Original Articles Effects of a Quality Improvement Program in Acute Care Evaluated by Patients, Nurses, and Physicians yyy Elisabeth Hansson, RN, RNT, MSc,* Bengt Fridlund, RNAN, RNT, PhD,* and Inger Hallström, RSCN, PhD*,†
y
From the *Department of Health Sciences, Lund University, and † The Vardal Institute, Lund University, Lund, Sweden. Address correspondence and reprint requests to Elisabeth Hansson, Department of Health Sciences, Lund University, P.O. Box 157, SE-221 00 Lund, Sweden. E-mail:
[email protected]. The Department of Health Sciences, Lund University, and Lund University Hospital provided financial support. 1524-9042/$32.00 © 2006 by the American Society for Pain Management Nursing doi:10.1016/j.pmn.2006.06.002
ABSTRACT:
Hospitalized patients have a right to be assessed for pain and receive optimal pain management, but inadequate pain management is still widely prevalent. Different quality improvement (QI) approaches have been used to improve patients’ experience of pain, with varied results. The aim of this study was to implement a QI program and evaluate its effects on pain management routines in acute care, experienced by patients, nurses, and physicians. A cross-sectional intervention study was conducted with patients, nurses, and physicians assigned to a control group for 2 months in 2002 and an intervention group for 2 months in 2003. Implementation of a QI program, including pain policy and an educational program for the health care professionals, was evaluated with questionnaires. Nurses’ assessment of pain at rest and pain with movement with pain rating scales increased (p < .001) after the intervention. Nurses participating in the educational program increased their knowledge of pain and changed their pain management routines and search habits for pain-related websites on the Internet. The number of patients assessed with pain rating scales increased (p ⴝ .011), but no difference was seen in patients’ experience of pain severity, interference with function, or use of nonpharmacologic methods. Younger and higher-educated patients seemed to prefer a more active participation role in their pain management than did older and low-educated patients. In future research it would be valuable to illuminate patients’ possibilities to participate in pain management further and test different implementation methods with varied length of educational programs including follow-ups for the whole team and the patients in acute care. © 2006 by the American Society for Pain Management Nursing Pain Management Nursing, Vol 7, No 3 (September), 2006: pp 93-108
94
Hansson, Fridlund, & Hallström
Inadequate pain assessment and pain management practices are problems in acutely hospitalized patients (Cohen et al., 2003; Costantini, Viterbori, & Flego, 2002). The patients’ different medical conditions, ages, reasons for pain, and pain treatment make pain management a challenging task for health care professionals. Effective pain control is important because it affects all dimensions of function and quality of life, and unrelieved pain may contribute to the development of chronic pain (Desbiens et al., 1997; Serlin, Mendoza, Nakamura, Edwards, & Cleeland, 1995). Relief of pain is a human right (IASP, 2004), and patients have a right to be assessed, regularly reassessed, and educated about pain (JCAHO, 2001), and to participate in care decisions (Fallberg, 2000). Guidelines and education to support professionals with knowledge and routines (American Pain Society [APS], 1995; Gordon et al., 2005; Spri and Swedish Physicians Association, 1997; Spri & Swedish Society of Nursing [SSF], 1999; Swedish Association for Anaesthesia and Intensive Care Medicine [SFAI], 2005) should therefore be included in the hospital’s quality improvement (QI) program and continuously evaluated. The aim of this study was to implement a QI program and evaluate effects on pain management routines in acute care, as experienced by patients, nurses, and physicians.
BARRIERS TO OPTIMAL PAIN MANAGEMENT Several barriers impede pain relief, including concerns of the patients themselves, their family members, health care professionals, and health care systems (Cohen et al., 2003; Drayer, Henderson, & Reidenberg, 1999; Ward et al., 1993). Patients have expressed a variety of reasons for not reporting pain. Patients with cancer-related pain do not want to distract the doctors from attending to their cancer; they want to be “good patients” and believe that their pain cannot be alleviated (Ward et al., 1993) or wait to be asked about their pain rather than request pain relieves themselves (Manias, Botti, & Bucknall, 2002). Fears of addiction among health care professionals also exist despite evidence that hospitalized patients have a low risk of becoming addicted because of medically administrated opioids (Drayer et al., 1999). A poor correlation between pain behavior and pain intensity can mislead staff who rely on pain behavior to assess pain intensity, and attitudes that patients exaggerate the intensity of pain still exist (Drayer et al., 1999). Nurses as well as physicians tend to overestimate low levels of pain and underestimate high levels of pain (Sjostrom, Haljamae, Dahlgren, & Lindstrom, 1997). Analgesics are given infrequently, in inadequate doses or lower doses than
prescribed (Dahlman, Dykes, & Elander, 1999; WattWatson, Stevens, Garfinkel, Streiner, & Gallop, 2001), and underuse of nonpharmacologic methods is described (Clarke et al., 1996). Difficulties in engaging and obtaining support from physicians and nurses in leading positions, who have the main responsibility for translating pain policy into practice in the ward, are also described as barriers (Dalton et al., 1999; Francke et al., 1997).
