Pharmacologic management of analgesia and sedation is desired and expected in the critical care setting. However, excessive sedation may be harmful.1–3 Routine assessments of sedation4–6 and of sedation and analgesia7 offset these harmful effects. Protocolized interruption and adjustment of sedation2,8 seem beneficial in some intensive care units (ICUs), but not in clinical environments where nurse assessment–driven sedation is part of routine care.9
Sedative3 and analgesic use1 is also linked to delirium. Delirium in the ICU setting worsens outcome and independently predicts mortality.1,10 Sedative-induced coma is a powerful risk factor for ICU delirium occurrence,1 and such coma independently decreases survival.
We hypothesized that teaching ICU staff to use a protocol in which the clinical features of pain, agitation, and delirium are distinguished and drive pharmacologic and nonpharmacologic intervention would reduce medication use. We further hypothesized that these reductions in potentially inappropriate medication administration would reduce the rates of iatrogenic coma and delirium and improve clinical outcomes.
Data were collected prospectively on all consecutive patients in a single tertiary care adult ICU during 2 periods: PRE (August 2003 to February 2004: no protocol was available) and POST (April 2005 to November 2005: a protocol was available). Between the 2 periods, a multidisciplinary team piloted, taught, and implemented management protocols for systematic management of analgesia, sedation, and delirium. Daily symptom assessments were performed routinely in the PRE group. Pharmacologic interventions were prescribed by physicians but not routinely titrated; opiate and sedative perfusions were usually prescribed at a fixed rate without systematic adjustments, and there were no systematic nonpharmacologic interventions.
All adult patients admitted for >24 hours were entered into the study. Moribund patients (i.e., patients expected to die within 24 hours or less) were excluded. The ICU admits all types of surgical and medical patients and is a regional referral bone marrow transplant center. The hospital does not provide cardiac surgery or trauma care.
At least once per 8-hour shift, throughout the ICU stay, in both the PRE and POST groups, each patient was evaluated for pain, sedation, and delirium in the following manner.
The Numeric Rating Scale (NRS)11 was used, ranging from 0 (no pain) to 10 (most pain) as reported by the patient to the caregiver. When the patient was unable to report pain, the nurse provided a subjective assessment of behavioral pain12,13 and documented it according to the NRS from 0 (no apparent pain) to 10 (severe pain). The 3- to 12-point behavioral pain scale (BPS) converted to an NRS of 0 (BPS points ranging from 3 to 5); of 10 (BPS 10–12); and 5 for intermediate values. For the NRS and for each scale, the value was tabulated for each day spent in the ICU, and then averaged over the number of days the patient spent in the ICU. “Worst” scores were also considered but not shown because they were not significantly different than the average scores.
We used the Richmond Agitation and Sedation Scale (RASS),14,15 a diagnostic tool validated for ICU use, from comatose (score of −5) to combative (score of +4). Patients with RASS scores of −4 or −5, who were receiving sedatives or opiates, with no neurological, metabolic, or structural abnormality to justify this level of sedation, were considered to have medication-induced (or iatrogenic) coma if such a RASS level was noted at any one time (or more frequently) during ICU admission.
A checklist evaluation (Intensive Care Delirium Screening Checklist [ICDSC] and nursing implementation instruction [Appendix 1]) was used where the level of sedation permitted (i.e., RASS score between −3 and +4). Changes in wakefulness and attention directly attributable to sedative medication were not “scored” as positive ICDSC points. Patients with a diagnostic threshold ICDSC score (i.e., ≥4) at any assessment during their ICU stay were considered to have clinical delirium.16 Patients with a maximal ICDSC score >0 but <3 at any time during their ICU stay were considered to have subsyndromal delirium.17 An ICDSC score of 0 was considered absence of delirium.
