Do Barriers to Pediatric Pain Management as Perceived by Nurses Change over Time?

Do Barriers to Pediatric Pain Management as Perceived by Nurses Change over Time?

Original Article Do Barriers to Pediatric Pain Management as Perceived by Nurses Change over Time? --- - From the *Children’s Hospital of Wisconsin;...

271KB Sizes 0 Downloads 9 Views

Original Article Do Barriers to Pediatric Pain Management as Perceived by Nurses Change over Time? ---

-

From the *Children’s Hospital of Wisconsin; †Medical College of Wisconsin, Milwaukee, Wisconsin. Address correspondence to Michelle L. Czarnecki, MSN, RN-BC, CPNP, Jane B. Pettit Pain and Palliative Care Center, Children’s Hospital of Wisconsin, P.O. Box 1997, MS 792, Milwaukee, WI 53201. E-mail: [email protected] org Received March 3, 2012; Revised November 13, 2012; Accepted December 23, 2012. 1524-9042/$36.00 Ó 2014 by the American Society for Pain Management Nursing http://dx.doi.org/10.1016/ j.pmn.2012.12.003

Michelle L. Czarnecki, MSN, RN-BC, CPNP,* Katherine S. Salamon, MS,* Jamie J. Thompson, BSN, RN, CPN,* and Keri R. Hainsworth, PhD†

ABSTRACT:

For decades, nurses (RNs) have identified barriers to providing the optimal pain management that children deserve; yet no studies were found in the literature that assessed these barriers over time or across multiple pediatric hospitals. The purpose of this study was to reassess barriers that pediatric RNs perceive, and how they describe optimal pain management, 3 years after our initial assessment, collect quantitative data regarding barriers identified through comments during our initial assessment, and describe any changes over time. The Modified Barriers to Optimal Pain Management survey was used to measure barriers in both studies. RNs were invited via e-mail to complete an electronic survey. Descriptive and inferential statistics were used to compare results over time. Four hundred forty-two RNs responded, representing a 38% response rate. RNs continue to describe optimal pain management most often in terms of patient comfort and level of functioning. While small changes were seen for several of the barriers, the most significant barriers continued to involve delays in the availability of medications, insufficient physician medication orders, and insufficient orders and time allowed to pre-medicate patients before procedures. To our knowledge, this is the first study to reassess RNs’ perceptions of barriers to pediatric pain management over time. While little change was seen in RNs’ descriptions of optimal pain management or in RNs’ perceptions of barriers, no single item was rated as more than a moderate barrier to pain management. The implications of these findings are discussed in the context of improvement strategies. Ó 2014 by the American Society for Pain Management Nursing

INTRODUCTION Providing optimal pain management has been a health care goal for decades yet pediatric nurses (RNs) continue to report a myriad of barriers that interfere with their ability to accomplish that goal (Byrd, Gonzales, & Parsons, 2009; Czarnecki et al., 2011; Ely, 2001; Gimbler-Berglund, Ljusegren, & Enskar, 2008; Simons & Pain Management Nursing, Vol 15, No 1 (March), 2014: pp 292-305

Do Barriers to Pediatric Pain Management Change?

Roberson, 2002; Van Hulle Vincent, 2005). While an important step to improving pain management, simply identifying barriers is not enough to leverage improvement. Once barriers are identified, implementation of improvement strategies and reassessment of barriers is needed. Byrd, Gonzales, and Parsons (2009) called for studies describing improvement strategies along with a subsequent reassessment of barriers in multiple organizations as a beneficial ‘‘next step’’ for researchers. Despite this, no studies were found describing barriers to pediatric pain management identified by RNs in a single organization at more than one point in time, nor were any studies found in which barriers were assessed across multiple organizations. This study is a sequel to a study previously conducted in our organization and is part of a multi-site study; as Children’s Hospital of Wisconsin (CHW) was the only site with two assessment times, only results from CHW are reported here. Our first study (Czarnecki et al., 2011) was conducted by an internal interdisciplinary team from the Joint Clinical Practice Council (JCPC) with the primary goal of identifying barriers to pain management from a nursing perspective. Although pain management is an interdisciplinary process, the team decided to focus on nursing initially, with the goal to survey physicians (MDs) and parents in the future. Two hundred seventy-two RNs responded, representing a 28% response rate. The barriers identified as most significant (based on overall means of a 0–10 scale with 0 indicating ‘‘not a barrier’’ and 10 indicating ‘‘a major barrier’’), included insufficient/inadequate MD orders, insufficient time to pre-medicate patients before procedures, insufficient pre-medication orders available before procedures, a low priority given to pain management by medical staff, and parents’reluctance to have patients receive medication. Additional barriers identified through written comments included delays in medications being available from pharmacies, deficiencies with Acute Pain Service (APS) coverage, and concerns regarding MDs not ordering APS consults when RNs felt it would be helpful. No difference was found in how well respondents felt they could overcome barriers based on level of education or experience in that study, unlike the results of Van Hulle Vincent and Denyes (2004), who reported a positive correlation between level of experience and ability to overcome barriers. The barriers identified were not unique, but our first study (Czarnecki et al., 2011) provided baseline data and allowed the JCPC and the APS to identify potential improvement strategies, several of which were subsequently implemented (Table 1). Although no direct cause-and-effect relationships can be inferred, the goal of this study was to reassess barriers 3 years later to learn if the barriers had changed over

293

time. Identification of changes or stability in RNs’ perceptions of barriers has the ability to contribute to improvement of pain management. Review of Literature A review of studies focusing on barriers to pediatric pain management from the pediatric RNs’ perspective was included with our first study (Czarnecki et al., 2011). Barriers discussed in that review included RNs’ knowledge deficits; poor communication with parents (Simons & Roberson, 2002); a lack of consistency in practice; parental fears regarding opioids; time constraints (including procedural preparation); working with children in general (Ely, 2001); insufficient pain medication orders provided by MDs (Ely, 2001; Gimbler-Berglund, Ljusegren, & Enskar, 2008; Van Hulle Vincent, 2005); children’s reluctance to report pain and take pain medicine; parents’ reluctance to have children receive medications; RNs’ concerns about side effects of medications (Van Hulle Vincent, 2005); issues with procedural pain management (Ely, 2001); and a lack of time, knowledge, and cooperation with parents and MDs (Gimbler-Berglund, Ljusegren, & Enskar, 2008). Since that review, three additional studies were found regarding barriers to pediatric pain management from a nursing perspective. Barriers to pain management in the newborn intensive care (NICU) environment were investigated by mailing surveys to RNs with current membership in the National Association of Neonatal Nurses (Byrd, Gonzales, & Parsons, 2009). The survey asked RNs to indicate to what extent they agreed or disagreed with each question and provided space for comments. Ninety usable surveys were returned representing a 30% response rate. In brief, barriers included concerns regarding MDs’ pain management practices, RNs’ and MDs’ reluctance to change practice, concerns about infant pain assessment, and inadequate education regarding assessment and management. In addition, only 45% of RNs agreed with the assertion that newborn pain is well controlled within their work areas. As the goal of that study was to identify barriers, other than the need for education, specific improvement strategies were not included. Interviews and focus groups were used to discuss barriers and identify improvement strategies in two other studies (Dowden, McCarthy, & Chalkiadis, 2008; Ellis et al., 2007). Both studies asked RNs to identify barriers to pediatric pain management and to make recommendations for improvement in their settings; neither study included reassessment of barriers after implementation of improvement strategies. Dowden, McCarthy, and Chalkiadis interviewed 454 staff members (57% were RNs), in a 310-bed pediatric hospital

294

TABLE 1. Improvement Strategies Barrier Identified in Our First Study  Insufficient MD orders.  Insufficient orders prior to procedures.  Insufficient time to premedicate prior to procedures.

