Specific Aims:
(1) To describe the goals of care and characteristics of clinical encounters (e.g. thoroughness, explanation, listening, respect, waiting time) valued most greatly by hospitalized patients, stratified by certain important patient characteristics, such as gender, race, preferences for shared decision-making, and health status.
(2) To compare outcomes (e.g. patient satisfaction, length of stay, and readmission rates) across two groups of hospitalized patients: (a) an intervention group cared for by a hospitalist or resident team that receives patient-specific information regarding their preferences, values, and goals of care; and (b) a usual care group whose physicians do not receive such information.

Decisional Dilemma:
The literature on the impact of hospitalists on patient satisfaction is scant. However, given their expanding role across the health care system, a rigorous understanding of their effect on patient-centered outcomes is critical. In addition, since the hospitalist model of care will almost certainly continue to proliferate, identifying interventions that improve patient satisfaction with care in the inpatient setting is essential. Some literature has suggested that the hospitalist model of care does not have a deleterious impact on patient satisfaction despite the discontinuity of care it introduces,1 and may improve efficiency of care.2 However, this literature is based on associations between hospitalist staffing ratios and patient satisfaction data aggregated to the hospital level, not patient-level satisfaction directly linked to their inpatient physician during a specific hospitalization.1 Although the differences were not substantial, other literature does show that hospitalized patients cared for by their primary care physicians were more satisfied than those cared for by a hospitalist.3 Most importantly, no studies have evaluated targeted interventions to communicate the dimensions of care most valued by individual patients to their inpatient physician(s). Therefore, the proposed work would seek to measure the impact of an intervention through which inpatient physicians receive detailed information regarding the preferences and values of the patients for whom they are caring. A greater understanding of how such clarification and communication of values influences patient satisfaction may have implications for improving patient satisfaction among hospitalized patients, many of whom are vulnerable due to poor health status, poor health literacy, or other factors.

Gap Analysis:
Many clinical decisions are driven by physician preferences for certain types of care (e.g. surgical versus non-surgical) with insufficient attention to patient preferences, values, and goals.4 However, the extent to which a greater understanding of patient preferences among hospitalists would impact care and patient-centered outcomes is not known. Moreover, the identification and implementation of interventions to improve communication in the inpatient setting has been particularly challenging. For example, the Study to Understand Prognoses and Preferences for Outcomes and Risks of Treatments (SUPPORT) showed that the studied intervention did not change the proportion of patients or surrogates who reported a discussion about cardiopulmonary resuscitation nor the proportion of patients whose preferences vis-à-vis the occurrence of such a discussion were respected.5 Furthermore, there is some evidence to suggest that a poor understanding of patient preferences creates and/or exacerbates utilization and health disparities,6 especially since gender, employment status and health status impact patients’ likelihood of receiving care7 and the dimensions of care associated with greater satisfaction.8 The potential for greater inpatient physician knowledge of patient values in reducing disparities in care received and satisfaction has not been established. Various methods of values clarification have been studied in the outpatient setting.9 Despite this, there is no consensus on the most robust methods to determine legitimate patterns of attribution importance and patient preferences. Moreover, less is known about such methods of values clarification among hospitalized patients cared for by physicians with whom they have not had the benefit of a longitudinal relationship.

1. An intervention group cared for by a hospitalist or resident team that receives patient-specific information
regarding preferences, values, and goals of care
2. A usual care group whose physicians do not receive such information

Study Design:
We will approach currently hospitalized patients who have consented to enrollment in the Hospital Project, which is an
ongoing cohort study of hospitalized general medicine patients at the University of Chicago. This study includes inpatient
and follow-up post-discharge surveys, access to medical records, and administrative data for the hospitalization. If
patients enrolled in the Hospitalist Project also agree to be enrolled in the study being proposed here, they will be
randomly assigned to the intervention or usual care group. Prior to randomization, all enrollees will be asked to agree to
a post-discharge telephone survey to measure their satisfaction with care during the hospitalization. This survey will be
based on the Hospital Consumer Assessment of Healthcare Providers and Systems (HCAHPS) survey.
The intervention group will be interviewed as soon as possible after admission to obtain additional information to clarify
their values and preferences for care and communication with health care providers. The content of the interview
instrument will be based on: (1) goal selection using the goal attainment scaling (GAS) literature;10 (2) the domains of
care relevant to patients identified in the HCAHPS survey and the Community Tracking Study (e.g. thoroughness,
explanation, listening, respect, waiting time). GAS has been used across many different patient populations as a method
to select and scale patient goals, values, and preferences to align interventions with objectives. In this study, GAS would
be used for patient value clarification to enable physicians to tailor diagnostic and therapeutic decisions as well as
communication content and style to patient values and preferences.
We will develop and test a concise method of summarizing an individual patient’s goals and preferences for use by
physicians. This summary will then be communicated as rapidly as possible (through the hospital paging system) to each
intervention arm patient’s inpatient physician.
Based on the post-discharge surveys, we will then conduct comparisons of the two groups according to measures of
patient satisfaction with care during the hospitalization as the primary outcome. We will specify length of stay and
readmission rates as secondary outcomes.

