Self-reported health among immigrants from the former Soviet Union : quantitative and qualitative studies in Sweden

Abstract

Aims: The general aim was to study self-reported health and its determinants in immigrants from the FSU in Sweden. Study I: to analyze (i) whether there is an association between being born in the FSU and other countries of the former Soviet Bloc and poor self-reported general health among immigrants in Sweden and (ii) whether this association persists after adjusting for demographic, socioeconomic, and migration-related variables. Study II: to study (i) whether self-reported mental health (defined as self-reported psychiatric illness and psychosomatic complaints) was poorer among persons born in the FSU and other countries of the former Soviet Bloc currently living in Sweden than among Swedish-born persons and (ii) whether the demographic, socioeconomic, and migration-related characteristics of respondents could explain this. Study III: (i) to explore the self-reported health of Russian-speaking immigrants from the FSU in Sweden and (ii) to describe the influence of the culture of origin and of acculturation into the host country on the selfreporting of health among these immigrants. The aim of Study IV was to explore the immigrants own rating of the extent of their acculturation (self-reported integration) as a source of information in contrast to an external (researchers ) measurement of acculturation.

Methods: Both quantitative (Studies I and II) and qualitative (Studies III and IV) research approaches were used. Studies I and II were based on eight pooled, cross-sectional random samples of permanent Swedish residents (including immigrants with permanent residence permits or Swedish citizenship who arrived in Sweden after 1944) interviewed within the Swedish Annual Level of Living Survey. Unconditional multivariate logistic regression was employed to estimate the odds ratios of poor self-reported health (both general and mental) after the stepwise and simultaneous adjustment for demographic, socioeconomic and migration-related variables. Study I included men and women aged 25 to 84 (N = 36,084) born in Sweden (n = 35,711), Poland (n = 158), other East European countries (n= 161), or the FSU (n = 54) interviewed between 1993 and 2000. Study II included men and women (N = 35,844) aged 25 to 84, born in Sweden (n = 35,459), Poland (n = 161), other East European countries (n = 164), or the FSU (n = 60), interviewed between 1994 and 2001. Studies III and IV were based on the data collected in Stockholm and its suburbs in April-June 2005 by means of the semi-structured in-depth interviews. The study sample included 15 Russian-speaking immigrants from the FSU currently living in Sweden (5 men and 10 women), with varying marital and socioeconomic status, a mean age of 47.3 (25 - 70) years and a mean time spent in Sweden of 13.6 (3 - 36) years. The results were validated by means of mini-interviews by telephone with 15 new respondents. Study III utilized the grounded theory approach to analyze the data. Study IV utilized the content analysis technique Framework.

Results: Studies I and II showed that while immigrants from the FSU had higher odds of self-reporting poor general health than the Swedish-born host population, they did not have higher odds of reporting poor mental health. This was not true of immigrants from other countries of the former Soviet Bloc (Poland and other East European countries), who had higher odds of self-reporting both poor general health and poor mental health than the Swedish-born individuals. These findings remained after adjustment for demographic and socioeconomic variables. Study III showed that acculturation influenced the way Russianspeaking immigrants from the FSU communicated their health regardless of their actual health status. Their self-reported health (rated from very good to poor ) was associated with their choice of response models for the general health question (explicit, normalized, or implicit). This choice was determined by the extent of their acculturation (Integrated, Assimilated, Separated, or Marginalized) and consequent preference for manner of communication (Swedish or Russian). This suggested that it is essential to account for acculturation in the studies of immigrant health. Study IV described an easy way to measure acculturation by applying a variable called self-reported integration , which might be regarded as a proxy for acculturation. The self-reported integration is determined by the response to the question: In your opinion, have you become integrated in Sweden? The findings showed that the answer to this question given by the immigrants corresponded with the researcher s estimation of the immigrants extent of acculturation and possibly represented a respondent s holistic view of his or her acculturation, based on self-evaluation of both internal and formal criteria of integration.

Conclusions In the present thesis we found that being born in the FSU was associated with poorer self-reported general health, but not self-reported mental health, than in the Swedish-born host population. This finding remained after adjustments for demographic and socioeconomic variables. However, it was found that acculturation influenced health communication among Russian-speaking immigrants from the FSU in Sweden. The more acculturated a respondent was, the more his/her response model to the general health question and his/her self-reported health corresponded to the ones traditional of the host country, but did not necessarily correspond to the actual state of the respondent’s health. This suggests that it is essential to account for acculturation in studies of immigrant health and calls for attention when designing studies and interventions targeted at this population group. Self-reported health is widely used in medical and related research practice nowadays and underestimating the influence of culture and acculturation on health communication might have negative consequences in the planning, implementation, and evaluation of health care services. Self-reported integration might be an informative variable for measuring the extent of acculturation. This might have implications in immigrant research, but its use as a proxy for acculturation should be tested primarily in other settings and in a quantitative analysis.