1. Addition of a ‘‘PRN’’ acetaminophen dose to admission ordersets to expedite its availability during the admission process. 1. Targeted education and feedback offered regarding procedural pain management. 2. Development of a new policy and procedure regarding patient preparation and assessment during procedures. 3. Investigation of opportunities to utilize the electronic health record to communicate specific patient information regarding procedures (what works well). 4. Development of a program to increase utilization of topical analgesics prior to IV starts and lab draws. 5. A recommendation was made to develop an orderset to direct procedures (prescribers would enter the procedure, time to be done, what was needed in preparation, supplies, premedication, etc.) 6. An online module discussing procedural pain management for surgery resident orientation. 1. Efforts to increase documentation of family education and level of understanding regarding pain management.

 Issues with APS coverage identified through comments.

1. APS to page RNs when starting rounds to expedite prioritizing patient care needs. 2. Increase APN availability/coverage for the APS

 Reluctance to get APS involved

1. A suggestion was made based on the first survey to have ‘‘automatic’’ or nurse driven consults. 1. Survey physicians to gain insight to perceived barriers.

 A low priority given to pain management by physicians

Implemented? 1. Implemented on a limited basis. Will be considered during upcoming changes to the electronic health record (EHR). 1. Yes 2. Yes 3. Investigated but not implemented due to future changes in the EHR. The goal is to implement this with the new EHR. 4. Yes. IV policy and procedure was changed to allow for vapo-coolant spray as a standard practice without a prescriber order. 5. An order set has been drafted but not yet been implemented. 6. Yes Since the current study:  Lab now offers vapocoolant spray and sucrose (For infants) in addition to distraction techniques as comfort measures.  Feedback is offered to HCPs when opportunities exist to improve procedural pain management. 2. Yes. The interdisciplinary teaching sheet was revised with more specific cues regarding what to teach patients and families related to pain and pain management. 1. Yes Since this study, an additional strategy has been implemented in which the APNs meet monthly with the residents who cover for the APS at night to discuss expectations, questions, etc. 2. Yes, a second APN joined the APS shortly after the data collection portion of this study. 1. Not implemented due to scope of practice concerns. 1. Not implemented yet due to methodological constraints.

Czarnecki et al.

 Parents’ reluctance to have child receive pain medication.

Proposed Improvement Strategies after Our First Study

Do Barriers to Pediatric Pain Management Change?

in Australia representing a 27% overall response rate. The barriers identified in this study were similar to those reported in previous studies and included concerns surrounding procedural pain management, the need for more consistent pain management practices, and the need for increased staff education regarding the use of nonpharmacologic techniques. Improvement strategies included the implementation of guidelines to reduce procedural pain, new child-friendly procedure rooms, the inclusion of pain assessment as the fifth vital sign, education regarding the use of topical anesthetics and non-pharmacologic techniques (including sucrose for infants), increased clinical resources, and increased use of pain assessment scales; no reassessment of barriers any time after implementation of these strategies was included. Ellis et al. (2007) used focus groups and questionnaires to identify supports and barriers to best practice, as well as factors that impacted the implementation of a comprehensive pain management program designed to improve nursing pain management. Communication was viewed as a potential strength, as well as a potential barrier, while inconsistencies in practice were seen only as a barrier. RNs also identified concerns such as assessment and management of infants being more difficult, and at times, pain management not being seen as a priority. Several improvement strategies were recommended including prescription of ‘‘around the clock analgesia’’ instead of ‘‘as needed analgesia,’’ development of better pain assessment documentation systems, development of pocket cards with analgesic dosages, inclusion of pain assessment scores as part of shift to shift hand-off, development of a hospital wide, systematic approach to procedural pain management, and development of an interdisciplinary pain service including an Advanced Practice Nurse (APN); no reassessment of barriers after implementation of these strategies was provided. Together, previous studies indicate that RNs in a variety of settings perceive similar barriers to pediatric pain management. However, few studies provided recommendations regarding improvement strategies and none provided a reassessment of barriers at a subsequent point in time. Purpose The primary purpose of the current study was to reassess barriers to optimal pain management perceived by RNs and describe any changes over time. It was outside the scope of this study to measure the direct impact of improvement strategies or the actual effect of perceived barriers on patient care. In addition, the investigators sought to collect quantitative data regarding barriers identified through comments provided in our

295

first study (Czarnecki et al., 2011). As improvement strategies (Table 1) were implemented at various points between studies, the investigators expected some barriers to improve.

METHODS Study Design The current study was time point number 2 in a crosssectional design to assess the perceived barriers to optimal pain management over time in a pediatric teaching hospital. The original Human Rights and Review Board approval allowed for resurveying RNs every 2–3 years, and an amendment allowed for the change from a hard copy paper survey to an electronic format distributed via e-mail. The goal of using the electronic format was to increase anonymity and improve the response rate as well as facilitate analysis. The current study involved replication of the survey used 3 years earlier (Modified Barriers to Optimal Pain Management, Czarnecki et al., 2011). Once the survey was converted to an electronic format and tested by the investigators, an electronic link was e-mailed to RNs in all patient care areas. The e-mail specifically asked that only RNs respond to the survey by clicking on the link to an online surveying company (surveymonkey.com). Once RNs opened the link, a cover letter included the following information: (1) why they were being invited to complete the survey; (2) the survey was part of a research study aimed at improving pain management; (3) completion was voluntary and implied informed consent; (4) respondents were free to skip over any question(s) they did not wish to answer; (5) no identifying information was being collected; and (6) directions for completion of the survey. Members of the leadership team in each area and the APN for the APS provided verbal reminders to RNs of their eligibility and the importance of their feedback to the goal of improving pain management. The last page of the survey thanked RNs for their participation and offered a coupon for popcorn and soda, redeemable by printing the last page (no survey questions, responses, or identifying information were included on that page) and returning it to any member of their leadership team including their unit based APN. The survey link was closed 4 weeks after initiation. As JCPC membership rotates every 2–3 years and several members had left the council, this study was conducted by the APS and used for comparison and development of future improvement strategies. Setting CHW is a 295-bed pediatric teaching hospital in southeastern Wisconsin, which at the time of this study had

296

Czarnecki et al.

