PAIN MANAGEMENT/ORIGINAL RESEARCH

Lack of Association Between Press Ganey Emergency Department Patient Satisfaction Scores and Emergency Department Administration of Analgesic Medications Tayler M. Schwartz, BS; Miao Tai, MS; Kavita M. Babu, MD; Roland C. Merchant, MD, ScD

Study objective: We explore the relationship between Press Ganey emergency department (ED) patient satisfaction scores and ED administration of analgesic medications, including amount of opioid analgesics received, among patients who completed a patient satisfaction survey. Methods: We conducted a secondary data analysis of Press Ganey ED patient satisfaction surveys from patients discharged from 2 academic, urban EDs October 2009 to September 2011. We matched survey responses to data on opioid and nonopioid analgesics administered in the ED, demographic characteristics, and temporal factors from the ED electronic medical records. We used polytomous logistic regression to compare quartiles of overall Press Ganey ED patient satisfaction scores to administration of analgesic medications, opioid analgesics, and number of morphine equivalents received. We adjusted models for demographic and hospital characteristics and temporal factors. Results: Of the 4,749 patients who returned surveys, 48.5% received analgesic medications, and 29.6% received opioid analgesics during their ED visit. Mean overall Press Ganey ED patient satisfaction scores for patients receiving either analgesic medications or opioid analgesics were lower than for those who did not receive these medications. In the univariable polytomous logistic regression analysis, receipt of analgesic medications, opioid analgesics, and a greater number of morphine equivalents were associated with lower overall scores. However, in the multivariable analysis, receipt of analgesic medications or opioid analgesics was not associated with overall scores, and receipt of greater morphine equivalents was inconsistently associated with lower overall scores. Conclusion: Overall Press Ganey ED patient satisfaction scores were not primarily based on in-ED receipt of analgesic medications or opioid analgesics; other factors appear to be more important. [Ann Emerg Med. 2014;-:1-13.] Please see page XX for the Editor’s Capsule Summary of this article. 0196-0644/$-see front matter Copyright © 2014 by the American College of Emergency Physicians. http://dx.doi.org/10.1016/j.annemergmed.2014.02.010

INTRODUCTION Background Opioid analgesic prescribing has increased concurrently with a sharp increase in opioid-related addiction and death.1,2 Although the majority of patients who receive opioid analgesics long term will not become addicted,3 the surge in opioidrelated adverse events has led to an increased focus on responsible prescribing. The Food and Drug Administration identifies prescribers as essential to the reduction of opioid analgesic misuse.4 The majority of emergency department (ED) visits involve treatment of painful conditions.5 Clinical encounters in the ED about treatment of pain and provision of opioid analgesics can be challenging because decisions are affected by time constraints, lack of familiarity with the patient’s opioid use or misuse history, and the duality of ensuring patient satisfaction and safety. Patients may expect and prefer to receive opioid analgesics despite ED clinician misgivings about their use.6,7 Clinicians

Volume

-,

no.

-

:

-

2014

may feel compelled to prescribe them because of concerns about patient satisfaction.8 There are several commercially available surveys used to assess patient satisfaction, and results of these surveys might influence hospital reputation and profits. Many hospitals use Press Ganey ED patient satisfaction surveys (Press Ganey Associates, Inc., South Bend, IN) to evaluate elements of the patient experience of care. However, some have expressed concern that good patient satisfaction scores might not necessarily indicate better care and that an emphasis on patient satisfaction scores could compromise patient care.9 For example, Pham et al10 observed that physicians whose compensation was based on measures including patient satisfaction scores ordered more advanced imaging. In both ED11 and primary care12,13 settings, clinicians were more likely to prescribe antibiotics when they believed it was what patients or parents wanted. This perceived link between ED clinician concern to increase patient satisfaction scores and ED clinician actions compels us to

Annals of Emergency Medicine 1

Schwartz et al

Patient Satisfaction Scores and Emergency Department Administration of Analgesic Medications

Editor’s Capsule Summary

What is already known on this topic Physicians may overprescribe opioids in an attempt to improve patient satisfaction survey results. What question this study addressed This secondary data analysis examined the association of pain management practices, including opioid prescribing, with patient satisfaction survey measures in 2 New England emergency departments. What this study adds to our knowledge After controlling for other factors, patient satisfaction, as assessed through 1 commercially available survey, was not associated with the administration of analgesics, including opioid analgesics. How this is relevant to clinical practice Pain management decisions should be driven by patient and clinical factors, rather than concern for patient satisfaction survey results.

explore whether such factors are at play in the provision of opioid analgesics. Importance In some emergency medicine settings, compensation and metrics of care are linked to Press Ganey ED patient satisfaction scores. Although how well Press Ganey ED patient satisfaction scores reflect patient satisfaction is debatable, this metric is widely used to quantify this entity, and thus it is important to explore how in-ED analgesic prescribing affects these scores. If ED clinicians are responding to perceived beliefs about the relationship between opioid analgesic prescribing and Press Ganey ED patient satisfaction scores, there could be a dangerous incentive to overprescribe opioid analgesics. Although researchers have studied other correlates of patient satisfaction in EDs in detail,14-79 current research lacks an examination of the relationship between in-ED receipt of analgesic medications and patient satisfaction. Goals of This Investigation The aim of this study was to examine the relationship between Press Ganey ED patient satisfaction scores and in-ED receipt of analgesic medications, including the amount of opioid analgesics received, among survey respondents. We hypothesized that higher scores were associated with receipt of analgesic medications, receipt of opioid analgesics, and a greater amount of opioid analgesics received. We also were interested in how 2 Annals of Emergency Medicine

patient-reported pain scores, response to medications, and other factors might affect Press Ganey ED patient satisfaction scores.

MATERIALS AND METHODS Study Design and Setting We performed a secondary data analysis of Press Ganey ED patient satisfaction survey results and electronic medical records of patients discharged from 2 New England hospitals in the same hospital system: a Level I trauma center and a university-affiliated community hospital. During the study period at the Level I trauma center, the mean age of discharged patients was 41 years, 50.5% were female patients, 63.3% were white, 17.7% were black, 29.4% had private health care insurance, 41.8% had Medicare/Medicaid, and 28.8% had no health care insurance. At the community hospital, the mean age was 46 years, 56.4% were female patients, 71.6% were white, 15.7% were black, 36.3% had private health care insurance, 45.7% had Medicare/ Medicaid, and 18.1% had no health care insurance. The hospital institutional review board approved this study. Selection of Participants The study population was composed of patients discharged from 2 New England EDs in the 2010 and 2011 fiscal years (October 1, 2009, to September 30, 2011) who completed a Press Ganey ED patient satisfaction survey. We limited our analysis to data from patients who were age 18-years-old or older, who completed a survey about their first ED visit during the study period, and who completed the survey themselves. Press Ganey randomly distributes satisfaction surveys to a sample of patients discharged from these EDs; however, information about their sampling methodology is limited. The total number of patients who received a Press Ganey survey and the response rate are considered proprietary information and are therefore not available from the company. Although response rates greatly depend on the variables chosen for calculation, as a matter of reference, the response rate to Press Ganey surveys at 2 academic, urban hospitals reported in 2 studies was 15% to 17%.22,27 Methods of Measurement We obtained Press Ganey ED patient satisfaction survey responses and matched them to electronic medical record data, using unique patient identifiers. Press Ganey was not involved in the design of the study, data analysis, reporting of the results, or composition or review of this article. Because the nature of the data reflects patient perspectives and unanswered questions could reflect those perspectives, missing data were not imputed. Data obtained from the ED electronic medical records included medication orders, age, sex, race, health insurance status, time of arrival at the ED, time elapsed from ED arrival to time to consulting a physician, total length of stay in the ED, year and month of visit, and patient-reported pain levels. Data from this source were obtained from direct download from the electronic medical record database and hence were not Volume

-,

no.

-

:

-

2014

Schwartz et al

Patient Satisfaction Scores and Emergency Department Administration of Analgesic Medications

interpreted or abstracted by the research team. We chose to consider these data in our analyses according to existing research and the possibility that these factors could affect patient satisfaction scores.14-79 We classified medications ordered as analgesic medications, opioid analgesics, or other medications (Figure E1, available online at http://www.annemergmed.com). We considered caffeine, prochlorperazine, promethazine, and carbamazepine to be used as analgesic medications when the chief complaint of the patient was headache. Using these data, we identified patients who had received analgesic medications and specifically opioid analgesics in the ED. Of the patients who received opioid analgesics, we calculated the number of morphine equivalents administered according to established conversion factors (Figure E2, available online at http://www. annemergmed.com).80,81 The 36-question Press Ganey ED patient satisfaction survey includes 8 sections labeled Arrival, Nurses, Doctors, Tests, Family or Friends, Personal/Insurance Information, Personal Issues, and Overall Assessment. In these sections, patients responded to questions on a scale ranging from 1 to 5, 1 being “very poor” and 5 being “very good.” The survey also contains background questions, which inquire about the identity of the person filling out the survey, as well as patient-reported arrival time to and time spent in the ED. Outcome Measures The primary outcome was overall Press Ganey ED patient satisfaction scores. Secondary outcomes were scores from responses to 12 Press Ganey ED patient survey questions of specific interest that might reflect on patient perception of pain management in the ED. We measured the relationships of overall scores and the 12 survey question scores to receipt of analgesic medications, opioid analgesics, and amount of opioid analgesics received in morphine equivalents. These 12 questions pertained to the nurses’ and physicians’ “courtesy” and “concern to keep you informed about your treatment,” the “degree to which [the nurses and physician] took the time to listen to you,” the nurses’ “attention to your needs” and “concern for your privacy,” the physician’s “concern to keep you informed about your treatment” and “concern for your comfort while treating you,” “how well your pain was controlled,” “the likelihood of your recommending our ED to others,” and the “overall rating of care received during your visit.” Primary Data Analysis We used Stata (StataCorp, College Station, TX) for the statistical analyses. For all analyses, an a¼.05 level of significance was used. We compared the demographic data and survey responses of groups who received analgesic medications or opioid analgesics with those who did not by calculating the difference in means and the corresponding 95% confidence intervals (CIs). Data from the Press Ganey ED patient satisfaction surveys and electronic medical records were summarized as mean, medians, and percentages with corresponding ranges and interquartile ranges. Volume

-,

no.

