Health Status Perception of People Close to Retirement Age: Relationship with Mental Health and Healthy Habits


 Background: The ageing of today's society intensifies the need for the correct and healthy ageing of the elderly, in order to ensure their overall well-being. Practical strategies are needed to acquire healthy habits at this stage of life. The aim of this study is to analyse the life habits of subjects close to retirement age and the factors that could influence these habits (gender, physical and mental health). Methods: A national (Spain) observational, descriptive and cross-sectional study, in which people close to retirement age are surveyed. The online survey included matters regarding socio-demographic, family, work, leisure, social participation and health indicators. Results: The study comprised of 1,700 participants (581 working; 714 retired), average age = 63 years (DT 5.7); 52% women. Most reported having a satisfactory social life (90%), living in pairs (74%), not smoking (80%), following a Mediterranean diet (73%) and taking medicines daily (70%). Disability (WHODAS-12) was higher in men (8.2 vs 6.5, p<0.001) although women showed more health problems and depression (mild-severe) (27% vs 17%, p<0.001), with a healthier diet and lower physical/work activity. The multivariate model showed a significant association of health status (EQ-VAS) with disability level, number of chronic diseases, sleep habits, exercise, diet, and alcohol consumption. When the level of depression was introduced, age and being a woman were also found to be related. In the subgroup of those retired or working (n=1295), retirees reported a better health status, associated with less disability, fewer chronic illnesses, lower depression, good eating and sleeping habits, and exercise. Conclusions: The results show that retirement does not necessarily mean worse health, but rather an opportunity in life to reinforce favourable health activities and correct those lifestyle factors that deteriorate it. This, together with the differences observed according to gender in the perception of clinical and psychological health, will allow for the design of strategies that promote healthy ageing.


1
The ageing of today's society, as a result of the low fertility rate and the continued fall in the adult 2 mortality rate, is unprecedented in human history [1]. On a global scale, the growth rate in the 3 number of people over 60 is twice that of the growth rate of the general population [2]. In Spain, 4 more than one fifth of its population is over 60, with the life expectancy of this group exceeding 5 80 [3]. Addressing and adapting to this new demographic reality requires substantial changes in 6 political, economic, social and health paradigms [4].

7
Sooner or later, ageing is associated with a decrease in one's physical and mental capacity, as 8 well as an increased risk of chronic-degenerative diseases and comorbidities [5][6][7]. Without doubt, 9 ageing also represents a success of human progress, provided that it is accompanied by actions 10 promoting healthy and active ageing that promotes functional capacity, social contribution and 11 quality of life in old age [5,8]. Adopting healthy living habits during adulthood and old age can be 12 a protective factor against disease (physical and mental), disability and premature death [9,10].

13
Moreover, such a lifestyle helps ensure healthy ageing as defined by the World Health 14 Organization: the process of developing and maintaining the functional capacity that allows well-15 being in old age [11]. 16 Retirement is a critical stage of life that entails the making of crucial decisions regarding health 17 and healthy habits [12]. It may represent an opportunity to adopt healthier routines and living 18 habits; but it may also give rise to the continuance or even intensification of previously established 19 unhealthy behaviours [13]. For this reason, more studies are needed to analyse the determinants 20 and factors associated with healthy living habits at this vital time of life.

21
Psychological and emotional status may prove to be of great importance at this stage of life. Life-22 changing events such as inactivity due to retirement, the increased frequency of chronic diseases, 23 the assumption of new family and social roles (being grandparents), the loss of loved ones or the 24 feeling of being close to death can influence one's psychological state and require both a physical 25 and emotional adaptation [14]. When such adjustment is favourable, in the final stage of life more 26 emotional control and psychological maturity are perceived, with a moderation of positive affect 27 compared to younger people (the young and middle-aged) [15]. In this sense, the acquisition of 28 healthy habits such as the continued exercise of certain abilities can positively influence this 1 process of growing maturity [14]. In Spain, more research is needed on the relationships between 2 retirement, life habits, physical health and psychological health. Such information would allow for 3 the design of useful programs and strategies for the promotion of healthy lifestyles, with the 4 ensuing individual and social benefits they suppose.