a 236-bed capacity. Services offered include inpatient acute and critical care; medical, surgical and ambulatory services; a level 1 emergency room/trauma center; and a level 3C NICU. Patient care is directed by an attending level MD with a combination of APNs, fellows, residents, and/or senior medical students. An APS, consisting of an attending anesthesiologist, 2 full-time APNs, and at times, an anesthesia fellow, is available as a resource to RNs at their request (the second APN started with the APS following the data collection portion of the current study) and for patient care consultation and/or management at the primary team’s request. The APS is covered by an anesthesia resident with attending level anesthesia back up when the APNs are not available. At the time of this study, residents were responsible for coverage on evenings, nights, and weekends. Participants A total of 1,150 RNs were employed at the time of the study and were sent the electronic invitation to participate. RNs working in any nursing role (staff nurse, manager, APN, clinic nurse, etc.) in any patient care area were included. Nursing assistants, nurse interns, and nursing students were excluded. No personal identifying information was collected, and all data were kept secure through the online surveying company and the use of password protection once analysis began. Measures The survey (Modified Barriers to Optimal Pain Management) used in our first study (Czarnecki et al., 2011) was adapted with permission from Van Hulle Vincent (2005). The investigators chose that survey as a starting point because that study involved a similar setting in the Midwest and identified barriers previously found in the literature and believed to be common in pediatric organizations. Van Hulle Vincent’s tool asked RNs to rate 13 potential barriers, such as inadequate or unavailable MD orders, a low priority given to pain by various disciplines, limitations in the RNs knowledge, or ability to assess pain and potential barriers posed by patients and parents. The investigators added supplementary questions based on the results of previous studies to gain insight into additional factors potentially impacting pain management practices in our organization, culminating in a 35-question survey for our first study. For the current study, RNs were asked to complete a 39-question survey (available upon request). Modifications made for the current study were based on written comments reported in our first study (Czarnecki et al., 2011). Specifically, two questions were added

asking respondents to rate processes that may have impacted the availability of medications from pharmacy. A question was added asking RNs to identify when issues with APS coverage most often occurred (multiple choices were allowed as well as ‘‘N/A, I have not experienced issues with coverage’’). Lastly, a question was added asking RNs to rate the degree to which reluctance by the MDs to get the APS involved was a barrier to care. The original survey created by Van Hulle Vincent (2005) reached an internal consistency of 0.86; once modified for our first study, the tool reached an internal consistency of 0.85, and after the addition of 4 questions for the current study, the measure reached an internal consistency of 0.87. As with our first study (Czarnecki et al., 2011), in addition to rating each of 21 potential barriers from 0 (‘‘not a barrier’’) to 10 (‘‘a major barrier’’), RNs were asked to describe what optimal pain management would look like, rate the impact of the APS on patient care from 0 (‘‘negative impact’’) to 10 (‘‘positive impact’’) or N/A if the respondent works in an area that does not routinely utilize the APS, the quality of pain management in our organization from 0 (‘‘the worst possible’’) to 10 (‘‘the best possible’’), their personal pain management practices from 0 (‘‘very conservative’’) to 10 (‘‘very liberal’’), how many times per week they encounter children in pain from 0 (‘‘none of my patients’’) to 10 (‘‘all of my patients’’), how well they are able to overcome barriers from 0 (‘‘not at all able’’) to 10 (‘‘very able’’), and how much of an impact improving barriers would have on their ability to provide quality pain management from 0 (‘‘none at all’’) to 10 (‘‘a great impact’’). RNs were asked to list any barriers not specifically identified and were allowed space for comments (both open-ended questions). They were also asked how they learned about pain management and what sources they used in decision making, information that could be useful in designing future improvement strategies. Lastly, RNs were asked to provide demographic information (i.e., role, gender, race, unit, highest educational level achieved, and categorical choices for years of pediatric nursing experience) to aid in describing the respondents and identifying differences within and between patient care areas. Data Analysis Descriptive statistics, including frequencies, frequency distributions, means, and Pearson correlations, were conducted. ANOVA analyses were conducted to determine if there were differences among demographic variables and the variables under study. One-sample t tests were used to compare the mean for each barrier collected in our first study (Czarnecki et al., 2011) with the mean for each barrier in the current study to

297

Do Barriers to Pediatric Pain Management Change?

determine if there were any changes. Descriptions of optimal pain management, as well as additional barriers and comments identified in the narrative sections of the survey were grouped into major concepts by the research team, and then checked by two independent coders to ensure accuracy. The concepts reported in our first study were used as a starting point and expanded upon as needed to include all responses from the current study. As a final step, similar concepts were combined, resulting in 13 descriptive groups. For example, patient comfortable, pain score < comfort goal, and pain score < 4/10 were combined into the overall concept of ‘‘patient comfortable.’’

RESULTS Demographics Table 2 shows the demographic information for the 442 respondents, representing a 38% response rate for the current study, as well as the demographic

information for respondents of our first study (Czarnecki et al., 2011). For the current study, the majority of surveys returned were from Caucasian (96.8%) females (97.8%), working on inpatient care units (76.6%) in staff nursing roles (88.4%). More than half of the respondents had less than 10 years of experience; 32.8% had more than 15 years of experience; and 75.7% had baccalaureate degrees. RNs working in a wide variety of patient care areas throughout the hospital were represented (acute and intensive care units, Clinical Resource Unit (float pool), ambulatory clinics, Radiology, Transport, Hematology/Oncology/ Transplant, Day Surgery/Post Anesthesia Recovery Unit/Operating Room, and ‘‘other’’). Ten percent of respondents did not identify their primary area of work. Data obtained from a self-report data base showed that at the time of the current study, RNs were employed an average of 10.2 years compared to an average of 8.8 years during the time of our first study (results from our first study not previously reported).

TABLE 2. Demographic Information of Respondents Current Study (Overall N ¼ 442) Variables Gendera Female Male Raceb Caucasian African American Hispanic E. Indian Other Educationc Associate Degree Diploma Baccalaureate Master of Science Rolec Staff RN Supervisor Manager APN Other Experience (yrs)d <2 2–<5 5–<10 10–<15 15–<20 >20

Original Study (Overall N ¼ 272) n (%)

400 (97.8) 9 (2.2) 392 (96.8) 3 (0.7) 3 (0.7) 1 (0.2) 6 (1.5) 49 (11.8) 21 (5.0) 315 (75.7) 31 (7.5) 366 (88.4) 13 (3.1) 4 (1.0) 18 (4.3) 13 (3.1) 66 (15.9) 84 (20.2) 86 (20.7) 43 (10.4) 49 (11.8) 87 (21.0)

Current: an ¼ 409, bn ¼ 405, cn ¼ 414, dn ¼ 415. Original: an ¼ 232, bn ¼ 253, cn ¼ 265.

Variables Gendera Female Male Raceb Caucasian African American Hispanic E. Indian Asian Other Educationc Associate Degree Diploma Baccalaureate Master of Science Rolec Staff RN Supervisor APN Other Experience (yrs)c <2 2–<5 5–<10 10–<15 15–<20 >20

n (%) 226 (83.1) 6 (2.6) 249 (91.5) 1 (0.4) 1 (0.4) 1 (0.4) 1 (0.4) 5 (1.8) 28 (10.3) 16 (5.9) 205 (75.4) 16 (5.9) 229 (84.2) 15 (5.5) 8 (2.9) 13 (4.9) 54 (19.9) 53 (19.5) 53 (19.5) 27 (9.9) 24 (8.8) 54 (19.9)

298

Czarnecki et al.