-

:

-

2014

Figure. Study enrollment diagram.

For the primary analysis, we first performed univariable polytomous logistic regression analyses to determine the relationships of receipt of analgesic medications, opioid analgesics, and number of morphine equivalents; demographic characteristics; hospital; and temporal covariates (age, sex, race, health insurance status, hospital location, total time spent in [ED length of stay] and time of arrival to the ED, and year and month of visit) with overall Press Ganey ED patient satisfaction scores. We chose reference groups by lowest overall scores, temporal order, or increasing values. The base groups were those with the lowest overall scores. Demographic, hospital, and temporal factors significant in the univariable polytomous logistic regression analysis from the primary analysis were used as covariates in all subsequent multivariable models. Overall scores were rightward skewed, reflecting the high scores reported by most patients (Figure E3, available online at http://www. annemergmed.com). We categorized overall scores into quartiles because this arrangement permitted distributing the data more evenly across categories (which increased power to detect differences), scores were not normally distributed despite various transformations, and likelihood ratio analysis demonstrated that this form fit the data well (data not shown). We constructed separate multivariable polytomous logistic regression models to assess the relationship of quartiles of overall scores to receipt of pain medications, receipt of opioid analgesics, and number of morphine equivalents received. We estimated odds ratios and corresponding 95% CIs. Annals of Emergency Medicine 3

Schwartz et al

Patient Satisfaction Scores and Emergency Department Administration of Analgesic Medications Table 1. Description of study population.

Demographic Characteristics

No Analgesic Analgesic Medications Medications Received, Received, n[2,445 n[2,304

Age, mean, y Sex, % Female Male Ethnicity/race, % Asian Black, Hispanic Black, non-Hispanic White, Hispanic White, non-Hispanic Other Not stated Health insurance, % Privately insured Health maintenance organization Governmental Worker’s compensation Uninsured Hospital location, % Community hospital Level I trauma center Temporal factors, % Arrival time 7:01 AM–3 PM 3:01 PM–11 PM 11:01 PM–7 AM Year 2009 2010 2011 Length of ED stay, mean, min

56.5

51.1

59.5 40.5

63.0 37.0

0.5 0.6 5.5 4.6 85.3 3.4 0.1

No Analgesic Medications-Analgesic Medications Received, D (95% CI) 5.38 (4.34 to 6.42)

No Opioid Analgesics-Opioid Analgesics Received, D (95% CI)

55.2

50.8

5.38 (4.34 to 6.42)

3.51 (6.28 to 0.74) 3.51 (0.74 to 6.28)

61.2 38.8

61.0 39.0

0.19 (2.86 to 3.23) 0.19 (3.23 to 2.86)

0.5 0.9 7.9 6.4 79.9 4.4 0.0

0.01 0.30 2.42 1.76 5.46 1.03 0.04

(0.40 to 0.42) (0.78 to 0.19) (3.84 to 0.99) (3.06 to 0.46) (3.30 to 7.61) (2.14 to 0.07) (0.10 to 0.18)

0.6 0.6 6.2 4.9 83.8 3.8 0.1

0.4 1.0 7.6 6.9 80.0 4.1 0.0

0.24 0.40 1.37 2.04 3.81 0.33 0.09

(0.17 to 0.65) (0.98 to 0.18) (2.99 to 0.24) (3.55 to 0.52) (1.37 to 6.25) (1.56 to 0.89) (0.01 to 0.19)

48.8 2.2 35.5 3.6 9.9

51.2 2.3 29.7 3.7 13.1

2.34 0.09 5.77 0.13 3.21

(5.18 to 0.51) (0.95 to 0.76) (3.11 to 8.43) (1.20 to 0.93) (5.02 to 1.39)

49.8 2.4 34.0 3.5 10.3

50.4 2.0 29.6 4.0 14.1

0.55 0.43 4.46 0.52 3.82

(3.67 to 2.57) (0.47 to 1.33) (1.58 to 7.34) (1.72 to 0.68) (5.91 to 1.73)

53.8 46.2

42.2 57.8

11.55 (8.73 to 14.38) 11.55 (14.38 to 8.73)

52.0 48.0

39.0 61.0

12.99 (9.92 to 16.05) 12.99 (16.05 to 9.92)

48.9 37.8 13.1

45.7 37.8 16.4

3.26 (0.42 to 6.10) 0.05 (2.81 to 2.71) 3.32 (5.34 to 1.30)

48.1 38.1 13.6

45.6 37.1 17.3

2.59 (0.52 to 5.70) 0.95 (2.10 to 3.97) 3.71 (6.00 to 1.41)

13.5 52.5 34.0 269.2

11.7 53.7 34.7 332.9

13.2 52.3 34.5 285.9

11.1 54.9 34.0 333.9

1.78 1.13 0.65 63.61

We used ED length of stay as recorded in the electronic medical record because the data for this covariate were complete compared with patient-reported time in the ED from the survey. We expressed ED length of stay in quintiles because these data were skewed and could not be transformed to a normal distribution, the relationship between time (both actual and reported) spent in the ED and overall scores was not linear, and quintiles mirrored the 5-point scale survey responses. Likelihood ratio analyses indicated that this form fit the data well (data not shown). In secondary analyses, we conducted multivariable polytomous logistic regression analyses evaluating the relationships between receipt of medications and the scores of the 12 survey questions of specific interest (nurses, physicians, personal issues, and overall ratings) in a similar manner to and adjusting for the same covariates as the primary analyses. Because there were few responses in the 2 lowest response levels (levels 1 and 2 of 5 levels) for these 12 questions, we merged the 2 lowest levels to allow more power. We also evaluated the effect of patient-reported pain levels on the relationships assessed in the primary analyses. We used the maximum, mean, median, first and last pain level, and the difference between first and 4 Annals of Emergency Medicine

No Opioid Opioid Analgesics Analgesics Received, Received, n[3,345 n[1,404

(0.10 to 3.67) (3.97 to 1.71) (3.35 to 2.05) (77.83 to 49.39)

2.10 2.63 0.52 47.92

(0.10 to 4.11) (5.73 to 0.48) (2.43 to 3.48) (62.22 to 33.63)

last pain levels (continuous and categorical) to explore the effect of pain levels on the relationships of receipt of analgesic medications, opioid analgesics, and number of morphine equivalents to overall scores. We considered these different forms of pain levels because when they were obtained varied and could be related to receipt of analgesic medications, their effect of these outcomes might not be similar, and because they were not normally distributed (Figure E4, available online at http:// www.annemergmed.com). Nursing staff obtained pain levels on a scale from 0 (no pain) to 10 (worst possible pain). Pain levels were not available for all patients, nor were first and last pain level necessarily ascertained at the beginning or end of the ED visit. We verified through regression analyses that pain level was related to receipt of pain medications, opioid analgesics, and number of morphine equivalents (data not shown). These analyses were considered secondary because some patients did not answer these questions or because the data were otherwise incomplete. Sensitivity Analyses We conducted a series of sensitivity analyses for our outcomes and covariates of interest. First, because Press Ganey overall Volume

-,

no.

-

:

-

2014

Schwartz et al

Patient Satisfaction Scores and Emergency Department Administration of Analgesic Medications

Table 2. Press Ganey ED patient satisfaction survey responses. No Analgesic Analgesic Medications Medications Received, Received, n[2,445, n[2,304, X X A. Arrival 1. Waiting time before staff noticed your arrival 2. Helpfulness of the person who first asked you about your condition 3. Comfort of the waiting area 4. Waiting time before you were brought to the treatment area 5. Waiting time in the treatment area, before you were treated by a physician 6. Convenience of parking Arrival subtotal B. Nurses 1. Courtesy of the nurses 2. Degree to which the nurse took the time to listen to you 3. Nurses’ attention to your needs 4. Nurses’ concern to keep you informed about your treatment 5. Nurses’ concern for your privacy Nurse subtotal C. Physicians 1. Courtesy of the physician 2. Degree to which the physician took the time to listen to you 3. Physician’s concern to keep you informed about your treatment 4. Physician’s concern for your comfort while treating you Physician subtotal D. Tests Laboratory 1. Courtesy of person who took your blood 2. Concern shown for your comfort when your blood was drawn Radiology 1. Waiting time for radiology test 2. Courtesy of the radiology staff 3. Concern shown for your comfort during your test Test subtotal E. Family or friends 1. Courtesy with which family or friends were treated 2. Staff concern to keep family or friends informed about your status during your course of treatment 3. Staff concern to let a family member or friend be with you while you were being treated Family or friends subtotal F. Personal/insurance information 1. Courtesy of the person who took your personal/insurance information 2. Privacy you felt when asked about your personal/insurance information 3. Ease of giving your personal/insurance information

Volume

-,

no.