5
For this reason, the present study was carried out with the aim of analysing life habits in a 6 population close to retirement age, as well as their relation to physical and mental health, taking 7 into account how gender may also be of relevance in this regard.

9
Study design and population 10 A descriptive, observational and cross-sectional study on a national level, approved by the 11 research ethics committee of the Universidad Francisco de Vitoria (Madrid).

12
The sample was obtained geographically by means of proportional multistage stratified sampling.

13
The sample size was 1,700 individuals, taken from a target population of 10,506,015 people aged 14 between 55 and 75 years of age (+/-10 years relative to the average retirement age in Spain) in 15 2020. In order to participate, after being informed of the purposes of the study, their informed 16 consent was obtained.

17
The data were collected through a structured and online survey of around 20 minutes, including 18 questions regarding socio-demographic, family, work, leisure, social participation, and health 19 indicators. The survey was conducted by professional interviewers specifically trained for the 20 project. The survey participants constituted a sample similar to the Spanish population in terms 21 of age, gender and geographical distribution. As a prerequisite, all participants needed access to 22 the Internet and had to be familiar with its use, being accustomed to answering online surveys.

23
For analysis, the variables were grouped as follows:

12
A general description of the total study sample (n=1700), by sex, was made, followed by an 13 analysis of the sample of active workers compared to those retired (n=1295). Socio-demographic, 14 family, academic, social and leisure variables as well as health indicators by gender and work 15 status (active, unemployed, family and home care, pensioner and retiree) were compared, using 16 the chi-squared test for categorical variables. Moreover, analysis of variance (ANOVA) was 17 performed for quantitative variables such as age, number of chronic pathologies, PHQ-9, or 18 perceived health status (EQ-VAS) reported by participants. Non-parametric equivalent tests were 19 used when the variables did not follow a normal distribution. Bonferroni correction was used for 20 multiple peer comparisons.

21
A multivariate linear regression model was performed with self-perceived health status (EQ-VAS)

22
as a dependent variable. The socio-demographic variables, disability level, comorbidities, life 23 habits, depression level, and health indicators were considered as independent variables. Two 24 independent regression models were created -first with the entire sample and subsequently for 25 the group of active workers and retirees. Tests were conducted to ensure that the basic 26 assumptions of linearity, normality, and independence of the multiple linear regression model 27 were met.

28
The minimum level of significance was set at p<0.05 and version 20 of the IBM SPSS for Windows 1 statistical package was used.

3
Characteristics of respondents 4 Total sample 5 Data were obtained from 1,700 participants, of whom 581 (34.2%) were active workers and 714 6 (42%) retired. Table 1 reflects the socio-demographic characteristics and life habits of the total 7 sample by gender, including data on chronic diseases and medication. The mean age of the 8 sample is 63 years old (DT 5.7 years), with 34% still working. The level of education is mainly at 9 a secondary school (50%) or university (42%) level of studies. The majority (90%) report having 10 a satisfactory or quiet social life, living in a couple (74%), not smoking (80%); They report following 11 a Mediterranean diet (73%), carrying out physical activity more than three days a week (52%), 12 almost 70% of the sample consume medicines daily and almost 50% consume more than three 13 medicines a day.

15
The sample includes a similar and comparable number of women and men. The percentage of 16 women living with a partner is significantly lower than that of men (67% vs 82%). There is a high 17 level of participants who report having university studies, 42%, the percentage being slightly lower 18 in men than in women (38 vs 46%, p<0.001). In contrast, the percentage of women who report 19 only having received primary education is significantly higher (9.4 vs 5.2%). The percentage of 20 working women is lower than that of men (30.4 vs. 38.2%). However, the percentage of women 21 who consider themselves family and home carers is much higher (13.9 vs. 0.7%, p<0.001). In 22 general, the percentage of women who report having health problems is higher than that of men.