Nurses are trained18 to routinely assess and record (Appendix 2) results of the validated assessment tools described above. During the protocol implementation period (February 2004 to April 2005), standardized prescription sheets for NRS-driven analgesic administration, RASS-driven sedative administration, and ICDSC-driven antipsychotic administration were created by a focus group of intensivists, ICU nurses, and pharmacists. The prescription sheets were shown to all ICU staff nurses, pharmacists, and physicians for review, and then piloted. Feedback was solicited for clarity, ease of use, and team preferences, and the prescription sheet was reformatted accordingly by the focus group before being implemented in the final form evaluated in this study (Appendix 3). All new ICU nurses received education on screening for and differentiating pain, sedation requirements, and delirium.18 The prescription sheet was approved by the pharmacy and medical records committees before implementation.
Before the second period of observation, portable radios and CD players were installed at each ICU patient's bedside.19,20 Nurses were encouraged to offer patients the following interventions: patients were asked whether they would prefer to listen to music20,21 or receive sedative medication to alleviate anxiety. In addition, all patients with hallucinations or other delirium features were reassured that the symptoms were “not real” and were routinely reoriented22,23; this was not done systematically in the PRE group.
Physicians were invited to prescribe a protocol for analgesia (Appendix 3). This protocol incorporated options for routine coanalgesia with acetaminophen and nonsteroidal antiinflammatory drugs (NSAIDs) (where there were no contraindications) and titrated administered medication to self-reported NRS levels (aiming for an NRS score of 0/10). A separate protocolized prescription for sedation was also made available, which consisted of adjusting sedatives to RASS goals specified by the ICU physician (ranging from 0 to −3) (Appendix 3). The pharmacologic management of delirium included haloperidol or olanzapine, which could be prescribed based on the ICDSC score observed by the nurse (e.g., pro re nata administration of haloperidol when the ICDSC score exceeded 3/8) (Appendix 3).
All opiates were considered morphine-equivalent doses24 and all benzodiazepines considered lorazepam equivalents24 to compare total doses of drug used. All other medications (antipsychotics, acetaminophen, propofol, and NSAIDs) were considered individually. Use of coanalgesia (i.e., NSAID or acetaminophen) was compared between cohorts. In addition, the percentage of patients in whom benzodiazepines and opiates were administered was compared between groups. Prescription patterns were evaluated by changes in SD of drug dose variability between physicians and within individual physician prescriptions.
The discharge from hospital category was established based on the patient's care provision immediately after hospital discharge: going home without help, home with help (clinical nursing or other care), rehabilitation in a convalescent center, and placement in long-term care. We describe “going home” as the ability to return to independent living (without help).
Demographic data were expressed as means ± SD for continuous variables and frequencies and percentages for categorical variables. Continuous variables were compared among the PRE and POST groups, and in the POST subgroup analysis, using t tests. For continuous variables with skewed distribution, nonparametric Mann-Whitney tests were used. Categorical variables were compared among the PRE and POST groups, and in the POST subgroup analysis, using χ2 tests. Mortality was studied using survival analysis and compared between groups with the log-rank test. All analyses were done with SAS version 9.1 (SAS Institute Cary, NC) and conducted at the 0.05 significance level. Because we aimed to compare the PRE and POST groups, analysis within the POST subgroups was considered strictly exploratory.
The IRB (Hôpital Maisonneuve-Rosemont, Comités Scientifique et Éthique, Centre de Recherche Guy Bernier) approved the study. Because all information was anonymously abstracted from bedside medical records for quality assurance purposes, individual patient consent was waived by the institutional ethics committee.
We enrolled 610 patients in the PRE group and 604 patients in the POST group; 81 patients (38 in PRE and 43 in POST) had >1 admission, and only the first admission was included in the analysis. Patient mix, baseline renal and hepatic function, medical staff, and clinical practice were unchanged between the PRE and POST period. For delirium evaluation, 67 patients in the PRE group and 46 in the POST group were excluded, because their level of consciousness precluded ICDSC measures; the level of consciousness that precludes assessment of delirium corresponds to a RASS score of −4 or −5 and, exceptionally (in <5% of the cases), a RASS score of −3 in an uncooperative patient. All other variables (sedation, analgesia, and outcomes) were measured in patients in whom delirium assessment was not possible. All patients could thus be evaluated for RASS levels, and patients could self-report NRS levels >80% of the time. All other variables (analgesia and outcomes) were measured in patients in whom delirium assessment was not possible.
