Quality of life is a complex construct that comprises many different aspects (Higgs et al., 2003; Walker & Lowenstein, 2009). Also the variables that constitute it may vary depending on the sample studied, the characteristics of the context where the participants live, the design, and the methodological peculiarities of the study (Smith et al., 2004). As a result there is a sizeable number of quality of life models: objective indicators models, subjective indicators models, need satisfaction models, psychological models, physical functioning models, social functioning models, environmental models, and idiosyncratic models (Bowling, 2004).

At the same time the overlap between the quality of life and subjective indicators of quality of life must also be underlined (Bowling, 2004; Bowling & Gabriel, 2007). These indicators, like life satisfaction for example, have also been used as criteria to measure successful aging. Thus, quality could be integrated into the successful aging paradigm, as another criterion for aging successfully (Bowling, 2004; Bowling & Iliffe, 2011).

According to Rowe and Kahn (1997) successful aging can be achieved when a person maintains an excellent health status and eludes illness; has a high rate of physical and mental activity, and is socially engaged. This coincides with the World Health Organization (WHO) proposal of active aging, which is defined as “the process of optimizing opportunities for health, participation, and security in order to enhance quality of life as people age” (WHO, 2002). Therefore, active aging emerges as a mean to achieve good quality of life in old age.

Some of the factors that are directly related to quality of life coincide with the determinants of active aging. These variables can be categorized into three different groups: a biomedical category that includes health and functional status; a psychological group comprising cognitive functioning, emotional functioning, and personality; and a social one that comprehends social relations, social support, social engagement, the environment, and its characteristics.

Physical health and functional status have a major impact on quality of life (Blane, Netuveli & Montgomery, 2008; Bowling, 2004; Bowling & Windsor, 2001; Low & Molzahn, 2007; Michalos, Hubley, Zumbo, & Hemingway, 2001; Walker & Lowenstein, 2009). Some health measures that influence the functional status, like the respiratory system or body mass index are associated with quality of life (Blane et al., 2008), especially if said functional status is impaired, then it negatively affects quality of life (Netuveli, Wiggins, Hildon, Montgomery, & Blane, 2005). Subjective health is another important predictor (Smith et al., 2004) explaining up to 35 percent of the variance in quality of life (Michalos et al., 2001).

Among the psychological predictors of quality of life, cognitive functioning stands out as a relevant factor. For example, older people who experience an abrupt deterioration of cognitive capacities are more impaired when it comes to perform daily activities and tasks, and have a higher mortality risk compared to those with good cognitive functioning (Yaffé et al., 2010). Thus, an excellent cognitive state is essential for quality of life, autonomy, and productivity (Desai, Grossberg, & Chibnall, 2010). Likewise, emotional disorders and negative emotional states can be detrimental for quality of life, for example depression and loneliness are associated with low quality of life (Blane et al., 2008; Smith et al., 2004).

Social relations, social support, the resources available in the neighboring contexts, accessibility, satisfaction with the neighborhood, and social participation are also prominent determinants of quality of life (Bowling, 2004; Bukov, Maas, & Lampert, 2002; Ekström, Dahlin, & Elmstahl, 2008; Netuveli et al., 2005; Smith et al., 2004). Social contacts are a source of satisfaction and quality of life for older people (Cheng, Chan, & Phillips, 2004; Instituto de Mayores y de Servicios Sociales [IMSERSO], 2009), especially when they afford emotional support (Low & Molzahn, 2007). In addition, a social context that provides social interaction opportunities allows goal attainment and leisure activities which increases quality of life (Low & Molzahn, 2007). Being satisfied with the neighborhood and living in a context free of physical obstacles also contributes to high quality of life (Netuveli et al., 2005; Smith et al., 2004).

Numerous programs developed in the gerontological context were aimed at maintaining and improving cognitive functioning, or at preserving a good physical status. Meanwhile only a limited number of interventions include other aspects that have also been found to improve quality of life, like leisure activities (Cardenas, Henderson, & Wilson, 2009).

Regarding interventions that include cognitive or physical activities, review studies (e.g. Jean, Bergeron, Thivierge, & Simard, 2010; Tardif & Simard, 2011) highlight their utility, since they promote a good mental state, well-being, and quality of life. Nevertheless, there is also evidence that interventions that combine both cognitive stimulation and physical activity have a stronger impact on both cognitive and emotional states than single-component programs (Fabre, Chamari, Mucci, Massé-Biron, & Préfaut, 2002; Rey, Canales, & Táboas, 2011; Valencia et al., 2008). Consequently, it is necessary to develop interventions that include physical, cognitive, and leisure activities (Cardenas et al., 2009; Hui & Rubenstein, 2006).

Promoting health and quality of life among rural older people has received little attention (Milne, Hatzidimitriadou, & Wiseman, 2007). In Spain, rural aging has not been studied in depth (Monreal et al., 2009; Triadó, Villar, Solé, & Osuna, 2003) despite the fact that, as Rodríguez-Rodríguez (2004) states, aging in a village or in a city is essentially different.

