Presented at the Australasian College for Emergency Medicine, Annual Scientific Meeting, Sydney, November 2011; and the 6th Mediterranean Emergency Medicine Congress, Kos, Greece, September 2011.
Factors Associated With High Levels of Patient Satisfaction With Pain Management
Article first published online: 4 OCT 2012
© 2012 by the Society for Academic Emergency Medicine
Academic Emergency Medicine
Volume 19, Issue 10, pages 1212–1215, October 2012
How to Cite
Shill, J., Taylor, D. McD., Ngui, B., Taylor, S. E., Ugoni, A. M., Yeoh, M. and Richardson, J. (2012), Factors Associated With High Levels of Patient Satisfaction With Pain Management. Academic Emergency Medicine, 19: 1212–1215. doi: 10.1111/j.1553-2712.2012.01451.x
Awarded The Australasian College for Emergency Medicine Morson Taylor Award for the best project protocol by a Fellow of the College.
The authors have no relevant financial information or potential conflicts of interest to disclose.
Supervising Editor: James R. Miner, MD.
- Issue published online: 15 OCT 2012
- Article first published online: 4 OCT 2012
- Received April 6, 2012; revision received May 30, 2012; accepted June 12, 2012.
ACADEMIC EMERGENCY MEDICINE 2012; 19:1212–1215 © 2012 by the Society for Academic Emergency Medicine
Objectives: The objective was to determine, among emergency department (ED) patients, the factors associated with a high level of satisfaction with pain management.
Methods: This was a prospective cohort study in a single ED. Consecutive adult patients, with triage pain scores of ≥4 (numerical rating scale = 0 to 10), were enrolled. Variables examined included demographics, presenting complaint, pain scores, nurse-initiated analgesia, analgesia administered, time to first analgesia, specific pain communication, and whether “adequate analgesia” was provided (defined as a decrease in pain score to <4 and a decrease from the triage pain score of ≥2). The level of patient satisfaction with their pain management (six-point scale: very unsatisfied to very satisfied) was determined by a blinded investigator 48 hours post discharge. Logistic regression analyses were undertaken.
Results: Data were complete for 476 patients: mean (±standard deviation [SD]) age was 43.6 (±17.2) years, and 237 were males (49.8%, 95% confidence interval [CI] = 45.2% to 54.4%). A total of 190 (39.9%, 95% CI = 35.5% to 44.5%) patients were “very satisfied” with their pain management, and 207 (43.5%, 95% CI = 39.0% to 48.1%) patients received adequate analgesia. Three variables were associated with the patient being very satisfied: the provision of adequate analgesia (odds ratio [OR] = 7.8, 95% CI = 4.9 to 12.4), specific pain communication (OR = 2.3, 95% CI = 1.3 to 4.1), and oral opioid administration (OR = 2.0, 95% CI = 1.1 to 3.4). Notably, the provision of nurse-initiated analgesia to 211 patients (44.3%, 95% CI = 39.8% to 48.9%) and the short time to analgesia (median = 11.5 minutes; interquartile range [IQR] = 2.0 to 85.8 minutes) were not associated with being very satisfied.
Conclusions: The receipt of adequate analgesia (as defined) is highly associated with patient satisfaction. This variable may serve as a clinically relevant and achievable target in the pursuit of best-practice pain management.
Objetivos: Determinar, entre los pacientes de un servicio de urgencias (SU), los factores asociados con un alto nivel de satisfacción con el manejo del dolor.
Métodos: Estudio de cohorte prospectivo en un único SU. Se incluyeron pacientes adultos consecutivos, con una puntuación de dolor en el triaje≥4 (escala numérica: 0 a 10). Las variables analizadas incluyeron variables demográficas, motivo de consulta, puntuaciones del dolor, analgesia iniciada por el enfermero, analgesia administrada, tiempo hasta la primera analgesia, comunicación específica del dolor y si la analgesia administrada fue la adecuada (definida como un descenso en la puntuación del dolor a < 4 y un descenso de la puntuación del dolor del triaje de ≥2). Se determinó el nivel de satisfacción del paciente con el manejo del dolor (escala de seis puntos: muy insatisfecho a muy satisfecho) por un investigador enmascarado a las 48 horas tras el alta. Se llevó a cabo un análisis de regresión logística.