Groups were comparable in all important baseline characteristics (Table 1) with the exception of the Acute Physiology and Chronic Health Evaluation (APACHE) scores (17.1 PRE vs 18.1 POST, P = 0.03). Patient mix, medical staff, and clinical practice were unchanged between the PRE and POST periods.
Analgesia and Analgesic Use
The POST cohort had superior analgesia (Table 2) while receiving significantly lower mean doses of opiates. An equivalent amount of patients received acetaminophen (Table 2). Opiates were administered to fewer patients in the POST group, and in those who received opiates, the mean dose was 4-fold less (Table 2). The percentage of patients given coanalgesia with acetaminophen did not differ between the groups, nor did the mean daily dose (Table 2). Median doses of morphine equivalents decreased from 45.17 (minimum = 0.23, maximum = 3419.33) to 9.90 (minimum = 0.05, maximum = 501.9). Coanalgesia with an NSAID was not different between the 2 groups (8.6% vs 8.2%, P = 0.8).
Sedation and Sedative Use
There were no differences in sedation scores (median RASS score −0.17 vs 0) and self-reported anxiety between the PRE and POST groups (Table 2). RASS values were abnormally distributed and comparable in both PRE (Shapiro-Wilk = 0.88, P < 0.000; Kolmogorov-Smirnov = 0.173, P < 0.01) and POST (Shapiro-Wilk = 0.814, P < 0.0001; Kolmogorov-Smirnov = 0.23, P < 0.01) cohorts. Mean daily doses of benzodiazepines were slightly lower. Propofol was given to fewer patients but in similar doses (Table 2). Agitation occurred infrequently. Whether all comers or only patients in whom delirium could be assessed were considered, fewer episodes tended to occur in the POST group (data not shown). Sedation levels were not significantly different between mechanically ventilated and nonmechanically ventilated patients.
Delirium and Antipsychotic Use
The number of patients manifesting subsyndromal delirium (i.e., ICDSC score >0 but <4) was significantly less in the POST group. The rate of delirium was similar (34.7% PRE vs 34.2% POST; P = 0.9). More patients remained cognitively intact, i.e., had an ICDSC score of 0 (32.5% PRE vs 41.2% POST; P = 0.004). The mean occurrence of agitation as a feature of the ICDSC score occurred in 0.16 (SD 0.28) of PRE patients and 0.15 (SD 0.27) of POST patients (not significant) (median 0, interquartile range 0.25; and median 0, interquartile range 0.33 for PRE and POST groups, respectively). Drug management with antipsychotics did not differ between the PRE and POST groups. The mean daily dose of analgesics and sedatives calculated from admission to the first detection of delirium in the PRE and POST groups combined was similar to the average ICU analgesic and sedative dose administered to patients with delirium.
Nursing documentation of analgesia, sedation, and delirium was done with similar frequency in the 2 groups. Almost all patients (99%) were exposed to music daily in the POST group, but the subjective effect on the patient was not recorded. In the POST cohort, we observed that nurse inquiries as to patient comfort made them less likely to administer fentanyl or other opiate analgesics; nurses informally reported having a greater understanding of individual patient needs.