-

:

-

2014

No Analgesic Medications-Analgesic Medications Received, D (95% CI)

No Opioid Opioid Analgesics Analgesics Received, Received, n[3,345, n[1,404, X X

No Opioid Analgesics-Opioid Analgesics Received, D (95% CI)

4.5

4.4

0.11 (0.05 to 0.16)

4.5

4.3

0.11 (0.05 to 0.17)

4.4

4.3

0.10 (0.04 to 0.15)

4.4

4.3

0.12 (0.06 to 0.18)

3.9 3.9

3.8 3.7

0.11 (0.03 to 0.18) 0.16 (0.08 to 0.24)

3.9 3.9

3.7 3.7

0.12 (0.03 to 0.20) 0.21 (0.12 to 0.30)

3.8

3.6

0.16 (0.09 to 0.24)

3.8

3.6

0.15 (0.07 to 0.23)

2.9 3.9

3.0 3.7

0.07 (0.19 to 0.05) 0.18 (0.12 to 0.24)

3.0 3.8

3.0 3.7

0.06 (0.19 to 0.08) 0.18 (0.11 to 0.25)

4.6 4.5

4.5 4.4

0.10 (0.05 to 0.15) 0.10 (0.05 to 0.16)

4.5 4.4

4.4 4.3

0.10 (0.04 to 0.15) 0.09 (0.03 to 0.15)

4.4 4.3

4.3 4.2

0.12 (0.06 to 0.17) 0.12 (0.06 to 0.18)

4.4 4.3

4.3 4.2

0.11 (0.05 to 0.18) 0.10 (0.03 to 0.16)

4.4 4.3

4.3 4.2

0.05 (0.01 to 0.12) 0.14 (0.08 to 0.20)

4.4 4.3

4.4 4.1

0.02 (0.04 to 0.08) 0.14 (0.07 to 0.21)

4.6 4.5

4.5 4.4

0.07 (0.02 to 0.12) 0.09 (0.04 to 0.15)

4.5 4.5

4.5 4.4

0.08 (0.02 to 0.13) 0.10 (0.04 to 0.16)

4.4

4.3

0.10 (0.04 to 0.16)

4.4

4.3

0.04 (0.02 to 0.11)

4.4

4.3

0.06 (0.01 to 0.12)

4.4

4.3

0.04 to (0.02 to 0.11)

4.3

4.2

0.12 (0.06 to 0.19)

4.3

4.2

0.12 (0.05 to 0.19)

3.8 3.7

3.9 3.8

0.06 (0.19 to 0.07) 0.06 (0.19 to 0.07)

3.8 3.7

4.0 3.9

0.20 (0.34 to 0.07) 0.22 (0.35 to 0.08)

3.5 3.8 3.7

3.9 4.2 4.2

0.41 (0.54 to 0.29) 0.41 (0.53 to 0.29) 0.42 (0.55 to 0.30)

3.5 3.8 3.8

4.0 4.3 4.3

0.50 (0.61 to 0.38) 0.49 (0.60 to 0.37) 0.47 (0.59 to 0.35)

4.1

4.1

0.03 (0.11 to 0.05)

4.1

4.2

0.14 (0.22 to 0.07)

3.8

4.1

0.28 (0.40 to 0.16)

3.8

4.1

0.32 (0.43 to 0.21)

3.6

3.9

0.30 (0.42 to 0.18)

3.6

4.0

0.40 (0.52 to 0.28)

3.8

4.1

0.30 (0.42 to 0.18)

3.8

4.2

0.38 (0.50 to 0.26)

3.8

4.0

0.19 (0.29 to 0.08)

3.8

4.1

0.24 (0.34 to 0.14)

4.4

4.3

0.06 (0.01 to 0.12)

4.4

4.3

0.06 (0.01 to 0.12)

4.3

4.3

0.05 (0.02 to 0.11)

4.3

4.3

0.02 (0.05 to 0.09)

4.4

4.3

0.04 (0.03 to 0.10)

4.3

4.3

0.02 (0.05 to 0.09)

Annals of Emergency Medicine 5

Schwartz et al

Patient Satisfaction Scores and Emergency Department Administration of Analgesic Medications Table 2. Continued. No Analgesic Analgesic Medications Medications Received, Received, n[2,445, n[2,304, X X Personal/insurance information subtotal G. Personal issues 1. How well you were kept informed about delays 2. Degree to which staff cared about you as a person 3. How well your pain was controlled 4. Information you were given about caring for yourself at home (eg, receiving medications, receiving follow-up care) 5. Degree of safety and security you felt in the ED Personal issues subtotal H. Overall assessment 1. Overall rating of care received during your visit 2. Likelihood of your recommending our ED to others Overall assessment subtotal Overall ED patient satisfaction

No Analgesic Medications-Analgesic Medications Received, D (95% CI)

No Opioid Opioid Analgesics Analgesics Received, Received, n[3,345, n[1,404, X X

No Opioid Analgesics-Opioid Analgesics Received, D (95% CI)

4.3

4.2

0.09 (0.03 to 0.15)

4.2

4.2

0.06 (0.01 to 0.13)

3.6

3.5

0.12 (0.03 to 0.21)

3.6

3.5

0.07 (0.03 to 0.16)

4.2

4.1

0.12 (0.05 to 0.18)

4.2

4.1

0.08 (0.01 to 0.15)

3.7 4.3

3.9 4.2

0.10 (0.04 to 0.17) 0.10 (0.04 to 0.17)

4.3 4.3

4.2 4.2

0.20 (0.30 to 0.11) 0.08 (0.01 to 0.15)

4.4

4.3

0.11 (0.05 to 0.17)

4.3

4.3

0.03 (0.03 to 0.10)

4.0

3.8

0.20 (0.14 to 0.27)

4.0

3.8

0.15 (0.08 to 0.23)

4.3

4.2

0.14 (0.08 to 0.20)

4.3

4.2

0.11 (0.04 to 0.19)

4.3

4.1

0.16 (0.09 to 0.24)

4.3

4.1

0.13 (0.05 to 0.21)

4.2 4.2

4.0 4.1

0.20 (0.13 to 0.28) 0.14 (0.09 to 0.20)

4.1 4.2

3.9 4.1

0.17 (0.08 to 0.26) 0.13 (0.07 to 0.19)

X, Mean; D, change.

patient satisfaction score alternatively could be expressed in percentiles (0 to 25th, 26th to 50th, 51st to 75th and 76th to 100th), we repeated our primary analyses with scores in this form. Second, we repeated the primary analyses by hospital to evaluate whether the relationships observed differed by institution. Third, we considered different measurements of ED visit times, duration, and perspectives of time as alternative covariates to ED length of stay from the electronic medical record: patient-reported duration of time spent in the ED per the Press Ganey survey response to this question, the difference between electronic medical record and patient-reported ED length of stay, and waiting time from ED arrival to consult a physician per the electronic medical record. We categorized the difference between actual and patient-reported duration of time in the ED as underestimated, overestimated, or accurate; these data were not normally distributed and can be considered relative rather than precise. We considered a difference that was equal to or greater than 60 minutes to be underestimated, a difference that was equal to or less than 60 minutes to be overestimated, and a difference of less than 60 to greater than 60 minutes to be accurately estimated. We repeated our primary analyses, using these alternative temporal covariates.

RESULTS Characteristics of Study Participants The Figure provides an accounting of those included in the study. Of the total study population (4,749 patients), 2,304 (48.5%) received analgesic medications (Table 1). Of the

6 Annals of Emergency Medicine

patients who received analgesic medications, 1,404 (60.9%) received opioid analgesics, which comprised 29.6% of the entire sample. Patients who received pain medications were more likely to be younger, female, Black/non-Hispanic or white/ Hispanic, uninsured, and evaluated at the Level I trauma center, and more likely to arrive at the ED late at night and have a longer length of ED stay (Table 1). Patients who received opioid analgesics were more likely to be younger, white/ Hispanic, and uninsured, evaluated at the Level I trauma center, evaluated later in the day, and to have a longer length of ED stay. Survey Results by Analgesic Medication Received Table 2 depicts the mean responses to each of the 36 questions and the 8 sections (Arrival, Nurses, Doctors, Tests, Family or Friends, Personal/Insurance Information, Personal Issues, and Overall Assessment) of the Press Ganey ED patient satisfaction survey, and the overall mean of all responses. Patients who received analgesic medications had lower overall scores than those who did not receive analgesic medications; had lower scores for 6 of the 8 sections except for the tests section, for which they had similar scores; and had greater scores for the Family or Friends section. Patients who received opioid analgesics had lower scores than those who did not receive opioid analgesics for 6 sections, except for the Personal/ Insurance Information section, for which they had similar scores, and the Tests and Family or Friends section, for which they had greater scores.

