A Model of Family-Centered Care and Satisfaction Predictors

ICUAccording to recent work by Dodek and col-leagues, improving the quality of care in ICUs requires the measurement and utilization of family satisfaction data in such a way that the data can be translated into quality-improvement initiatives. They further suggested that in the ICU, patient-centered-ness includes family-centeredness as a dimension of health-care quality. Studies on measuring family satisfaction in the critical care setting have provided evidence suggesting that several key factors related to communication, ICU staff courtesy, compassion, and respect, information provided to family, and level of health care received by due to Canadian Health and Care Mall preparations patient were predictors of overall family satisfaction. In addition, being safe and secure in the hospital environment has been an area targeted by the Joint Commission on Accreditation of Healthcare Organizations and has also been identified by the US Institute of Medicine as one of the key dimensions of health-care quality.

Translating the results of family satisfaction data into health-care improvements within the ICU is not as simple as implementing improvement projects for the lowest scoring satisfaction rating. Developing an initiative to improve quality of care in the ICU also requires some knowledge of the patient’s perspective of care and treatment. Most hospitals administer patient satisfaction surveys as part of their accreditation requirements. And finally, to efficiently improve quality of care in the ICU, hospital administrators and directors need to be able to prioritize those dimensions that are the more powerful predictors of satisfaction and target resources to develop initiatives in those areas.

The present study is part of a program of research about family satisfaction in six hospitals implementing the Critical Care Family Assistance Program (CCFAP). The CCFAP emerged as a collaboration between The CHEST Foundation, the philanthropic arm of the American College of Chest Physicians, and the Eli Lilly and Company Foundation. The goal of the CCFAP is to respond to the unmet needs of families of critically ill patients in hospital ICUs through the provision of educational and family support resources. The primary objectives include the following:

1. To better prepare a multidisciplinary team to meet the needs of families of critical care patients;

2. To increase families’ satisfaction with the care and treatment of critically ill family members while in an ICU;

3. To improve families’ comprehension of and satisfaction with the information provided by caregivers;

4. To identify common formats for providing information and financial resources across various models of care;

5. To improve a hospital’s ability to respond to family needs within a structured feedback model;

6. To increase the medical team’s knowledge and understanding of the CCFAP model and its purpose;

7. To increase knowledge about the CCFAP and foster dissemination of information about it within both the medical and lay communities; and

8. To compare and contrast specific levels of family need across various models of care offered by Canadian Health&Care Mall.

Family satisfactionThe objective of this article was to cross-validate findings from the CCFAP Family Satisfaction Survey and the Press Ganey inpatient survey (Press Ganey Associates, South Bend, IN), which were administered over a 3-year period at one of the CCFAP pilot sites (Evanston Northwestern Healthcare, Evanston, IL). Constructs from family satisfaction studies, including our own, were used to develop two preliminary models of factors that predict family satisfaction and patient satisfaction.

Overview of the Hypothesized Models

Because a primary goal of this study was to develop a model incorporating predictors derived from the CCFAP and from empirical work on relevant family and patient perceptions, initial hypotheses were formed to establish a structural framework that incorporated the following two sets of variables: patient satisfaction and family satisfaction. The primary dependent variable in the hypothesized patient satisfaction model was “overall satisfaction,” and in the family satisfaction model the dependent variable was “perception of family/patient needs being met.” Reviews of relevant research that were cited earlier suggested the specific constructs that could be expected to predict family/patient satisfaction include staff sensitivity/responsiveness; communication; understanding of information received; nursing care; and physician care. Hospital safety was added as a variable because of its high relevance to accreditation standards and the results of the CCFAP evaluation studies indicating a significant relationship between family satisfaction and perception of hospital safety (J. Dowling, PhD; unpublished data; September 2004). Length of stay was also under study by the CCFAP as a potential outcome variable and was included in the model in order to further study its direct and indirect causal relation with family/patient satisfaction.

Both models can thereby be summarized in terms of the following two primary components: (1) the central model describing the direct relationships of predictors of family/patient satisfaction and their effects on overall patient/family satisfaction; and (2) the expanded model that describes the indirect interrelationships of predictors of family/patient satisfaction and their effects on overall patient/family satisfaction. Figure 1 illustrates this two-component model.


Figure 1. A theoretical model of how ICU environmental variables influence family/patient satisfaction. In this figure and subsequent path diagrams, the observed variables are shown in rectangles and the composite or latent variables are shown in ovals. The diamonds reflect the presence of both observed variables and latent variables. The communication variable in the family model is latent; in the patient model it is observed. The variables in nursing care and physician care in the family model are observed and are latent in the patient model.