Patient Satisfaction with an Emergency ... - Wiley Online Library

12 downloads 27966 Views 59KB Size Report
support, physical comfort, and special needs, than did the inpatient group. ... of Public Health, University of Illinois at Chicago, Chicago, IL. (RJR, GLA); Center ...
178

ASTHMA UNIT

Rydman et al. • PATIENT SATISFACTION WITH ASTHMA OBSERVATION UNIT

Patient Satisfaction with an Emergency Department Asthma Observation Unit ROBERT J. RYDMAN, PHD, REBECCA R. ROBERTS, MD, GARY L. ALBRECHT, PHD, ROBERT J. ZALENSKI, MD, MICHAEL MCDERMOTT, MD

Abstract. Objective: To compare levels of patient satisfaction between the diagnostic and treatment protocols in an ED-based asthma observation unit (AOU) and those with standard inpatient hospitalization. Methods: This was a prospective, randomized, controlled trial with a sample of 163 patients presenting to the ED with acute asthma exacerbations over a 30-month period. Eligible patients were those who could not resolve their symptoms after three hours of standard ED therapy. Patients were then randomly assigned to an ED-based AOU (experimental group) or to customary inpatient care (control group). Patient satisfaction and problems with care processes were assessed by standardized instrumentation at discharge in both groups. Results: The AOU patients scored higher than those randomized to the inpatient hospitalization protocol on four summary ratings of patient satisfaction measures: received service wanted, recommendation of the service to others, satisfaction with the service, and overall

A

STHMA is a chronic condition with acute episodes necessitating continuous and coordinated care to prevent exacerbations, control symptoms, monitor treatment, and reduce chronic airway inflammation. Inconsistent or inadequate self-management of asthma, combined with a lack of or nonstandardized community-based primary care, often results in episodic treatment of chronic conditions in hospital EDs.1 As a result of this lack of early intervention and/or coordinated care, the From the Division of Health Policy and Administration, School of Public Health, University of Illinois at Chicago, Chicago, IL (RJR, GLA); Center for Health Services Research, University of Illinois at Chicago, Chicago, IL (RJR, RRR, RJZ); Departments of Emergency Medicine, Cook County Hospital, Rush University, Chicago, IL (RJR, RRR, RJZ); and Department of Emergency Medicine, Wayne State University, Detroit, MI (MM). Received July 1, 1998; revision received November 4, 1998; accepted November 10, 1998. Presented at the SAEM annual meeting, Denver, CO, May 1996. Supported by the Agency for Health Care Policy and Research (Grant HHSHS 07103). Address for correspondence and reprints: Robert J. Rydman, PhD, Department of Emergency Medicine, Cook County Hospital, 1900 West Polk Street, 10th Floor, Chicago, IL 60612. Fax: 312-633-8189; e-mail: [email protected]

satisfaction. The AOU patients reported fewer total number of problems with care received, and fewer specific problems with communication, emotional support, physical comfort, and special needs, than did the inpatient group. However, the AOU patients reported more problems regarding their knowledge of financial costs and liabilities for their service than did the inpatients. Conclusion: Patients were more satisfied and had fewer problems with rapid diagnosis and treatment in the AOU than they did with routine inpatient hospitalization. Since AOUs represent a new ambulatory service modality, patients would benefit from greater awareness of the costs and coverage for AOUs as compared with hospital inpatient care. These findings have important implications for the future short- and long-term success and feasibility of ED-based AOUs. Key words: patient satisfaction; observation units; asthma; asthma randomized trial. ACADEMIC EMERGENCY MEDICINE 1999; 6:178 – 183

rise in preventable hospitalizations for chronic conditions such as asthma is startling. This is especially true for large urban areas. In a comparison of Canada and the United States, there were marked differences in ambulatory care sensitive (ACS) admissions between major cities, with the United States having much higher rates than Canada. ACS conditions as described by the Institute of Medicine are those diagnoses that are theoretically manageable in ambulatory care settings, but often are managed in hospital EDs. They also have a higher than usual chance for an inpatient admission disposition, and have been coined as ‘‘preventable hospitalizations.’’ Within the United States, large cities average ACS admission rates 3.7 times higher in poor areas than in high-income areas.2 During a study period between 1982 and 1991, the rates for ACS admissions for asthma went up 45%.2 The use of asthma observation units (AOUs) in EDs is an innovative strategy designed to deal with the increased number of asthma cases and the major problem of preventable hospitalizations.3 – 5 The goal of AOUs is to develop rapid diagnostic and treatment protocols that will

