While general population surveys have largely failed so far to identify the negative health consequences of the current economic crisis, this study highlights that in Barcelona, people seeking help from Caritas and facing serious housing problems have a much worse health status than the general population, even when compared to those belonging to the most deprived social classes. This is especially so for users of the HMS, who specifically seek help for dealing with unbearable housing payments. Moreover, even within a small sample of subjects, all of whom are facing some kind of housing problem, housing conditions are related to poor mental health, even after controlling for other adverse life conditions such as unemployment or being undocumented. These findings are consistent with the growing body of literature showing the association between poor housing conditions, mainly housing affordability, and poor physical and mental health [4, 5, 15, 26, 27].
It is worth mentioning that the population assisted by Caritas has a high vulnerability profile, shown by their socio-demographic and housing characteristics as well as by their health outcomes. Difficulties in affording housing costs among this population is most likely due to the fact that the majority of interviewees are unemployed, mainly among men who tend to be the breadwinners. In addition, more than one third of women are single mothers, who have to face with both the responsibility of house and children care as well as with being in a precarious situation, as other studies have shown [28, 29]. Nonetheless, it is also worth highlighting that the sample was not only made up of the more excluded populations, but also included other groups that do not tend to use Caritas’s services, such as those from less-deprived socials classes.
Within the two Caritas groups, women had worse self-reported health, slept less, were more depressed or anxious, used medications more frequently, and had worse quality of life, which has been described elsewhere [28, 30, 31]. Additionally, users of the HMS had considerably worse health than those from the social services living in substandard housing. HMS users are mostly families approaching Caritas for the first time searching for help to deal with a mortgage or rental payment problem. Such struggles have been shown to increase psychosocial stress and anxiety, and homeowners frequently view these difficulties as a personal failure [4, 8, 32].
Our study found, across all groups, that those with affordability problems reported worse mental health, mainly among those who worried about not having sufficient money for food or those who had at some point slept in a shed, shelter, car, or on the street. The poor mental health outcomes suffered by this population, especially by the HMS group, are consistent with a number of studies that demonstrate the poor health outcomes of people who experience homelessness  or foreclosure [9, 34]. Additionally, across almost all groups, housing satisfaction was associated with mental health such that those with lower satisfaction reported worse mental health. However, we did not detect strong associations between poor mental health and dwelling or neighbourhood conditions. This might be attributable to the lack of variability in housing and neighbourhood characteristics among the study population, since the majority of study participants experience serious substandard living circumstances, therefore making it difficult to tease out associations with health outcomes in this relatively small population.
In men we found that overcrowding was associated with better mental health. This was surprising given that some studies indicate that overcrowding can result in worse physical and mental health outcomes due to higher rates of infectious transmission and other poor living conditions suffered by those living in overcrowded homes and the stress associated with being unable to pay one’s bills which could force one to move in with others [35, 36]. However, the fact that these men live in an overcrowded home could indicate that they have a social safety network to fall back on in difficult times. Such social support could lead to improved mental health. Future studies should confirm this hypothesis. This association was not observed among women, which could be explained by the social construction of gender . The role of breadwinner, assigned to men, represents a source of pressure in times of economic difficulties. Similarly, the association with the emotional link with the dwelling was observed only among men, since the male figure is responsible for providing his family with an adequate housing.
The study had a few limitations. Our sample size was relatively small due to the difficulties in reaching these populations representing multiple types of housing instability, both those who live in substandard housing and those who struggle to pay housing costs. However, our sample was large enough to analyse the life conditions and health status of a very difficult to reach population, often underrepresented in general health surveys. Perseveration and close follow up of individuals was needed, for which the implication of Caritas in the study was essential. The economic and legal instability of these families, their proneness to mobility, and their precarious situation were important barriers to participation in the study. Incentives were not used in this study. Differences between participants and non-respondents could not be analysed, although most likely non-respondents were those in more precarious situation and, therefore, most likely to have worse health status. Additionally, the cross-sectional nature of the baseline survey prohibits us from drawing causal inferences from the findings, although cohort studies done in other countries have found similar associations [32, 34]. Also, all participants were drawn from the Caritas registry and as such may not accurately represent all of those living in substandard housing in Barcelona, only those who are seeking assistance. Finally, participants were younger than overall Barcelona residents and mostly foreign-born. Since age is related to prevalence of most health problems, age-standardized prevalence was used instead. Among the strengths, participants responded to three thorough questionnaires, with most of the items consisting in validated instruments. Additionally, this study is the first to compare the health outcomes of a vulnerable population with housing instability with those from the overall population of Barcelona, and therefore contributes to the scarce literature analysing the impact of housing on health in Spain, a country characterised by having a housing system with limited access to adequate housing.