Volume

-,

no.

-

:

-

2014

Schwartz et al

Patient Satisfaction Scores and Emergency Department Administration of Analgesic Medications

Table 3. Overall Press Ganey ED patient satisfaction scores primary analysis: univariable polytomous logistic regression. Outcome: Overall Press Ganey ED Patient Satisfaction Scores, OR (95% CI) Total Population, n[4,749 Q1

Covariates

Analgesic medications Base vs no analgesic medications received Opioid analgesics vs no Base opioid analgesics received Morphine equivalents received Age Base Sex Female Ref Male Base Race/ethnicity White, non-Hispanic Ref White, Hispanic Base Black, non-Hispanic Base Black, non-Hispanic Base Asian Base Other Base Insurance Private Ref Health maintenance Base organization Governmental Base Uninsured Base Worker’s compensation Base Hospital Level I trauma center Ref Community hospital Base Length of ED stay, quintile First Ref Second Base Third Base Fourth Base Fifth Base Arrival time 7:01 AM–3 PM Ref 3:01 PM–11 PM Base Base 11:01 PM–7 AM Year 2009 Ref 2010 Base 2011 Base Month January Ref February Base March Base April Base May Base June Base July Base August Base September Base October Base November Base December Base

Q2

Q3

Received Opioid Analgesics, n[1,421 Q4

Q1

Q2

Q3

Q4

0.80 (0.68–0.94) 0.79 (0.67–0.92) 0.61 (0.52–0.71)

0.82 (0.69–0.97) 0.84 (0.71–0.99) 0.67 (0.56–0.80)

Base 0.96 (0.94–0.98) 0.97 (0.95–0.99) 0.96 (0.93–0.98) 1.02 (1.02–1.03) 1.03 (1.03–1.04) 1.04 (1.03–1.04) Base 1.02 (1.01–1.03) 1.04 (1.03–1.05) 1.04 (1.03–1.04) Ref Ref Ref Ref Ref Ref Ref 1.20 (1.01–1.41) 1.12 (0.94–1.31) 1.36 (1.15–1.61) Base 0.81 (0.61–1.09) 0.79 (0.59–1.06) 1.03 (0.76–1.39) Ref 0.98 (0.69–1.38) 1.09 (0.80–1.49) 1.36 (0.57–3.25) 0.31 (0.10–0.97) 1.20 (0.80–1.81)

0.86 0.92 0.87 0.23 1.02

Ref (0.60–1.22) (0.67–1.27) (0.34–2.27) (0.06–0.80) (0.67–1.56)

Ref 0.81 (0.56–1.15) 0.73 (0.52–1.02) 0.54 (0.18–1.61) 0.37 (0.13–1.05) 0.86 (0.56–1.33)

Ref Base Base Base Base Base

Ref 1.24 (0.69–2.20) 1.35 (0.81–2.24) 0.74 (0.18–3.13) N 2.35 (1.07–5.13)

Ref 1.35 (0.77–2.35) 0.87 (0.50–1.51) 1.16 (0.33–4.05) N 1.86 (0.83–4.16)

Ref 1.11 (0.61–2.03) 0.74 (0.41–1.36) 0.26 (0.03–2.28) 0.66 (0.12–3.64) 1.72 (0.74–3.99)

Ref Ref Ref Ref Ref Ref Ref 0.94 (0.59–1.51) 0.45 (0.26–0.80) 0.47 (0.26–0.83) Base 0.94 (0.39–2.28) 0.29 (0.08–1.03) 0.48 (0.15–1.53) 1.31 (1.09–1.57) 1.15 (0.95–1.38) 1.43 (1.19–1.72) Base 1.17 (0.84–1.64) 1.09 (0.78–1.52) 1.33 (0.94–1.87) 1.09 (0.85–1.40) 0.85 (0.65–1.10) 0.79 (0.60–1.03) Base 1.30 (0.85–1.99) 1.08 (0.70–1.66) 1.10 (0.70–1.74) 10.3 (0.67–1.61) 0.96 (0.62–1.49) 1.12 (0.73–1.72) Base 0.84 (0.39–1.78) 0.89 (0.43–1.84) 0.97 (0.45–2.07) Ref Ref Ref Ref Ref Ref Ref 1.89 (1.60–2.23) 2.90 (2.45–3.43) 3.05 (2.58–3.61) Base 1.68 (1.23–2.30) 2.81 (2.07–3.81) 2.93 (2.14–4.02) Ref 0.68 (0.51–0.92) 0.62 (0.47–0.82) 0.43 (0.32–0.56) 0.39 (0.30–0.51)

Ref 0.80 (0.60–1.07) 0.55 (0.42–0.73) 0.36 (0.27–0.48) 0.34 (0.26–0.45)

Ref 0.59 (0.45–0.77) 0.31 (0.23–0.40) 0.20 (0.15–0.26) 0.13 (0.10–0.18)

Ref Base Base Base Base

1.06 0.89 0.67 0.60

Ref (0.54–2.10) (0.46–1.71) (0.36–1.27) (0.32–1.12)

0.75 0.60 0.49 0.43

Ref (0.39–1.43) (0.33–1.12) (0.27–0.90) (0.24–0.79)

Ref 0.55 (0.30–1.00) 0.37 (0.20–0.66) 0.24 (0.13–0.42) 0.13 (0.07–0.23)

Ref Ref Ref Ref Ref Ref Ref 0.79 (0.66–0.94) 0.92 (0.78–1.10) 0.67 (0.56–0.80) Base 1.08 (0.79–1.48) 0.94 (0.69–1.28) 0.85 (0.61–1.18) 0.70 (0.55–0.89) 0.69 (0.54–0.89) 0.67 (0.53–0.85) Base 0.88 (0.58–1.33) 0.72 (0.47–1.09) 0.93 (0.62–1.41) Ref Ref Ref Ref Ref Ref Ref 1.00 (0.78–1.28) 1.26 (0.98–1.63) 1.18 (0.92–1.51) Base 0.63 (0.40–0.99) 0.91 (0.56–1.48) 0.89 (0.54–1.48) 1.09 (0.84–1.41) 1.24 (0.95–1.62) 1.12 (0.86–1.45) Base 0.78 (0.48–1.27) 0.94 (0.56–1.55) 0.99 (0.58–1.68) Ref 0.82 (0.57–1.19) 0.85 (0.58–1.23) 0.76 (0.52–1.12) 0.83 (0.57–1.22) 0.84 (0.57–1.23) 0.84 (0.57–1.24) 0.82 (0.56–1.21) 0.77 (0.49–1.21) 0.98 (0.67–1.42) 0.88 (0.59–1.31) 0.76 (0.53–1.11)

Ref 0.79 (0.53–1.16) 0.98 (0.67–1.43) 0.92 (0.62–1.35) 0.89 (0.60–1.32) 0.95 (0.64–1.41) 1.16 (0.79–1.71) 1.14 (0.78–1.66) 1.24 (0.81–1.90) 0.93 (0.63–1.38) 1.17 (0.79–1.75) 0.68 (0.46–1.00)

Ref 0.59 (0.40–0.87) 0.85 (0.58–1.24) 0.90 (0.61–1.31) 0.99 (0.67–1.44) 1.05 (0.72–1.53) 0.79 (0.53–1.18) 0.76 (0.52–1.12) 0.80 (0.51–1.26) 0.77 (0.52–1.14) 1.10 (0.74–1.64) 0.95 (0.66–1.37)

Ref Base Base Base Base Base Base Base Base Base Base Base

Ref 0.85 (0.46–1.56) 0.67 (0.35–1.29) 0.80 (0.43–1.49) 0.57 (0.29–1.11) 0.83 (0.43–1.61) 0.51 (0.26–0.99) 0.47 (0.23–0.94) 0.50 (0.23–1.09) 1.10 (0.58–2.08) 0.82 (0.40–1.67) 1.09 (0.58–2.06)

Ref 0.89 (0.45–1.75) 1.14 (0.59–2.22) 1.19 (0.62–2.28) 0.73 (0.36–1.51) 0.90 (0.43–1.85) 1.32 (0.69–2.54) 1.76 (0.93–3.33) 1.32 (0.64–2.73) 1.10 (0.54–2.23) 1.87 (0.93–3.77) 1.02 (0.50–2.08)

Ref 0.51 (0.24–1.09) 0.96 (0.48–1.91) 0.91 (0.46–1.81) 1.03 (0.52–2.04) 1.28 (0.64–2.55) 0.89 (0.44–1.78) 0.87 (0.43–1.77) 0.87 (0.40–1.92) 0.91 (0.44–1.90) 1.71 (0.84–3.49) 1.72 (0.88–3.35)

Q, Quartile; OR, odds ratio; Base, base outcome.

Volume

-,

no.