ACADEMIC EMERGENCY MEDICINE • March 1999, Volume 6, Number 3

quickly address patient needs and offset inpatient care costs while offering equivalent or superior biomedical, health, quality-of-life, and patient satisfaction outcomes. Recent research indicates that AOUs can provide more effective treatment alternatives at lower cost for refractory asthmatic populations.3 – 5 This study examines the patient satisfaction outcomes of being treated in an AOU vs the more routine inpatient hospitalization. Patient satisfaction is an integral component of contemporary medical care because satisfied patients are more likely than unsatisfied patients to listen to and follow doctors’ orders, adhere to their treatment regimens, use referrals, consume fewer services, and remain in a coordinated system of care and are less likely to file malpractice claims.6 – 9 Satisfaction is also important under managed care plans where members select physician groups and treatment plans with specific coverages. As an integral component of clinical outcomes research, patient satisfaction has been driven by the rise of consumerism in Western medical practice and by the emphasis on containing costs and managing care to achieve cost-effective outcomes.10 This research uses a conceptual approach and measurement of patient satisfaction developed and tested in a large national survey to a randomized trial of ambulatory and inpatient interventions for asthma.11,12 The results have immediate implications for patient care and the direction of future outcomes research on ED-based observation units.13 – 15

METHODS Study Design. This study evaluated patient satisfaction in a prospective, randomized, controlled clinical trial comparing an ED-based AOU with customary inpatient care for management of acute asthma exacerbations. This study was approved by the institutional review boards of Cook County Hospital and the University of Illinois at Chicago, and written informed consent was obtained from all participants. Study Population and Setting. The study sample for this prospective clinical trial evaluating asthma management was constituted from acute asthmatic persons presenting to the ED during randomly selected shifts over a 30-month period. Acute asthma patients who met objective criteria for hospital admission after three hours of standard ED therapy (n = 163) were randomized to an experimental group receiving AOU care or a control group receiving inpatient admission. Complete eligibility criteria have been published elsewhere.16

179

Study Protocol. Patients were excluded from the study if: 1) the peak expiratory flow rate (PEFR) deteriorated or failed to improve (to 40% of predicted) after aggressive ED treatment for up to three hours; 2) the PaCO2 was >45 torr; 3) there was fever >101⬚F; 4) there was evidence of chronic obstructive pulmonary disease (COPD) secondary to smoking or a restrictive lung disease; 4) the patient was pregnant; 5) the patient lacked a home, lacked access to a phone, had psychiatric disorders, or had any other reason that would impair his or her ability to cooperate with follow-up treatment and reassessments; or 6) the patient refused medication.16 Patients received standard aggressive asthma therapy including adrenergic agonists administered by hand-held nebulizers and corticosteroids. Complete treatment protocol are described in a previous publication.16 The inpatient protocol was based on national guideline recommendations and indicated three aggressive adrenergic agonists administered by hand-held nebulizers and IV steroids. Discharge criteria were the same as for the AOU. Measures. To perform a check on the randomization procedure, baseline measures were taken after determination of patient eligibility, consent, and randomization. These included: social demographic characteristics, patient autonomy in care, social support, and perception of health variables. These measures are routinely used in the health outcome literature and are known to be correlated with summary satisfaction with care outcomes.17,18 Trained researchers who were not involved in direct patient care interviewed the study patients. The same interviewers took outcome measurements of patient satisfaction and problems with health care at the time of discharge for inpatient controls and at the end of the diagnostic protocol for the AOU patients. The global outcome measures assessed quality of service, receipt of desired services, fulfillment of needs by the hospital, recommendation of the hospital’s service to others, satisfaction with service received, effective handling of the health care problem, and overall satisfaction. Patients rated these items on a fourpoint Likert scale ranging from 1 (lowest) to 4 (highest). Intermediate outcome measures included 50 process-of-care problem areas grouped into the nine general domains of physical comfort, emotional support, communication, management of pain, special needs and preferences, family involvement, discharge preparation, patient education, and financial information/perceived burden. From these data, ten new study variables were created in each of the nine domain groups, plus a ‘‘total problems with care’’ composite variable. These