The unbalanced distribution of resources offered to older people that live in rural and urban areas is noteworthy in Spain. While the elderly from urban contexts are provided with more resources and have more possibilities to be socially engaged and to take part in leisure activities, rural older adults count on a limited offer of this kind of activities (García-Sanz, 1997; Rodríguez-Rodríguez, 2004; Wilcox, Bopp, Oberrecht, Kammerman, & McElmurray, 2003). As a consequence, older adults from rural areas have a reduced probability to age successfully.

Conversely, Spanish aging polices do not give specific recommendations about older people from rural areas and the number of interventions designed specifically for them is sparse (García-González & Rodríguez-Rodríguez, 2005). Therefore, it seems that intervention programs that aim at maintaining quality of life in Spanish rural contexts are particularly necessary. Rodríguez-Rodríguez (2004) states that these kind of programs should focus on two aspects: on the one hand, promoting active aging and, on the other hand, fostering proximity services that provide formal support to older adults with high disability rates and to their caregivers.

Until recently, the quality of life intervention programs developed in rural areas in Spain (García-González & Rodríguez-Rodríguez, 2005; Monreal & Vilà, 2008) have specifically concentrated on community social intervention; one of the programs we identified aimed at establishing a reference centre for social intervention in a rural area (Monreal & Vilà, 2008), and the other one (Rodríguez-Rodríguez, 2004) had the goal to create new services and resources in rural communities, such as accessible transport, delivery of technical aids, and the establishment of older adults associations that would develop cultural activities. None of them included aspects related with participants’ physical or psychological functioning. For this reason, we acknowledge the need to design an intervention program that contemplates physical status, cognitive function, and leisure activities.

Considering all it has been previously discussed, the goal of this study is to analyze the effectiveness of an intervention program aiming at improving quality of life through cognitive stimulation, physical, and crafts activities in a group of community-dwelling older adults living in a rural context in Orense, Galicia, Spain.

The sample comprised 86 older adults (78 participants in the intervention group; 8 participants in the control group) aged 65 and older from 11 villages of a rural area with high rates of depopulation and aging in Orense, Spain. The control group was considerably reduced due to the terms and conditions of the sponsoring organisms that could only provide finance for the intervention program to be implemented for a limited period of time.

This research was approved by the committee and head of the foundation that sponsored the study, Fundación Barrié de la Maza, and the village councils which collaborated in the study. Also, prior to the start of the intervention the participants gave their informed consent.

Demographic characteristics included gender, age, educational level, income, and living arrangements.

Mini-Examen Cognoscitivo (MEC) (Lobo, Esquerra, Gómez Burgada, Sala, & Seva, 1979) Spanish version of Mini-Mental State Examination comprises 11 items that screen cognitive impairment by assessing 5 cognitive areas: orientation (temporal and spatial), attention and calculation, word recall, language, and visuospatial abilities. The maximum score for the Spanish version that has been used in this research is 35 points and any score lower than 24 point suggests cognitive impairment, specifically mild (23-21 points), moderate (20-11 points) and severe (≤ 10 points). According to Lobo et al. (1999) the MEC correlates with the Wechsler Adults Intelligence Scale (WAIS) and with clinical examinations of intelligence.

Clock Drawing Test (CDT) (Cacho, García, Arcaya, Vicente, & Lantada, 1999) is a widely used instrument that screens cognitive impairments. It is divided in two parts, first, the patient is asked to draw a clock with the dial, all the numbers, and set the hands at ten past eleven. After that, the subject has to copy the drawing of the clock. Normal cognitive functioning is assumed if scores are 6 or higher for the first part and of 8 or higher for the copy.

WHOQOL-BREF (WHOQOL Group, 1998) is an abbreviated version of the WHOQOL-100 that comprises a total of 26 Likert items with 5 response options (from 1= nothing to 5=completely). The items are grouped into four domains: physical health, psychological health, social relationships, and environment. The WHOQOL-BREF has a correlation of 0.90 withthe original version of 100 items and has good discriminant, content validity, and test-retest reliability. In the present study, the Spanish version (Lucas, 1998) showsgood reliability (α =0.80 to 0.82).

The cognitive stimulation workshop was designed to promote good cognitive functioning and included the following activities: calculation, attention, reasoning, spatial orientation, time orientation, linguistic-factors related exercises, memory, seriation, planning and categorization tasks, and information on healthy habits that prevent cognitive impairment.

The crafts workshop was aimed at maintaining and improving dexterity and manual skills, and at encouraging creativity and promoting social interaction. The group of elderly met twice a week to decorate pottery and recycled objects, and to make jewerly.

