Introduction
Previous research has shown that there is an inverse association between socioeconomic status (SES) and the incidence and prevalence of coronary heart disease (CHD) [
1,
2] and also between SES and mortality [
3,
4]. Less is known about how SES affects patients’ quality of life (QoL) after diagnosis of CHD. Low QoL is a characteristic of aged people of low SES that is not completely explained by their health status [
5]. Furthermore, the adverse consequences of CHD in older patients are particularly profound in specific domains of QoL. The impact of heart failure has been found to determine a greater decline in functional status in older patients when compared to a younger group [
6]. A prospective study among older patients also showed that the impact of CHD has enduring negative consequences particularly for physical functioning, while its impact on social and psychological domains was significant but less pronounced [
7]. In the present study, we build upon these results by examining a potential differential vulnerability effect of SES on the course of different domains of QoL after the emergence of CHD. That is, we evaluate whether older patients with low SES are worse off in terms of QoL once they develop CHD.
Previous research has suggested that low SES may be related to an unfavorable course in several health outcomes (e.g., perceived general health, long-term disability, mobility decline, social isolation, emotional reaction) [
8,
9]. The extent to which patients experience these unfavorable outcomes is partly dependent upon the availability of psychosocial coping resources, such as social support and control [
10‐
12]. This finding supports the notion that SES might be related to adverse QoL outcomes after the onset of CHD. Given that adjustment to CHD requires behavioral change, psychosocial resources might be very important in the response to CHD. It might be expected that CHD patients of low SES are more prone to decline in QoL once the disease has been established. Nevertheless, there are no clear indicators describing the impact of SES on QoL in the course of CHD.
In order to study adequately the impact of CHD on the course of QoL, it is important to take into account both the premorbid level of functioning and the multiple assessments after diagnosis. This is of particular relevance if we want to find out whether the differences in QoL due to SES are already present before the incidence of CHD or are a consequence of the disease.
This prospective study examines differences between patients of differing SES in relation to QoL at various assessment points, including a premorbid assessment. We analyze longitudinal changes reported in QoL from the premorbid assessment until 1 year after diagnosis, comparing the course of adaptation in two groups of patients with low and high SES. The emergence of CHD is associated with a greater impairment in functional status, particularly physical functioning [
7], than in other QoL domains (such as psychological distress). We, therefore, predict that SES will have a specific impact on functional status up to 1 year after CHD diagnosis. Furthermore, on the basis of indications found in the literature, we expect that patients with high SES will report better outcomes in response to CHD, principally in physical functioning.
Discussion
The results of this prospective study show that differences between socioeconomic groups are present at different time periods, prediagnosis and postdiagnosis, in five domains of QoL. In all cases, high SES was associated with more favorable outcomes than low SES. Furthermore, after diagnosis, CHD patients with high SES exhibit higher levels of social functioning, role functioning, and physical functioning than patients with low SES. Differences in social and role functioning may result from a differentiated response to the disease, while differences in physical functioning could be related to the premorbid situation. Confirming our expectations, high SES predicts better outcomes in the functional domains of QoL up to 1 year after the diagnosis of CHD. Differences in psychological distress were limited to depressive feelings at the premorbid measurement, therefore, unrelated to CHD.
The novel aspect of our research is the inclusion of patients’ premorbid data. This approach allowed us to investigate the specific mechanisms of the process of adjustment to CHD, which otherwise would have remained undiscovered, leading to potentially erroneous interpretations.
A possible limitation of the study is the high level of subject attrition because we selected only those patients who replied through the four assessments. Patients with poorer health may have dropped out of the study, biasing our sample toward CHD patients with milder disease or lower anxiety and depression. That might also explain the paucity of differences between socioeconomic groups in the psychological domains. The participants in our study might have responded more quickly and actively to the condition, enhancing their chances of survival. As we illustrated in the methodological section, the study participants were younger and socially more active than nonparticipants.
