Is preexisting mental illness associated with lower patient satisfaction for older trauma patients? A cross-sectional descriptive study

This prospective cross-sectional study included trauma patients aged ≥55 years, admitted between November 1, 2016-December 31, 2017, across two level I trauma centers. Included patients were then categorized into either PMI+ (patients who had ≥1 diagnosis of mental illness prior to the trauma admission) or PMI- (patients without a diagnosis) based on information contained in their medical records. Further details on the inclusion and exclusion criterion, and the processes and details for administering the surveys (including their theoretical framework and initial findings) are outlined in a previous study [17]. Briefly, eligible patients (trauma admissions ≥55 years old) were approached by the study coordinators at each hospital to be enrolled. If the patient was unable to consent, they could be enrolled by proxy. Patients with severe cognitive impairment as well as those who did not meet state trauma registry inclusion criteria were excluded. Consented patients were administered the FAMCARE-P13 (Family Satisfaction with Advanced Cancer Scale, Patient), prior to discharge. The FAMCARE-P13 survey measures the degree of patient, satisfaction with information provided, availability of hospital staff, physical care management, and psychosocial care and has a possible score range of 1–65 (13 questions), with higher scores reflecting higher levels of satisfaction [18‐20]. The survey questions also can be grouped into the following conceptual structures: Information giving, Availability of care, Physical care, and Psychosocial care [21].

These analyses were conducted in the context of a larger prospective study. Because this was an analysis from an ongoing study, a formal sample size calculation was not performed. Patients were provided written informed consent prior to being enrolled and surveyed. This study was carried out according to the STROBE-guidelines on the reporting of observational studies and was approved by institutional review boards at the participating centers.

According to the American College of Surgeons’ (ACS) Guidelines for level I trauma centers, hospitals are required to screen at least 80% of all trauma admissions using screening, brief intervention, and referral to treatment (SBIRT), to identify those with problematic mental illnesses [22]. The exact process for the screening and evaluation for mental illness is left up to each trauma center. Across the hospitals in this study, a team of social workers and a clinician are usually responsible with screening patients for alcohol, substance use, and/or mental health concerns, and discerning the need for a brief or formal intervention depending on the severity of the concern.

The following covariates were collected on each patient from the trauma registry: sex, age (55–65, > 65), race, injury severity score (ISS, 1–15, ≥16), hospital length of stay (LOS), ICU LOS, cause of injury (fall vs. high acuity vs. sport vs. other), hospital discharge destination (home/health vs. skilled nursing facility vs. long term care vs. hospice), the presence of the following comorbid conditions: a history of mental illness, smoking, dementia, diabetes, chronic obstructive pulmonary disease (COPD), hypertension, functionally dependent health status, anticoagulant use, congestive heart failure (CHF), alcoholism, and an advanced directive limiting care.

Covariates collected from patient EMRs included: specific mental illness diagnosis, medical insurance, mental illness management (behavioral health assessment, psychiatric consultation, discharge plans), mental illness medication, and the history of the diagnosis (longstanding mental illness: ≥2 years or newer diagnosis < 2 years), traumatic brain injury (TBI) diagnosis.

The primary outcome variable was overall mean (SD) satisfaction score for patients from the FAMCARE-P13 survey and the secondary outcome was patient satisfaction by survey question.

Satisfaction and in-hospital patient characteristics were compared univariately by the patient’s PMI status. Descriptive statistics were analyzed using Chi-squared tests or Fisher’s exact tests for categorical variables. Student’s t-tests, one-way ANOVAs, Wilcoxon two-sample tests, or Kruskal Wallis tests were used for continuous variables, as appropriate. Data are displayed as means and standard deviations (SD), medians and interquartile ranges (IQR), or proportions.

Generalized linear models were used to determine whether differences in patient satisfaction existed between the PMI+ and PMI- patients. Entry and exit criterion were set to p = 0.2 and p = 0.1, respectively, and manually entered into the model. Any significant interactions between PMI status and other covariates on patient satisfaction were explored. PMI status was the primary exposure variable and thus was automatically included in each model. SAS 9.4 (Cary, NC) was used for all analyses. Two-tailed tests with an alpha < 0.05 were used for all tests.