180

ASTHMA UNIT

Rydman et al. • PATIENT SATISFACTION WITH ASTHMA OBSERVATION UNIT

TABLE 1. Patient Characteristics by Treatment Group EDOU* (Experimental) Group (n = 81)

Inpatient (Control) Group (n = 82)

55% 45%

59% 41%

0.50 0.50

37 (⫾10) yr 36 yr

36 (⫾10) yr 35 yr

0.627 0.69

Proportion of admissions during preceding year

24%

19%

0.456

No. ED visits during preceding six months — median (range)

2 (1 – 5)

2 (1 – 4)

0.368

27 (8.1%)

28 (8.8%)

0.30

Proportion of smokers

38%

33%

0.502

Proportion with a smoker in the domicile

39%

47%

0.494

Ethnicity Black Other

82% 18%

81% 19%

0.312 0.312

High autonomy — mean (⫾SD)§ Low autonomy — mean (⫾SD)‫ن‬

1.70 (⫾0.52) 1.98 (⫾0.62)

1.65 (⫾0.49) 1.89 (⫾0.60)

0.567 0.328

Social support — mean (⫾SD)

1.12 (⫾0.18)

1.12 (⫾0.18)

0.983

Gender Men Women Age Mean (⫾SD) Median

PEFR‡ — mean (% predicted)

p-value†

*EDOU = ED observation unit. †␹ 2, Student’s t-test, or Wilcoxon rank-sum test. ‡PEFR = peak expiratory flow rate. §1, yes; 2, no. ‫ن‬1, yes; 2, no.

instruments and measures have been widely used, validated, and reported in the literature.4,11,12,17,19 In addition, asthma patients in both study arms were also asked to provide their perceptions on the length of hospital stay using a five-point Likert scale that ranged from 1 (substantially shorter) through 3 (about right) to 5 (substantially longer). Data Analysis. We initially generated univariate descriptive statistics for all of the major study variables. For categorical level variables, frequencies and 95% confidence intervals were calculated. For continuous level variables, means and standard deviations and tests of normality were computed. To test the main study hypothesis concerning experimental and control group differences in patient satisfaction and problems with care processes, we used Student’s t-tests and Wilcoxon rank-sum tests.

RESULTS One hundred sixty-three consenting asthma patients were sampled for the study and were ran-

domly assigned to the AOU experimental group (n = 81) or inpatient control group (n = 82). Their demographic characteristics and clinical histories are summarized in Table 1. Patients randomized to each study arm were statistically comparable. In general, patients were predominately male (experimental = 55% vs control = 59%), and black (experimental = 82% vs control = 81%). The mean age was 37 years for the experimental group vs 36 years for the control group. Both experimental and control groups had: 68% with at least 12 years of education, 36% currently employed, a mean number of four people per household, and a median perception of health slightly above the scale midpoint of 3.0. At the baseline, experimental and control patient scores revealed no statistical difference with regard to the perceived level of autonomy and social support. Table 1 also demonstrates statistical equivalence between the experimental and control groups on clinical performance and exposure variables. Thirty-eight percent of the AOU patients vs 33% of the inpatients were active smokers. The AOU group had 39% with a smoker in the home vs 33% of the inpatients. The AOU patients had a mean

181

ACADEMIC EMERGENCY MEDICINE • March 1999, Volume 6, Number 3

predicted PEFR of 27% (SD ⫾ 8.1) vs 28% (SD ⫾ 8.8) for the inpatients. In terms of clinical condition and experience, 24% of the AOU patients vs 19% of the inpatients had a hospital admission in the preceding year. Patients in both study arms had a median of two visits to the ED in the preceding six months. Taken together, these findings suggest that our randomization procedures were successful. The data in Table 2 present the results for the test of the major study question: Are there differences in global patient satisfaction between the AOU and inpatient hospitalization groups? The patients in the AOU scored higher after diagnosis and treatment than did those in the inpatient group on all seven global satisfaction measures. Statistically significant differences were demonstrated on four of these satisfaction measures: received service wanted, recommendation of the service to others, satisfaction with the service, and overall satisfaction. Intermediate study outcomes measured by total numbers of problems with care, communication, financial information, emotional support, special