-

:

-

2014

Annals of Emergency Medicine 7

Outcome: Overall Press Ganey ED Patient Satisfaction Scores, OR (95% CI) Analysis for Receipt of Analgesic Medications, n[4,749 Covariates Analgesic medications vs no analgesic medications received Opioid analgesics vs no opioid analgesics received Morphine equivalents received Age Sex Female Male Hospital Level I trauma center Community hospital Length of ED stay, quintile First Second Third Fourth Fifth Arrival time 7:01 AM–3 PM 3:01 PM–11 PM 11:01 PM–7 AM

Q1

Q2

Q3

Q4

Base

1.02 (0.86–1.20)

1.10 (0.93–1.31)

0.98 (0.82–1.17)

Analysis for Receipt of Opioid Analgesics, n[4,749 Q1

Q2

Q3

Q4

Base

1.03 (0.86–1.23)

1.18 (0.98–1.42)

1.10 (0.91–1.34)

Analysis for Number of Morphine Equivalents Received, n[1,421 Q1

Q2

Q3

Q4

Base

0.97 (0.95–0.99)

0.99 (0.96–1.01)

0.99 (0.96–1.02)

Base

1.02 (1.02–1.03)

1.03 (1.03–1.04)

1.04 (1.03–1.04)

Base

1.02 (1.02–1.03)

1.03 (1.03–1.04)

1.04 (1.03–1.04)

Base

1.02 (1.01–1.03)

1.03 (1.02–1.04)

1.04 (1.03–1.05)

Ref Base

Ref 1.19 (1.01–1.41)

Ref 1.11 (0.93–1.32)

Ref 1.35 (1.13–1.62)

Ref Base

Ref 1.19 (1.00–1.41)

Ref 1.11 (0.93–1.32)

Ref 1.35 (1.13–1.61)

Ref Base

Ref 0.85 (0.63–1.15)

Ref 0.83 (0.61–1.13)

Ref 1.05 (0.76–1.45)

Ref Base

Ref 1.46 (1.22–1.74)

Ref 2.10 (1.76–2.52)

Ref 1.80 (1.50–2.17)

Ref Base

Ref 1.46 (1.22–1.74)

Ref 2.11 (1.77–2.53)

Ref 1.81 (1.51–2.18)

Ref Base

Ref 1.46 (1.05–2.02)

Ref 2.24 (1.62–3.08)

Ref 1.97 (1.40–2.76)

Ref Base Base Base Base

0.67 0.60 0.42 0.40

Ref (0.58–1.05) (0.39–0.70) (0.27–0.49) (0.28–0.51)

Ref 0.57 (0.42–0.75) 0.28 (0.21–0.38) 0.19 (0.14–0.26) 0.14 (0.10–0.19)

Ref Base Base Base Base

0.67 0.60 0.42 0.40

Ref (0.49–0.90) (0.45–0.80) (0.32–0.56) (0.30–0.54)

Ref 0.77 (0.58–1.04) 0.52 (0.39–0.69) 0.36 (0.27–0.48) 0.37 (0.28–0.50)

Ref 0.56 (0.42–0.74) 0.28 (0.21–0.37) 0.19 (0.14–0.25) 0.13 (0.10–0.18)

Ref Base Base Base Base

1.11 0.95 0.76 0.73

Ref (0.56–2.19) (0.49–1.84) (0.40–1.47) (0.38–1.40)

Ref 0.80 (0.41–1.55) 0.65 (0.34–1.23) 0.58 (0.31–1.09) 0.54 (0.29–1.03)

Ref 0.55 (0.30–1.04) 0.36 (0.19–0.67) 0.26 (0.14–0.47) 0.14 (0.07–0.27)

Ref Base Base

Ref 0.86 (0.72–1.03) 0.86 (0.67–1.10)

Ref 1.03 (0.86–1.24) 0.92 (0.71–1.19)

Ref 0.77 (0.64–0.93) 0.96 (0.74–1.25)

Ref Base Base

Ref 0.86 (0.72–1.03) 0.86 (0.67–1.10)

Ref 1.03 (0.86–1.24) 0.92 (0.71–1.19)

Ref 0.77 (0.64–0.93) 0.96 (0.74–1.24)

Ref Base Base

Ref 1.09 (0.79–1.50) 1.00 (0.66–1.52)

Ref 1.00 (0.72–1.39) 0.92 (0.59–1.42)

Ref 0.92 (0.65–1.31) 1.27 (0.82–1.97)

Ref (0.49–0.90) (0.45–0.80) (0.32–0.56) (0.30–0.54)

0.78 0.53 0.36 0.38

Patient Satisfaction Scores and Emergency Department Administration of Analgesic Medications

8 Annals of Emergency Medicine

Table 4. Overall Press Ganey ED patient satisfaction scores primary analysis: multivariable polytomous logistic regression.

Volume -, -

: -

2014

Schwartz et al

no.

-,

no.

Schwartz et al

Volume

Table 5. Twelve survey questions of specific interest: multivariable polytomous logistic regression analysis.* Analysis for Receipt of Analgesic Medications, OR (95% CI)

-

: -

2014

Analysis for Number of Morphine Equivalents Received, OR (95% CI)

n¼4,699 n¼4,679 n¼4,567 n¼4,647 n¼4,580

Base Base Base Base Base

0.82 0.96 0.77 0.96 1.19

(0.54–1.25) (0.67–1.39) (0.56–1.07) (0.72–1.28) (0.82–1.73)

0.86 0.76 0.78 0.88 1.14

(0.60–1.23) (0.56–1.05) (0.59–1.04) (0.69–1.14) (0.82–1.58)

0.81 0.84 0.80 0.93 1.11

(0.57–1.15) (0.62–1.14) (0.61–1.05) (0.73–1.19) (0.81–1.53)

Base Base Base Base Base

0.87 0.84 0.72 1.13 1.17

(0.57–1.33) (0.58–1.21) (0.51–0.99) (0.84–1.52) (0.80–1.73)

0.82 0.75 0.65 0.92 1.02

(0.57–1.17) (0.55–1.03) (0.49–0.87) (0.71–1.20) (0.72–1.43)

0.82 0.80 0.75 1.02 1.11

(0.58–1.17) (0.59–1.09) (0.57–0.98) (0.80–1.32) (0.79–1.55)

n¼1,410 n¼1,407 n¼1,406 n¼1,401 n¼1,383

Base Base Base Base Base

1.02 1.03 1.01 1.02 1.07

(0.97–1.08) (0.98–1.08) (0.97–1.08) (0.98–1.06) (1.01–1.13)

1.01 1.03 1.03 1.03 1.05

(0.96–1.05) (0.99–1.08) (0.99–1.08) (0.99–1.07) (0.99–1.11)

1.01 1.03 1.03 1.03 1.05

(0.97–1.06) (0.98–1.07) (0.98–1.07) (0.99–1.07) (0.99–1.11)

n¼4,684 n¼4,657 n¼4,644 n¼4,614

Base Base Base Base

0.84 0.82 0.76 0.70

(0.56–1.26) (0.57–1.17) (0.55–1.04) (0.51–0.97)

0.81 0.77 0.73 0.71

(0.57–1.15) (0.56–1.05) (0.55–0.96) (0.53–0.93)

0.83 0.81 0.80 0.82

(0.59–1.17) (0.60–1.09) (0.62–1.04) (0.62–1.07)

Base Base Base Base

0.81 0.86 0.75 0.62

(0.53–1.23) (0.59–1.24) (0.54–1.04) (0.44–0.87)

0.70 0.75 0.80 0.74

(0.49–1.01) (0.54–1.02) (0.61–1.06) (0.55–0.98)

0.75 0.78 0.86 0.84

(0.53–1.06) (0.58–1.06) (0.66–1.13) (0.64–1.10)

n¼1,406 n¼1,400 n¼1,398 n¼1,398

Base Base Base Base

0.97 0.99 1.02 1.00

(0.92–1.01) (0.94–1.04) (0.97–1.06) (0.96–1.05)

0.96 1.00 1.00 0.98

(0.92–0.99) (0.96–1.04) (0.96–1.04) (0.95–1.02)

0.96 0.99 1.00 0.98

(0.93–0.9) (0.95–1.03) (0.96–1.03) (0.95–1.02)

n¼3,952 Base 0.95 (0.72–1.26) 0.86 (0.68–1.09) 0.86 (0.68–1.08) Base 0.79 (0.60–1.04) 0.77 (0.60–0.97) 0.88 (0.70–1.11) n¼1,352 Base 0.99 (0.96–1.03) 1.00 (0.97–1.03) 1.00 (0.97–1.03) n¼4,631 Base 0.84 (0.63–1.12) 0.92 (0.71–1.18) 1.00 (0.79–1.27) Base 0.81 (0.60–1.10) 0.84 (0.65–1.09) 1.02 (0.80–1.30) n¼1,379 Base 0.98 (0.94–1.02) 1.00 (0.97–1.03) 0.99 (0.96–1.02) n¼4,573 Base 0.98 (0.74–1.29) 0.91 (0.72–1.15) 1.00 (0.80–1.24) Base 0.81 (0.60–1.09) 0.92 (0.72–1.17) 1.06 (0.85–1.33) n¼1,369 Base 1.02 (0.98–1.06) 1.03 (0.99–1.06) 1.00 (0.97–1.03)

Level, Response category level. *Models adjusted for age, sex, hospital, length of ED stay, and arrival time.

