Research design

We followed steps corresponding to an ex post facto simple retrospective design (^{Montero & León, 2002}).

Participants

An intentional, non-probabilistic sampling strategy was used on older adults attending adult day-care centers and workshops reporting to the Government of the City of Buenos Aires and of Greater Buenos Aires (Argentine Republic). A snowball effect was also used.

The sample was comprised of 150 mixed-gender, self-sufficient older adults over 60 years old, of middle socioeconomic status, residing in a major urban center (Autonomous City of Buenos Aires and Greater Buenos Aires, Argentine Republic), with stable and formal employment prior to retirement.

In table 1 we can find the arithmetic means, standard deviations and frequency and percentage counts, according to the level of measurement, of socioeconomic profile characteristics of the total sample. According to the summarized version of the Gino Germani Social and Economic Status Index, (Grimson et. al., 1972), in the answers older adults gave on education and occupation it is observed that all sample subjects have a middle socioeconomic status.

Table 1: Sociodemographic profile of the total sample (n=150) 

As for the profile of retirees in the sample, average retirement age is 65.29 years old (sd=4.86 years), and average time spent in retirement is 9.31 years (sd= 5.57). As for occupation prior to retirement, they declare being “employees” (26%) “independent professional/businessman” (22%) or having engaged in “several trades” (23%), and having worked in said activities for “over 15 years” (91%).

Instruments

- Socio-Demographic Background and Retirement Profile Questionnaire. An ad hoc questionnaire was designed, with 12 open and closed questions with fixed alternatives, to obtain information on socio-demographic characteristics (gender, age, nationality, marital status, level of educational attainment, prior occupation, housing, children and religious affiliation) and retirement characteristics (retirement age, time spent in retirement, number of years spent working in prior occupation).

- Health Status Questionnaire. To find out more about the presence of psychophysiological disorders related to retirement, an ad hoc questionnaire was designed consisting of 5 closed questions with fixed alternatives, evaluating if the subjects perceived any changes in their health related to retirement. In case of affirmative answers, disorders were classified according to their type: skin, urinary, endocrine, musculoskeletal, gastrointestinal, cardiovascular, respiratory, immunological and reproductive (Ortego, López, Alvarez & Aparicio, 2012), the moment of appearance and the existence of a medical history linked to development of disorder.

- Index of Psychological Well-Being. This technique was selected with the purpose of estimating retiree’s assessment of his/her life and himself/herself in general. The Index was structured by ^{Stefani (2002}), using version-A of the Life Satisfaction Index by ^{Neugarten, Havighurst and Tobin (1961}), adapted locally by Eva Muchinik (1984). The Index consists of 11 items that clearly define three of the five dimensions proposed by Neugarten: “zest for life”, “congruence between desired and achieved goals” and “optimistic mood tone”. The ^{Bachman & O’Malley version for older adults - RSE-B - (1977}) of the ^{Rosenberg Self-Esteem Scale (1965}) was also included, consisting of 10 items.

Therefore, the Index of Psychological Well-Being consists of 21 items: 13 direct statements -positively oriented- (items 1, 3, 5, 6, 8, 9, 10, 12, 13, 14, 16, 18 and 20) and 8 reversed statements -negatively oriented- (items 2, 4, 7, 11, 15, 17, 19 y 21). When administering the questionnaire, subjects were asked to score them according to their level of agreement through a 3-point Likert scale: “disagree”, “undecided” and “agree”. Positive statements were assigned a value of 2 for “agree”, 1 for “undecided”, and 0 for “disagree”. Negative statements were assigned a reversed value: 0 for “agree”, 1 for “undecided” and 2 for “disagree”.

The total score, as well as the score for each sub-scale (Life Satisfaction Index and Self-Esteem Scale), is obtained by calculating the average values assigned to each answer. In this way, the subject is placed in a 0-2 continuum, where a low score indicates, respectively, low levels of psychological well-being, low life satisfaction or negative self-esteem, and a high score indicates high levels of psychological well-being, high life satisfaction or positive self-esteem. This instrument was adapted to the local medium, obtaining results that indicate it has adequate psychometric qualities for this type of study: Cronbach’s a=0.75 (^{Stefani, 2002}); it was also chosen because it is easy to administer and to evaluate.

Procedure

Steps were taken to obtain authorization to interview older adults attending day-care centers and workshops reporting to the Government of the City of Buenos Aires and other areas of influence. A copy of the research project along with the informed consent was sent to the authorities of these institutions. Once the corresponding permits were obtained, the evaluation protocol was administered to older adults through an individual interview lasting approximately 40 minutes.

Statistical analysis

Percentages, arithmetic mean and standard deviation were calculated to describe study variables, according to their level of measurement.