needs, and physical comfort are presented in Table 3. There are statistically significant differences between the AOU and inpatient hospitalization groups on six of these variables. The patients in the AOU study arm reported fewer total number of problems with care and fewer problems with communication, emotional support, physical comfort, and special needs than did the inpatient group. However, the inpatients reported fewer problems relative to understanding their financial liability and costs of care they received. No difference was found between the experimental and control groups on perceived length of stay. Both groups averaged a mean of 3.0 (i.e., about right), as contrasted with scale anchor points of 1.0 (substantially shorter than expected) to 5.0 (substantially longer than expected).

DISCUSSION Our study results have important and far-reaching implications for emergency medicine. It seems clear that ED-based diagnostic and treatment units can provide care that is equivalent to, and

TABLE 2. Global Satisfaction Outcomes by Treatment Group* EDOU† (n = 81) Mean (⫾SD) Service quality Received service wanted Hospital met needs Recommended service Satisfied with service Treatment effective Satisfaction level

3.17 3.41 3.56 3.73 3.48 3.78 3.59

(⫾0.71) (⫾0.77) (⫾0.65) (⫾0.57) (⫾0.72) (⫾0.45) (⫾0.65)

Inpatient (n = 82) Mean (⫾SD) 2.92 3.08 3.35 3.45 3.21 3.63 3.33

(⫾0.93) (⫾0.94) (⫾0.76) (⫾0.71) (⫾0.88) (⫾0.68) (⫾0.84)

95% CI t-test

p-value‡

1.88 2.46 1.83 2.79 2.07 1.68 2.21

0.04 0.01 0.05 0.00 0.03 0.09 0.03

EDOU 3.01, 3.24, 3.42, 3.61, 3.32, 3.68, 3.45,

Inpatient

3.32 3.58 3.70 3.85 3.64 3.88 3.73

2.71, 2.88, 3.19, 3.30, 3.02, 3.48, 3.15,

3.12 3.28 3.51 3.60 3.40 3.78 3.51

*Scale ranges from 1 (lowest, strongly disagree, most dissatisfied) to 4 (highest, strongly agree, most satisfied). †EDOU = ED observation unit. ‡p-values are for Student’s t-test or the Wilcoxon rank-sum test.

TABLE 3. Problems with Care Processes, by Treatment Group* EDOU† (n = 81) Mean (⫾SD) Total problems with care Communication Financial information Emotional support Physical comfort Special needs Patient education Family involvement Pain Discharge preparation

1.13 1.24 1.75 1.04 0.72 1.22 1.23 1.50 1.58 1.42

(⫾0.18) (⫾0.29) (⫾0.32) (⫾0.09) (⫾0.35) (⫾0.16) (⫾0.31) (⫾0.48) (⫾0.22) (⫾0.27)

Inpatient (n = 82) Mean (⫾SD) 1.20 1.39 1.59 1.10 0.95 1.29 1.22 1.59 1.51 1.47

(⫾0.17) (⫾0.25) (⫾0.35) (⫾0.21) (⫾0.36) (⫾0.17) (⫾0.30) (⫾0.45) (⫾0.27) (⫾0.28)

95% CI t-test

p-value‡

2.24 3.29 ⫺2.53 2.21 3.50 2.18 0.12 0.95 1.31 1.14

0.03 0.00 0.02 0.03 0.0006 0.03 0.91 0.35 0.19 0.26

EDOU 1.09, 1.21, 1.68, 1.02, 0.64, 1.19, 1.16, 1.40, 1.53, 1.36,

1.17 1.33 1.82 1.06 0.80 1.25 1.30 1.60 1.63 1.48

Inpatient 1.16, 1.34, 1.51, 1.05, 0.87, 1.25, 1.16, 1.49, 1.45, 1.41,

1.24 1.44 1.67 1.15 1.03 1.33 1.28 1.69 1.57 1.53

*Generally scores ranged from low to high, with higher scores indicating larger numbers of adverse events for each cluster and for all problems with care processes combined. Scale ranges from 1 to 2, with higher mean scores indicating larger numbers of adverse events. For example, 1 = yes, 2 = no (question: ‘‘Did your provider adequately explain the limits of your financial obligation for your care?). †EDOU = ED observation unit. ‡p-values are for Student’s t-test or the Wilcoxon rank-sum test.

