Since coronavirus disease 2019 (COVID‐19) entered the Netherlands, the older adults (aged 70 or above) were recommended to isolate themselves, resulting in less social contact and possibly increased loneliness. The aim of this qualitative study was to explore independently living older adults’ perceptions of social and emotional well‐being during the COVID‐19‐related self‐isolation, and their motivation to expand their social network in the future. Semi‐structured phone interviews were held with 20 community‐dwelling adults (age range 56–87; 55% female) between April and June 2020 in the Netherlands. The interviews were audio recorded and transcribed verbatim. Open coding process was applied to identify categories and themes. Participants said to use more digital technologies to maintain contacts and adapt to the government measurements. Most participants missed the lack of social contacts, while some participants had no problems with the reduced social contacts. The emotional well‐being of most participants did not change. Some participants felt unpleasant or mentioned that the mood of other people had changed. Participants were not motivated to expand their social network because of existing strong networks. The relatively vital community‐dwelling older adults in this study were able to adapt to the government recommendations for self‐isolation with limited negative impact on their socio‐emotional well‐being.

Several approaches have been proposed for latent class modeling with external variables, including one-step, two-step, and three-step estimators. However, very little is known yet about the performance of these approaches when direct effects of the external variable to the indicators of latent class membership are present. In the current article, we compare those approaches and investigate the consequences of not modeling these direct effects when present, as well as the power of residual and fit statistics to identify such effects. The results of the simulations show that not modeling direct effect can lead to severe parameter bias, especially with a weak measurement model. Both residual and fit statistics can be used to identify such effects, as long as the number and strength of these effects is low and the measurement model is sufficiently strong.

Introduction:Care integration in primary elderly care is suboptimal. Validated instruments are needed to enable the implementation of integrated primary care. We aimed to assess construct validity of the Rainbow Model of Integrated Care measurement tool (RMIC-MT) for healthcare professionals working in an integrated primary elderly care setting in the Netherlands. Methods:In a cross-sectional study, the RMIC-MT, a 36-item questionnaire covering all domains of the Rainbow Model of Integrated Care (RMIC), was sent out to local networks of primary elderly care professionals. Confirmatory factor analysis with maximum likelihood estimation was used for the validation of the factor structure of the RMIC-MT. Model fit was assessed by the chi-square test and fit indices. Results:The RMIC-MT was completed by 323 professionals, primarily general practitioners, community nurses, practice nurses, and case managers. Confirmatory factor analysis and corresponding fit indices showed moderate to good fit, thereby confirming a nine factor model with a total of 36 items. Conclusions:The RMIC-MT is promising for the primary elderly care setting in the Netherlands. It can be used for evaluating integrated care initiatives in a primary care setting, thereby contributing to implementation of integrated primary elderly care.

No abstract