Resultados: Los datos se completaron en 476 pacientes: la edad media fue 43,6 años (desviación estándar: ±17,2 años), y 237 fueron hombres (49,8%, IC 95% = 45,2% a 54,4%). Ciento noventa (39,9%, IC 95% = 35,5% a 44,5%) pacientes estuvieron “muy satisfechos” con el manejo del dolor, y 207 (43,5%, IC 95% = 39,0% a 48,1%) de los pacientes consideraron que recibieron una “analgesia adecuada”. Tres variables se asociaron con que el paciente estuviera muy satisfecho: el proporcionar una “adecuada analgesia” (OR 7,8, IC 95% = 4,9 a 12,4), la comunicación específica del dolor (OR 2,3, IC 95% = 1,3 a 4,1), y la administración de opioides por vía oral (OR 2,0, IC 95% = 1,1 a 3,4). Particularmente, no se asoció con estar muy satisfecho ni la administración de analgesia iniciada por el enfermero a 211 pacientes (44,3%, IC 95% = 39,8% a 48,9%) ni el tiempo corto hasta la administración de la analgesia (mediana 11,5 minutos, RIC 2,0 a 85,8 minutos).
Conclusiones: El recibir una “adecuada analgesia” (como se ha definido previamente) está asociado en un alto grado con la satisfacción del paciente. Esta variable puede servir como un objetivo asequible y clínicamente relevante en el ejercicio de la mejor práctica clínica en el manejo del dolor.
One important goal of pain management is a high level of patient satisfaction with that management. Initiatives to improve emergency department (ED) pain management have included staff training,1 nurse-initiated analgesia,2 time-to-analgesia key performance indicators,3 and mandatory recording of pain scores.3 Although well intended, the endpoints of these initiatives do not necessarily result in increased patient satisfaction.
“Adequate analgesia” has been defined as providing a “reduction in the triage pain score of ≥2 and to a level <4” (verbal numerical rating scale, range = 0 to 10).2 It has been reported that there is a significant association between the provision of adequate analgesia and patients being very satisfied with their pain management (odds ratio [OR] = 2.1).2 In this study, we examined the factors associated with a high level of patient satisfaction with their pain management. If the receipt of adequate analgesia is confirmed as an important factor, it would have the potential to serve as a valuable, clinically relevant, and achievable target.
This was a prospective cohort study in a convenience sample of ED patients. The study was authorized by the hospital’s Human Research Ethics Committee, and patients gave informed verbal consent after admission to an ED cubicle.
Study Setting and Population
The setting was a university-affiliated teaching hospital with an adult/pediatric ED. It has an annual census of 70,000 patients and is located in a metropolitan area. The study ran from July 2010 to April 2011.
Consecutive eligible patients were enrolled during 6-hour periods when researchers were available, generally between 08:00 and 22:00 hours, 7 days a week. Inclusion criteria were age ≥18 years and a triage pain score ≥4 (moderate/severe pain4) on a 0-to-10 verbally administered numerical rating scale.5 Exclusion criteria were suspected cardiac chest pain, triage category 1 (immediate management required), inability to communicate a pain score as determined by the ED staff (dementia, delirium, significant illness, poor English), treatment by a study investigator, or inability to follow up.
Standard ED care was provided. No change in pain management practice occurred to reflect a “real-world” scenario. The treating ED staff were not aware of the study hypothesis or outcome measure. The researchers had no discussions with either the patients or staff about the level of pain or its management.
Eligible patients were identified at triage and enrolled shortly thereafter, at a time when other staff were not interacting with the patients. All ED data were collected prospectively and comprised patient demographics (age, sex, ethnicity), the presenting complaint (reason for analgesia), whether nurse-initiated analgesia was provided, analgesia administered (none, simple [paracetamol ± codeine, nonsteroidal anti-inflammatory drugs], oral opioids [oxycodone], or intravenous [IV] opioids [morphine, fentanyl]), and time to first analgesia. In accordance with standard practice for the ED, vital signs and the initial pain score were recorded by the triage nurse (time [t]0), and subsequent pain scores by the cubicle nurse every 30 minutes thereafter (t0 to t8) for 4 hours (if the patient was still in the ED). Time to first analgesia was the time to first analgesia administration. This and time to adequate analgesia were measured from triage (t0).
A second investigator, blinded to the ED data, collected the remaining data approximately 48 hours after ED discharge. Patients were either visited in the ward or telephoned. Only two questions, from the Patient Outcomes Questionnaire,6 were asked. The first was “How satisfied or dissatisfied are you with the results of your pain treatment [in the ED], overall?” Responses on a six-point ordinal scale (very satisfied, satisfied, slightly satisfied, slightly dissatisfied, dissatisfied, very dissatisfied) were recorded. The second question related to specific pain communication: “[During your stay in the ED], did your doctor or nurse make it clear to you that they consider treatment of pain very important and that you should be sure to tell them when you have pain?” The study outcome of interest was those factors significantly associated with patients being very satisfied with their ED pain management.