182

ASTHMA UNIT

Rydman et al. • PATIENT SATISFACTION WITH ASTHMA OBSERVATION UNIT

seemingly more satisfying to patients than, traditional inpatient hospitalization for people who present with acute asthma. This means that patients who meet study criteria (i.e., 20 – 65% of all asthmatic patients presenting to various EDs depending on casemix) now being referred to inpatient hospital services would perceive themselves as better served by an ED-based AOU.16 A companion study also revealed that AOU patients as compared with inpatients had: superior quality of life seven days postdischarge, equivalent rates of relapse necessitating emergency or primary care over eight weeks of follow-up, and lower costs by remaining in the ED for their acute episodes.4 Patient satisfaction results found in this study provide insight into the nature and types of differences in patient satisfaction and problems with care processes between the experimental and control groups. An examination of our major study outcomes revealed that the AOU group reported greater satisfaction and fewer problems with care than did those in the inpatient control group. Four sets of patient satisfaction variables were statistically significant and the others showed strong trends toward significance (p < 0.1). Five sets of process-of-care problems were fewer in the AOU group, including total numbers of problems. It was remarkable that the staff in a busy inner-city ED could outperform the hospital inpatient service in these areas. In another light, these study results also suggest that patients who might avoid or delay treatment because of lack of access to primary care, or the time, expense, and inconvenience of inpatient care, may be much more amenable to visiting the ED for a limited length of time to alleviate their symptoms and receive advice and medications for a 12-hour treatment period. Providing improved and more accessible care for the population of individuals with asthma is a major public health goal of Healthy People 2000.20 Therefore, offering a critical diagnostic and treatment service in the ED meets a long-standing public health goal as well as the cost-containment goals of providing health care in the most accessible and least restrictive environment. Physicians, policy makers, insurers, and managed care representatives should not underestimate the appeal and attraction of this new alternative source of treatment. Patients will continue to present at places where they are comfortable, trust the service as being effective, and know that they will be engaged regardless of payment ability. Patients also did not perceive their lengths of stay as being too short or too long. The ED, in this instance, provides them with a very attractive choice of access, quality, and cost savings.

The field of emergency medicine should carefully consider providing AOUs in those locations where they make most sense, epidemiologically and economically. Initally, urban EDs with high patient volumes seem to be a logical place to begin extending such coverage. However, health care systems operating under capitated reimbursement systems, or seeking to restructure or lower fixed cost expenditures, would also be good AOU candidates. This study provides insights for replicating successful AOUs with regard to what patient care consumers find to be their most important features, namely, communication, emotional support, physical comfort, special needs, and assurance of financial coverage.

LIMITATIONS AND FUTURE QUESTIONS It is important to stress both careful planning of the research design and selection of valid and appropriate instruments for measuring patient satisfaction with health care. Previous research has shown that there are serious threats to validity in this work. Patient factors such as age, sex, perceptions of poor health, high levels of autonomy, low levels of social support, and depressed affect are known to influence the reporting of satisfaction with care in characteristically different ways.10 – 14,18 Attending to these potential confounding factors, we used a successful randomized experimental design in which patient casemix, age, sex, level of autonomy, level of social support, and level of depressed affect were controlled by the design. We attended to issues of measurement appropriateness and validity by selecting a set of instruments that had been extensively used and validated in a large national study on patient satisfaction.11,12 One limitation of this study is the use of a single institution for patient sampling and the conduct of the clinical trial interventions. Ideally, this study would have been performed at multiple institutions to rule out site-specific rival explanations for patient preferences, especially with regard to type of patients served and institutional reputation. The reality is that this research was adequately funded through federal granting agencies, but not to the point of multihospital involvement. The field will have to rely on future replicative studies in other locations to ensure generalization of trends found here. Researchers are encouraged to include patient satisfaction measures in future AOU outcome studies to establish a body of knowledge for the ED-based observation medicine industry.

