In this study, an individual socioeconomic status index (ISESI) was built and validated for a population aged between 50 and 69 based on information available in the Patient Information System (PIS) of the Region of Valencia (RV), Spain. The ISESI made it possible to analyse inequalities in RV colorectal cancer screening programme (CRCSP) participation.
A multivariate methodology was used to give a weight to the variable categories that make up the ISESI, representing the statistical relationships between these categories. This methodology made it possible to reduce and combine the wide range of socioeconomic variables that were available in the PIS, including nationality, employment status, disability status, type of healthcare coverage, risk of vulnerability and FS. As a result, a qualitative and quantitative index was built in order to allow and facilitate the analysis of health inequalities.
It should be noted that the coefficient of variability explained in the ISESI is low, as shown in the results. This could be due to the fact that the type of information available in the PIS has the purpose of establishing the healthcare coverage rights, type of healthcare coverage and contribution to prescription charges of people registered as living in the RV. Therefore, as these rights are greatly dependent on employment status and family income, the information available in the PIS and, therefore, the information used to develop the ISESI, is focused on these characteristics. However, despite this, the percentage obtained is greater than the expected variance if random data were used, and therefore they give the ISESI validity and representativity.
The ISESI variables with the highest weight were employment status and risk of vulnerability, followed by nationality and healthcare coverage and, lastly, disability and FS. This indicates that the ISESI created in this study mainly characterises the population in accordance with their employment status and how this affects their social, economic, and healthcare vulnerability.
By comparing the ISESI with other SES variables and performing external validation with variables such as personal income or household income, in addition to employment status, we confirmed that the ISESI characterises the population according to socioeconomic characteristics, based on their employment and income status. In contrast, it does not appear to be related with variables traditionally used to measure SES, such as educational level or occupational social class [13]. As commented above, this is due to the type of information available in the PIS. Nonetheless, it should be noted that despite this limitation, one of the most significant advantages of the PIS is that the information is systematically collected and updated on a regular basis and coded in a uniform manner. Therefore, it is an official, stable and publicly funded information system that undergoes regular quality control [11].
The multidimensional character of the concept of SES and the growing importance of assessing health inequalities has led to the creation of SES indices using information available in various sources. There are several initiatives related to the construction of socioeconomic status indices to measure health inequalities at both the national and international level [7, 8, 14–16]. In Spain, the greatest success has been seen in the development of area-level indices based on housing census data [7, 17]. One of the most commonly used indices in the Spanish context was constructed from socioeconomic indicators at the census-section level, specifically with information on occupation type and indicators related to employment status, resulting in an ecological index [7]. These indices have at times been used to analyse the impact of area-level inequalities in cancer screening, using them from an individualised perspective [18, 19]. The ISESI created by this study has great potential as it is an individual SES index, which complements the use of ecological indices. Furthermore, and considering that the PIS is population-based, it should be noted that the ISESI is available for the entire population of the RV registered in the PIS, which is an advantage of this index as compared to other socioeconomic status indicators. Our results show that combining different socioeconomic characteristics in an index to measure inequalities in CRCSP is better than using each of the population’s socioeconomic characteristics independently. These results are in line with other studies that combine several socioeconomic characteristics in a single individual index to analyse health inequalities in the adult population [20].
Some authors state that the type of socioeconomic indicator and its influence on health seems to have a different effect depending on the health problem under analysis [21–24]. One specific study shows that socioeconomic status measured in terms of income has the most significant effect on all health indicators in old age [21]. Another study shows that educational level creates inequalities in all-cause mortality, while socioeconomic variables affect cardiovascular illnesses and cancer [23]. A study performed in the UK found that the most deprived neighbourhoods presented worse conditions in terms of waiting time, repeat hospitalisation and dying in hospital than the least deprived neighbourhoods [24].
This index has been created to assess inequalities in CRC screening, among other uses. Consequently, it was developed with information on a population group aged between 50 and 69, considering the age of the target population of these programmes. The index can be incorporated to analyse inequalities in CRCSP result indicators or can be used as an SES adjustment variable. Nonetheless, the same methodology could be replicated to create indices adapted to the target populations of other public health programmes in the RV, such as the early detection of breast and cervical cancer programmes, or programmes for sexual and reproductive health, active ageing or gender violence prevention.
An initial approach to using this ISESI to identify inequalities in CRC screening has demonstrated that the population situated in Q1, i.e., with the best socioeconomic conditions, and in Q4, i.e., with the worst socioeconomic conditions, were less likely to participate than those in intermediate quartiles (Q2 and Q3). These results are in line with other studies performed in Spain [10, 18, 19, 25]. Specifically, Buron (2017) found that inequalities in CRC screening uptake in Catalonia seem to be concentrated primarily in the most disadvantaged groups, followed by the least disadvantaged ones [17]. Studies performed in the context of European screening programmes showed a participation gradient with the lowest percentages seen in the most disadvantaged social strata in the case of both men and women [26–28].
Some of the variables used to construct the ISESI, such as employment status, were used to identify inequalities in European CRCSP participation [29–31]. The results of these studies are inconsistent, as some conclude that there is no relationship between employment status and participation [29] while others do find such a relationship [31], with a trend towards lower participation in employed people compared to unemployed or retired people. In our study, we saw this trend in retired people but not in unemployed people. In addition, several studies associate income level—a variable that showed a strong correlation with the ISESI in external validation—with inequalities in participation [32–34]. They also show that the probability of participation falls as income level decreases, in line with the results of our analyses. Finally, educational level—a variable that was not used to create this index due to unavailability—has been positively linked to CRCSP participation [35, 36].
Analysing social inequalities in CRCSP participation is a complex phenomenon that requires the use of multiple and varied socioeconomic indicators in order to study these inequalities in more detail. The resulting ISESI and its inclusion in the RV CRCSP information system could help provide a better understanding of inequalities in CRC screening.