The purpose of this paper was to investigate the size and the extent of social inequalities in quality of life and health behaviours in Limassol.
This is a cross-sectional survey using primary data.
Multistage sampling was performed in four stages - parishes and communities, neighbourhoods (specific streets as a starting point from all parishes), homes (households), and finally, people.
- For municipalities and communities, a stratified random sampling was used in terms of urban areas, community size and generally accepted socio-economic indicators, to ensure that municipalities/communities from the entire range of socioeconomic classification were selected. Specifically, parishes and communities were ranked in terms of the percentage of the population with university education, an indicator which is one of the most commonly accepted socio-economic indicators and was the only one available at the stage of sampling from the open access files of the Statistical Service of Cyprus.
- Neighborhoods: In order to select neighborhoods from the whole range of socio-economic scale as well as to achieve wider geographical coverage, it was decided to include streets from all the parishes of the city of Limassol.
- Households: The households that participated in the study were selected in a systematic way. In the city, the field researcher started from one end of the road and moved along its entire length, selecting every third-fourth house, on the left and right side of the road alternately. Where it was not possible to recruit 10 people by the end of the selected street (e.g. refusal to participate, no one is home, etc.), the researcher completed the sample with households from the wider neighborhood - which is defined in the case of study as all lanes beginning / ending or crossing the selected street until the next intersection. In the case of an apartment building, a similar practice of systematic sampling based on the floors and the apartments was applied, e.g. left apartment on the 1st floor, right apartment on the 3rd floor accordingly avoiding repeating the same pattern always.
- People: quota sampling was used at household level (50:50, male female alternately in each household). The medical history of each person who participated in the research was not asked, as long as the person could answer the questions in the questionnaire.When the researcher had the required numbered, she stopped, but she was trying to get people from the beginning, middle and end of the road.
The population of the city of Limassol was divided into quarters, depending on its socio-economic status and specifically on the percentage of the population with university education.
The estimation of sample was based on power analysis. The result size in this population group in the city of Limassol is f = 0.20-0.25 (the which corresponds to a moderate effect size). The sample size of 188-300 individuals provides 90% statistical power to detect a difference in statistical significance level of 5%. The minimum desired sample in urban areas was set at 450 people. The sample size ensures a similar level of statistical power to detect such a degree social gradient in the quality of life in both genders separately, at least in urban areas (225 men and 225 women). Finally, it should be noted that, due to the nature of the study, which is based on multi-stage random sampling of neighbourhoods/communities and households, it was taken into account that the sample size should also ensure the greatest possible geographical coverage and include neighbourhoods with diverse socioeconomic background. Therefore, in terms of the objectives of the study, it is preferable in such cases that the sample consists of as many neighbourhoods/communities as possible rather than selecting many people from a small number of neighbourhoods and communities, so that the sample is more representative of the whole range of socio-economic disadvantage, which is expected to be concentrated in the area. The minimum number of people per neighbourhood / area was set at 10 people.
Also, it should be mentioned that the specific sample size is considered satisfactory as it provides statistical accuracy of ± 5 percentage points for the 95% confidence interval in the estimation of percentages (eg smoking).
In order to ensure a satisfactory response rate, postal communication (distribution of an open letter to all homes on the preselected roads) informed prospective participants that a university researcher would visit them for a short interview in the next few days. Information was also given on the importance and contribution of the study with the request to participate.
- People aged 45 to 65, to include middle age adults as the economically active population who has completed their studies, has a family, has income, and therefore has been integrated into their own professional, income and social status. The choice of the 45-65 age group was based on two popular hypotheses, which seek to explain the mechanisms that lead to the effect of socio-economic status on health in old age. These hypotheses are: the "cumulative disadvantages / advantages" hypothesis and the "age as a leveler" hypothesis (Kingston et al., 2015). Health inequalities appear to be smaller at younger ages, wider at middle and early older ages, and smaller again at later ages (House, Lantz & Herd, 2005; Dannefer, 2003).
- People who can speak and read in Greek
- Permanent residents in Limassol (or permanent resident or residence in Limassol for 5 years)
Demographic / socio-economic characteristics and lifestyle characteristics
Variables related to personal characteristics, such as demographic characteristics (age, gender, marital status, area where they reside), socio-economic characteristics (level of education, annual income, occupation), as well as lifestyle characteristics (smoking, alcohol, physical activity).
The level of self-assessment of the individual's health was measured on a 5-point scale (Likert scale), ranging from excellent health to poor. Individuals were asked to evaluate their health as: 1= excellent, 2 = very good, 3 = good, 4 = moderate and 5 = poor.