Additionally, 75.8% of the RNs at the time of the current study had baccalaureate degrees compared to 74.4% at the time of our first study. Taken together, these data suggest that the samples in both studies are representative of the RNs employed at CHW during the time of each study. c2 analyses were used to determine if there were differences between the demographic characteristics of the participants in our first study (Czarnecki et al., 2011) as compared with the current study. Analyses revealed no significant differences between the two samples with regard to years of experience (c2 ¼ 3.05, p > .05), highest level of education achieved (c2 ¼ 1.03, p > .05), role of the RN (c2 ¼ 7.16, p > .05), ethnicity (c2 ¼ 2.52, p > .05), and gender (c2 ¼ 0.001, p > .05). There were significant differences between the two samples in terms of department/unit of work (c2 ¼ 220.58, p < .001). This difference is primarily due to the larger sample size in the current study, resulting in several units being represented in this study (EDTC, Radiology, Transport, Day Surgery, Medical Neuroscience, CRU), but not specifically identified in the previous study. The majority of respondents (58.5%) reported having fairly liberal (7–10/10) pain management practices, with 3.2% reporting more conservative practices (0–3/10). Of note, the tool did not provide any descriptive information for this question; respondents used their own opinions to interpret ‘‘conservative or liberal.’’ The mean for all respondents was 6.81 (1.8). The majority of respondents (77.8%) reported working with patients in pain frequently (M ¼ 6.08  2.73 for all respondents), and current pain management practices were rated positively (6–10/10) by 81.7% of respondents (M ¼ 6.93  1.51 for all respondents). Respondents (87%) reported ‘‘assessment of the infant or child’’ as the primary source used in making decisions about pain management followed by ‘‘child or family request’’ (56.3%) and ‘‘physician order’’ (29.1%). When asked how they learned about pain management, ‘‘nursing school’’ was the answer selected most often (32.7%) followed by ‘‘CHW orientation’’ (22.6%), ‘‘other RNs on the unit’’ (19.8%), and ‘‘the pain service APN’’ (19.8%). Optimal Pain Management Respondents were asked to describe optimal pain management using as many descriptors as desired. One hundred ninety-five (44.1%) respondents provided 1–7 descriptions each. The most common description, provided by 146 (75%) respondents, involved patients’ level of comfort (i.e., pain score < 4/10 or # patient’s comfort goal, patient comfortable, no complaints of pain, no objective signs of pain such as grimacing or

TABLE 3. Current Study: Mean and SD of Barriers to Pain Management Barriers 1. Delays in orders being processed by the pharmacy* 2. Delays in orders being delivered by the pharmacy* 3. Inadequate or insufficient MD medication orders 4. Insufficient time allowed to premedicate prior to procedures 5. Insufficient pre-medication orders prior to procedures 6. Reluctance of MD to get pain service involved* 7. Low priority given to PM by medical staff 8. Parents’ reluctance to have children receive medication 9. My concern about side effects of medications 10. Patients’ reluctance to report/rate pain 11. Patients’ reluctance to take pain medications 12. Current documentation format 13. Competing demands on my time 14. Insufficient resources to provide guidance 15. My concern about children becoming tolerant to analgesics 16. Limitations in my knowledge of PM 17. Low priority given to PM by nursing staff 18. Low priority given to PM by nursing management 19. Limitations in my ability to assess pain 20. My concern about children becoming addicted 21. Low priority given to PM by me

M (SD) 4.91 (3.00) 4.90 (2.96) 4.64 (2.73) 4.34 (2.80) 4.34 (2.79) 4.28 (3.13) 3.91 (2.78) 3.60 (2.32) 3.26 (2.30) 3.24 (2.52) 2.87 (2.50) 2.80 (2.57) 2.76 (2.69) 2.59 (2.29) 2.40 (2.32) 1.98 (1.90) 1.84 (1.92) 1.70 (2.07) 1.39 (1.61) 1.20 (1.64) 0.67 (1.14)

PM ¼ pain management. Bolding indicates the barriers with the 5 highest and 5 lowest means. Barriers are sorted from highest to lowest means, not in questionnaire sequence. Means are based on a 0–10 scale with 0 indicating ‘‘not a barrier’’ and 10 indicating ‘‘a major barrier’’. *Indicates potential barriers added to this study based on comments from respondents of the original study.

crying). Of those comments, 23% described pain specifically in terms of the patients’ pain assessment score or ‘‘comfort goal’’ (a pain number goal identified by the point at which the patient feels comfortable with activity, sleeping, ambulating, etc.). Patients’ level of functioning (i.e., ability to do activities of daily living such as repositioning, ambulating, playing, sleeping, and physical therapy) was used to describe optimal pain management by 133 (68%) respondents. Seventy

299

Do Barriers to Pediatric Pain Management Change?

TABLE 4. One-Sample t tests Comparing Pain Barriers Across Studies Current Study Original Study

Inadequate or insufficient MD medication orders Insufficient time allowed to pre-medicate prior to procedures Insufficient pre-medication orders prior to procedures Low priority given to PM by medical staff Parents’ reluctance to have children receive medication My concern about side effects of medications Patients’ reluctance to report/rate pain Patients’ reluctance to take pain medications Current documentation format Competing demands on my time Insufficient resources to provide guidance My concern about children becoming tolerant to analgesics Limitations in my knowledge of PM Low priority given to PM by nursing staff Low priority given to PM by nursing management Limitations in my ability to assess pain My concern about children becoming addicted Low priority given to PM by me

M (SD)

M (SD)

Mean Difference

4.64 (2.73) 4.34 (2.80)

4.98 (2.67) 4.57 (2.62)

0.34 0.23

2.55 (420) .011 1.66 (419) NS

4.34 (2.79) 3.91 (2.78) 3.60 (2.32) 3.26 (2.30) 3.24 (2.52) 2.87 (2.50) 2.80 (2.57) 2.76 (2.69) 2.59 (2.29) 2.40 (2.32)

4.92 (2.81) 4.17 (2.59) 3.49 (2.36) 2.90 (2.06) 3.00 (2.56) 2.72 (2.52) 2.98 (2.63) 3.19 (2.61) 2.75 (2.35) 2.12 (1.98)

0.58 0.26 0.11 0.36 0.24 0.15 0.18 0.43 0.16 0.28

4.21 (413) 1.91 (423) 0.93 (419) 3.33 (441) 1.91 (415) 1.25 (416) 1.41 (414) 3.35 (438) 1.42 (414) 2.52 (439)

1.98 (1.90) 1.84 (1.92) 1.70 (2.07) 1.39 (1.61) 1.20 (1.64) 0.67 (1.14)

1.83 (1.79) 1.95 (1.80) 1.55 (1.82) 1.34 (1.51) 1.13 (1.75) 0.53 (0.96)

0.15 0.11 0.15 0.05 0.07 0.14

1.62 (426) NS 1.23 (425) NS 1.48 (423) NS 0.59 (424) NS 0.83 (440) NS 2.46 (426) .014

t(df)

p

.000 NS NS .001 NS NS NS .001 NS .012

PM ¼ pain management; NS ¼ not significant. Barriers are sorted from highest to lowest means for the current study, not in questionnaire sequence. Means are based on a 0–10 scale with 0 indicating ‘‘not a barrier’’ and 10 indicating ‘‘a major barrier’’.