ACADEMIC EMERGENCY MEDICINE • March 1999, Volume 6, Number 3

CONCLUSION This study sought to examine differences in patient satisfaction among patients with asthma presenting to the ED who were randomized to an extended treatment AOU vs customary hospital inpatient care following nonresolution of their exacerbation within three hours of standard ED therapy. Results indicated AOU patients were more satisfied and had fewer problems with care than hospital inpatients. Patient satisfaction outcomes provide an added dimension of hospital performance information, which may tip the scale in selecting medical care services when biomedical and cost comparisons may not provide a clear choice. For this reason, coupled with the good business practice of providing valuable sociological feedback to the American polity on medical innovations, it is a worthwhile endeavor. References 1. Albrecht GL, Slobodkin D, Rydman RJ. The role of emergency departments in American health care. Res Soc Health Care. 1996; 13:287 – 313. 2. Billings J, Anderson GM, Newman LS. Recent findings on preventable hospitalizations. Health Aff. 1996; 15:239 – 49. 3. McDermott MF, Murphy DG, Zalenski RJ, et al. A comparison between emergency diagnostic and treatment unit and inpatient care in the management of acute asthma. Arch Intern Med. 1977; 157:2055 – 62. 4. Rydman RJ, Isola ML, Roberts RR, et al. Emergency department observation unit (EDOU) versus hospital inpatient care for a chronic asthma population. Med Care. 1998; 36: 599 – 609. 5. McCarren M, McDermott M, Zalenski, RJ, et al. Prediction

183

of relapse within eight weeks after an acute asthma exacerbation in adults. J Clin Epidemiol. 1998; 51:107 – 18. 6. Waitzkin H. Doctor – patient communication: clinical implications of social scientific research. JAMA. 1984; 252:2442 – 6. 7. Waitzkin H. Information giving in medical care. J Health Soc Behav. 1985; 26:81 – 93. 8. Waitzkin H. The Politics of Medical Encounters: How Patients and Doctors Deal with Social Problems. New Haven, CT: Yale University Press, 1991. 9. Waitzkin H. Patient – doctor communication in cross national perspective. Med Care. 1996; 34:641 – 71. 10. Sitza J, Wood N. Patient satisfaction: a review of issues and concepts. Soc Sci Med. 1997; 45:1829 – 43. 11. Cleary PD, Edgman-Levitan S, McMullen W, Delbanco TL. The relationship between reported problems and summary evaluations of hospital care. QRB Qual Rev Bull. 1992; 18(2): 53 – 9. 12. Cleary PD, Edgman-Levitan S, Roberts M, et al. Patients evaluate their hospital care: a national survey. Health Aff. 1991; 10:254 – 67. 13. Donabedian A. Quality assessment and assurance: unity of purpose, diversity of means. Inquiry. 1988; 25:173 – 92. 14. Jenkinson C (ed). Measuring Health and Medical Outcomes. London: UCL Press, 1994. 15. Rossi OV, Kinnula VL, Huhti E. Emergency room visits for acute attacks of asthma: characterization of patients and visits. Respiration. 1991; 58:21 – 5. 16. Beveridge RC, Grunfeld AF, Hodder RV, Verbeek PR, for the CAEP/CTS Asthma Advisory Committee. Guidelines for the emergency management of asthma in adults. Can Med Assoc J. 1996; 155:25 – 37. 17. Rydman RJ, Zalenski RJ, Roberts RR, et al. Patient satisfaction with an emergency department chest pain observation unit. Ann Emerg Med. 1997; 29:109 – 15. 18. Abramowitz S, Cote AA, Berry E. Analyzing patient satisfaction: a multi-analytic approach. QRB Qual Rev Bull. 1987; 13:122 – 30. 19. Ware JE, Synder MK, Wright WR. Defining and measuring patient satisfaction with medical care. Evaluation and Program Planning. 1983; 6:247 – 63. 20. US Public Health Service. Healthy People 2000: National Health Promotion and Disease Prevention Objectives, Clinical Preventive Services, 529-548. Washington, DC: US Department of Health and Human Services, US Public Health Service, 1990; DHHS publication no. (PHS) 91-50212.