Design and settings

This cross-sectional study took place between 2010 and 2016. As part of a project, a study was conducted on the quality of life and health status of adults. The study protocol was approved by the Commission for Ethics of Scientific Research of the University of Health and Sport Sciences in Wroclaw (2009, amended in 2015). The measurements took place in the Laboratory of Biokinetics Research having the Quality Management System Certificate -PIV-EN ISO 9001: 2009 (No. PW 48606-10E). The study complied with the ethical requirements for human experimentation according to the Declaration of Helsinki. Recruitment for the study began on October 5, 2009, and ended on September 30, 2016. Participants were informed about the purpose, methods of the study, the procedures used, and the possibility of withdrawing from the study at any stage. All subjects who participated in the study provided written informed consent. The study did not include strictly medical questions, revealing confidential characteristics about the subjects’ health. The research team assured each participant of the confidentiality of the data. The research was retrospectively registered on the ISRCTN platform as 18225729.

Subjects

Our research included a voluntary sample of older adults aged 60‒90 from the Lower Silesian Voivodeship in southwestern Poland. Participants volunteered for the study advertised in the local media. Invitations were also sent to senior citizens’ centers. A total of 1562 people registered for the study. Of this group, 340 people did not meet the age criterion, 11 people had medical contraindications, and 103 people did not complete the study. In the end, the study group comprised 1108 people, including 314 men and 794 women, aged between 60 and 90 years.

Inclusion criteria included (1) age 60–90 years, (2) ability to move independently, (3) no medical contraindications, (4) good verbal contact, and (5) voluntary written consent to participate in the study. The ability to move independently and the absence of medical contraindications were essential because the subjects also performed a physical fitness test, the results of which have been published in other papers. Exclusion criteria included (1) cancer, acute trauma, and infection, (2) febrile conditions, (3) recent myocardial infarction, (4) other medical contraindications to physical exercise, and (5) lack of consent to participate in the study.

Methods

The Polish version of the WHOQoL-BREF questionnaire [27,28] was used to assess quality of life. The scale contains 26 questions. The first two, concerning subjective assessment of quality of life and health assessment, are assessed separately. The following questions are grouped into four domains – somatic, psychological, social, and environmental. The somatic domain includes feeling pain, the need for medical treatment for everyday functioning, having energy, and being satisfied with sleep, ability to perform your daily living activities, and capacity to work. The psychological domain concerns enjoying life, a sense of meaning in life, concentration, acceptance of bodily appearance, satisfaction with oneself and experiencing negative feelings. The social domain includes satisfaction with personal relationships, sex life and support from friends. The environmental domain includes a safety, a healthy environment, having money, availability of information, opportunities for leisure activities, and satisfaction with housing conditions, health services, and transport. Each question is scored on a 5-point scale from 1 to 5, where 1 is the lowest score and 5 is the highest. The exceptions are two questions from the somatic domain concerning pain feeling and the need for medical treatment, and one question from the psychological domain concerning the experience of negative feelings, where the coding is reversed. Domain scores are determined by calculating the arithmetic mean of the questions contained in each domain. The higher the domain score, the higher the quality of life. The raw scores were transformed according to the 0–100 key to correspond to the WHOQoL-100 scale scores [27]. The WHOQOL-BREF has been shown to have good to excellent psychometric properties of reliability and performs well in preliminary tests of validity [28]. According to Jaracz et al. [29], who measured the validity and reliability of the WHOQoL scale in the Polish population, a high validity ranging between 0.62–0.76 for the physical domain, 0.55–0.78 for the psychological domain, 0.68–0.85 for the social domain, and 0.58–0.68 for the environmental domain was found. Acceptable internal consistency was shown, with Cronbach’s alpha coefficients being greater than 0.7 for all domains except for the social domain [29].

Patients also completed a short form regarding age, gender, marital status (the response ‘in a relationship’ identified those who were married or living in a civil partnership), number of children born (for women), education, number of chronic diseases, as well as cigarette smoking and alcohol consumption. Smoking and alcohol consumption were recorded as binary variables (yes/no), without further quantification. Information about self-assessed health was collected via self-reported questionnaires, with participants provided assistance.