(36%) respondents described comfort in terms of access to safe, effective, timely pain relief, and 13 (6%) identified physiological data (i.e., vital signs) as a measure of optimal pain management, with most of those responses coming from RNs working with intubated or critically ill children (i.e., intensive care and perioperative settings). Other miscellaneous descriptions included comments regarding parent satisfaction, side effects being under control, education, documentation, a multidisciplinary approach, and appropriate procedural pain management. Barriers to Optimal Pain Management Respondents were asked to rate each of 21 potential barriers to optimal pain management from 0–10 (Table 3). Of the 18 barriers rated in both studies, the most significant barriers (as indicated by the highest means) in the current study included: inadequate/ insufficient MD medication orders (M ¼ 4.64  2.73), insufficient time allowed to pre-medicate before procedures (M ¼ 4.34  2.80), insufficient premedication orders available before procedures (M ¼ 4.34  2.79), a low priority given to pain management by medical staff (M ¼ 3.91  2.78), and parents’ reluctance to have children receive medication (M ¼ 3.60  2.32). Respondents in the current study gave

somewhat similar ratings to the questions regarding inadequate/insufficient MD orders and a low priority given to pain management by the medical staff (r ¼ .62, p < .001). In addition, respondents gave marginally similar ratings to the questions regarding time (competing demands on time and insufficient time to pre-medicate before procedures) (r ¼ .27, p < .001). Lastly, respondents rated the questions regarding a low priority given to pain management by medical staff and insufficient orders before procedures somewhat similarly (r ¼ .45, p < .001). Results of the three barriers added to the survey used in the current study included delays in orders being processed by pharmacy (M ¼ 4.91  3.00), delays in medications being delivered by pharmacy (M ¼ 4.90  2.96), and the reluctance of MDs to get the APS involved (M ¼ 4.28  3.13). Respondents were asked to identify any additional barriers not captured in the quantitative section; 123 responses were provided. No single concept was reported by more than 4% of the entire sample; comments included concerns regarding medication management (e.g., insufficient orders being available, opioids being weaned too quickly, other shifts not administering analgesics, outpatient management of medication prescriptions), assessment (e.g., how to

300

Czarnecki et al.

TABLE 5. Mean, SD, and Percentage of ‘‘Extreme’’ Responses Reported by Nurses in the Current Study Barriers

M (SD)

Delays in orders being processed by the pharmacy Delays in orders being delivered by the pharmacy Inadequate or insufficient MD medication orders Insufficient time allowed to pre-medicate prior to procedures Insufficient pre-medication orders prior to procedures Reluctance of MD to get pain service involved Low priority given to PM by medical staff Parents’ reluctance to have children receive medication My concern about side effects of medications Patients’ reluctance to report/rate pain Patients’ reluctance to take pain medications Current documentation format Competing demands on my time Insufficient resources to provide guidance My concern about children becoming tolerant to analgesics Limitations in my knowledge of PM Low priority given to PM by nursing staff Low priority given to PM by nursing management Limitations in my ability to assess pain My concern about children becoming addicted Low priority given to PM by me

4.91 (3.00) 4.90 (2.96) 4.64 (2.73) 4.34 (2.80) 4.34 (2.79) 4.28 (3.13) 3.91 (2.78) 3.60 (2.32) 3.26 (2.30) 3.24 (2.52) 2.87 (2.50) 2.80 (2.57) 2.76 (2.69) 2.59 (2.29) 2.40 (2.32) 1.98 (1.90) 1.84 (1.92) 1.70 (2.07) 1.39 (1.61) 1.20 (1.64) 0.67 (1.14)

% of Nurses Rating % of Nurses Rating as ‘‘0–2/10’’ as ‘‘8–10/10’’ 24.9% 25.6% 24.0% 30.9% 30.0% 32.4% 36.5% 35.5% 40.2% 40.8% 47.7% 53.0% 54.2% 55.4% 58.9% 66.2% 70.2% 72.4% 79.0% 81.6% 91.3%

22.7% 22.1% 15.7% 15.1% 16.2% 18.8% 12.0% 6.3% 4.5% 5.5% 5.0% 6.0% 6.7% 2.6% 3.2% 0.6% 1.7% 1.9% 0.4% 0.4% 0.2%

PM ¼ pain management. Bolding indicates the barriers with the 5 highest and 5 lowest means. Responses based on a 0–10 scale with 0 indicating ‘‘not a barrier’’ and 10 indicating ‘‘a major barrier’’.

handle patients perceived as overrating their pain, concerns about assessment tools and infant assessment), the bar coding system, communication, difficulties getting the APS involved (e.g., reluctance by some services to get APS involved, RNs wanting to be able to consult the APS without MDs being upset), and the APS needing better coverage on weekends and off shifts. Comparison of Barriers Differences between the barriers reported in the current study and our first study (Czarnecki et al., 2011) are shown in Table 4. A significant decrease was found between the means reported in the current study compared to those reported in our first study for the following barriers: inadequate or insufficient MD medication orders, insufficient pre-medication orders before procedures, and competing demands on the respondent’s time. A significant increase was found in the following barriers: the respondent’s concern about side effects of medications, the respondent’s concern about children becoming tolerant to analgesics, and a low priority given to pain management by the respondent. Because even the most significant responses in both studies resulted in means <5/10, the investigators

looked at the frequency of the more ‘‘extreme’’ responses (0–2/10 and 8–10/10) to glean more insight from these findings (Table 5, results from our first study, Czarnecki et al., 2011, not previously reported). For those potential barriers perceived to be less significant, up to 91% of respondents rated them ‘‘extremely low’’ (0–2/10). Conversely, for those barriers perceived to be more significant, less than 23% rated them ‘‘extremely high’’ (8–10/10). The Acute Pain Service The APS was rated positively (6–10/10) by the majority (238/322; 74%) of respondents and the mean for all respondents was 7.86 (2.24). Because not all areas use the APS (i.e., ambulatory clinics, diagnostic imaging areas, EDTC, perioperative areas), as instructed on the survey, when asked to rate the impact of the APS on patient care, a subgroup of respondents (n ¼ 96) chose ‘‘N/A,’’ indicating they worked in an area that does not routinely use the APS (the investigators verified that those responses came from RNs working in areas that do not routinely use the APS). During our first study (Czarnecki et al., 2011), deficiencies in APS coverage were identified (e.g., ‘‘difficulty reaching the person on call,’’ ‘‘residents

Do Barriers to Pediatric Pain Management Change?