23
There are also differences in certain health habits such as physical exercise, which is practised 24 more by men (18% low physical activity in men versus 25% in women, p=0.02) and the perception 25 of sex life (p<0.001), or following a healthy diet, which is more common in women (p=0.003) (table 26 1).

27
Perception of clinical and mental health 1 No significant gender differences were observed in the average health status reported by 2 participants using an analogue quality of life (EQ-VAS) scale (Table 2). The worst state of health 3 was reported by pensioners and the best by the unemployed (71.5 vs. 79.6; p<0.001), who 4 reported a level of health comparable to those working, with no differences found between men 5 and women (Table 2). Regarding the disability scale measured by the WHODAS-12, men 6 reported a higher value than women (8.2 vs 6.5, p<0,001), slightly higher being the percentage 7 of women who reported depression (mild to severe) compared to men (27 vs 17%, p<0,001).

8
When analysing health habits with the multivariate model, a strong association was found 9 between the subject's health status and disability level, the number of chronic diseases, sleeping 10 habits, exercise, diet and alcohol consumption (Table 3). Health status was also significantly 11 associated with sleeping well, exercising three or more times a week, drinking some wine or beer 12 at mealtimes and following a Mediterranean diet with fruit and vegetables, legumes and lean 13 meats on a regular basis (Table 3). When analysing the level of depression among subjects, the 14 percentage of variance the model found rose from 33.3% to 36.3%, with age and sex appearing 15 as new variables associated with the perception of health status (Table 3).

16
Active workers and the retired 17 Table 4 reflects the characteristics of the subsample of active workers and those retired (n=1295).

18
Compared with those working (3%; n=17), a higher percentage of retirees were widowers (7.2%; 19 n=51) and declared themselves to be consuming five or more medications on a daily basis 20 (p<0.001). The degree of depression was lower in retirees (p=0.001). Data on the self-perceived 21 levels of clinical and psychological health of workers and those retired are reflected in Table 5.

22
Overall, retirees showed better overall health and lower levels of depression than those working.

23
Linear regression model 24 Table 6 reflects the results of the linear regression models of perceived health status (EQ-VAS), 25 according to employment status. In the subgroup of those in active employment, the best 26 perceived health status was associated with being a woman, having a lower degree of disability, a lower number of chronic diseases, a lower degree of depression, and exercising three or more 1 times a week (explained variance 33.9%). In the subgroup of those retired, the best perceived 2 health status was associated with lower disability, fewer chronic diseases and lower degree of 3 depression, a healthy diet (Mediterranean diet), the consumption of fruit and vegetables, legumes 4 and lean meats on a regular basis, drinking some wine or beer at mealtimes, sleeping well and 5 exercising 3 or more times a week (explained variance 35.9%).

7
The present study provides data on the socio-demographic and clinical profile of a countrywide 8 sample of the adult and elderly population, in a time of life close to or recently after (between 55 9 and 75 years) the average retirement age. It also provides information on factors that influence 10 perceived health status.

11
Of the 1,700 respondents, the proportion of women and men was very similar, despite the fact 12 that women are the majority population group in the elderly, and even more so the older the age 13 [23]. As the sample was made up of those that volunteered to take part in a computer survey, it

18
With regard to the employment status, significant gender differences were observed: the 19 percentage of active or retired was higher in men compared to women. In contrast, there was

24
The self-perceived health status was higher than recently described in Spain by Janssen et al. for 25 the age group 55-64-year-olds [24]. This may be due, in part, to the fact that these data [24] came 26 from a study published by König et al., in 2009 [25], when the economic and social situation in 27 Spain was worse due to the global financial crisis of 2008 -a circumstance that could interfere 28 with the self-perceived state of health at that time. In any case, the effect of certain selection bias 1 related to the origin of the sample population and its particularities with respect to the general 2 Spanish population of the same age group cannot be ruled out.

