Changes in older people’s quality of life in the COVID-19 era: a population-based study in Finland

This study utilized data from the FinSote National survey (www.​thl.​fi/​finsote) conducted in 2017–2018 and 2020–2021 in the Finnish Institute for Health and Welfare. The FinSote National survey is a nationally representative study assessing health, wellbeing and service use of the Finnish population by different population groups and regions [21]. The survey is based on a random sampling design, which is stratified by age and region and drawn from the Digital and Population Data Services Agency. In the sampling phase, those persons who have been included in the sample in previous years are excluded, and therefore, the study samples do not contain the same individuals across years. Data are collected from three age groups, 20–54, 55–74, and + 75-year-olds, by postal and online questionnaires, which are available in four different languages (Finnish, Swedish, Russian, and English). Data are collected every other year. In 2017–2018, the data were collected between October 2017 and March 2018, and in 2020, between September 2020 and February 2021. Hereinafter, we use 2018 and 2020 to describe the time points accordingly to the year in which most of the responses were received.

In the present study, we report results only for participants aged 75 years or older. In 2018, totally 9781 people belonging to this age group responded to the survey, rendering a response rate of 57%. In 2020, the respective numbers were 9919 and 59%.

Quality of life was assessed in 2018 and 2020 with the validated EUROHIS-QOL-8 scale, which is a shortened 8-item version of the WHOQOL-BREF [22, 23]. The items forming a one-factor structure are: (1) How would you rate your quality of life? (2) How satisfied are you with your health? (3) Do you have enough energy for everyday life? (4) How satisfied are you with your ability to perform your daily living activities? (5) How satisfied are you with yourself? (6) How satisfied are you with your personal relationships? (7) Have you enough money to meet your needs? and (8) How satisfied are you with the conditions of your living place? The items are answered by using individualized five-point Likert scales (e.g., from “not at all” to “completely”) with higher scores indicating a more positive response. In the present study, we utilized the EUROHIS-QOL-8 index, which is calculated by summing all the individual item responses and then dividing the sum by the number of responded items. The index ranges between 1 and 5 with higher scores indicating greater QoL.

Quality of life-related factors included keeping in touch with friends and relatives, feeling lonely, feeling optimistic about the future, daily exercise, sleeping difficulties, and alcohol use. These items were assessed in 2020 from about 80% of the participants (while the remaining 20% received questions related to digitalization). Changes in QoL-related factors were assessed with the question “Have the coronavirus pandemic or the subsequent restrictive measures affected your everyday life?” and the response options were “No influence”, “Yes, decreased”, “Yes, increased” and “Not applicable” [21]. Since the last response option was guided to be chosen only in case if the item was not a part of the participant’s life at all, we decided to exclude the “Not applicable” responses from the analyses (see Table 3). Further, we converted the remaining response options to reflect a positive, negative, or no change.

We used sex, age group, educational level, urbanization of place of residence, economic deprivation, living alone, IADL difficulty and use of Internet as factors representing different population groups. All these variables were categorized into two categories. Age, sex and living area (urbanization) were drawn from the Digital and Population Data Services Agency. Age was categorized into 75–84 vs. + 85 years old. Urbanization was initially trichotomous describing people who live (1) in the city, (2) in an urban area, or (3) in the countryside, but we subsequently combined the first two categories. Educational level, economic deprivation, living alone, IADL difficulty, and use of Internet were self-reported and collected with the postal questionnaires. Educational level was calculated based on self-reported years of full-time education and categorized initially into three: low education (approximately 0–8 years), moderate education (approx. 9–12 years) and high education (approx. 13 years or more). We then combined the moderate and high education categories. Economic deprivation was assessed with a question “Have you within the past 12 months ever…” including three items: (1) feared that you will run out of food before you can get money to buy more, (2) been unable to buy medicines because you did not have any money, and (3) not visited a doctor because you did not have any money? Response options were “yes” and “no”, and people who responded “yes” to at least one of these three items were categorized as reporting economic deprivation. Living alone was assessed with a question with two response options “yes” and “no”. Difficulties in instrumental activities of daily living (IADL) were assessed with four individual items: (1) light housework (vacuum cleaning, washing dishes, making beds, doing laundry, etc.), (2) minor repairs around the home (replacing a light bulb or a smoke alarm battery, etc.), (3) day-to-day financial transactions (paying bills, withdrawing cash, etc.) and (4) shopping for food. Response options were “yes, with no difficulty”, “yes, with some difficulty”, “yes, but with great difficulty”, “no, I cannot”. Persons who chose either of the last two response options at least to one of the four items were categorized as having IADL difficulty. Finally, Internet use was assessed with a question “Do you use Internet for the following: e-services (e.g., My Kanta, MyTax, the Social Insurance Institution of Finland [Kela])”. Response options were “I use it independently”, “I use it with another person’s help or someone else uses it on my behalf” and “I don’t use”. People who reported using Internet independently or with the help of another person were categorized as using Internet.

All analyses were conducted with the IBM SPSS Statistics version 27. In the analyses, we utilized the complex samples method and weight coefficients for the random sample to represent the whole Finnish older population. The differences in mean QoL scores between 2018 and 2020 were tested with t test. Effect sizes were calculated using Cohen’s D to communicate the practical significance of results obtained from the t test [24]. The Kolmogorov–Smirnov test was used to test the normality of the distribution in the QoL scores. Differences in self-evaluated changes in QoL-related factors during the COVID-19 pandemic, i.e., keeping in touch with friends and relatives, feeling lonely, feeling optimistic about the future, daily exercise, sleeping difficulties and alcohol use, were analyzed using crosstabs and tested with chi-square test. All analyses were stratified by sex, age group, educational level, urbanization, economic deprivation, living alone, IADL difficulty and Internet use.