preferring to wait until morning to make changes’’). During this study, respondents were asked to identify more precisely when issues were occurring (i.e., weekday days, weekday nights, weekends, etc. or ‘‘N/A–I have not had issues with coverage’’) and were allowed to choose as many responses as applicable. Results were then categorized to allow multiple answers to be combined (i.e., weekday days and weekday nights; weekday nights and weekends, etc.). The investigators found that several respondents who had either not answered or answered ‘‘N/A’’ for the question regarding the impact of the APS on patient care responded to this question regarding when issues with APS coverage most often occur. Because RNs working in areas that do not utilize the APS should not have issues with coverage, those responses were removed, resulting in 317 responses. To have included all responses would have resulted in a higher frequency of ‘‘N/A–I have not had issues with coverage’’ which would have portrayed a more positive, but less accurate representation of coverage issues. Of the 317 responses analyzed, 157 (49.5%) reported having no issues with APS coverage; 60 (18.9%) identified weekday nights and weekends; 58 (18.3%) identified weekends; 30 (9.5%) identified weekday nights; 9 (2.8%) identified weekday days; and 3 (0.9%) identified ‘‘all times’’. Despite the barriers identified throughout the study, 327 (78%) respondents reported being able to overcome identified barriers and ultimately provide quality pain management for their patients (M ¼ 7.1  2.21 for all respondents). Using one-way ANOVAs, no relationship was found between years of experience (F(5,409) ¼ .57, p > .05) or educational level (F(3,412) ¼ .06, p > .005) and the reported ability to overcome barriers. Respondents reported being optimistic that improving identified barriers would positively impact their ability to provide quality pain management (M ¼7.53  2.60 for all respondents).

DISCUSSION CHW resurveyed RNs regarding perceived barriers to pediatric pain management to learn what, if anything, had changed over the 3-year period between surveys. To the best of our knowledge, this study represents the first assessment of barriers to pediatric pain management in the same organization, using the same tool, at more than one point in time. Because a myriad of factors can influence how one responds to survey questions at any given point in time, the investigators felt it important to resurvey RNs to see if the barriers identified as most significant in our first study (Czarnecki et al., 2011) reemerged, or if others surfaced as most significant.

301

This study had a 38% response rate, an increase from 28% in our first study (Czarnecki et al., 2011). Possible explanations for this include the possibility that more RNs received the electronic invitation in the current study than received the verbal and written invitation in our first study; RNs may have preferred the electronic survey over the paper survey used in our first study; or perhaps, RNs learned of the incentive offered for participation and chose to participate for that reason. The use of electronic surveys and reminders for other types of studies has yielded mixed results, and the effectiveness of electronic surveys remains inconclusive (McCluskey & Topping, 2011). The 38% response rate is comparable to other studies investigating pediatric pain management (Byrd, Gonzales, & Parsons, 2009; Dowden, McCarthy, & Chalkiadis, 2008; Ellis et al., 2007), and the current sample represents RNs working in a wide variety of settings throughout the hospital. Results suggest that respondents were optimistic that improving barriers would positively impact their ability to provide optimal pain management, results similar to other studies evaluating pediatric pain management (Czarnecki et al., 2011; Ellis et al., 2007; Ely, 2001). Demographics While the majority of respondents had less than 10 years of experience, 33% had greater than 15 years of experience. No noteworthy differences were found between respondents in the current study compared with those in our first study (Czarnecki et al., 2011); both groups were comprised primarily of female RNs with baccalaureate degrees functioning in a staff nursing role. No difference was found in how well respondents felt they could overcome barriers based on level of education or experience in either of our studies. These results differ from Van Hulle Vincent and Denyes (2004), who reported a positive correlation between experience level and the RNs’ ability to overcome barriers. One might expect better problem-solving skills by RNs with more education and/or more experience. However, in both of our studies, the means of even the most significant barriers were <5/10, and current pain management practices were rated positively by most respondents. Cumulatively, the consistency of the results found in both of our studies might suggest that, overall, RNs do not perceive many insurmountable barriers to meeting their patients’ pain management goals. As it was outside the scope of this study to observe or measure actual patient care, it is unknown how these perceived barriers actually impact patient care. RNs were asked what sources they used in making decisions about pain management. Assessment of the

302

Czarnecki et al.

child and child/family request were the two most frequent responses in both of our studies. The third most frequently reported source varied between studies with the current study showing MD orders compared to reports from the previous RN in our first study (Czarnecki et al., 2011). These results are similar to other studies reporting patient assessment (Czarnecki et al., 2011; Gimbler-Berglund, Ljusegren, & Enskar, 2008; Van Hulle Vincent, 2007), parental request (Czarnecki et al., 2011) or input (GimblerBerglund, Ljusegren, & Enskar, 2008; Van Hulle Vincent, 2007), information from other RNs about the patient (Gimbler-Berglund, Ljusegren, & Enskar, 2008; Twycross, 2008), and MDs input/orders (GimblerBerglund, Ljusegren, & Enskar, 2008) as important components in guiding pediatric pain management. RNs also were asked how they personally learned about pain management. Results consistent across both of our surveys include other RNs on the unit and the APN for the APS. Together, these results highlight the importance of education and role modeling on the development of pediatric pain management practices, results similar to other studies (GimblerBerglund, Ljusegren, & Enskar, 2008; Twycross, 2008). Optimal Pain Management Understanding how RNs define optimal pain management may shed light on perceived barriers. Descriptions of optimal pain management did not change over time, although there was a substantial decrease in the number of respondents providing descriptions (44% of respondents in the current study compared to 84% in our first study). One possible explanation may be that this question was moved from the beginning of the survey during our first study (paper format) to the end of survey in the current study (electronic format). Because the electronic survey was most likely accessed at work, RNs may have gotten called away and not been able to finish. Alternatively, they may have simply tired of answering questions. The two most common descriptors for describing optimal pain management were the patients’ level of comfort and patients’ level of functioning, the same two concepts most often reported in our first study (Czarnecki et al., 2011). A small increase was seen in the number of respondents using pain assessment scores or comfort goals as descriptors. No specific strategies were implemented to address this between surveys, and although the increase was small, it is encouraging. As vital signs are ‘‘neither sensitive nor specific as indicators of pain’’ according to the Agency for Health Care Policy and Research (1992), it was also encouraging that fewer respondents used physiological data as a measurement of pain. This decrease may be

due to a slight decrease in the number of respondents working in the ICU areas where this indicator may be appropriate, or perhaps an improvement in respondents’ understanding of the role vital signs play in pain assessment. Overall, it does not appear that RNs’ descriptions of optimal pain management changed over time, which suggests some consistency in the goals that RNs are trying to achieve when providing pain management to children. Comparison of Barriers The consistency seen in barriers rated as most significant in both studies indicates stability in the issues RNs perceive as interfering with their ability to provide optimal pain management. In addition to being consistent across our two studies, inadequate or insufficient MD medication orders (Ely, 2001; Ellis et al., 2007; Schafheutle, Cantrill, & Noyce, 2001; Twycross, 2002; Van Hulle Vincent, 2005; Van Hulle Vincent & Denyes, 2004), procedural pain management (Dowden, McCarthy, & Chalkiadis, 2008; Ely, 2001), a low priority given to pain management by medical staff (Van Hulle Vincent, 2005), and parents’ reluctance to have children receive medication (Ely, 2001; Simons & Roberson, 2002; Van Hulle Vincent, 2005) have been reported elsewhere. Cumulatively, results of these studies may provide areas of potential impact when designing future improvement strategies. There was a significant decrease in the means for two of the five highest barriers (inadequate or insufficient MD medication orders and insufficient premedication orders before procedures). Although these decreases were statistically significant, the investigators do not feel they are clinically significant on their own. However, these improvements may be related to improvements in RNs’ perception of ‘ a low priority given to pain management by the medical staff’’ as the mean for that barrier also decreased. The third barrier to show an improvement regarded competing demands on the respondents’ time. Again, on its own, this decrease was not considered clinically significant but may be related to the improvement reported in some of the issues causing a delay in pain management (i.e., inadequate or insufficient orders, insufficient time to premedicate before procedures). The similarities in ratings for the questions regarding a low priority given to pain management by medical staff, inadequate/insufficient MD orders and insufficient orders before procedures may suggest that RNs associate the availability of orders with a certain level of priority given to pain management by MDs. Improvement Strategies Table 1 identifies proposed improvement strategies, many of which were implemented between our two

