“What Are My Injuries?” Health Literacy and Patient Comprehension of Trauma Care and Injuries

Abstract

Background

Trauma patients often have complex injuries treated by multidisciplinary providers with wide-ranging expertise. We hypothesized that trauma patients would frequently incorrectly identify both their injuries and care teams. We also hypothesized that low health literacy level would be correlated with low levels of comprehension about injuries or care teams.

Materials and Methods

We performed a prospective study of adult trauma inpatients >18 years. Participants were surveyed to report on 1) injured body regions 2) their care teams, and 3) health literacy via a validated survey. Self-reported injuries and care teams were compared to the patient’s medical record. We also studied whether health literacy was associated with patient knowledge of injuries and care teams.

Results

Fifty participants were surveyed; thirty-two percent could not identify ≥50% of their injuries. Patients reliably identified injuries to the head, but injuries to other body areas were often misidentified. Forty-two percent of patients were not able to identify ≥50% of their medical teams, and 28% could not identify ≥75% of their medical teams. Patients often did not recognize teams such as Nutrition, Physical/Occupational Therapy, or Social Work as part of their care. Thirteen participants reported adequate health literacy. Health literacy was not related to participant knowledge of injuries or care teams (both p=0.9).

Conclusion

Many trauma inpatients were unable to correctly identify their injuries and care teams despite a range of self-reported health literacy scores.

Introduction

According to the Centers for Disease Control, 2.8 million people were hospitalized due to injuries in 2015, making traumatic injuries one of the most significant contributors to morbidity and mortality and a burden on the healthcare system.(1) Prior studies have demonstrated that trauma patients suffer consequences of their injuries long after discharge,(2) and injury care plans commonly extend into the outpatient setting, making outpatient follow-up crucial. Unfortunately, trauma patient clinic follow-up is often poor.(3) Adherence with follow up is multifactorial, but complexity of discharge plan and relationship with the trauma team have been shown to directly correlate with improved treatment plan adherence and subsequent follow up.(4, 5) In order to ensure patients follow up with the appropriate teams, it is essential that patients understand not only the breadth and scope of their injuries but also which care teams are involved in their care for follow-up.

Trauma is by definition unexpected, often complex, and can involve the specialty care of a variety of medical specialists including surgeons, intensivists, therapists, dieticians, and social workers. Factors such as dynamic care plans can be confusing to even the savviest trauma patient. Trauma patients may also suffer physiologic reasons to have difficulty understanding care plans due to traumatic brain injury or delirium while admitted.(6) Finally, the care of trauma inpatients often is multidisciplinary, resulting in the trauma patients interacting with several specialists per day.(7) Health literacy, or a person’s ability to “obtain, process, and understand basic health information and services,” is poorly studied in trauma patients, but it is estimated that more than one third of US adults had limited health literacy.(8) Additionally, a trauma admission represents a stressful and challenging time in a patient’s life. For these and many other factors, the in-hospital treatment may feel like a blur to the patient.

We sought to further understand the level of understanding that trauma patients have about their traumatic injuries as well as the make-up of the multidisciplinary teams caring for them. We also hoped to gain a better understanding of the health literacy level of our patient population and determine if there was an association between health literacy and patient comprehension of their care. We hypothesized that trauma patients would frequently incorrectly identify injuries and care teams. We also hypothesized that low health literacy level would be correlated with low levels of comprehension about injuries or care teams.

Methods

We performed a prospective study involving adult inpatients admitted to the trauma service from June 2019 to October 2019. Inclusion criteria included English speaking patients age >18 years, hospital length of stay >2 days, and appropriate mental status. Patients were approached after being hospitalized for at least two days, on a regular nursing floor and not the intensive care unit, to ensure that the trauma team had performed daily rounds on the patient at least one time. As a patient’s mental status can fluctuate, eligible patients were identified daily via consultation with floor nurses and therapy teams as being cognitively appropriate for study approach. For inclusion, patients were required to be alert and oriented to person, place, and time, had to verbalize understanding of the project, and be awake and able to converse with staff to complete the survey. Patients were given a memo about the research study and could withdraw participation at any time. For patient confidentiality reasons and the small size of the study, specific demographic information was not collected.

