Reasons for unmet needs for health care: the role of social capital and social support in some western EU countries

Abstract

This paper focuses on the demand side factors that determine access to health care and analyses the issues of unmet needs for health care and the reasons thereof in western EU countries. A probit model is estimated from a sample of the whole population, accounting for the possibility of individual selection in unmet needs for health care (UN) (selection equation). Expanded probit models (including the inverse Mills ratio) are then used on the reasons for unmet needs (RUN) with social capital and social support as determinants and using the European Union Statistics on Income and Living Conditions dataset from 2006. In the RUN equations, the findings show that females, large households, people with low income and financial constraints, the unemployed and those in poor health have a higher probability of declaring unmet needs due to economic costs. Additionally, people in tertiary education, those with high income and the employed have a higher probability of not visiting a doctor when needed due to time constraints. Furthermore, the frequency of contact with friends and the ability to ask for help are correlated with a lower probability of unmet needs due to economic costs, while the frequency of contact with relatives is correlated with a lower probability of unmet needs due to time constraints and distance. However, the ability to ask for help is also correlated with a higher probability of not having medical care due to time constraints and the wait-and-see approach.

Appendices

Appendix 1: Demographic and socioeconomic characteristics in the unmet needs (UN) equation

With regard to demographic and socioeconomic characteristics (Table 6), gender and marital status are not statistically significant. The same applies to household size and country of origin.

The probability of reporting unmet needs is negatively and significantly correlated with age. The youngest age groups show a positive correlation with the probability of reporting unmet needs. By contrast, older people present a negative association with the likelihood of declaring unmet needs when they feel they need examination or treatment (respectively, by 0.6, 1.1 and 1.6%.

The probability of not visiting a doctor when needed is weaker among individuals with higher education than individuals with lower education. Moreover, individuals with tertiary education have a higher likelihood of declaring unmet needs than those with secondary education. Following the literature a possible explanation is that individuals with tertiary education have greater time constraints which may lead them to postpone medical visits and treatment (Chaupain-Guillot and Guillot 2015).

The likelihood of forgoing medical examination or treatment is correlated with the economic situation of the household. Individuals living in higher-income households with home ownership are less likely to report unmet medical treatment. Individuals who have arrears on utility bills and are unable to cope with unexpected financial expenses present, respectively, a 4.5 and 2.4% higher probability of declaring unmet needs for health care. Hence, a poor economic situation in the household emerges as a burden in healthcare access. Furthermore, the likelihood of declaring unmet healthcare needs is also positively correlated with the employed, the unemployed and the inactive. While for those employed a feasible explanation of unmet needs may be that they have “time constraints”, for the unemployed unmet needs may be due to the economic burden (Lee et al. 2015). Finally, the probability of declaring unmet needs has a strong positive relationship with poor perceived health. Individuals who perceive that they are in poor or very poor health are more likely to declare unmet medical needs (the opposite holds for individuals who perceive good and very good health). Having a chronic condition is also positively correlated with the probability of experiencing an unmet medical need, as is the fact of being hampered in daily activities because of health problems. These results reveal that less healthy people have multiple or recurrent care needs but they might decide to forgo or delay some examinations or treatment due to the economic burden involved.

Finally, looking at the country-fixed effect, taking Germany as a reference category, we show that France and Portugal have a higher probability of unmet needs (6.3 and 4.7%, respectively), while the Netherlands and Greece a lower likelihood (4.2 and 4.1%, respectively).

Appendix 2: Demographic and socioeconomic characteristics in the reasons for unmet needs (RUN) equation

With regard to demographic and socioeconomic characteristics (Table 7), “female” is found positively associated with Expensive and negatively related to Wait and see. This shows that women are more likely to experience unmet needs due to cost but they are less willing to wait and see when they need to visit a doctor. Marital status is negatively correlated with Distance, indicating that a spouse may provide marital support when needed. Age is shown negatively associated with the likelihood of declaring unmet needs due to cost (70 years and over) and time availability (50 years and over). These results seem to indicate that older people are less likely to face cost and time barriers than younger people.

Household size is found to be positively associated to Expensive and negatively correlated with Distance. These findings appear to indicate that living in a large family generates two opposite effects: it increases the household costs of health care and it decreases the distance-related costs for accessing health care. Being born in EU countries is found to have a positive correlation with Expensive and Distance while a negative relationship with Wait and see. Thus, individuals born in EU countries are more likely to declare unmet needs due to cost and distance but less likely due to personal attitudes.

Tertiary education and household income are both found negatively correlated with Expensive and positively associated to No time. Hence, people with more individual and household economic resources are less likely to experience unmet needs due to economic constraints. However, more time spent on procuring economic resources means less time available for seeing a doctor when needed. These explanations also seem to support results on the employed, which are negatively related to Expensive and positively to No time. Furthermore, tertiary education and home ownership are found negatively related, respectively, to Distance and Expensive. Financial constraints, i.e. utility arrears and unexpected expenses, are found positively correlated with the probability of unmet needs due to costs and negatively related to the likelihood of not visiting a doctor when needed due to the lack of time available. Being unemployed is shown to be positively correlated with a higher likelihood of declaring unmet needs due to cost and lack of time and with a smaller probability of unmet needs due to proximity. “Inactive” is found to be associated with Distance and No time, respectively, with negative and positive sign.

In terms of health status, self-perceived poor health and chronic conditions are associated with a higher probability of having unmet needs due to economic cost. Self-perceived poor health is also related to a lower probability of declaring unmet needs due to lack of time and personal attributes. The last result is also found for chronic conditions. Finally, limitations in ADLs are shown to be associated with a higher likelihood of needs being unmet due to a wait-and-see approach.

The findings indicating that females, low income, financial constraints, the unemployed and those with poor health status are more likely to declare unmet needs due to cost (accessibility) are in line with previous studies (Cavalieri 2013; Fjaer et al. 2017).

Looking at country-fixed effects, taking Germany as a reference category, Spain and the UK have, respectively, a 27.2 and 25.0% lower probability of unmet needs due to economic cost. Denmark and Spain show a higher probability of declaring unmet needs due to time constraints (respectively 58.1 and 22.7%) while the UK exhibits less probability (0.70%). Portugal and Italy present a lower likelihood of unmet needs due to the wait-and-see approach (16.9 and 16.4%, respectively).