Our pilot data2 indicated that 40% of patients who did not receive adequate analgesia were very satisfied with their pain management. To demonstrate that 55% of patients who do receive adequate analgesia are very satisfied, a sample size of at least 382 patients was needed (ratio of received to did not receive adequate analgesia 1:1, power 0.8, two-sided, α = 0.05).
Univariate data analysis and stepwise logistic regression were undertaken using Stata 1999 (StataCorp, College Station, TX). Factors showing no trends and clearly not associated with patients being very satisfied were not entered into the regression models.
Of 570 patients who met the inclusion criteria, 16 declined participation, 57 were lost to follow-up, 19 had delayed follow-up, and two had incomplete data. Data on the remaining 476 patients were analyzed (see Table 1). Mean (±standard deviation [SD]) patient age was 43.6 (±17.2) years. The mean (±SD) pain score at triage was 6.5 (±1.8). The median times with interquartile range (IQR) to first analgesia and adequate analgesia were 11.5 minutes (IQR = 2.0 to 85.8 minutes) and 120 minutes (IQR = 60 to 180 minutes), respectively. A total of 207 (43.5%, 95% confidence interval [CI] = 39.0% to 48.1%) patients received adequate analgesia.
|Variable||N in subgroup||n (%) very satisfied||OR (95% CI)|
|Univariate Analysis||Multivariate Analysis|
|Adequate analgesia given|
|Yes||207||132 (63.8)||6.4 (4.3–9.6)||7.8 (4.9–12.4)|
|“Specific pain communication” given|
|Yes||376||164 (43.6)||2.2 (1.3–3.6)||2.3 (1.3–4.1)|
|Oral opioid administration|
|Yes||133||65 (48.9)||1.7 (1.1–2.5)||2.0 (1.1–3.4)|
|IV opioid administration|
|Yes||66||28 (42.4)||1.1 (0.7–1.9)||1.0 (0.5–2.1)|
|Triage pain score|
|4||76||32 (42.1)||1.5 (0.6–3.5)||2.6 (0.9–7.9)|
|5||86||37 (43.0)||1.5 (0.6–3.6)||2.0 (0.7–5.6)|
|6||85||34 (40.0)||1.3 (0.6–3.2)||2.2 (0.8–6.4)|
|7||91||37 (40.7)||1.4 (0.6–3.3)||2.3 (0.8–6.5)|
|8||73||27 (37.0)||1.2 (0.5–2.9)||1.5 (0.5–4.5)|
|9||35||13 (37.1)||1.2 (0.4–3.3)||1.5 (0.5–4.9)|
|Nurse initiated analgesia|
|Yes||211||81 (38.4)||0.9 (0.6–1.3)||0.8 (0.5–1.4)|
|Analgesic drug given at all|
|Yes||350||145 (41.4)||1.3 (0.8–1.9)||0.9 (0.4–1.7)|
|20–29||106||44 (41.4)||2.0 (0.7–5.9)||3.3 (0.9–11.8)|
|30–39||98||37 (37.8)||1.7 (0.6–5.1)||3.0 (0.8–10.9)|
|40–49||69||27 (39.1)||1.8 (0.6–5.6)||3.0 (0.8–11.2)|
|50–59||81||32 (39.5)||1.8 (0.6–5.6)||2.5 (0.7–9.0)|
|60+||103||45 (43.7)||2.2 (0.7–6.5)||3.5 (1.0–12.6)|
|Female||239||89 (37.2)||0.8 (0.6–1.2)||0.8 (0.5–1.3)|
|Italian||36||17 (47.2)||1.3 (0.7–2.6)||1.3 (0.6–3.0)|
|Greek||27||10 (37.0)||0.9 (0.4–2.0)||0.8 (0.3–2.1)|
|European other||25||8 (32.0)||0.7 (0.3–1.7)||0.6 (0.2–1.7)|
|Asian||23||7 (30.4)||0.6 (0.3–1.6)||0.4 (0.1–1.1)|
|Other||29||12 (41.4)||1.0 (0.5–2.2)||1.4 (0.5–3.4)|
|Abdominal pain||152||59 (38.8)||0.9 (0.6–1.5)||0.8 (0.5–1.4)|
|Headache||50||19 (38.0)||0.9 (0.5–1.7)||1.0 (0.4–2.2)|
|Chest pain||25||10 (40.0)||1.0 (0.4–2.3)||1.3 (0.5–3.7)|
|Other||104||43 (41.3)||1.0 (0.6–1.7)||1.1 (0.6–1.9)|
|Time to first analgesia (min)|
|0–30||194||76 (39.2)||1||Not done|
|31–60||42||23 (54.8)||1.9 (1.0–3.7)|
|61–90||33||13 (39.4)||1.0 (0.5–2.2)|
|91–120||27||9 (33.3)||0.8 (0.3–1.8)|
|120+||54||24 (44.4)||1.2 (0.7–2.3)|
|Time to adequate analgesia (min)|
|0–30||16||14 (87.5)||2.2 (0.4–12.6)||Not done|
|31–60||45||29 (64.4)||0.6 (0.2–1.7)|
|61–90||33||21 (63.6)||0.6 (0.2–1.8)|
|91–120||22||12 (54.5)||0.4 (0.1–1.3)|
|121–150||21||11 (52.4)||0.4 (0.1–1.2)|
|151–180||21||11 (52.4)||0.4 (0.1–1.2)|
|181–210||24||15 (62.5)||0.5 (0.2–1.8)|
Overall, 190 (39.9%, 95% CI = 35.5% to 44.5%) patients were very satisfied with their pain management. The univariate analysis (see Table 1) revealed that the only factors significantly associated with being very satisfied with pain management were the receipt of adequate analgesia, specific advice about pain, and an oral opioid. In the multivariate analysis, these factors remained significantly associated with being very satisfied. The strongest association was with the receipt of adequate analgesia (OR = 7.8). No association was observed for any other factor. Although not significant, there was a negative association between the triage pain score and a high level of satisfaction.