According to the presence/absence of psychophysiological disorders related to retirement, a comparison was established between: a) total average score in the Index of Psychological Well-Being and its components (Life Satisfaction and Self-Esteem), by calculating the arithmetic mean, standard deviation and Student’s t-test, and b) the scores the subjects chose for items in the Life Satisfaction Index and Self-Esteem Scale, respectively, through multivariate analysis of variance.

Psychophysiological Disorders

Table 2 and 3 show calculated frequencies and percentages of psychophysiological disorders present in the sample. There we find that 59% of the sample did not report the presence of a psychophysiological disorder. In turn, those that reported new psychophysiological disorders related to retirement, said that 39% were “cardiovascular” and 27% were “musculoskeletal”, and that 82% of them took place “during the first year” of retirement and 71% occurred without a previous medical history of the disorder.

Table 2: Presence of psychophysiological disorders in the sample (n=150) 

Table 3: Description of psychophysiological disorders (n=62) 

Psychological Well-being

Arithmetic means and standard deviations were calculated for the sample of total scores obtained in the Psychological Well-Being Index and its subscales. The total sample exhibited a psychological well-being average score (m=1.52; sd=.32), that is reflected in life satisfaction (m=1.33; sd=.39) and positive self-esteem (m=1.73; sd=.33), indicating a favorable trend.

Psychological Well-being according to psychophysiological disorders

Table 4 presents means and standard deviations of total scores in the Psychological Well-Being Index for those who expressed absence or presence of psychophysiological disorders, as well as the results of the Student’s t-test to compare both groups. In global terms, we can observe (considering all index items and their respective subscales) a statistically significant difference for psychological well-being (t ₍₁₄₈₎₌ 3.462; p=.001), indicating that those that did not develop psychophysiological disorders present, with a higher probability, higher scores of psychological well-being than those who got sick (m ₁ =1.60 sd ₁ =.30 vs. m ₂ =1.42 sd ₂ =.34).

Table 4: Psychological well-being and its components, according to presence of psychophysiological disorder 

Tables 5 and 6 show the means and standard deviations of scores obtained in the Life Satisfaction Index and the Self-Esteem Scale respectively, for subgroups conformed by those who stated either absence or presence of psychophysiological disorders. Also, each table displays the results of the multivariate analysis of variance (MANOVA), considering the items that make up these questionnaires both as a whole and separately.

Regarding the first analysis in MANOVA, results indicate a significantly different pattern of responses in the Life Satisfaction Index and the Self-Esteem Scale according to absence or presence of psychophysiological disorders (Life satisfaction: T_Hott=.144; FHott_(11;138)=1.812; p=.05; Self-Esteem: T_Hott=.255; FHott_(10;139)=3.539; p=.000). Therefore, said MANOVA analysis confirms results previously obtained regarding the Psychological Well-Being Index, in that those interviewees that stated not having developed psychophysiological disorders related to retirement scored higher in life satisfaction and self-esteem.

When considering each item separately, resulting “F” values in MANOVA show a pattern where the means in the group that did not present psychophysiological disorders are generally higher than those in the group that did present psychophysiological disorders. Also, statistically significant differences can be observed in some items of the Life Satisfaction Index (Table 5: in 6 out of 11 items) and of the Self-Esteem Scale (Table 6: in 6 out of 10 items), which would explain the global differences described before.

As can be observed in table 5, older adults that reported an absence of psychophysiological disorders tend to present higher scores in life satisfaction, compared to those who reported a presence of psychophysiological disorders; they state to a greater degree not feeling old and tired (m ₁ =1.33 vs. m ₂ =.85), not feeling that what they do is so boring or monotonous (m ₁ =1.53 vs. m ₂ =1.19), feeling as happy as when they were young (m ₁ =1.26 vs. m ₂ =.97), not feeling that their life could be happier than what it is now (m ₁ =1.00 vs. m ₂ =.69), feeling that these are the best years of their life (m ₁ =1.11 vs. m ₂ =.85), and feeling that the things they do are as important as they ever were (m ₁ =1.76 vs. m ₂ =1.56).

Table 5: Life Satisfaction Index, according to presence of psychophysiological disorders 

In addition, table 6 indicates that those older adults that did not develop psychophysiological disorders, compared to those who did, generally tend to obtain higher self-esteem scores, reporting a belief that they: possess several positive qualities (m ₁ =1.93 vs. m ₂ =1.63), do things as well as most other people their age (m ₁ =1.84 vs. m ₂ =1.52), have a lot to be proud of (m ₁ =1.64 vs. m ₂ =1.39), show a more positive attitude towards themselves (m ₁ =1.73 vs. m ₂ =1.50), can do a job well (m ₁ =1.86 vs. m ₂ =1.71) and, lastly, that their life is useful (m ₁ =1.65 vs. m ₂ =1.44).

Table 6: Self-Esteem Scale, according to presence of psychophysiological disorders