Table 6. Overall Press Ganey ED patient satisfaction scores: multivariable polytomous logistic regression analysis adjusted for pain level.* Form of Pain Level Variable Used as Covariate in Models

Annals of Emergency Medicine 9

First Last Mean Maximum Change in pain level (continuous) Change in pain level (categorical)

Outcome: Overall Press Ganey ED Patient Satisfaction Scores, OR (95% CI) Analysis for Receipt of Analgesic Medications Q1

Q2

n¼2,432 n¼956 n¼2,432 n¼2,432 n¼956

Base Base Base Base Base

1.25 1.35 1.25 1.24 1.28

n¼956

Base

1.18 (0.78–1.78)

Q3

Q4

Analysis for Receipt of Opioid Analgesics Q1

Q2

(0.95–1.64) 1.28 (0.98–1.69) 1.11 (0.84–1.48) Base 1.28 (0.96–1.71) (0.88–2.06) 1.31 (0.86–2.00) 1.49 (0.95–2.33) Base 1.15 (0.74–1.79) (0.95–1.63) 1.28 (0.98–1.67) 1.13 (0.86–1.50) Base 1.28 (0.96–1.71) (0.95–1.63) 1.30 (0.99–1.71) 1.14 (0.72–1.68) Base 1.27 (0.95–1.69) (0.86–1.91) 1.12 (0.75–1.67) 1.10 (0.72–1.68) Base 1.10 (0.73–1.65)

1.10 (0.73–1.78)

1.18 (0.76–1.82)

Base

Q3 1.40 1.29 1.40 1.41 1.08

Q4

Q1

(1.05–1.88) 1.31 (0.96–1.79) n¼743 (0.83–2.00) 1.27 (0.79–2.05) n¼327 (1.05–1.86) 1.34 (0.99–1.82) n¼743 (1.06–1.89) 1.35 (0.99–1.83) n¼743 (0.72–1.63) 0.91 (0.59–1.42) n¼327

1.01 (0.67–1.54) 1.06 (0.70–1.61)

*Models adjusted for age, sex, hospital, length of ED stay, and arrival time, in addition to pain level.

Analysis for Number of Morphine Equivalents Received

0.96 (0.61–1.51)

Q2

Q3

Base 0.96 (0.93–0.99) 0.97 Base 0.97 (0.92–1.02) 0.95 Base 0.97 (0.93–1.00) 0.97 Base 0.96 (0.93–0.99) 0.97 Base 0.96 (0.92–1.01) 0.95

n¼327 Base 0.96 (0.92–1.01)

(0.93–1.00) (0.90–1.00) (0.93–1.00) (0.93–1.00) (0.90–0.99)

Q4 0.98 0.99 0.99 0.98 0.98

(0.94–1.02) (0.93–1.04) (0.95–1.03) (0.94–1.02) (0.92–1.03)

0.95 (0.90–0.99) 0.98 (0.92–1.03)

Patient Satisfaction Scores and Emergency Department Administration of Analgesic Medications

Nurses Courtesy Listening Attention to needs Information Concern for privacy Physicians Courtesy Listening Information Concern for comfort Personal issues Pain control Overall Overall rating Likelihood to recommend

Analysis for Receipt of Opioid Analgesics, OR (95% CI)

Schwartz et al

Patient Satisfaction Scores and Emergency Department Administration of Analgesic Medications Primary Analysis: Overall Patient Satisfaction Scores Associations In the univariable analyses (Table 3), receipt of analgesic medications, opioid analgesics, or receipt of a higher number of morphine equivalents in the ED was associated with lower overall Press Ganey ED patient satisfaction scores. Patients who were younger and evaluated at the Level I trauma center ED were more likely to give lower overall scores, and among the total population specifically (n¼4,749), those who were female patients, spent more time in the ED, and arrived later in the day were also more likely to give lower overall scores. Insurance status, race, and year and month of visit were not associated with overall scores. When accounting for the other covariates, among the total study population (n¼4,749), receipt of analgesic medications or opioid analgesics was no longer associated with overall scores (Table 4). Decreasing patient age, female sex (but not consistently across quartiles), presentation to Level I trauma center ED, and longer ED length of stay remained associated with lower overall scores. Among patients who received opioid analgesics, there was a slight relationship between greater morphine equivalents and lower scores only in the second quartile, and otherwise only decreasing age and presenting to the Level I trauma center ED were associated with lower overall scores. Secondary Analysis Table 5 shows the results of the multivariable polytomous logistic regression analysis for the 12 survey questions of specific interest. No consistent patterns emerged between receipt of medications and scores for each question, except that patients who received analgesic medications and opioid analgesics had lower ratings of physician concern for comfort, those who received opioid analgesics had lower ratings of nurse attention to needs, and those who received higher morphine equivalents had lower ratings of physician courtesy. Table 6 shows the results of the multivariable analyses of patient-reported pain levels as an additional cofactor in different iterations of the primary analyses. The addition of patientreported pain levels in their various forms did not change the relationships observed in the primary analyses. Sensitivity Analysis When the primary analyses were repeated with overall Press Ganey ED patient scores as percentiles instead of quartiles, no relationship existed between receipt of analgesics, opioid analgesics, or amount of morphine equivalents received in the univariable and multivariable analyses (Table E1, available online at http://www.annemergmed.com). When the primary analyses were repeated by hospital, the results did not differ by institution, although the slight relationship between morphine equivalents and lower scores disappeared (Tables E2 and E3, available online at http://www.annemergmed.com). Table E4 (available online at http://www.annemergmed.com) shows the results of the analyses of the different measurements of patient-reported and electronic medical record recorded time 10 Annals of Emergency Medicine

spent in ED as covariates in repeated iterations of the primary analyses. There was no pattern of relationships between receipt of analgesic medications or opioid analgesics and overall scores, but there were trends of lower overall scores and greater morphine equivalents received when the difference between reported and actual time and actual time waiting to consult a physician were used as covariates in the models.

LIMITATIONS Several limitations must be considered in interpreting our results. There are both known and unknown limitations associated with use of Press Ganey ED patient satisfaction survey results. Although Press Ganey surveys are described as reliable and valid on the company Web site, there is no published research that independently assesses these aspects, to our knowledge. The nature of their sampling methods might create a selection bias toward inclusion of ED patients with lower-acuity conditions, and exclusion of those with limited literacy skills, those who do not communicate easily in English, those without a permanent address, and the institutionalized. In addition, there is no compensation associated with return of the Press Ganey survey, and the overall return rate is believed to be low, yet sample size and return rate data are both proprietary information and cannot be evaluated. Regardless, Press Ganey is a commonly used standard for quantifying patient satisfaction, and there are no other widely applied measurements used in these EDs as points of comparison. Furthermore, the primary aim of this investigation was to examine the relationship of Press Ganey ED patient satisfaction scores to receipt of analgesic medications in the ED because that relationship has led to concern about clinician prescribing practices. Although it is possible the data and findings may not be representative of other EDs, we included 2 different institutions to obtain a greater diversity of patients in an attempt to increase external validity. Press Ganey ED patient satisfaction scores were skewed toward higher scores, as is common in ED patient satisfaction surveys39 and expected from hospitals aiming to provide patients with quality health care experiences. It is commonly believed that patients with very positive or very negative experiences are more likely to complete satisfaction surveys, but this aspect cannot be examined with these data. Furthermore, although we could not explore this aspect in this study, oligoanalgesia among patients who received opioid analgesics may have contributed to the findings in this study, or at least to patient responses on the surveys.82 Higher acuity among patients who received opioid analgesics also might have contributed to the findings and responses, but research suggests that higher acuity patients actually tend to be more satisfied than lower acuity patients.20,83 Our study included only patients discharged from the ED, and our results might not apply to those who were admitted to the hospital after their ED visit. Admitted patients likely have greater disease severity and different experiences in regard to analgesic medications and patient satisfaction. There may have Volume

-,

no.

-

:

-

2014

Schwartz et al

Patient Satisfaction Scores and Emergency Department Administration of Analgesic Medications

been additional factors that contributed to patient satisfaction scores for which we were unable to account. We also cannot quantify the effect of recall bias for patients whose visits were not fresh in their memories. We cannot account for the effect of mood while the patient was completing the survey, nor are we able to calculate the level of each patient’s personal opioid tolerance, which could likely have an effect on his or her perception of pain. We also could not evaluate the effect of analgesics prescribed for use after discharge. Although we acknowledge that such prescriptions could affect Press Ganey ED patient satisfaction scores, we would be unable to perform any substantive assessment about the percentage of opioid analgesic prescriptions that were filled or the number of prescribed doses that were received by patients who completed the surveys in this type of study. Given the marked limitations of the available data about discharge prescriptions, we focused on in-ED analgesic medication and opioid analgesics administration only.