Do Barriers to Pediatric Pain Management Change?

studies, as well as additional strategies implemented since the time of the current study. Although the concern raised regarding inadequate/insufficient MD orders was not limited to the APS, the APS has taken steps to improve this barrier. For example, the APS routinely enters orders for the progression from patient controlled analgesia (PCA) to oral analgesics that can be initiated by the RN when the patient tolerates clear liquids. An additional order is provided that directs the RN to discontinue the PCA when the patient tolerates oral analgesics and is comfortable during required activity. Most recently, a conditional order was added to the APS order sets for PCAs and epidurals that allows RNs to activate a naloxone drip to treat pruritus. Such provisional orders decrease the time required for the RNs to seek out such orders and may impact both RNs’ perceptions of inadequate orders and perhaps the improvement seen in RNs’ perception of time constraints interfering with pain management. Lastly, the members of the APS continue to make a concerted effort to increase communication with both the RNs caring for patients and members of the primary team to address comfort concerns from multiple perspectives in a timely manner. One of the strategies implemented between the two studies that may have had an impact on the RNs’ perceptions of the barriers related to procedural pain management was implementation of a house-wide policy regarding procedural pain management. This patient care policy outlines the expectations for pain assessment and management associated with procedures. Education regarding the expectations set forth in that policy was provided to RNs and prescribers through on-line modules regarding procedural sedation. As procedural pain management was still a significant concern based on this study, additional improvement strategies have been implemented. For example, policies were revised and education provided to allow lab technicians to use vapocoolant spray or sucrose to decrease the pain associated with lab draws. Internal quality improvement data show that over half the patients at the main campus and nearly all patients at a satellite location are being offered some type of comfort measure before lab draws, and efforts are underway to increase the consistency of use even further. We have also implemented the J tip (National Medical Products, Inc., Irvine, CA) as a method of administering buffered lidocaine before IV placement as an alternative to topical lidocaine or vapocoolant spray. Together these strategies should, hopefully, improve procedural pain management for our patients and decrease the barriers perceived by our RNs. The means for only 3 barriers (concerns about side effects, concerns about tolerance, and a low

303

priority given to pain management by the respondent) increased. As with those barriers showing improvement, although these changes were statistically significant, the investigators did not believe the changes were clinically significant. As these barriers were not reported as significant barriers in our first study (Czarnecki et al., 2011), no improvement strategies were implemented to specifically address them; it is not clear what impacted these results. Interestingly, the questions added to the current study regarding the processing of orders and the delivery of medications by pharmacy resulted in the highest means in the current study and means higher than any of the barriers in our first study (Czarnecki et al., 2011). At the time of this study, once an order was entered into the computerized physician order entry (CPOE) system, pharmacy had to process orders into another system, and then provide the medication, two potential opportunities for delay. Since the time of this study, a pharmacy interface has been implemented that significantly reduces the time required for processing/delivering medication orders. Anecdotally, internal quality improvement monitoring indicates a significant decrease in the frequency of events reported regarding delays in pain medication availability. Additional barriers identified through narrative comments focused on medication management, specifically surrounding inappropriate weaning strategies and insufficient orders, results similar to our first study (Czarnecki et al., 2011). The repetition of these comments may help focus future improvement strategies in our organization and support the concerns raised by RNs caring for neonates in the NICU setting (Byrd, Gonzales, & Parsons, 2009). In addition, quantitative data regarding this barrier could be collected in future surveys regarding barriers to pediatric pain management. No one concept was reported by more than 4% of respondents, and none of the concepts were particularly new or surprising, perhaps suggesting that the tool had captured most potential barriers. Barriers identified to be least significant were also similar to the results of our first study (Czarnecki et al., 2011); all but one were the same. The overall sense of pain management as being important to RNs, as well as to nursing leadership, was also reported by Van Hulle Vincent (2005) and is echoed by RNs discussing pediatric pain management in the NICU setting (Byrd, Gonzales, & Parsons, 2009) and RNs discussing pediatric post-operative pain management practices (Ellis et al., 2007). As organizational culture may be important to improving practice (Lewthwaite et al., 2011; Twycross, 2008), it is reassuring that RNs continue to report a perception that pain management is important to nursing staff and leadership.

304

Czarnecki et al.

Of the barriers rated as least significant in our first study (Czarnecki et al., 2011), only one (a low priority given to pain management by the respondent) showed a significant worsening in the current study. Although this difference was statistically significant, the investigators do not consider the change to be clinically significant as this barrier continues to be the lowest rated barrier of all. The results of this question must be interpreted with caution as the likelihood of an RN indicating that pain management is not a priority may be quite low when the study was conducted by a team involving the APS. In addition, the priority that RNs assign to pain management has not been found to influence a number of practice elements (i.e., assessment, use of non-pharmacologic techniques) (Twycross, 2008). The uncertainty surrounding the significance of this particular question may suggest removal from future surveys. Results indicating that RNs perceived pain management as important to other RNs but less important to the medical staff is consistent with results of other studies regarding pediatric pain management (Byrd, Gonzales, & Parsons, 2009; Czarnecki et al., 2011; Dowden, McCarthy, & Chalkiadis, 2008; Van Hulle Vincent, 2005). Of note, although not statistically significant, ‘‘a low priority given to pain management by medical staff’’ was one of the barriers to show improvement in the current study compared to our first study (Czarnecki et al., 2011, p.158) and may be related to the improvement seen in the barrier regarding insufficient MD orders as these two questions were rated similarly by respondents. Impact of APS At the time of this study, an average of over 1,100 referrals were made to the APS annually, resulting in an average of over 5,100 days of service per year. The APS was rated as having a positive effect on patient care in this study, which is consistent with results of several other studies (Czarnecki et al., 2011; Kitowski & McNeil, 2002; Mackintosh & Bowles, 1999; Mackintosh & Bowles, 2003; Miaskowski, Crews, Ready, Paul, & Ginsberg, 1999; Shapiro, Zohar, Kantor, & Fredman, 2004; Stadler, Schlander, Braeckman, Nguyen, & Boogaerts, 2004). Based on the results of our first study, a question was added that identified deficits in coverage most often occurred during the times the APNs were not available. Because the second full-time APN started with the APS soon after the data collection portion of this study, the impact of that strategy has not yet been measured. Anecdotal comments from staff have been positive, and since the current study, APN coverage has been expanded further to include some weekends and evenings. In