The rate of medication-induced coma was reduced from 20.5% to 8.7% (P < 0.0001). ICU stay decreased from 6.3 to 5.35 days (P = 0.009) and LOS from 55 to 27 days (P < 0.0001). Despite similar sedation levels, duration of mechanical ventilation was reduced from 7, 51 to 5, 93 days (P = 0.01). Overall, the percent of patients able to go home increased from 68.2% to 74.8% (P = 0.049). Thirty-eight percent of all patients returning home in the PRE group had had a prior ICDSC = 0% vs 52% of patients in the POST group. Patients with delirium had a similar LOS in the PRE and POST groups (from 10.8 ± 11.3 days to 9.3 ± 8.2 days, P = 0.139). In both groups, patients with an ICDSC = 0 were more likely to go home than patients with subsyndromal delirium and delirium (P < 0.0001 and P = 0.0019, respectively). A total of 83.6% of patients with an ICDSC score of 0/8 returned home in the PRE cohort, as did 81.8% of the POST cohort. Of the patients returning home, 47.9% in the first cohort had an ICDSC score of 0/8, and 52.5% in the second cohort had an ICDSC score of 0/8.
Age, a feature associated with mortality in our patient population,1 was comparable at baseline in the PRE and POST cohorts; the other feature associated with mortality, APACHE score, was higher and statistically, but not clinically, significant. The 30-day mortality risk in the PRE cohort was 29.4% vs 22.9% in the POST cohort (log-rank test, P = 0.009) (Fig. 1).
Subgroup Analysis of the POST Cohort
All POST patients were systematically assessed by nurses, but whereas 52% had pharmacologic management of analgesia and sedation prescribed directly by protocol during the entire ICU stay, 42.6% did not, and 5.4% had periods with and without prescribed protocol. Within the POST group, patients who received protocolized prescriptions were sicker (APACHE score 18.2 ± 7.6 vs 16.3 ± 7.3, P = 0.007), older (64.5 ± 14.4 years vs 60.9 ± 16.6 years, P = 0.007), more likely to be surgical patients (62.8% vs 34.8%, P < 0.0001), and more likely to be mechanically ventilated for >24 hours (53.4% vs 29.4%, P < 0.0001). Mean opiate doses were not different between prescription-managed patients and nonprescription-managed patients (mean dose calculations included only those patients who received any opioid). The proportion of patients receiving opiates was significantly lower in the nonprescription-managed group (Table 3), as was the proportion of patients receiving benzodiazepines. Protocolized prescription patients were more likely to receive coanalgesia with acetaminophen and an NSAID. The rate of iatrogenic coma decreased in both prescription-managed and nonprescription-managed patients.
In this study, we evaluated the management of specific, distinct, and measurable clinical entities: pain, agitation, and delirium. The approach described was multidisciplinary, and involved teaching nurses and physicians about protocolized management of sedatives and analgesics prescribed by physicians, as well as nonpharmacologic care provided by nurses (music and reassurance). The effect of this combined approach is poorly described in the literature. Only 1 study in a trauma population has shown its benefit.25
Increased awareness of the necessity for drugs, aided by protocol-driven care, was associated with improvement in pain control and a large reduction in opiate administration. Opiate doses were administered in narrower ranges in the POST group, suggesting a decrease in prescription variability. The benefits of analgesia driven by patient need have long been recognized in postoperative patients.24 Our results suggest that a similar approach is beneficial in medical-surgical ICU patients.
Thirty-seven percent of patients in the POST cohort received no opiates and had acceptable pain control. The Society of Critical Care Medicine recommendations favoring opiate perfusion as the recommended analgesic approach in critical care26 may thus not be ubiquitously applicable. Simultaneous assessments of pain, sedation, anxiety, and delirium may have reduced opiate administration for perceived discomfort by identifying symptoms from potentially confounding but distinct clinical entities.
We systematically provided all patients who so desired with music in the POST group. The effect of music, and of other “soft” relaxation interventions such as reassurance, cannot be measured objectively. This study suggests no harm and potential benefit for this approach in the critically ill.
