefore dependent on the socioeconomic status of their partner. Repeating the analysis for the two other health indicators What is known about the topic? included in EU-SILC led to similar results, thus confirming the Previous studies have found a complementary explana- importance of combining the educational attainments from both tory effect of the inclusion of partner’s characteristics to better partners to better explain health inequalities among partnered understand health inequalities among individuals who live older Spaniards. The only exception was the male model for chronic with a partner, even after controlling for own characteristics. conditions and limitations, as results were not statistically signif- icant. This seems to support results from elsewhere that show an What does this study add to the literature? educational gradient in chronic conditions and limitations among older women but not among men where the effect of education is We contribute to this existing body of research by testing less strong or significant in fewer chronic conditions.31,32 It would the effect of combining information from both partners (edu- certainly be interesting for future research to explore these sex cational level and age) rather than analysing individual and differences further as well as consider different types of chronic partner’s characteristics separately. We recommend targeting besides vulnerable groups like single-person or -parent house- diseases. holds in health policy also couples (and their families) from low Our study has shown that within an apparently advantageous socioeconomic status groups. population group in terms of health status, namely married or cohabiting couples, there are significant socioeconomic differences as higher educated individuals and those without economic prob- lems have better heath than their lower educated and poorer counterparts. Therefore, a relevant contribution for public policy is Editor in charge that not only typical vulnerable groups like single-person, single- parent or low-income households should be targeted, but also María del Mar García-Calvente. retired low-educated partnered individuals, even if their partner is higher educated. This is not only because their probability of not Transparency declaration having good health does not improve if the latter is the case, already 44% of 65-81 year-olds do not live with a partner, among whom The corresponding author on behalf of the other authors guar- health is significantly worse.3 Better targeted health education and antee the accuracy, transparency and honesty of the data and health promotion33 and ensuring adequate income in old age for information contained in the study, that no relevant information low-educated individuals are thus pertinent in the quest to reduce has been omitted and that all discrepancies between authors have social health inequalities. Also noteworthy is the effect of part- been adequately resolved and described. ner’s health on own health. This result is particularly relevant for health professionals who are in daily contact with patients as health Authorship contributions inequalities are larger at the household level than at the individual level.21 The Spanish Society for Family and Community Medicine The first author conceived and designed the study and analysed already recommends through their regularly updated Programme the data. The second author contributed to the literature review and of Preventive and Health Promotion Activities (PAPPS in its Span- the methodology used in the analysis. Both authors interpreted the ish acronym) general practitioners to assess the motivation and results and drafted and revised the manuscript and approved the availability of the patient and his or her partner or other family final version for publication. members to change health-damaging behaviour and establish a joint action plan.34 Our result therefore reinforces the importance of involving the patient’s partner as this would ease implementing Funding specific interventions to achieve moderate habit changes (and may also benefit from it). This study belongs to the multi-country project “Care, Retire- One weakness of our study is the lack of long-term longitudinal ment & Wellbeing of Older People Across Different Welfare information about individuals’ trajectories to study possible mech- Regimes” (CREW), founded from the Spanish Ministry of Economy, anisms of selection prior to partner co-residence. The data source Industry and Competitiveness (PCINN-2016-005). This manuscript (EU-SILC) also lacks information on health behaviour, including is also part of the project INTERSOC-HEALTH (RTI2018-099875-J- fruit, vegetable, smoking and alcohol consumption, and physi- I00; PI: JG) founded by the Spanish Ministry of Science, Innovation cal activity, that may provide additional clues as to why higher and Universities. This work was supported by the Spanish Min- educated individuals have better health than their lower edu- istry of Economy and Competitiveness under the “Ramón y Cajal” cated counterparts. Additionally, combining individual and partner program (RYC-2013-14851) for Jeroen Spijker and the “Juan de la categories reduced the number of cases in each combination of edu- Cierva” program (FJCI-2015-25066) for Jordi Gumà. Financial sup- cational level, resulting in fewer significant associations, an issue port was also received from the Catalan Government under the that would be resolved with a larger sample size. We must also CERCA Program. keep in mind possible long-term influences of the economic crisis in Spain (2008-2014) at the time of the study (2015), as this may Conflicts of interest have amplified the sex-differences we found in the influence of the partner’s socio-economic profile on individual’s health. Finally, we None. Descargado para Anonymous User (n/a) en Autonomous University of Barcelona de ClinicalKey.es por Elsevier en marzo 24, 2021. Para uso personal exclusivamente. No se permiten otros usos sin autorización. Copyright ©2021. Elsevier Inc. Todos los derechos reservados. 