DISCUSSION After accounting for patient, temporal, and hospital-related factors, we identified no relationship between Press Ganey ED patient satisfaction scores and the receipt of analgesic medications or opioid analgesics. We did find a small but inconsistent relationship between higher number of morphine equivalents received and lower scores, but this finding was not present at all levels of scores and was not stable across the different analyses we performed. We did observe a consistent relationship between increasing age and treatment at the community hospital and higher Press Ganey ED patient satisfaction scores. We also found that men and patients who had spent a shorter time in the ED generally gave higher scores. We found that patients who had a longer length of ED stay gave lower scores. Existing research on predictors of Press Ganey ED patients satisfaction scores suggests that satisfaction with physician and nurse care, satisfaction with information provided,20,40 and waiting time being less than expected12,38 are predictors of overall satisfaction score.18,19 Although some studies found total length of ED stay18,26 and older age14-16,39 to be predictive of scores, others found that these factors were not predictive.16,18,19,22,40 Additionally, studies have reported relationships between shorter patient-reported waiting time to consult a physician,16,37 shorter electronic medical record-reported length of ED stay,26,41 and higher patient satisfaction scores, although Sun et al20 found no such relationships. We did not observe that including pain levels in various forms as having an effect on the relationship of receipt of analgesic medications, opioid analgesics and number of morphine equivalents on overall scores. Other studies have also evaluated the relationship between pain levels and patient satisfaction scores. Downey and Zun35 found that a decrease in pain level is related to high patient satisfaction scores, whereas Blank et al42 found that it is not related. Kelly49 found that first, discharge, or change pain score was unrelated to patient satisfaction scores. Volume

-,

no.

-

:

-

2014

Factors associated with patient satisfaction scores in EDs are likely to vary by study sample, which likely accounts for some of the differences between our findings and those from other studies. These differences also can be attributed to the different covariates assessed and methods of statistical analysis used across studies. Other unconsidered confounders could affect patient satisfaction, such as triage category28,32 or level of ED crowding.17,24 Discrepancies between our and other studies also may be due in part to our large sample size, which enabled us to detect subtle associations. For example, even though we found an association between increasing age and patient satisfaction, the effect size was small across all quartiles of scores (although we used age as a continuous variable). Studies that include few patients usually have less ability to detect such associations. Future research into the relationship of opioid analgesics and patient satisfaction should explore different settings, covariates, and measures of patient satisfaction. In conclusion, when measured through the use of Press Ganey surveys, overall ED patient satisfaction scores among ED patients are not primarily based on in-ED receipt of opioid analgesics or analgesic medications; other factors appear to be more important. This suggests that ED clinicians can administer analgesic medications in the ED according to clinical and patient factors without being concerned about Press Ganey ED patient satisfaction scores. Supervising editor: Knox H. Todd, MD, MPH Author affiliations: From the Alpert Medical School of Brown University (Schwartz); the Department of Biostatistics (Tai) and Department of Epidemiology (Merchant), School of Public Health, Brown University; and the Department of Emergency Medicine, Alpert Medical School of Brown University, Providence, RI (Tai, Babu, Merchant). Author contributions: KMB conceived of the study. TMS drafted the article. KMB and RCM obtained funding, contributed revisions, and coordinated the analysis and production of the article. TMS, KMB, and RCM designed the study. TMS and MT managed the data. MT analyzed the data. TMS takes responsibility for the paper as a whole. Funding and support: By Annals policy, all authors are required to disclose any and all commercial, financial, and other relationships in any way related to the subject of this article as per ICMJE conflict of interest guidelines (see www.icmje.org). The authors have stated that no such relationships exist and provided the following details: The Brown Summer Research Assistantship and the University Emergency Medicine Foundation Junior Faculty Research Development Grant supported this study. Publication dates: Received for publication August 8, 2013. Revision received January 8, 2014. Accepted for publication February 7, 2014. Presented at the American College of Emergency Physicians Scientific Assembly Research Forum, October 2013, Seattle, WA. Address for correspondence: Roland C. Merchant, MD, ScD, E-mail [email protected].

Annals of Emergency Medicine 11

Schwartz et al

Patient Satisfaction Scores and Emergency Department Administration of Analgesic Medications REFERENCES 1. Paulozzi LJ, Budnitz DS, Xi YL. Increasing deaths from opioid analgesics in the United States. Pharmacoepidemiol Drug Saf. 2006;15:618-627. 2. Centers for Disease Control and Prevention. Vital signs: overdoses of prescription opioid pain relievers—United States, 1999–2008. MMWR Morb Mortal Wkly Rep. 2011;60: 1487-1492. 3. Fishbain DA, Cole B, Lewis J, et al. What percentage of chronic nonmalignant pain patients exposed to chronic opioid analgesic therapy develop abuse/addiction and/or aberrant drug-related behaviors? a structured evidence-based review. Pain Med. 2008;9:444-459. 4. United States Food and Drug Administration. Attention prescribers: FDA seeks your help in curtailing the US opioid epidemic. United States Food and Drug Administration. Silver Spring, MD; 2013. 5. Cordell WH, Keene KK, Giles BK, et al. The high prevalence of pain in emergency medical care. Am J Emerg Med. 2002;20: 165-169. 6. Zgierska A, Miller M, Rabago D. Patient satisfaction, prescription drug abuse, and potential unintended consequences. JAMA. 2012;307: 1377-1378. 7. Lembke A. Why doctors prescribe opioids to known opioid abusers. N Engl J Med. 2012;367:1580-1581. 8. Emergency physician’s monthly survey. November 20, 2009. Available at: http://www.epmonthly.com/whitecoat/2009/11/patientsatisfiction. Accessed January 7, 2014. 9. Darnall BD, Schatman ME. Autonomy vs paternalism in the emergency department: the potential deleterious impact of patient satisfaction surveys. Pain Med. 2013;14:968. 10. Pham HH, Landon BE, Reschovsky JD, et al. Rapidity and modality of imaging for acute low back pain in elderly patients. Arch Intern Med. 2009;169:972-981. 11. Ong S, Nakase J, Moran GJ, et al. Antibiotic use for emergency department patients with upper respiratory infections: prescribing practices, patient expectations, and patient satisfaction. Ann Emerg Med. 2007;50:213-220. 12. Mangione-Smith R, McGlynn EA, Elliott MN, et al. The relationship between perceived parental expectations and pediatrician antimicrobial prescribing behavior. Pediatrics. 1999;103:711-718. 13. Cockburn J, Pit S. Prescribing behaviour in clinical practice: patients’ expectations and doctors’ perceptions of patients’ expectations—a questionnaire study. BMJ. 1997;315:520-523. 14. Hayes RP, Baker DW. Methodological problems in comparing Englishspeaking and Spanish-speaking patients’ satisfaction with interpersonal aspects of care. Med Care. 1998;36:230-236. 15. Boudreaux ED, O’Hea EL. Patient satisfaction in the emergency department: a review of the literature and implications for practice. J Emerg Med. 2004;26:13-26. 16. Hedges JR, Trout A, Magnusson AR. Satisfied Patients Exiting the Emergency Department (SPEED) study. Acad Emerg Med. 2002;9:15-21. 17. Tekwani KL, Kerem Y, Mistry CD, et al. Emergency department crowding is associated with reduced satisfaction scores in patients discharged from the emergency department. West J Emerg Med. 2013;14:11-15. 18. Hall MF, Press I. Keys to patient satisfaction in the emergency department: results of a multiple facility study. Hosp Health Surv Adm. 1996;41:515-532. 19. Yarnold PR, Michelson EA, Thompson DA, et al. Predicting patient satisfaction: a study of two emergency departments. J Behav Med. 1998;21:545-563. 20. Sun BC, Adams J, Orav EJ, et al. Determinants of patient satisfaction and willingness to return with emergency care. Ann Emerg Med. 2000;35:426-434.