Sedation levels as measured with the RASS in our patients, and the percentage of patients not receiving any benzodiazepines or propofol, reflect local practice of sedating patients as little as possible. Administered sedatives were not very different between the 2 cohorts, in contrast to publications supporting the benefits of protocolized sedation interruption and titration.2,8 Few of these publications describe intervention versus control sedation (e.g., RASS) levels. The patients in a recent broadly cited publication27 promoting protocols, aptly named “wake up and breathe,” were randomized into the trial with baseline RASS levels of −4. These patients would be considered oversedated in our practice. In contrast, protocol-directed sedation was no better than usual care in Australian ICUs with experienced nurses given responsibility for many aspects of sedation practice; protocol and nonprotocol RASS levels were not provided in that study. Our RASS levels were similarly low in the PRE and POST groups despite titrating sedatives routinely in only just over half of the second cohort. Because sedation practice varies widely across cultures and continents, the benefit of protocolized sedation titration may vary depending on implementation rates, but possibly also on baseline sedation levels. Our work suggests that light sedation does not increase agitation or anxiety. The loss of ICU stay recall is associated with psychological trauma.22 Given the amnesic properties of benzodiazepines and the benefits of sedative interruption,28 longitudinal studies may better answer which depth of targeted sedation favors best patient outcome.
Delirium and Antipsychotic Use
A spectrum of cerebral dysfunction, which might include delirium, and coma upon infusion of opiates and sedatives, has been found in the literature. Its biological mechanisms and the links among contributory factors are the subject of debate. Delirium (34.7% PRE vs 34.2% POST, P = 0.9) was not different in PRE and POST cohorts, suggesting that either iatrogenic coma does not cause delirium despite the strong association between the 2 clinical entities or the reduction in iatrogenic coma “unmasked” patients who were subsequently found to have delirium. Subsyndromal delirium rates were reduced significantly. In parallel, patients who remained cognitively intact were far more likely to go home. These findings suggest that subsyndromal delirium may be modifiable. There is an important association between subsyndromal delirium and the likelihood of short- or long-term dependency on institutionalized care. Teaching protocolized ICU management of analgesia, sedation, and delirium may thus translate into meaningful improvements in outcome for patients and their families.
Iatrogenic coma was dramatically reduced in the POST group. This variable has been independently associated with a higher mortality and a worse outcome,1 and it is associated with delirium in patients given benzodiazepines3 and independent of administered drugs1; however, reducing the rate of iatrogenic coma did not reduce the incidence of delirium. Two explanations are possible. One is that there is no causal link. The other is that patients sensitive to the effects of sedation (the previously “iatrogenically comatose”) are also more likely to develop delirium, and that these patient characteristics were unmasked by their availability to be evaluated when not comatose. With the rigorous titration of medication to effect in the POST group, the rate of iatrogenic coma was reduced from 20.5% to 8.7%. However, even this represents a higher rate than we had expected, especially given the reduction in drug doses. This finding suggests that, in addition to whether medication doses are adjusted to patient needs, other patient characteristics may put patients at risk for developing iatrogenic coma.
Mortality and Other Outcomes
The 30-day mortality risk in the PRE cohort was 29.4% vs 22.9% in the POST cohort (log-rank test, P = 0.009), despite APACHE scores of 17 (PRE) and 18 (POST) on admission. The pre-post design of the study precludes any attribution of causality between patient management and mortality and is further complicated by a heterogeneous population that included bone marrow transplant patients (who accounted for many of the long hospital LOS). At the very least, our findings suggest that the protocol does not confer any risk. The association of the protocol with a decrease in 30-day mortality, in addition to dependency benefits, may be worth validating with a randomized controlled trial.
The ability to go home is of critical importance to patients and families. The decrease in need for dependent care upon discharge to a nursing home or long-term facility was associated with a decrease in subsyndromal delirium. Subsyndromal delirium carries an intermediate risk, compared with delirium and normal cognition, for ICU LOS. It is associated with as long a hospital stay as delirium and does not seem to be associated with increased mortality.16 Return home is as unlikely after subsyndromal delirium as it is for delirium.16 Whether there is a link between protocolized ICU management of analgesia, sedation, and delirium and the capacity that a patient has to return home remains to be tested in a prospective randomized manner.