Study Population and Setting:
The University of Chicago Medical Center (UCMC) is an academic medical center located on the south side of Chicago,
where many patients are vulnerable due to low socioeconomic status, low educational attainment, and poor health
literacy/numeracy. The proposed study would enroll patients hospitalized on the general internal medicine teaching
service or hospitalist services at UCMC who consent to be interviewed for the Hospitalist Project and also agree to be
enrolled in the proposed study. Of hospitalizations eligible for enrollment in the Hospitalist Project, approximately 70%
consent to be interviewed (12.8% refuse, while 17.3% and 0.3% are discharged or die prior to being approached,
respectively).11 The population of patients enrolled has the following characteristics: mean age 57.1, 59.9% female
gender, 75.7% African American race, 55.3% high school graduate or less education, 23.4% insured through Medicaid,
53.6% general self-assessed status of fair or poor (as opposed to excellent, very good, or good).11

Sample Size and Power:
Among a sample of patients from the Community Tracking Study Household Survey, 72.0%, 76.4%, and 75.3% of patients
reported excellent or very good patient ratings of examination thoroughness, physician explanation, and physician
listening, respectively (mean = 74.6).8 Among this sample, 64.2% were very satisfied with their care. Using published
data on GAS suggesting a 33.6% improvement in goal achievement consistent with patient preferences,10 we estimate
this as the maximum possible treatment effect obtainable through our proposed intervention. Assuming a treatment
effect of this magnitude, we would only require 62 patients in each arm of the study to detect such a difference
between the two groups with 80% power and 95% confidence. A more conservative treatment effect estimate of 10%
improvement would require 833 patients in each arm. Thus, we will target enrolling 1666 patients (N [total] = 1666,
N1 = 833, N2 = 833) in each arm of the proposed study.

References:
1. Chen LM, Birkmeyer JD, Saint S, Jha AK. Hospitalist staffing and patient satisfaction in the national Medicare population. J Hosp Med. 2013;8(3):126-131.
2. Lindenauer PK, Rothberg MB, Pekow PS, Kenwood C, Benjamin EM, Auerbach AD. Outcomes of care by hospitalists, general internists, and family physicians. N Engl J Med. 2007;357(25):2589-2600.
3. Seiler A, Visintainer P, Brzostek R, et al. Patient satisfaction with hospital care provided by hospitalists and primary care physicians. J Hosp Med. 2012;7(2):131-136.
4. Birkmeyer JD, Reames BN, McCulloch P, Carr AJ, Campbell WB, Wennberg JE. Understanding of regional variation in the use of surgery. Lancet. 2013;382(9898):1121-1129.
5. A controlled trial to improve care for seriously ill hospitalized patients. The study to understand prognoses and preferences for outcomes and risks of treatments (SUPPORT). The SUPPORT Principal Investigators. JAMA. 1995;274(20):1591-1598.
6. Katz JN. Patient preferences and health disparities. JAMA. 2001;286(12):1506-1509.
7. Tak HJ, Hougham GW, Ruhnke A, Ruhnke GW. The effect of in-office waiting time on physician visit frequency among working-age adults. Soc Sci Med. 2014;118:43-51.
8. Tak H, Ruhnke GW, Shih YC. The Association between Patient-Centered Attributes of Care and Patient Satisfaction. Patient. 2015;8(2):187-197.
9. Pignone MP, Howard K, Brenner AT, et al. Comparing 3 techniques for eliciting patient values for decision making about prostate-specific antigen screening: a randomized controlled trial. JAMA Intern Med. 2013;173(5):362-368.
10. Rockwood K, Howlett S, Stadnyk K, Carver D, Powell C, Stolee P. Responsiveness of goal attainment scaling in a randomized controlled trial of comprehensive geriatric assessment. J Clin Epidemiol. 2003;56(8):736-743.
11. Tak HJ, Ruhnke GW, Meltzer DO. Association of patient preferences for participation in decision making with length of stay and costs among hospitalized patients. JAMA Intern Med. 2013;173(13):1195-1205.