198 J. Gumà, J. Spijker / Gac Sanit. 2021;35(2):193–198 Appendix A. Supplementary data 15. Tucker-Seeley RD, Harley AE, Stoddard AM, et al. Financial hardship and self-rated health among low-income housing residents. Health Education & Behavior. 2013;40:442–8. Supplementary data associated with this article can be found, in 16. Drefahl S. How does the age gap between partners affect their survival? Demog- the online version, at doi:10.1016/j.gaceta.2019.07.005. raphy. 2010;47:313–26. 17. Schulz R, Sherwood PR. Physical and mental health effects of family caregiving. J Soc Work Educ. 2008;44 Suppl 3:105–13. References 18. Davidson K, Arber S, Ginn J. Gendered meanings of care work within late life marital relationships. Canadian Journal on Aging/La Revue Canadienne du Vieil- 1. Doblhammer G, Gumà J. Gender differences in the relationship between house- lissement. 2000;19:536–53. hold position and health in twelve European countries: are they associated with 19. Meyler D, Stimpson JP, Peek MK. Health concordance within couples: a system- the value climate? In: Doblhammer G, Gumà J, editors. A demographic perspec- atic review. Soc Sci Med. 2007;64:2297–310. tive on gender, family and health in Europe. Switzerland: Springer; 2018. p. 20. Umberson D, Crosnoe R, Reczek C. Social relationships and health behavior 103–31. across the life course. Annu Rev Sociol. 2010;36:139–57. 2. Koball HL, Moiduddin E, Henderson J, et al. What do we know about the link 21. Monden CW. Partners in health? Exploring resemblance in health between part- between marriage and health? J Fam Issues. 2010;31:1019–40. ners in married and cohabiting couples. Sociol Health Illn. 2007;29:391–411. 3. Waite L, Gallagher M. The case for marriage: why married people are happier, 22. Vila L-E, Mora J-G. Changing returns to education in Spain during the 1980s. healthier and better off financially. New York: Broadway Books; 2000. Economics of Education Review. 1998;17:173–8. 4. Kravdal Ø. A broader perspective on education and mortality: are we influenced 23. Sanchez L, Hall CS. Traditional values and democratic impulses: the gender by other people’s education? Soc Sci Med. 2008;66:620–36. division of labor in contemporary Spain. J Comp Fam Stud. 1999:659–85. 5. Monden CW, van Lenthe F, De Graaf ND, et al. Partner’s and own education: 24. Mood C. Logistic regression: why we cannot do what we think we can do, and does who you live with matter for self-assessed health, smoking and excessive what we can do about it. Eur Sociol Rev. 2010;26:67–82. alcohol consumption? Soc Sci Med. 2003;57:1901–12. 25. Azen R, Traxel N. Using dominance analysis to determine predictor importance 6. Waite L. Does marriage matter? Demography. 1995;32:483–507. in logistic regression. J Educ Behav Stat. 2009;34:319–47. 7. Lin N. Building a network theory of social capital. In: Social capital. Abingdon, 26. Gumà J, Arpino B, Solé-Auró A. Determinantes sociales de la salud de distintos UK: Routledge; 2017. p. 3–28. niveles por género: educación y hogar en España. Gac Sanit. 2019;33:127–33. 8. Holmes CJ, Zajacova A. Education as “the great equalizer”: health benefits for 27. Gumà J, Spijker J. Are couples really a homogeneous cluster in terms of health in black and white adults. Social Science Quarterly. 2014;95:1064–85. Europe? Paper presented at the European Population Conference. Mainz. 2016. 9. Duncan GJ, Daly MC, McDonough P, et al. Optimal indicators of socioeconomic 28. Bauer JM, Sousa-Poza A. Impacts of informal caregiving on caregiver employ- status for health research. Am J Public Health. 2002;92:1151–7. ment, health, and family. J Popul Ageing. 2015;8:113–45. 10. Steenland K, Henley J, Thun M. All-cause and cause-specific death rates by edu- 29. Larrañaga I, Martín U, Bacigalupe A, et al. Impacto del cuidado informal en la cational status for two million people in two American Cancer Society cohorts, salud y la calidad de vida de las personas cuidadoras: análisis de las desigual- 1959-1996. Am J Epidemiol. 2002;156:11–21. dades de género. Gac Sanit. 2008;22:443–50. 11. Pincus T, Callahan LF. Associations of low formal education level and poor health 30. Solsona M. The second demographic transition from a gender perspective. Inno- status: behavioral, in addition to demographic and medical, explanations? J Clin vation: The European Journal of Social Science Research. 1998;11:211–25. Epidemiol. 1994;47:355–61. 31. Suárez AL, González JE, Robles MB, et al. Prevalence of obesity, diabetes, hyper- 12. Takagi D, Kondo N, Takada M, et al. Differences in spousal influence on smoking tension, hypercholesterolemia, and metabolic syndrome in over 50-year-olds in cessation by gender and education among Japanese couples. BMC Public Health. Sanlucar de Barrameda, Spain. Rev Esp Cardiol (English ed.). 2008;61:1150–8. 2014;14:1184. 32. Morcillo Cebolla V, de Lorenzo-Cáceres Ascanio A, Domínguez Ruiz de León P, 13. Escolar Pujolar A. Determinantes sociales frente a estilos de vida en la diabetes et al. Desigualdades en la salud autopercibida de la población española mayor mellitus de tipo 2 en Andalucía: ¿la dificultad para llegar a fin de mes o la de 65 años. Gac Sanit. 2014;28:511–21. obesidad? Gac Sanit. 2009;23:427–32. 33. Ridde V. Reducing social inequalities in health: public health, community health 14. Patel V, Kirkwood BR, Pednekar S, et al. Risk factors for common mental or health promotion? Promot Educ. 2007;14:63–7. disorders in women: population-based longitudinal study. Br J Psychiatry. 34. Córdoba GR, Camarelles FG, Muñoz ES, et al. Recomendaciones sobre el estilo de 2006;189:547–55. vida. Actualizacón PAPPS 2018. Atencion Primaria. 2018;50:29–40. Descargado para Anonymous User (n/a) en Autonomous University of Barcelona de ClinicalKey.es por Elsevier en marzo 24, 2021. Para uso personal exclusivamente. No se permiten otros usos sin autorización. Copyright ©2021. Elsevier Inc. Todos los derechos reservados.