Patients who consented to participate were asked to complete a survey to identify their injuries and care teams (Figure 1). Participants were given the option of filling out the survey alone or having the researcher read the questions aloud. Researchers did not review charts prior to administering surveys to prevent unintentional encouragement towards correct answers. The survey form was divided into two parts, one asking, “What injuries do you have?” and the other asking, “Which teams are involved in your care?” Participants were asked to identify their injuries across 11 categories: brain, eye, liver, lung, spleen, skin, kidney, blood vessel, digestive tract, other, and broken bone. The other category was left blank so participants could write in an answer not otherwise included on the survey. Pictographs were included to help participants with identification. Participants were also asked to identify their care teams in 13 categories: Trauma, Orthopedics, Neurosurgery, Ophthalmology, Plastic Surgery, Oral-Maxillofacial Surgery (OMFS), Ear/Nose/Throat (ENT), Vascular Surgery, Physical Therapy, Occupational Therapy, Speech Therapy, Nutrition, and Social Work.

Figure 1.

Patient Survey

For analysis, injury locations were combined by body region. Head injuries included brain and skull injuries. Face included eye and facial injuries. Thorax included lung and rib injuries. Abdomen/pelvis included liver, spleen, kidney, GI, and pelvic floor injuries. Spine represented spinal cord injuries and fractures of the spine. Upper extremities included left and right clavicles, scapulae, arms, forearms, and hand injuries. Lower extremities included left and right thigh, leg, and foot injuries. External was used to identify skin injuries. Blood vessel included vascular injuries. Other was anything written in by the patient in the blank slot provided on the form. The total number of possible injuries in each body region were summed; the number of true positives, true negatives, false positives, and false negatives was calculated for each body region and care team.

For analysis of the survey, percent correct and percent incorrect were determined, and a Cohen’s kappa (κ) for agreement was calculated. Cohen’s kappa values were interpreted as <0.5 demonstrates poor agreement, 0.5-0.8 demonstrates moderate agreement, and >0.8 demonstrates excellent agreement. To score the survey, participant answers were compared to the documentation in their electronic medical record. For injuries, correct was defined as injury present and correctly identified by the patient, as well as injury not present and correctly left unmarked. Incorrect was defined as injury present but unmarked, or injury not present but marked by the patient. For care teams, correct was defined as care team involved and correctly marked or care team uninvolved and correctly unmarked. Incorrect was defined as care team involved and omitted and care team not involved and incorrectly identified.

Participants also completed a validated health literacy questionnaire. The BRIEF health literacy screening tool consists of four written questions graded across 5-point Likert scales: 1) How often do you have someone help you read hospital materials, 2) How confident are you filling out medical forms by yourself, 3) How often do you have problems learning about your medical condition because of difficulty understanding written information, and 4) How often do you have a problem understanding what is told to you about your medical condition.(9, 10) This screening tool is validated for self-reporting health literacy in a general population of adults aged 18-64. Scores for this tool range from 2 through 20, and were grouped as recommended for this assessment, with 2-12 indicating inadequate health literacy, 13-16 indicating marginal health literacy, and 17-20 indicating adequate health literacy. Non-parametric Kruskal-Wallis tests were utilized to assess whether the 3-groups of health literacy level were associated with correct identification of injuries or care teams.

Data were analyzed using Stata/MP 16.1 (College Station, TX). This study was approved by our hospital system’s Institutional Review Board.

Results

Eighty-nine patients were approached and 50 consented to participate. The most correctly identified injures included: upper extremities, face, head, lower extremities, and abdomen/pelvis, with more than 90% correctly identified (Table 1). Injured body regions that were correctly identified by patients with high Cohen’s Kappa scores (κ =0.5-0.8) included: head, thorax, upper extremity, lower extremity, and spine, demonstrating moderate agreement (Table 3). Only head injuries had κ > 0.8, or excellent agreement (Table 3). The most missed injuries regions were external (32%, κ=0.18), spine (20%, κ=0.51), and blood vessel (18%, κ=0.29). Thirty-two percent of patients were not able to identify 50% or more of their injuries. Eight percent were not able to identify 75% or more.

Table 1.