APPROACHES TO IMPROVING PAIN ASSESSMENT AND PAIN MANAGEMENT Different approaches to improve the quality in pain management have been done with varied results. The implementation of APS standard (APS, 1991), further developed to APS guidelines (1995), in the hospital improved nurses’ pain knowledge and attitudes scores, pain assessment, and documentation but had less effect on patient outcomes (Bookbinder et al., 1996; Comley & DeMeyer, 2001; Dahl et al., 2003). Quality assurance programs including a nurse-based anesthesiologist-supervised model increased postoperative pain relief (Bardiau, Taviaux, Albert, Boogaerts, & Stadler, 2003) and had effects over a long time on pain management routines in surgical wards (Stomberg, Wickstrom, Joelsson, Sjostrom, & Haljamae, 2003). Guidelines, in combination with education, decreased pain levels in patients with postoperative pain (Harmer & Davies, 1998; MacLellan, 2004) and hospitalized patients with varied pain conditions (de Rond et al., 2001), but results with no changes in patient outcomes are also described (Dahl et al., 2003). Incorporating pain assessment into observation charts (Harmer & Davies, 1998) and viewing pain assessment as the fifth vital sign increased assessment and documentations of pain overall (Cohen et al., 2003; de Rond, de Wit, & van Dam, 2001; de Rond, de Wit, van Dam, & Muller, 2000). A pain education program for health care professionals on its own showed no significant improvement in terms of getting nurses to ask patients about pain or to use rating scales (Francke et al., 1997), but may influence pain treatment (Dahlman et al., 1999) and improve documentation (Dalton et al., 1996). However, the design of the studies and the content of the programs differ, which make it difficult to evaluate the results of the studies. Few studies (Dalton et al., 1999) have involved physicians in educational programs, but several recommend it (Bookbinder et al., 1996; de Rond et al., 2000; Francke et al., 1997). A pain education program for patients and their family caregivers has
Effects of a Quality Improvement Program
demonstrated benefits for relief of cancer pain (Ferrell, Rhiner, & Ferrell, 1993). Information to the patient constitutes the basis for the potential to participate in care decisions (Sahlsten, Larsson, Lindencrona, & Plos, 2005), so there is a need for health care professionals to be up-to-date with the latest evidence-based care. One approach to support nurses is the potential of the Internet as a fast and efficient way of accessing evidence-based knowledge (Tod, Harrison, Morris Docker, Black, & Wolstenholme, 2003). Supporting health care professionals with search planning and developing their own collection of useful web-based resources is needed to improve nurses’ regular use of the Internet in searching for new knowledge in clinical settings (Morris-Docker, Tod, Harrison, Wolstenholme, & Black, 2004).
METHODS Design and Settings A cross-sectional intervention study was designed, with patients, nurses, and physicians assigned to a control group for 2 months in 2002 and an intervention group for 2 months in 2003. The study took place in five acute medical and surgical wards with 110 beds and one emergency department at a university hospital in southern Sweden. Two of the wards were for short-term acute medical and surgical care (1–2 days), and three wards were for longer acute hospital stay (9 –10 days). The study was planned and implemented in compliance with common ethical principles used in clinical research and was approved by the Ethics Committee of Lund University, Sweden. All patients and health care professionals gave informed consent, and their confidentiality was ensured. Routine Care at the Clinic There was no general policy regarding pain assessment, pain management, and follow-up of outcome variables of pain treatment. Prescription for pain relief medication was individual for patients with continuous and/or ondemand doses, and nonpharmacologic methods were available. General prescriptions for pain medication (as on-demand analgesic doses, e.g., acetaminophen for general pain treatment and strong opioids for postoperative pain treatment) were used on the wards but not in the emergency department. A nurse-based, anesthesiologistsupervised organization at the hospital was available for consultation. No regular pain education was offered to team members. However, training for health care professionals in specific illness-related areas such as education in heart failure, acute abdomen, and diabetes was given continuously.
95
The Quality Improvement Program The QI program included the development and implementation of a pain policy and a pain educational program for health care professionals, proceeding in the following steps: 1. Nurses and physicians who showed a special interest in developing pain management were selected from each of the six wards for a QI group in cooperation with the manager of the wards. 2. The QI group and the management group at the clinic developed a pain policy to guide future practice. The pain policy included regular pain assessment for all patients both at rest and with movement and maximum pain intensity levels less than 3 at rest within 1 hour after treatment for acute pain. It was decided that pain should be documented under a pain heading in each patient’s chart and continuous measurement with Visual Analogue Scale (VAS) or Numeric Rating Scale (NRS) should be documented together with assessment of other vital signs. Physicians were encouraged to prescribe scheduled analgesic plus on-demand analgesic to patients in pain. The QI group was responsible for implementing the pain policy on the wards and auditing the medical records four times per year to follow the progress and give information to colleagues. 3. Members of the QI group were offered an educational program based on a written pain education manual including 15 chapters developed by Swedish pain experts (Smärtgräns, 2000) and use of basic Internet search knowledge. The program was conducted for 8 days over a 6-month period and included case studies and discussions with staff from the nurse-based anesthesiologist-supervised pain service at the hospital and 2 days of training in Transcutaneous Electric Nerve Stimulation (TENS) including pain physiology underlying nonpharmacologic methods. Participants read the manual and formulated questions for discussion in the QI group to search for answers on the Internet. Guidelines on how to assess and record pain were developed by the QI group during the education. Nurses in the QI group were trained to supervise future education groups. 4. A website was developed to support the QI group with updated documents and useful web-based resources for evidence-based knowledge and was used during the educational program. 5. When the educational program ended, nurses from the QI group started five educational groups offered to all health care professionals. These programs were based on the same written pain education manual (Smärtgräns, 2000), except for the 2 days of TENS education. Each group met for 3 hours every 2 weeks for a total of nine times. 6. A 3-hour education session including information about the pain policy, a case discussion with the nurse
96
Hansson, Fridlund, & Hallström
Control group - January -February 2002 Patients n=400
Included patients: n=217 (54%) Median age (IQR): 74 (20) Female: 50%
Excluded patients: n=183 (46%) Median age (IQR): 77 (25) Female 45% –73 confused –17 could not communicate –15 too ill –42 hospitalized