addition, each month one of the APNs for the APS meets with the anesthesia residents responsible for overnight coverage. During these sessions, expectations are reviewed, case examples shared, and questions answered. A decrease in the number of concerns raised through event reporting and verbal comments regarding the APS has been noted during both APN and non-APN coverage times. A potential barrier regarding MD reluctance to get the APS involved was identified. Based on the results, it appears that although RNs find the APS beneficial to patients, some perceive reluctance from MDs to consult the APS. Potential reasons may include concerns of cost, disagreement between the RNs and MDs regarding patient assessments, or MDs wanting to manage patients’ pain or withdrawal symptoms independently. Interestingly, RNs are able to call the APNs for the APS to request informal consultation (i.e., discuss assessments, possible therapies that might be appropriate, etc.), but rarely is this service/resource utilized. Data from the MDs is needed to further explore the RNs’ perception of reluctance, and perhaps reminding more RNs about the role of the APNs would be beneficial. Limitations Several limitations exist with the comparison of the results between the current study and our first study (Czarnecki et al., 2011). Regarding methodology, participants, time of year, and shift worked were not controlled and may have impacted the results. The order of questions, method of survey distribution (paper vs. electronic), and use of an incentive varied across the two studies, and the survey link was sent via e-mail, so the possibility exists that staff other than RNs may have participated in the study; the investigators had no way of verifying participants’ licensure or role. As with any self-report survey, respondent bias is possible. Although the surveys were completely anonymous, RNs may not have felt comfortable reporting certain barriers. The implementation of strategies and study design did not allow for direct measurement of the improvement strategies implemented between studies. Lastly, the homogeneity of the sample may limit the ability to generalize our results to some other organizations. Nonetheless, the investigators believe the responses are reasonably consistent across time and add strength to the results of both studies.

CONCLUSION This is the first study to report barriers that pediatric RNs perceive to providing optimal pain management in one organization at two different points in time

Do Barriers to Pediatric Pain Management Change?

using the same survey tool. Data suggest little change in how RNs describe optimal pain management or RNs’ perceptions of barriers despite the initiation of improvement strategies. Knowing these barriers ‘‘withstood the test of time’’ and may represent more than mere idiosyncrasies may help organizations focus improvement strategies and patient care policy statements toward those areas most likely to impact the most common and meaningful barriers. As pediatric pain management is an interdisciplinary endeavor,

305

future studies investigating the perception of barriers from the perspective of other health care professionals, as well as parents, are needed. Acknowledgments We would like to thank Dr. Steven J. Weisman for his review of this manuscript and acknowledge the Jane B. Pettit Pain and Palliative Care Center and the nurses at Children’s Hospital of Wisconsin who participated in the survey for their ongoing support of this project.

REFERENCES Agency for Health Care Policy and Research [AHCPR] (1992). Acute pain management in infants, children, and adolescents: Operative and medical procedures. Clinical Practice Guideline. (AHCPR Publication No. 92–0020). Rockville, MD: AHCPR. Byrd, P. J., Gonzales, I., & Parsons, V. (2009). Exploring barriers to pain management in newborn intensive care units: A pilot survey of NICU nurses. Advances in Neonatal Care, 9(6), 299–306. Czarnecki, M. L., Simon, K., Thompson, J. J., Armus, C. L., Hanson, T. C., Berg, K. A., Petrie, J. L., Xiang, Q., & Malin, S. (2011). Barriers to pediatric pain management: A nursing perspective. Pain Management Nursing, 12(3), 154–162. Dowden, S., McCarthy, M., & Chalkiadis, G. (2008). Achieving organizational change in pediatric pain management. Pain Research & Management, 13(4), 321–326. Ellis, J. A., McCleary, L., Bouin, R., Dube, K., Rowley, B., MacNeil, M., & Cooke, C. (2007). Implementing best practice pain management in a pediatric hospital. Journal for Specialists in Pediatric Nursing, 12(4), 264–277. Ely, B. (2001). Pediatric nurses’ pain management practice: Barriers to change. Pediatric Nursing, 27(5), 473–480. Gimbler-Berglund, I., Ljusegren, G., & Ensk€ar, K. (2008). Factors influencing pain management in children. Paediatric Nursing, 20(10), 21–24. Kitowski, T., & McNeil, H. (2002). Evaluation of an acute pain service. Journal of PeriAnesthesia Nursing, 17(1), 21–29. Lewthwaite, B. J., Jabusch, K. M., Wheeler, B. J., SchnellHoehn, K. N., Mills, J., Estrella-Holder, E., & Fedorowicz, A. (2011). Nurses’ knowledge and attitudes regarding pain management in hospitalized adults. The Journal of Continuing Education in Nursing, 42(6), 251–257. Mackintosh, C., & Bowles, S. (1999). Evaluation of a nurseled acute pain service. Can clinical nurse specialists make a difference? Journal of Advanced Nursing, 25(1), 30–37. Mackintosh, C., & Bowles, S. (2003). The effect of an acute pain service on nurses’ knowledge and beliefs

about post-operative pain. Journal of Clinical Nursing, 9(1), 119–126. McCluskey, S., & Topping, A. E. (2011). Increasing response rates to lifestyle surveys: A pragmatic evidence review. Perspectives in Public Health, 13(2), 89–94. Miaskowski, C., Crews, J., Ready, L. B., Paul, S. M., & Ginsberg, B. (1999). Anesthesia-based pain services improve the quality of postoperative pain management. Pain, 80(1-2), 23–29. Schafheutle, E. I., Cantrill, J. A., & Noyce, P. R. (2001). Why is pain management suboptimal on surgical wards? Journal of Advanced Nursing, 33(6), 728–737. Shapiro, A., Zohar, E., Kantor, M., & Fredman, B. (2004). Establishing a nurse-based anesthesiologist-supervised inpatient acute pain service: Experience of 4,617 patients. Journal of Clinical Anesthesia, 16(6), 415–420. Simons, J., & Roberson, E. (2002). Poor communication and knowledge deficits: Obstacles to effective management of children’s postoperative pain. Journal of Advanced Nursing, 40(1), 78–86. Stadler, M., Schlander, M., Braeckman, M., Nguyen, T., & Boogaerts, J. G. (2004). A cost-utility and cost-effectiveness analysis of an acute pain service. Journal of Clinical Anesthesia, 16(3), 159–167. Twycross, A. (2002). Educating nurses about pain management: The way forward. Journal of Clinical Nursing, 11(6), 705–714. Twycross, A. (2008). Does the perceived importance of a pain management task affect the quality of children’s nurses’ post-operative pain management practices? Journal of Clinical Nursing, 17(23), 3205–3216. Van Hulle Vincent, C. (2005). Nurses’ knowledge, attitudes, and practices: Regarding children’s pain. MCN American Journal of Maternal Child Nursing, 30(3), 177–183. Van Hulle Vincent, C. (2007). Nurses’ perceptions of children’s pain: A pilot study of cognitive representations. Journal of Pain Symptom Management, 33(3), 290–301. Van Hulle Vincent, C., & Denyes, M. J. (2004). Relieving children’s pain: Nurses’ abilities and analgesic administration practices. Journal of Pediatric Nursing, 19(1), 40–50.