Survey Results of Injured Body Region

Body Regions	Body Parts	Total	True +	False −	False +	True −	Total Incorrect	Percent Incorrect	Cohen's Kappa
Head	Skull	5	4	1	0	45	1	2%	0.88
	Brain	9	6	3	0	41	3	6%	0.77
Face	Face	0	0	0	2	48	2	4%	0.00
	Eye	2	2	0	2	46	2	4%	0.65
Thorax	Lung	17	7	10	0	33	10	20%	0.48
	Rib(s)	17	12	5	1	32	6	12%	0.72
Abdomen and Pelvis	Liver	2	1	1	1	47	2	4%	0.48
	Spleen	1	0	1	6	43	7	14%	0.00
	Kidney	1	1	0	3	46	3	6%	0.38
	Digestive Tract	6	3	3	1	43	4	8%	0.56
	Pelvis	6	5	1	1	43	2	4%	0.81
Spine	Spine	16	9	7	3	31	10	20%	0.51
Upper Extremities	Right Clavicle	1	0	1	0	49	1	2%	0.00
	Left Clavicle	1	1	0	1	48	1	2%	0.66
	Right Scapula	1	0	1	0	49	1	2%	0.00
	Left Scapula	0	0	0	0	50	0	0%	1.00
	Right Arm	3	2	1	2	45	3	6%	0.54
	Left Arm	3	1	2	1	46	3	6%	0.37
	Right Hand	2	2	0	1	47	1	2%	0.79
	Left Hand	1	1	0	0	49	0	0%	1.00
Lower Extremities	Right Leg	5	4	1	3	42	4	8%	0.44
	Left Leg	6	2	4	2	42	6	12%	0.00
	Right Foot	2	1	1	3	45	4	8%	0.16
	Left Foot	3	3	0	0	47	0	0%	1.00
External	Skin	14	3	11	5	31	16	32%	0.09
Blood Vessel	Blood Vessel	8	3	5	4	38	9	18%	0.29
Other	Other System	5	2	3	11	34	14	28%	0.18
	Other Bone	2	1	1	2	46	3	6%	0.21
None	None	8	7	1	3	39	4	8%	0.00

Of care teams, patients correctly identified OMFS, Plastic Surgery, and Ophthalmology more than 90% of the time (Table 2). A κ = 1 was obtained for both Trauma Surgery and OMFS, although OMFS was not consulted on any patients in this sample and no patients identified OMFS as a provider, resulting in perfect correlation. Ophthalmology and Orthopedics scored 0.5-0.8 for moderate agreement, and Plastic Surgery scored > 0.8, or excellent agreement. The teams most frequently omitted were Nutrition (72%, κ=0.00), Occupational Therapy (60%, κ=0.05), and Social Work (60%, κ=0.09). Forty-two percent of patients were not able to identify 50% or more of their medical teams. Twenty-eight percent of patients were not able to identify 75% or more of their medical teams.

Table 2.

Survey Results of Care Teams

Team	True Positive	False Negative	False Positive	True Negative	Total Incorrect	Percent Wrong	Cohen’s Kappa
Trauma	42	8	0	0	8	16%	1.00
Oral-Maxillo Surgery	0	0	0	50	0	0%	1.00
Plastic Surgery	6	1	0	43	1	2%	0.91
Orthopedics	17	10	3	24	13	26%	0.59
Ophthalmology	2	3	1	45	4	8%	0.54
Neurosurgery	9	12	3	30	15	30%	0.42
Vascular Surg	2	7	0	41	7	14%	0.32
Speech Therapy	1	5	2	44	7	14%	0.23
Social Work	15	30	0	5	30	60%	0.09
ENT	2	9	3	40	12	24%	0.09
Physical Therapy	31	18	0	1	18	36%	0.06
Occupational Therapy	18	30	0	2	30	60%	0.05
Nutrition	2	34	2	14	36	72%	0.00

All fifty patients completed the health literacy survey. Health literacy surveys revealed that 20 patients (40%) had scores between 2-12, indicating inadequate health literacy, 17 patients (34%) had scores indicating marginal health literacy, and 13 patients (26%) had scores indicating adequate health literacy. Self-reported health literacy was not associated with the percent of injuries correctly identified by a patient (p=0.9) or the percent of care teams correctly identified by a patient (p=0.9) in a 3-group Kruskal-Wallis test.

Discussion

Many patients were unable to correctly identify their injuries or care team members. Injuries to the head were correctly identified the most often, and most patients identified the trauma team as a part of their care team. Ancillary services appeared to be the most poorly recognized, such as Nutrition, Occupational Therapy, Physical Therapy, and Social Work. It is possible that patients may not have realized that these services are independent of their admitting care team, the trauma team. Self-reported health literacy was not associated with a patient’s ability to identify their injuries or care teams.