As Table 2 shows, the mean QoL index score among all study participants was around 3.7 (range 1–5) in 2018 and 3.8 in 2020, indicating a slight increase between these 2 years (p < 0.001 obtained from t test). Similar increases were observed for all population groups, except for those reporting economic deprivation. Among those who reported economic deprivation, QoL was at a lower level to start with and over time, decreased a little (p = 0.036). Despite the statistical significance in all models, the effect sizes calculated using Cohen’s D ranged from 0.11 to 0.25 [23].

Changes in quality of life related factors were assessed with the question “Have the coronavirus pandemic or the subsequent restrictive measures affected your everyday life?” Majority of respondents reported having less interaction with friends and family due to the COVID-19 pandemic (Table 3). Negative changes were more often reported among women than men, those with higher education compared to those with lower education, those living with someone versus living alone, those having a good financial situation compared to those who suffered from economic deprivation, and those who used Internet to run errands versus those who did not (Fig. 1; p < 0.001 for all except for economic deprivation, for which p = 0.03, obtained from chi-square test). Almost half of the respondents reported also having felt lonelier during the pandemic (Table 3). Feelings of loneliness were reported more often among women, those over 85 years old, those who lived alone, those encountering economic deprivation, not using Internet to run errands and those manifesting severe IADL difficulties (Fig. 1; p < 0.001 for all, except for Internet use, for which p = 0.007, obtained from chi-square test). For feeling optimistic about the future, not as many negative changes were reported. Instead, about half of the respondents perceived that COVID-19 pandemic did not affect their feelings about the future and about a fifth reported having increased optimism (Table 3). However, negative changes were more often reported among women, those encountering economic deprivation, those with higher education, living alone, and using Internet to run errands (p < 0.001 for all except for living alone, for which p = 0.002, obtained from chi-square test).

Although most respondents felt that the pandemic had no effect on their sleeping, one fifth still reported that sleeping difficulties and nightmares increased due to the pandemic (Table 3). Negative changes were more commonly reported by women, those encountering economic deprivation, those over 85 years old, those living alone, those with severe IADL difficulties and those not using Internet to run errands (Fig. 2; p < 0.001 for all, except for age and Internet use, for which p = 0.006 and p = 0.002, respectively, obtained from chi-square test). For daily exercise, both increases and decreases were observed: while slightly more than one fourth had decreased the amount of daily exercise, more than one in a ten had increased it (Table 3). Negative changes were more commonly reported by women, persons over 85 years old, those living alone, those manifesting severe IADL difficulty, those encountering economic deprivation, and those having higher education (Fig. 2; p < 0.001 for all except for education, for which p = 0.01, obtained from chi-square test). Also, those living in an urban or semi-urban area more often reported on decreased exercise levels due to the pandemic than those living in a rural-like area (p < 0.001, obtained from chi-square test). For alcohol use, nearly all participants reported that the pandemic had no effect on it (Table 3). The few negative changes were more commonly observed for men than women (Fig. 2; p < 0.001, obtained from chi-square test).

The present study was a large population-based, nationally representative survey, which included data on many socioeconomic and social background factors and allowed analyzing different population groups separately. Overall, there is little information from nationally representative samples on how the COVID-19 pandemic has affected to the quality of older people’s life, taken different population groups into account. The oldest old (over 85 years; 28% in the weighted sample) were well-represented within the study compared to many earlier studies on older adults. All analyses were conducted using weighted coefficients, so the data were nationally representative, and the findings generalizable to the whole Finnish older population, and likely to other western countries with similar COVID-19 situation. While the existing COVID-19 literature on older people’s QoL has focused on reporting the situation amid the first wave of the pandemic (e.g., [6, 17, 26]), the present study could utilize data collected at a time when the control measures had been in place for over a half a year to a year.

The major limitation in our study, however, is that it is a repeated cross-sectional survey and not a longitudinal investigation. Hence, although we could compare QoL-scores from year 2018 and 2020 and speculate the plausible changes, it is impossible to draw conclusions on causality and on the primary factors explaining the changes. Secondly, the EUROHIS-QoL 8-item index used here is composed of eight items representing psychological, physical, social, and environmental aspects of live. The concept of quality of life is multidimensional in nature, and thus the unidimensional nature of the index may not reveal a change in a specific area of life. The EUROHIS-QoL 8-item index is, however, recommended for use in public health research and has a good validity [40, 41]. In addition, as the data were collected with postal and online questionnaires, it is possible that self-reporting bias, such as social desirability bias and recall bias [42], have affected the results to some extent. On the other hand, having these two data collection methods has likely diminished the possibility that older people who use Internet, and perhaps are in better function, are over-represented. Concerning statistical analysis, it should be noted that we did not make corrections for the p values although many tests were conducted. However, given the large sample size and p value being < 0.001 in majority of the tests, this is not likely to change the results. Finally, although the study was population-based, it must be noted that older people with the poorest health and function, e.g., with an advanced cognitive impairment, were not well presented in the current study and hence, the findings of this study may not be generalizable to them.