The Shapiro-Wilk test verified the distribution of quantitative variables (S1 Table). Due to the lack of normality of distribution for all analyzed variables, non-parametric methods were used. For quantitative variables, the median and quartile deviation (Med ± SQ) assessed with the Mann-Whitney U test. The percentage frequency factors were calculated for the categorical data and the χ2 test of independence was performed; interactions between all categorized variables were checked using log-linear analysis. To assess the differences in quality of life parameters depending on social factors, a difference analysis was performed using the Mann-Whitney U test if the social grouping factor had two categories, or the Kruskal-Wallis test if the grouping factor had more categories. Relationships between quantitative variables such as age, the number of chronic diseases, and the quality of life assessment were assessed using Spearman's rank correlation coefficient (ϱ). The statistical significance of the results was assumed at p < 0.05. All analyses were performed using TIBCO Statistica® 13.3.0 (StatSoft Poland).

Characteristic of participants

The quality of life of the respondents (57%) was good. Older age was recorded in men (68.4 ± 3.8 years) than in women (65.6 ± 3.8 years), in unmarried people (67.2 ± 4.3 years) than in those living in a relationship (65.8 ± 3.8 years), in those with higher education (67.5 ± 4.2 years) than with secondary education (65.5 ± 3.8 years), and in those rating their health status neutrally rather (67.2 ± 4.6 years) than well (65.6 ± 3.4 years) and not smoking (66.5 ± 3.8 vs. 64.6 ± 3.4 years).

The somatic domain was differentiated by all other factors in addition to gender. Higher values in this domain were obtained by participants living in a relationship (75.0 ± 9.0 vs. 69.0 ± 9.0, p = 0.034), with higher education (75.0 ± 6.0 vs. 69.0 ± 9.0, p < 0.001), and rating their health status as good (81.0 ± 6.0 vs. 66.0 ± 3.0 and 50.0 ± 9.5, p < 0.001). Those who consumed alcohol (75.0 ± 6.0 vs. 69.0 ± 9.0, p = 0.005) and/or smoked cigarettes (81.0 ± 12.5 vs. 75.0 ± 9.0, p = 0.027) scored significantly higher than others in this domain.

Education, health status, and alcohol consumption differentiated the psychological domain in a similar way to the somatic domain. Additionally, men achieved higher values in psychological domain than women (69.0 ± 9.5 vs. 63.0 ± 6.5, p < 0.001).

The social domain was differentiated only by marital status and health status, indicating higher results in this domain among people in a relationship (75.0 ± 12.5 vs. 69.0 ± 9.5, p < 0.001) and good health (75.0 ± 6.0 vs. 69.0 ± 9.5 and 56.0 ± 15.5, p < 0.001).

In the environmental domain, higher scores were recorded in men than in women (69.0 ± 6.0 vs. 63.0 ± 9.5, p = 0.004), in people in a relationship than in single people (69.0 ± 6.0 vs. 63.0 ± 9.5, p < 0.001), and in people with higher education than with secondary and primary education (69.0 ± 6.0 vs. 63.0 ± 9.5 and 63.0 ± 6.5, p < 0.001). Respondents in good health had a better quality of life in all domains (Table 1, S2 Table).

Download:

• PNG
  larger image
• TIFF
  original image

Table 1. Characteristics of participants and differences in quality of life assessments (median ± interquartile deviation).

https://doi.org/10.1371/journal.pone.0349206.t001

Basic statistics of women and men can be seen in S3 –S5 Tables.

Main results

The relationships between the categorized variables were checked using log-linear analysis. Only two-way interactions were found to be significant for gender, marital status, education, health status, alcohol consumption, and cigarette smoking. No higher-order interactions were demonstrated. The statistics of the most favorable log-linear model are presented in Table 2. The frequencies estimated based on the obtained model did not differ significantly from the observed frequencies (p = 0.914).