3
Perceived health status was associated with lower disability and fewer chronic diseases, sleeping 4 well, increased weekly exercise, and healthy eating (Mediterranean diet) including a moderate 5 drinking of wine or beer during meals. Similar associations have been found in other studies, 6 where perceived health was found to be strongly associated with physical and functional health,

21
In accordance with other studies that reflect a higher consumption of vegetables and fish in 22 women and a higher consumption of processed and red meat in men [32], our results find a 23 significantly healthier diet in women. These results suggest a need to design health promotion 24 programs selectively aimed at improving men's eating habits, particularly training in the basic 25 culinary skills for those who live alone or prepare their own food.

26
Regarding the perceived level of health according to the working status, as is the case in the 27 general population, both in workers and in those retired, the best health status was associated 28 with a lower degree of disability. fewer chronic diseases and exercise three or more times a week.

1
In workers, this perception of having a good diet was also related with being a woman and being 2 retired. In both workers and those retired, the average level of disability was very low, with less 3 disability observed in retirees, although without statistically significant differences. This may be 4 due to the increase in weekly physical activity and to healthier eating habits (Mediterranean diet) 5 reported by retirees, as it could serve as a counterbalance to the deterioration that ageing

15
With ageing, biological changes occur in the human body affecting mood, physical condition and 16 social activity that influence one's perception of health [41]. In Spain, depression takes ninth place 17 in terms of chronic diseases and is also more prevalent in women [3,42]. This phenomenon is 18 reflected in our study: although the percentage of moderate to severe depression is low (less than 19 1 in 10 participants), it is higher in women. In this sense, it has been noted that, in time, women 20 are more likely than men to suffer from non-lethal disabling conditions such as depression [26, 21 43]. When depression was included in our analysis, the model showed women and being younger 22 were associated with a better perception of one's health status. It is logical that those women who 23 do not suffer from this type of illness have a better perception of their health status. [26,43].

24
In both workers and those retired, a better health status was associated with a lower degree of 25 depression, with a significantly lower degree of depression observed in the latter group. There 26 exist conflicting data on the prevalence of depression in retirement. Despite being related to the 27 loss of employment and the social life associated with work, retirement is also seen as an escape 28 from work, which is a source of daily stress, obligations and responsibilities. [44,45] In our case, 1 the most plausible explanation is that of a sense of freedom from work, as our retired population 2 had significantly more satisfactory social life and exercised more.

3
The fact that the sample analysed is composed of people who volunteer to conduct computer 4 surveys (panelists), implies a limitation of the study as it suggests a bias in the representativeness 5 of the overall population. Other limitations to consider are the cross-sectional design of the study, 6 which does not allow the analysis of causality, and that the mental health analysed is self-7 referenced and therefore not quantifiable as in the case of cognitive scales. 14 The study provides new and valuable information on the determinants of perceived health in the 15 adult and mature stages of life, highlighting the similarities and differences between workers and 16 those retired. Both before and after retirement, the best perceived health status is associated with 17 a lower degree of disability and depression, a lower number of chronic diseases, and with those 18 who exercise three or more times a week. In workers, it was also associated with being a woman 19 and in those who have retired and have a healthy diet and adequate sleep.

20
Perceived differences according to gender also need to be taken into account. The perception of 21 clinical and psychological health status proved to be similar among men and women, albeit with 22 more depression reported by women. Disregarding depression, being a woman and of a younger 23 age are also associated with a better perception of health status. Men do more physical exercise 24 than women, although women have a healthier diet. The social life of women is better, although 25 they profess to having a worse sex life.

26
It is important to act upon all identified factors in order to help improve health status and achieve 27 satisfactory ageing. To achieve this, the implementation of specific strategies for each group is

20
The study received a financial support financed by Direct line Aseguradora S.A. 21

22
All authors contributed to the study concept and design. Material preparation, data collection and