Physicians prescribed the pharmacologic protocol to sicker and older surgical patients in whom mechanical ventilation, ICU stay, and analgesic requirements may have been anticipated to last longer. Regardless of pharmacologic protocol prescription, overall practice was most altered with opiate administration, with no compromise in analgesia. Protocolization seemed to reduce variability in practice, as reflected by the changes in SDs of drug doses. Pharmaceutical protocol adherence was assumed based on the presence of standardized prescription sheets in the chart. Among “nonprescription protocol” patients, however, several were noted to include nonstandardized prescriptions that recommended adjustment of drug doses to patient symptoms.
Analysis of those who did and did not receive protocolized prescriptions (in the POST cohort) allowed us to make 2 important observations. The first was that practice such as prescribing opiates is altered by implementation of a protocolized approach regardless of whether a protocolized prescription is adhered to or not, probably because the efficacy and side effects of analgesics and sedatives are monitored more closely regardless of a drug administration prescription. The second is that nurses and physicians are likely to select patients for whom protocols are prescribed. The corollary of the first finding is that the benefit of protocolized care cannot, and should not, be measured by adherence to prescriptions. Both observations support that expert reasoning by critical care caregivers is altered when evaluation (of pain, sedation, anxiety, and delirium, in this case) is performed with a therapeutic intervention goal. The basis for this type of behavioral change is well described in theoretical frameworks such as Activity Theory.29
This was an observational, not a randomized study, and therefore has the associated limitations of this design. The study's pre-post design was chosen to best mimic real-life clinical situations, because of concern that a randomized model would be contaminated by culture change over time, and because observational studies can provide valuable evidence even when compared with randomized trials.30 The study's second cohort APACHE score was slightly higher, yet the study suggested that at the very least, the intervention was unlikely to cause harm. In addition, critically ill populations are heterogeneous; even cohorts as large as the PRE and POST groups may incorporate confounders that are not accounted for. The subgroups described within the second cohort (prescription protocol, POST cohort versus no prescription protocol, POST cohort) were not equivalent in terms of baseline characteristics. Any comparisons between them should thus be considered strictly exploratory. The presence of multiple interventions limits attributing outcome to one intervention (such as pharmacologic intervention in prescriptions, or nonpharmacologic intervention such as listening to music) or another. In addition, we were unable to reliably report on the use of reassurance (which may have been performed by some nurses in the PRE cohort) or on the utilization of music in the POST cohort. Sleep quality and length, day-night sleep-wake cycles, ambient light, and noise level were likely to be comparable in the 2 cohorts but were not measured directly. We did not assess overall and societal costs associated with discharge to a nursing home or other long-term facility. Quality of life, from a patient's or family's perspective, which may have portrayed burden of outcome, was not assessed. Because of our exclusion criteria, the results do not apply to moribund patients expected to die within 24 hours.
Teaching staff protocol–based ICU patient assessments targeting nonpharmacologic and pharmacologic management of pain, sedation, and delirium were associated with improved outcome, regardless of strict adherence to protocolized medication administration. These data suggest that individualization of care with therapeutic intent improves outcomes.
YS, ML, and DKA piloted and implemented the protocol. YS was involved in study conception and design, and data collection and analysis. FM, SA, DKA, and ML provided significant contributions to protocol implementation and evaluation. FM and SA were involved in critical review of the manuscript. BPK helped to review and analyze the data and redact the manuscript.
The authors thank the participating intensive care physicians at Maisonneuve-Rosemont Hospital (Drs. Jean Gelinas, Brian Laufer, Martin Légaré, and Jeanne Teitelbaum), and Marieve Cossette for the statistical review of the manuscript.