SF 36 Questionnaire - Quality of Life
Quality of life questionnaire SF-36v2 Standard, Greek Version was used to measure quality of life of research participants. The questionnaire was created in 1992 , and is used in several countries for the self-esteem of the Quality of Life and comparing the health status of different population groups.
The SF-36 scale is a tool used to measure the health level of a population. Its basic attribute is the simultaneous measurement and assessment of the level of physical and mental health. The grid of 36 questions includes eight measurement scales consisting of questions that represent the most measured health dimensions. These scales are: physical functioning (PF), role physical (RP), bodily pain (BP), general health (GH), vitality (VT), social functioning (SF), role emotional (RE), and mental health (MH). The first four (4) dimensions make up the physical health, while other mental health of the individual. These eight scales are evaluated with a score ranging from 0-100 each, where 0 represents the minimum possible value and at the same time the worst health, while the maximum score of the scale, the value 100 is excellent health. Where a score of less than 50 this means that the person's health is below the average . The generality of the SF-36 questions allows the adaptability of the questionnaire to each group of the population, while the Greek translation, as well as the entire questionnaire have been tested in repeated surveys in the health sector in Greece [23, 24] and Cyprus [25, 26].
IPAQ- International Physical Activity Questionnaire short form
The International Physical Activity Questionnaire short form (IPAQ) is a popular and frequently used questionnaire, which was developed in the late '90s by a multinational working group, supported by the World Health Organization, to be used for comparative evaluation of physical population activity different groups and nationalities . The IPAQ-short is also used by the European Union (Eurobarometer), while it has been tested in many international studies and is characterized by high reliability and satisfactory validity .
Social status and indicators of socio-economic disadvantage
The social status of the individual will be measured by the level of education, occupation, income.
Educational measurement refers to the highest level of education an individual attended (None / Primary, Secondary-Lower, Secondary-Upper, Undergraduate, Postgraduate studies).
Regarding the occupation, because the Cyprus Statistical Service classifies occupations only in the field of employment- which does not concern graduation, the Standard Occupational Classification of United Kingdom was used. Each participant was categorized according to his / her occupation.
The income indicator refers to the classification of individuals in relation to the monthly family income and was assessed according to the Cyprus Statistical Office standard through categories in which each participant was asked to choose one.
Data collection procedure
For the data collection the door-to-door survey method was employed. The survey was oral, in the form of an interview. The interviewer asked all participants the same set of questions and completed all sections of the questionnaire on paper during the interview. The interviewer visited each address once. 450 residents aged 45-64 (50:50 gender quota) across 45 randomly selected neighbourhoods (10 systematic random sample per area) from different city quarters, stratified by population density and proportion of adult residents with tertiary education.
The data was collected wherever the participant wanted, whether it was inside the house or at the door. The questionnaire filling time was about 10 minutes.
The data collection was mainly done in the afternoons, as during these hours were more likely to be at home the target group. In case of refusal to participate, the frequency (number of people who said they did not want to participate) was counted, to have an overview of the response rate (losses).
An opinion was sought from the Cyprus National Bioethics Committee and a reply has been received that this investigation does not fall within the remit of the CNBC for further bioethical evaluation. The participants, who wished to participate in the research, gave oral consent. In addition, anonymity and confidentiality of information were ensured, as only the researcher, research assistants and supervising professors had access. The questionnaires were destroyed after analyzing the data.
Statistical analysis was performed using the statistical package IBM SPSS Statistics 23 and the significance level was set at p <0,05.
The mean value (M) and standard deviation (SD) were used to describe quantitative variables. Data concerning the International Physical Activity Questionnaire short form (I-PAQ) was entered in the IBM SPSS Statistics 23 Statistical Package according to the questionnaire’s guide . The import of SF-36 questionnaire data was done using the special software Quality Metric Health Outcomes Scoring Software 5.0, which was provided free of charge by OPTUM. ANOVA was used statistical analysis in order to see the difference in both the physical dimension and the mental dimension of the tool, for the social status indicators (education, income and occupation). Linear regression was also used to calculate the correlation coefficient of movement at the different levels of the variables in relation to the quality of life (2 dimensions). The regression analysis was repeated as above, including the gender and the gender interaction with the social status indicator (education, income, occupation), in order to investigate the differential correlation of each variable with the quality of life as gender.
For the variable smoking, 3 categories were initially created (smokers, ex-smokers and non-smokers). Then, for a better evaluation of the smoking habit, the variable Pack-years of smoking was created, which was calculated with the following equation: pack-year of smoking = Years that someone smokes or smoked × (number of cigarettes per day) / 20. For the rest (non-smokers) the value is 0.