Download:

• PNG
  larger image
• TIFF
  original image

Table 2. Assessment of partial and marginal relationships for the model including interactions between gender, marital status, education, health status, alcohol consumption and cigarette smoking.

https://doi.org/10.1371/journal.pone.0349206.t002

The prevalence of each category of marital status, education, and alcohol consumption was associated with gender (Table 2). Detailed analysis of these interactions (results not reported in the table due to divisions according to many categorical variables, which reduces their readability) revealed that 81% of men and 49% of women were married/partnered (χ2 = 91.97, p < 0.001), more than half of men (55%) had a higher education, while more than half of women (53%) had a secondary education (χ2 = 30.88, p < 0.001). Those with less than secondary education comprised 9–10% of both gender groups. Alcohol consumption was observed by twice as many men as women (31% vs 14%, χ2 = 42.07, p < 0 .001), while cigarette smoking was identified in 6–7% of participants regardless of gender. Furthermore, a significant interaction between alcohol consumption and smoking was observed, with 77% of participants reporting neither alcohol consumption nor smoking, and 2% using both (χ2 = 13.92, p < 0.001). The majority of participants rated their health status as good irrespective of gender (poor health was indicated by only 4% of participants). Gender did not differentiate the number of chronic diseases present, with only 7% of participants indicating no chronic diseases and one in four participants indicating the presence of at least five different chronic diseases.

The relationship between health assessment and marital status and education was such that good health status was more frequently reported by those in a relationship than those who were single (61% vs 52%, χ2 = 9.27, p = 0.010) and by those with higher education than those with secondary and primary education (63% vs 54% and 43% respectively, χ2 = 19.03, p = 0.008).

Of the women surveyed, 92% were mothers, of whom 31% had 1 child, 53% two, and 16% three or more. The number of children correlated with current marital status (χ2 = 35.61, p < 0.001) and education (χ2 = 75.57, p < 0.001). Having two or more children was more common among women in a relationship (57%), while having one or no children was more common among single women (60–75%). More women with a higher education (45%) compared to the rest (34%) had no children or only one child. The number of children was not associated with self-rated health or substance abuse.

Spearman’s rank correlation coefficient was used to calculate associations between subjective quality of life scores in various domains and participants’ actual socio-demographic conditions. Only the number of chronic diseases and somatic domain scores correlated very weakly with age. The number of chronic diseases was negatively associated with quality of life, health scores, and all domains, with the somatic domain being the strongest and the social domain the least strong. Quality of life and health scores were positively correlated with each other and all domains. The strongest correlations were between health assessment with the somatic domain and the psychological domain with the environmental domain (Table 3).

Download:

• PNG
  larger image
• TIFF
  original image

Table 3. Spearman’s rank correlation coefficients (ϱ).

https://doi.org/10.1371/journal.pone.0349206.t003

Strength of the study

The study covers a large group of older adults, so the results obtained can be generalized to the area of southwestern Poland. The study was conducted in person (not via the Internet) with all the people surveyed by the same qualified team of researchers, which makes it possible to assess actual quality of life. The study used standardized questionnaires to assess quality of life, and therefore the results can be compared to studies by other authors in Europe and worldwide.

Health professionals and policymakers consider the search for mediating mechanisms to be important from a public health perspective. Understanding the potential determinants of changes in quality of life has important implications for the design and implementation of good practices promoting healthy lifestyles among older people [45]. Our study highlights the importance of modifiable factors, such as health status, number of chronic diseases and level of education, for the quality of life of the older adult population.

Limitation

The study also has limitations. These include, among others, the cross-sectional nature of the research, which makes it impossible to track the dynamics of changes in involutionary processes. As this study is cross-sectional, causal relationships between social factors and quality of life cannot be inferred. The older adults surveyed came from one region of Poland (the southwestern area), which may not give a complete picture of the situation in the whole country. However, the fact that the study focuses on the quality of life of participants from one region, will allow us to compare the quality of life and social situation of older populations in other regions. The next limitation is that the sample did not include older people with mobility difficulties and some medical limitations, which could also affect the quality of life results. Additionally, since no random sampling was used and only people volunteering for the study were examined. It can be assumed that more physically and socially active people applied to participate. Also, the large number of women surveyed compared to men may distort the results. Therefore, the study sample may not reflect the entire population of older people from southwestern Poland. Future research on the quality of life of older women and men should also include people with reduced mobility and other medical limitations so that the group in the study can more accurately reflect the state of the older adult population in Poland. The study sample should be randomized. This will make the results obtained on quality of life more complete.