12 Annals of Emergency Medicine

21. Sun BC, Adams JG, Burstin HR. Validating a model of patient satisfaction with emergency care. Ann Emerg Med. 2001;38:527-532. 22. Rhee KJ, Bird J. Perceptions and satisfaction with emergency department care. J Emerg Med. 1996;14:679-683. 23. Raper J, Davis BA, Scott L. Patient satisfaction with emergency department triage nursing care: a multicenter study. J Nurs Care Qual. 1999;13:11-24. 24. Pines JM, Iyer S, Disbot M, et al. The effect of emergency department crowding on patient satisfaction for admitted patients. Acad Emerg Med. 2008;15:825-831. 25. Patel PB, Vinson DR. Physician e-mail and telephone contact after emergency department visit improves patient satisfaction: a crossover trial. Ann Emerg Med. 2013;61:631-637. 26. Boudreaux ED, Ary RD, Mandry CV, et al. Determinants of patient satisfaction in a large, municipal ED: the role of demographic variables, visit characteristics, and patient perceptions. Am J Emerg Med. 2000;18:394-400. 27. Olsen JC, Olsen EC. Patient satisfaction in the emergency department and the use of business cards by physicians. J Emerg Med. 2012;42:317-321. 28. Boudreaux ED, Friedman J, Chansky ME, et al. Emergency department patient satisfaction: examining the role of acuity. Acad Emerg Med. 2004;11:162-168. 29. Bruce TA, Bowman JM, Brown ST. Factors that influence patient satisfaction in the emergency department. J Nurs Care Qual. 1998;13:31-37. 30. Burstin HR, Conn A, Setnik G, et al. Benchmarking and quality improvement: the Harvard Emergency Department Quality Study. Am J Med. 1999;107:437-449. 31. Carey RG, Seibert JH. A patient survey system to measure quality improvement—questionnaire reliability and validity. Med Care. 1993;31:834-845. 32. Carrasquillo O, Orav J, Brennan TA, et al. Impact of language barriers on patient satisfaction in an emergency department. J Gen Intern Med. 1999;14:82-87. 33. DeBehnke D, Decker MC. The effects of a physician-nurse patient care team on patient satisfaction in an academic ED. Am J Emerg Med. 2002;20:267-270. 34. Derose KP, Hays RD, McCaffrey DF, et al. Does physician gender affect satisfaction of men and women visiting the emergency department? J Gen Intern Med. 2001;16:218-226. 35. Downey LV, Zun LS. Pain management in the emergency department and its relationship to patient satisfaction. J Emerg Trauma Shock. 2010;3:326-330. 36. Mowen JC, Licata JW, McPhail J. Waiting in the emergency room: how to improve patient satisfaction. J Health Care Mark. 1993;13:26-33. 37. Cassidy-Smith TN, Baumann BM, Boudreaux ED. The disconfirmation paradigm: throughput times and emergency department patient satisfaction. J Emerg Med. 2007;32:7-13. 38. Boudreaux ED, D’Autremont S, Wood K, et al. Predictors of emergency department patient satisfaction: stability over 17 months. Acad Emerg Med. 2004;11:51-58. 39. Trout A, Magnusson AR, Hedges JR. Patient satisfaction investigations and the emergency department: what does the literature say? Acad Emerg Med. 2000;7:695-709. 40. Thompson DA, Yarnold PR, Williams DR, et al. Effects of actual waiting time, perceived waiting time, information delivery, and expressive quality on patient satisfaction in the emergency department. Ann Emerg Med. 1996;28:657-665. 41. Walsh M, Knott JC. Satisfaction with the emergency department environment decreases with length of stay. Emerg Med J. 2010;27:821-828. 42. Blank FS, Mader TJ, Wolfe J, et al. Adequacy of pain assessment and pain relief and correlation of patient satisfaction in 68 ED fast-track patients. J Emerg Nurs. 2001;27:327-334.

Volume

-,

no.

-

:

-

2014

Schwartz et al

Patient Satisfaction Scores and Emergency Department Administration of Analgesic Medications

43. Taylor C, Benger JR. Patient satisfaction in emergency medicine. Emerg Med J. 2004;21:528-532. 44. Clark CA, Pokorny ME, Brown ST. Consumer satisfaction with nursing care in a rural community hospital emergency department. J Nurs Care Qual. 1996;10:49-57. 45. Counselman FL, Graffeo CA, Hill JT. Patient satisfaction with physician assistants (PAs) in an ED fast track. Am J Emerg Med. 2000;18:661-665. 46. Durston W, Carl ML, Guerra W. Patient satisfaction and diagnostic accuracy with ultrasound by emergency physicians. Am J Emerg Med. 1999;17:642-646. 47. Fosnocht DE, Swanson ER, Bossart P. Patient expectations for pain medication delivery. Am J Emerg Med. 2001;19:399-402. 48. Hostutler JJ, Taft SH, Snyder C. Patient needs in the emergency department: nurses’ and patients’ perceptions. J Nurs Admin. 1999;29:43-50. 49. Kelly AM. Patient satisfaction with pain management does not correlate with initial or discharge VAS pain score, verbal pain rating at discharge, or change in VAS score in the emergency department. J Emerg Med. 2000;19:113-116. 50. D’Amore JZ, Mounessa JS, Rosen L, et al. Does a patient advocate during the discharge process improve patient safety and satisfaction in the emergency department? Ann Emerg Med. 2012;60:S98-S99. 51. Rega PP, Roberts SM, Khuder S, et al. The delivery of a health promotion intervention by a public health promotion specialist improves patient satisfaction in the emergency department. Acad Emerg Med. 2012;19:313-317. 52. Nerney MP, Chin MH, Jin L, et al. Factors associated with older patients’ satisfaction with care in an inner-city emergency department. Ann Emerg Med. 2001;38:140-145. 53. Baker DW, Hayes R, Fortier JP. Interpreter use and satisfaction with interpersonal aspects of care for Spanish-speaking patients. Med Care. 1998;36:1461-1470. 54. Bursch B, Beezy J, Shaw R. Emergency department satisfaction—what matters most. Ann Emerg Med. 1993;22:586-591. 55. Campanella HC, Campanella PM, Grayson K. Factors affecting Department of Defense patient satisfaction in a military emergency department. Mil Med. 2000;165:396-402. 56. Corbett SW, White PD, Wittlake WA. Benefits of an informational videotape for emergency department patients. Am J Emerg Med. 2000;18:67-71. 57. Krishel S, Baraff LJ. Effect of emergency department information on patient satisfaction. Ann Emerg Med. 1993;22:568-572. 58. Mack JL, File KM, Horwitz JE, et al. The effect of urgency on patient satisfaction and future emergency department choice. Health Care Manage Rev. 1995;20:7-15. 59. Mayer TA, Cates RJ, Mastorovich MJ, et al. Emergency department patient satisfaction: customer service training improves patient satisfaction and ratings of physician and nurse skill. J Healthc Manag. 1998;43:427-440. 60. Mazor SS, Hampers LC, Chande VT, et al. Teaching Spanish to pediatric emergency physicians—effects on patient satisfaction. Arch Pediatr Adolesc Med. 2002;156:693-695. 61. Raper JL. A cognitive approach to patient satisfaction with emergency department nursing care. J Nurs Care Qual. 1996;10:48-58. 62. Rhee KJ, Allen RA, Bird J. Telephone vs mail response to an emergency department patient satisfaction survey. Acad Emerg Med. 1998;5:1121-1123.

Volume

-,

no.

-

:

-

2014

63. Rhee KJ, Dermyer AL. Patient satisfaction with a nurse practitioner in a university emergency service. Ann Emerg Med. 1995;26: 130-132. 64. Richards CR, Richell-Herren K, Mackway-Jones K. Emergency management of chest pain: patient satisfaction with an emergency department based six hour rule out myocardial infarction protocol. Emerg Med J. 2002;19:122-125. 65. Rydman RJ, Roberts RR, Albrecht GL, et al. Patient satisfaction with an emergency department asthma observation unit. Acad Emerg Med. 1999;6:178-183. 66. Rydman RJ, Zalenski RJ, Roberts RR, et al. Patient satisfaction with an emergency department chest pain observation unit. Ann Emerg Med. 1997;29:109-115. 67. Schiermeyer RP, Tayal V, Butzin CA. Physician business cards enhance patient satisfaction. Am J Emerg Med. 1994;12:125-126. 68. Singer AJ, Sanders BT, Kowalska A, et al. The effect of introducing bedside TV sets on patient satisfaction in the ED. Am J Emerg Med. 2000;18:119-120. 69. Singer AJ, Thode HC. Determination of the minimal clinically significant difference on a patient visual analog satisfaction scale. Acad Emerg Med. 1998;5:1007-1011. 70. Sox CM, Burstin HR, Orav EJ, et al. The effect of supervision of residents on quality of care in five university-affiliated emergency departments. Acad Med. 1998;73:776-782. 71. Spaite DW, Bartholomeaux F, Guisto J, et al. Rapid process redesign in a university-based emergency department: decreasing waiting time intervals and improving patient satisfaction. Ann Emerg Med. 2002;39:168-177. 72. Stahmer SA, Shofer FS, Marino A, et al. Do quantitative changes in pain intensity correlate with pain relief and satisfaction? Acad Emerg Med. 1998;5:851-857. 73. Thompson DA, Yarnold PR. Relating patient satisfaction to waiting time perceptions and expectations: the disconfirmation paradigm. Acad Emerg Med. 1995;2:1057-1062. 74. Wissow LS, Roter D, Bauman LJ, et al. Patient-provider communication during the emergency department care of children with asthma. Med Care. 1998;36:1439-1450. 75. Lewis KE, Woodside RE. Patient satisfaction with care in the emergency department. J Adv Nurs. 1992;17:959-964. 76. Watson WT, Marshall ES, Fosbinder D. Elderly patients’ perceptions of care in the emergency department. J Emerg Nurs. 1999;25:88-92. 77. Mack JL, File KM, Horwitz JE, et al. Factors associated with emergency room choice among Medicare patients. J Ambul Care Mark. 1995;6:45-58. 78. Magaret ND, Clark TA, Warden CR, et al. Patient satisfaction in the emergency department—a survey of pediatric patients and their parents. Acad Emerg Med. 2002;9:1379-1388. 79. McMillan JR, Younger MS, DeWine LC. Satisfaction with hospital emergency department as a function of patient triage. Health Care Manage Rev. 1986;11:21-27. 80. American Cancer Society. Pain Management Pocket Tool. 2005. 81. Von Korff M, Saunders K, Ray GT, et al. De facto long-term opioid therapy for noncancer pain. Clin J Pain. 2008;24:521-527. 82. Rupp T, Delaney KA. Inadequate analgesia in emergency medicine. Ann Emerg Med. 2004;43:494-503. 83. Boudreaux ED, Friedman J, Chansky ME, et al. Emergency department patient satisfaction: examining the role of acuity. Acad Emerg Med. 2004;11:162-168.

Annals of Emergency Medicine 13