82 Chapter 3 variables: gender; year of birth; date of diagnosis of the tumour, tumour type (BC/CRC), and tumour stage. Within the combined dataset several new variables were determined: ‘number of times invited’, ‘number of times participated’, and ‘percentage participated after being invited’. For every 4-digit zip code a neighbourhood SES-score was set by the Netherlands Institute for Social Research (Sociaal en Cultureel Planbureau, SCP) on a continues scale in 2017.19 This score incorporates data on house value and income. We categorised this score into quartiles (1-4: the higher the number, the higher the SES), including all neighbourhoods in the Netherlands. Thereafter the 4-digit zip code for neighbourhoods of The Hague were assigned with a neighbourhood SES-score. Data analysis The subdivision of attendance groups for both CSPs was determined over the set time period: how many people were invited, how many people did participate, and how many people were registered with a cancer diagnosis. We distinguished invitees who always (100%), sometimes (>0% and <100%), and never (0%) participated after receiving an invitation. For further analysis we divided our data in ‘attenders’ and ‘non-attenders’. Attenders were defined as: invitees who participated in the CSPs in more than 50%, after being invited. Non-attenders were defined as: invitees who participated in 50% or less, after being invited. The proportion of attenders and non-attenders was presented descriptively, using counts and percentages. To test independent continuous variables, Mann-Whitney U and Kruskal-Wallis tests were conducted. For categorical independent variables, univariate regression analyses were performed with an α 0.05 and β 0.8. This resulted in odds ratios (ORs) per attendance group, with corresponding 95% confidence intervals (95% CIs). Likelihood Ratio tests were performed to test for the influence of each independent variable in the regression models. Our data was stored and analysed by making use of IBM SPSS (version 25). Patient and public involvement The development of the research question, study design and outcome measures were developed by a team of experienced primary care doctors and researchers, who also concerned patients’ and public’s interests. Patients were not directly involved in these processes. The results of this research work are going to be published open access and disseminated to whom is interested, among others primary care doctors and the Municipal Health Services.
RkJQdWJsaXNoZXIy MTk4NDMw