1. Ouimet S, Kavanagh BP, Gottfried SB, Skrobik Y. Incidence, risk factors and consequences of ICU delirium. Intensive Care Med 2007;33:66–73
2. Kress JP, Pohlman AS, O'Connor MF, Hall JB. Daily interruption of sedative infusions in critically ill patients undergoing mechanical ventilation. N Engl J Med 2000;342:1471–7
3. Pandharipande P, Shintani A, Peterson J, Pun BT, Wilkinson GR, Dittus RS, Bernard GR, Ely EW. Lorazepam is an independent risk factor for transitioning to delirium in intensive care unit patients. Anesthesiology 2006;104:21–6
4. De Jonghe B, Bastuji-Garin S, Fangio P, Lacherade JC, Jabot J, Appere-De-Vecchi C, Rocha N, Outin H. Sedation algorithm in critically ill patients without acute brain injury. Crit Care Med 2005;33:120–7
5. Brook AD, Ahrens TS, Schaiff R, Prentice D, Sherman G, Shannon W, Kollef MH. Effect of a nursing-implemented sedation protocol on the duration of mechanical ventilation. Crit Care Med 1999;27:2609–15
6. Brattebo G, Hofoss D, Flaatten H, Muri AK, Gjerde S, Plsek PE. Effect of a scoring system and protocol for sedation on duration of patients' need for ventilator support in a surgical intensive care unit. Br Med J 2002;324:1386–9
7. Chanques G, Jaber S, Barbotte E, Violet S, Sebbane M, Perrigault PF, Mann C, Lefrant JY, Eledjam JJ. Impact of systematic evaluation of pain and agitation in an intensive care unit. Crit Care Med 2006;34:1691–9
8. Schweickert WD, Gehlbach BK, Pohlman AS, Hall JB, Kress JP. Daily interruption of sedative infusions and complications of critical illness in mechanically ventilated patients. Crit Care Med 2004;32:1272–6
9. Bucknall TK, Manias E, Presneill JJ. A randomized trial of protocol-directed sedation management for mechanical ventilation in an Australian intensive care unit. Crit Care Med 2008;36:1444–50
10. Ely EW, Shintani A, Truman B, Speroff T, Gordon SM, Harrell FE Jr, Inouye SK, Bernard GR, Dittus RS. Delirium as a predictor of mortality in mechanically ventilated patients in the intensive care unit. JAMA 2004;291:1753–62
11. Puntillo KA, Miaskowski C, Kehrle K, Stannard D, Gleeson S, Nye P. Relationship between behavioral and physiological indicators of pain, critical care patients' self-reports of pain, and opioid administration. Crit Care Med 1997;25:1159–66
12. Payen JF, Bru O, Bosson J, Lagrasta A, Novel E, Deschaux I, Lavagne P, Jacquot C. Assessing pain in critically ill sedated patients by using a behavioral pain scale. Crit Care Med 2001;29:2258–63
13. Aïssaoui Y, Zeggwagh AA, Zekraoui A, Abidi K, Abouqal R. Validation of a behavioral pain scale in critically ill, sedated, and mechanically ventilated patients. Anesth Analg 2005; 101:1470–6
14. Sessler CN, Gosnell MS, Grap MJ, Brophy GM, O'Neal PV, Keane KA, Tesoro EP, Elswick RK. The Richmond Agitation-Sedation Scale: validity and reliability in adult intensive care unit patients. Am J Respir Crit Care Med 2002;166:1338–44
15. Ely EW, Truman B, Shintani A, Thomason JW, Wheeler AP, Gordon S, Francis J, Speroff T, Gautam S, Margolin R, Sessler CN, Dittus RS, Bernard GR. Monitoring sedation status over time in ICU patients: reliability and validity of the Richmond Agitation-Sedation Scale (RASS). JAMA 2003;289:2983–91
16. Bergeron N, Dubois MJ, Dumont M, Dial S, Skrobik Y. Intensive care delirium screening checklist: evaluation of a new screening tool. Intensive Care Med 2001;27:859–64