Exercise workshop was aimed at improving the well-being and physical health of participants, enhancing mobility, and autonomy. Each session began with warm up and stretching exercises, and continued with the activities and exercises chosen for that session. At the end of the session the intensity of the activity was reduced and stretching exercises were performed again. Seven different exercise routines were designed that were cyclically repeated until the intervention program finished. These exercises consisted of aerobics, fitness, dumbbell exercises, mate exercises and stretching, choreography using sticks and gymnastics balls.

The pre-test assessment was conducted during January 1^st 2011 using the questionnaire on demographic characteristics, the MEC, the CDT, and the WHOQOL-BREF. The intervention program was developed between March and December 2011, four hours a week for each workshop. After finishing the intervention program the MEC, the CDT, and the WHOQOL-BREF were administered for the post-test assessment. The assessment of each participant lasted between 45 and 60 minutes, and data collection and implementation of the intervention program were conducted by previously trained staff.

A cross-sectional correlational of non-equivalent groups design was used. Preliminary analyses were performed to test differences between the control and the intervention group with respect to their demographic characteristics and the pretest scores for the MEC, CDT, and the WHOQOL-BREF. Also, the effect of age and educational levels on the MEC, the CDT, and the WHOQOL-BREF dimensions were analyzed using T-tests and analysis of variance (ANOVA). In order to determine if there were significant differences between the intervention and the control group in posttest WHOQOL-BREF, MEC, and CDT regardless of covariates, several analysis of covariance (ANCOVA) were conducted. Changes in quality of life and cognitive functioning (i.e., improvement or deterioration) were considered significant if p < 0.05.

Table 1 shows the means and standard deviations of the study variables and the demographic characteristics for the control and the intervention group. In the intervention group the majority of the participants were female while in the control group the majority was male. The mean age of the participants in the intervention group was 68.77 ± 8.97 while the mean age for the participants in the control group was 76.29 ± 4.54, these differences in the mean age were statistically significant (t(83) = 2.183, p < 0.05, η2p = 0.05). Similar educational level and pre-test scores were observed for the participants in both the control and the experimental groups in the MEC, the CDT, and WHOQOL dimensions, as no statistical significant differences were found between them (p > 0.05).

TABLE 1 TABLE 1 Descriptive statistics for study variable by group Source: own work.

Note: MEC: Mini-Examen Cognoscitivo (Spanish version of Mini-Mental State Examination); CDT: Clock Drawing test; General QL: WHOQOL-BREF Item 1; *p < 0.05; **p < 0.01; *** p < 0.001.

Preliminary analysis revealed that educational level was associated with statistically significant differences in the pre-test MEC score (F(3) = 121.139, p < 0.001, η2p = 0.31); participants without formal education scored higher than those with primary incomplete education (p < 0.01) and participants with complete primary education scored higher than those with incomplete primary education (p < 0.001). Statistically significant differences were also found for the pre-test CDT score by educational level (F(2) = 11.262, p < 0.001, η2p = 0.21) between the highest scoring group with complete primary education and the lowest scoring group of participants with incomplete primary education (p < 0.001), and between participants without formal education and those with primary incomplete education (p < 0.05).

Educational level was also associated with statistically significant differences in the WHOQOL-BREF physical health domain (F(2) = 4.279, p < 0.05, η2p=.09). Participants with complete primary education scored significantly higher than those with incomplete primary education (p < 0.05). Also, educational level was linked with statistically significant differences in the WHOQOL-BREF environment dimension (F(2) = 3.357, p < 0.05, η2p = 0.08), participants without formal education scored significantly higher than those with incomplete primary education (p < 0.05).

Similarly educational level had a statistically significant effect on MEC post-test (F(2) = 10.573, p < 0.001, η2p = 0.24), participants with incomplete primary education obtained the lowest score when compared with participants without formal education (p < 0.05), and with those with complete primary education (p < 0.001). Statistically significant differences were also found for the post-test CDT score by educational level (F(2) = 3.46, p < 0.05, η2p = 0.09) between participants with complete primary education, who obtained the highest score, and those with incomplete primary education who obtained the lowest score (p < 0.05).

The ANCOVA analysis revealed that after removing the covariate influence of age, educational level, gender, and the pretest scores, statistically significant differences in the posttest scores were found between the participants in the intervention and the control group. Participants in the intervention group had significantly higher scores in posttest MEC (F = 5.853; p < 0.001), CDT (F = 4.141; p < 0.01), psychological health (F = 9.651; p < 0.001), and environment dimension (F = 15.010; p < 0.001). All ANCOVAs had very high power once covariate influence was removed.

After the nine months intervention program, significant differences in pre-test and post-test scores were found in the intervention group, both for the MEC (t(62) = -7.432, p < 0.001,η2p = 0.47) (see Figure 1) and the CDT (t(62) = -3.974, p < 0.001,η2p = 0.20), while in the control group the average scores were similar in the first and second measurement (see Table 1).

Figure 1.

* p < 0.001