The hospital has been described as a “complex socio-technical” system made up of individual providers from different departments or practices discussing complicated concepts using words and terms that are unfamiliar to non-medical individuals.(11) It can easily become overwhelming to a patient when meeting with multiple providers. Physicians may be partly accountable, as they often neglect to introduce themselves.(12, 13) Regular use of medical terminology may also exacerbate patient confusion. Soto-Arnáez et al found that inappropriate medical abbreviations are used in discharge reports and prescriptions,(14) and Links and colleagues documented surgeon use of medical jargon during surgical consultations.(15) It is clear that coordinating communication between care teams and patients about their injuries and plan of care is a multifactorial challenge. Even in development of our own questionnaire or survey, our team was unsure of which terminology was appropriate for use in the questionnaire, and debated whether “clavicle” or “collarbone” was more likely to be understood, or whether “fracture” or “broken bone” should be utilized. Providers should utilize common language that is easy to understand by the average trauma patient without including jargon, and more study should be performed to determine what language is best understood in these circumstances.

In our small study of 50 patients, only 13 self-reported through the survey as having “adequate health literacy.” In addition, we did not find that health literacy was associated with better understanding of their inpatient trauma care. Prior studies have also found low health literacy levels among trauma patients. In a study of trauma patients by Swartz et al, patients with low health literacy were noted to have lower injury recall as well as lower follow-up rates.(16) It is logical that if a patient does not understand that they have an injury in a specific body part, they would be much less likely to follow up for that injury. Prior research from our own institution showed that nearly half of our trauma patients were nonadherent to all or part of follow-up plans that were communicated at the time of discharge.(7) Our data showed that increased numbers of referrals made at discharge were associated with increased likelihood of nonadherence. While our pilot study did not find a difference between health literacy and knowledge about their injuries and care teams, our study underscores that improved or more standardized methods of communication that are understood by the patients would be desirable for all trauma patients.

In the prior study from our institution, involvement with a psychosocial trauma support program in our institution was associated with improved adherence, suggesting that patients who feel that their care teams care about them may be more likely to adhere to care plans. Unfortunately, in trauma this is often not the case. Trauma patients tend to rate communication with their teams as low, but patients with low health literacy are even more likely to note dissatisfaction with physician communication.(17) Low health literacy patients are also less likely to perceive that their concerns were being heard.(17) Unfortunately, these communication deficiencies potentially have long-term consequences. A survey of patients six months post-trauma noted that 54% were unable to recall operations performed and 72% were unable to recall the name of any physician from their hospital stay, which was not related to education or income level.(18)

Health literacy and patient education have been of growing interest in recent years, with a growing number of studies highlighting potential ameliorating strategies. Two studies on video education for discharge from the emergency department have shown promise with this medium, with improved patient understanding of discharge instructions.(19, 20) Our institution has also piloted use of video education in our trauma patients to teach opioid safety, but noted limited success and challenging implementation.(21) Another communication model is closed loop communication, where one party gives information, the receiving party repeats the information and asks for clarification where necessary, and the original party verifies understand. This model has been used successfully in aviation for years and is being examined for use in medical training for provider-to-provider communication.(22) A similar form of communication is known as the teach-back method, which involves patients repeating back what they understand in their own words so that providers can assess their understanding and provide clarification in real time. Studies in several different fields of medicine have demonstrated the effectiveness of this form of communication with patients.(23-25) Finally, to improve patient identification and recognition of care teams, Arora et. al. had team members introduce themselves using FACE cards, which included a photo and explanation of their role on the care team.(26) To have effective and durable solutions, it is clear that this issue requires resources and time, both of which are unfortunately limited.

There are several important limitations to our study. Our study only included 50 participants, and the subgroup analysis did not have sufficient power to draw conclusions. Bias was potentially introduced as patient selection was connected to the nursing and therapy staff identifying patients who were alert and able to participate in the survey. We conducted the survey in English only, and trauma patients who are non-English speaking are likely at high risk for misunderstanding and subsequent nonadherence due to poor communication. Our criteria included a hospital stay of greater than two days to help ensure that patients would have met the trauma team at least once but it is possible that not all teams had an opportunity to communicate with the patient by the time of the survey administration. This study was also performed at a single trauma center which may not be reflective of the health literacy of patients in different regions, nor does it capture how other institutions may differ in communication between care teams and patients. Despite these limitations, our study had important and significant findings that even at the time of admission to the hospital, patients did not understand their injuries or their care teams, suggesting that there are important opportunities to improve the communication with our patients.

Adherence to care plans following traumatic injury is an important step in early identification of complications and prevention against unplanned readmissions.(2) Our study found that many trauma inpatients were unable to correctly identify their injuries and care teams despite a range of self-reported health literacy scores. Communication during the hospital stay can be challenging for both the patients and the providers, but attention to improved patient comprehension is warranted. A standardized approach to ensure not only that education occurs but is understood by the patients may help enhance the recovery of trauma patients.