Table 1 reports any differences in characteristics between patients by PMI status. Compared to patients who were PMI-, PMI+ patients had significantly lower mean (SD) satisfaction (82.9% (13.5) vs. 78.7% (16.2), p = 0.04), were more likely to be in the age range of 55–64 vs. ≥65 years (21% vs. 35%, p = 0.02), were significantly more likely to be female (49% vs. 74%, p = 0.003), to have ≥2 comorbidities (36% vs. 92%, p < 0.001), and to have private insurance (30% vs. 45%, p = 0.03). There were no other significant differences in injury characteristics or outcomes found between groups.

The most common mental illness diagnosis reported among PMI+ patients was depression (70%), followed by anxiety (35%), and then dementia (8%), and in most (89%) patients, the diagnosis was longstanding (≥2 years) (Table 2). Fourteen (23%) PMI+ patients received a behavioral health assessment (hospital 1, n = 2 and hospital 2, n = 12); the most common trigger for the assessment was “Coping concerns with acute change in activities of daily living” (62%), while the remaining triggers included: “Safety concerns at home” (15%); “Evaluation outbursts in hospital”(8%); “Long history of depressive symptoms” (8%), and “Suicide attempt” (8%).

Satisfaction by PMI status was modified by hospital; therefore, each adjusted model was also hospital-stratified, and data are reported in Table 3 (hospital 1) and 4 (hospital 2). After adjustment for smoking status (yes vs. no), PMI+ patients at hospital 1 had significantly lower overall satisfaction, than PMI- patients, respectively (69.8% vs. 79.5%, p < 0.003, Table 3). Conversely at hospital 2, after adjustment for smoking status and hospital LOS, PMI+ patients did not have significantly lower satisfaction than PMI- patients (73.5% vs. 74.7%, p = 0.61, Table 4). Please refer to Additional file 1 for supplementary 1 table on differences in patient characteristics by hospital.

Table 5 outlines satisfaction by survey question between PMI+ and PMI- patients at hospital 1. Under Information giving, PMI+ patients had significantly lower mean satisfaction than PMI- patients with question 4, “Information provided about your prognosis (75.7% vs. 85.6, p = 0.03, Table 5) and question 6, “Information given about your procedures” (73.3% vs. 88.4%, p = 0.002). Interestingly, PMI+ patients also had significantly lower satisfaction than PMI- patients for all questions (1, 5, 7, 10, 11) related to Physical care. Please refer to Additional file 1 (supplementary table 2) for patient satisfaction by survey question at hospital 2.

There are a number of limitations in the study. First, the patients responding to the satisfaction surveys may not have been representative of the general population seen across the facilities. Patients who were “very dissatisfied” or “undecided” may have withheld from taking the survey. Second, healthcare surveys administered in person can sometimes artificially inflate the results because patients might believe that their scores will negatively affect the care received; however, by capturing results soon before hospital discharge and without disclosing them to treating staff, a more accurate and unbiased description of patient satisfaction was obtained. Third, because this was a cross-sectional study of two trauma centers, other hospitals with different populations and behavioral health management strategies may not observe the same satisfaction scores; nonetheless, the inclusion of a general, older adult trauma population may help other centers hoping to understand satisfaction and characteristics of trauma patients who are PMI+. Fourth, because data was used from a larger ongoing study, the surveys used were not previously evaluated in a trauma population, or for those who are PMI+ and additional work may be needed to verify the results using an instrument tailored for this population; however, Vogel and colleagues (2019) successfully measured patient and caregiver satisfaction using the FAMCARE surveys in a trauma population [17, 39]. Fifth, because the measurement of mental illness in this population was taken from patients who consented to take a survey, it is likely an underestimation of the true prevalence across these hospitals. Last, SBIRT components were not captured for PMI+ patients and thus compliance with ACS guidelines was not able to be measured.