In order to accurately describe the clinical profile of the patients, additional biomedical data should be incorporated in this study. For instance, the left ventricular ejection fraction (LVEF) is a sensitive measure of disease severity which can be employed complementarily to the NYHA. LVEF represents the fraction of blood pumped out of the left ventricle during each heartbeat, which is substantially limited in CHD patients. Because in the present dataset information on disease severity were only gathered through the NYHA classification, further studies on QoL in CHD, including records of patients’ LVEF and other clinical variables, would be recommendable.
Another limitation may have been the varying time intervals between baseline assessment and diagnosis in our sample, which ranged from 1 to 58 months. If patients of one socioeconomic group were registered for a diagnosis of CHD a long time after the first interview, they could report more changes in QoL compared to the others who entered the study shortly after the baseline. However, we calculated and compared the mean time between baseline and diagnosis for both socioeconomic groups and found no statistically significant difference between them (26 months for the low SES and 28 months for the high one). Additionally, we checked the correlations between the time elapsed since the diagnosis and the outcome variables after the diagnosis. There were no significant correlations with the outcome variables of interest. Therefore, we do not consider such a time factor to be a confounder in the present study.
Status inconsistency (i.e., discrepancies between the three indicators of SES) is associated with CHD and might have obscured results showing socioeconomic disparities in psychological domains [
26]. However, educational level, job prestige, and income are moderately correlated with each other (
r values between 0.37 and 0.42, all
p < 0.001). In addition, the factor loadings of the three indicators were similar, indicating that they contribute to SES in a similar way. In other words, although we cannot exclude the presence of status inconsistency, such a phenomenon should be of limited importance in the present study.
Similarly to status inconsistency, downward mobility (i.e., decrease in SES over time) is associated with health problems and incidence of CHD and might represent a possible bias in our study. Because SES was only measured at the premorbid assessment, we could not gauge downward mobility in the present research. However, considering that the mean age of the sample was 72 years (the youngest patient was 57 years old), implying that most of the participants were already retired, it is quite unlikely that such patients would have experienced a collapse in their career or retirement funds at this specific moment of their life. For these reasons, we exclude downward mobility as a possible limitation of the study.
Our results indicate that cross-sectional differences between SES after the onset of CHD are regulated by distinct mechanisms. Interestingly, the significant differences in depressive feelings between SES groups observed at prediagnosis (patients with low SES reported higher depressive feelings) disappeared after the emergence of the CHD and did not reappear throughout the study. The emergence of the disease appeared to modulate differences that were previously present between SES groups.
We observed differences in physical functioning 6 and 12 months after CHD diagnosis but, crucially, this difference existed premorbidly. This finding may be important for understanding the impact of CHD on physical functioning, especially in research which does not include premorbid measures. In the present study, we found that differences in physical functioning are not a direct consequence of CHD but rather the reestablishment of a premorbid situation. On the other hand, socioeconomic inequalities in social and role functioning only emerged 1 year after the diagnosis. In this case, the observed differences represent a differential response to the impact of the disease.
A few recommendations can be derived from the results of this study. The research on social inequalities in health among aged people is still scarce, while it should be of primary concern considering the substantial impact of this age group on health care use [
27]. More investments are needed to alleviate the unfavorable situation of disadvantaged groups on specific aspects of QoL, ensuring that health care services respond appropriately and promptly to the needs of different socioeconomic groups. Moreover, this study suggests that interventions for older patients with low SES should be primarily focused on domains such as role and social activities where differences according to SES are a direct consequence of CHD. More generally, the physical domain should be of particular clinical concern as well due to the fact that socioeconomic differences in physical functioning are already present at the premorbid stage and remain substantial after the onset of the disease.
Two main issues which could not be directly addressed in the present research should be integrated into future studies. First, more light should be shed on the role of psychological resources (such as control beliefs) in the explanation of socioeconomic differences in adaptation to CHD in old age [
10]. A higher risk of heart disease, associated with persons of low SES, is better explained by differences in control beliefs than by the classical coronary risk factors (e.g., smoking, hypertension) [
28]. One may wonder whether psychological resources could help to explain the differences in outcomes of QoL after the onset of CHD for different socioeconomic groups. Secondly, it would be useful to expand our approach to include research on other conditions, in order to determine which outcomes are disease-specific and which represent a more general response to the onset of diseases.