This study demonstrates a very strong association between the receipt of adequate analgesia (as defined) and a high level of patient satisfaction. This finding is consistent with our pilot study,2 and a review by Welch,7 who reported that pain management correlates positively with ED patient satisfaction. Others have also reported that the greater the decrease in the pain score, the greater the patient satisfaction.1,2
The clinical relevance of this finding is that, in attempting to maximize patient satisfaction, clinicians should strive to provide adequate analgesia. Indeed, adequate analgesia could provide a useful clinical target. Achieving this target would entail regular pain score monitoring and the administration of analgesia until the pain score decreases by at least two and to a level of less than four. This target is simple and should be easy to achieve. This study shows that it was achieved in almost one half of cases when standard pain management practice was undertaken. It would be reasonable to expect a substantial increase in this proportion if the target were proactively pursued.
The receipt of specific pain communication was also significantly associated with a high level of satisfaction. This finding is not surprising. Communication regarding pain experienced by patients has been shown to improve satisfaction levels.8 If this communication is provided, the patient is likely to perceive that the staff are interested in them. Patients who feel that staff are empathetic, reassuring, and understanding are more satisfied with their pain management.9
The administration of opioid drugs would be expected to decrease pain scores considerably, thereby increasing patient satisfaction. Our finding that oral opioid administration, unlike IV administration, was associated with a high level of satisfaction, is consistent with that of Fosnocht et al.,10 who reported that the oral route is preferred for pain medication delivery.
As nurse-initiated analgesia is one means of decreasing the time to analgesia, these two factors interact. In this study, it is notable that neither was related to patient satisfaction. These findings likely relate to the very short time to analgesia.
Selection bias may have been introduced by use of a convenience sample, refusal to participate, and loss to follow-up. Patients in extreme pain may not have been enrolled due to their inability to provide a pain score, and the findings may not be applicable to this subgroup. Recall bias may have been introduced by delaying the collection of satisfaction data by 48 hours. The pain scale employed was crude. However, the determination of whether or not an individual received adequate analgesia was made using intrapatient measurements (comparison of an individual’s scores over time).
Confounding variables may have included staffing and practice management changes. Awareness of the study and the regular collection of pain scores may have increased patient interest and satisfaction. However, data collection was minimally intrusive and there was no interference in pain management. As staff were aware that the study was examining pain management, the Hawthorne effect may have affected their practice. As the study was undertaken in a single institution, its external validity may be limited.
The receipt of adequate analgesia, specific communication regarding pain, and an oral opioid drug are associated with high levels of patient satisfaction with their pain management. Adequate analgesia has the potential to serve as a clinically relevant target. This study only found an association between the receipt of adequate analgesia and patient satisfaction. A multicenter trial is being undertaken to investigate the effect of an intervention comprising the use of adequate analgesia as a clinical target.