17. Ouimet S, Riker R, Bergeron N, Cossette M, Kavanagh B, Skrobik Y. Subsyndromal delirium in the ICU: evidence for a disease spectrum. Intensive Care Med 2007;33:1007–13
18. Devlin JW, Marquis F, Riker RR, Robbins T, Garpestad E, Fong JJ, Didomenico D, Skrobik Y. Combined didactic and scenario-based education improves the ability of intensive care unit staff to recognize delirium at the bedside. Crit Care 2008;12:R19
19. Prensner JD, Yowler CJ, Smith LF, Steele AL, Fratianne RB. Music therapy for assistance with pain and anxiety management in burn treatment. J Burn Care Rehabil 2001;22:83–8
20. Nilsson U, Rawal N, Unestahl LE, Zetterberg C, Unosson M. Improved recovery after music and therapeutic suggestions during general anaesthesia: a double-blind randomised controlled trial. Acta Anaesthesiol Scand 2001;45:812–7
21. Henry LL. Music therapy: a nursing intervention for the control of pain and anxiety in the ICU: a review of the research literature. Dimens Crit Care Nurs 1995;14:295–304
22. Hupcey JE. Feeling safe: the psychosocial needs of ICU patients. J Nursing Scholarsh 2000;32:361–7
23. Hupcey JE. The need to know: experiences of critically ill patients. Am J Crit Care 2000;9:192–8
24. Cammarano WB, Drasner K, Katz JA. Pain control, sedation and use of muscle relaxants. In: Hall JB, Schmidt GA, Wood LD eds. Principles of Critical Care. New York, NY: McGraw-Hill, 1998:87–109
25. Robinson BR, Mueller EW, Henson K, Branson RD, Barsoum S, Tsuei BJ. An analgesia-delirium-sedation protocol for critically ill trauma patients reduces ventilator days and hospital length of stay. J Trauma 2008;65:517–26
26. Curatolo M, Sveticic G. Drug combinations in pain treatment: a review of the published evidence and a method for finding the optimal combination. Best Pract Res Clin Anaesthesiol 2002;16:507–19
27. Girard TD, Kress JP, Fuchs BD, Thomason JW, Schweickert WD, Pun BT, Taichman DB, Dunn JG, Pohlman AS, Kinniry PA, Jackson JC, Canonico AE, Light RW, Shintani AK, Thompson JL, Gordon SM, Hall JB, Dittus RS, Bernard GR, Ely EW. Efficacy and safety of a paired sedation and ventilator weaning protocol for mechanically ventilated patients in intensive care (Awakening and Breathing Controlled trial): a randomised controlled trial. Lancet 2008;371:126–34
28. Jacobi J, Fraser GL, Coursin DB, Riker RR, Fontaine D, Wittbrodt ET, Chalfin DB, Masica MF, Bjerke HS, Coplin WM, Crippen DW, Fuchs BD, Kelleher RM, Marik PE, Nasraway SA Jr, Murray MJ, Peruzzi WT, Lumb PD. Clinical practice guidelines of the sustained use of sedatives and analgesics in the critically ill adult. Crit Care Med 2002;30:119–41
29. Kress JP, Gehlbach B, Lacy M, Pliskin N, Pohlman AS, Hall JB. The long-term psychological effects of daily sedative interruption on critically ill patients. Am J Respir Crit Care Med 2003;168:1457–61
30. Havnes A. Examination and learning: an activity-theoretical analysis of the relationship between assessment and educational practice. Assess Eval High Educ 2004;29:159–76
31. Concato J, Shah N, Horwitz RI. Randomized, controlled trials, observational studies, and the hierarchy of research designs. N Engl J Med 2000;342:1887–92
32. O'Connor PG, Schottenfeld RS. Patients with alcohol problems. N Engl J Med 1998;338:592–602