Background. Despite an increased awareness among clinicians regarding pain and pain management for infants undergoing surgery, pain associated with procedures performed outside the operating room may not be adequately managed.
Purpose. To examine the beliefs and self-described behavior of physicians and nurses regarding the management of procedural pain in newborn infants.
Methods. A survey was distributed to 467 clinicians (nurses and physicians) working in 11 level II and 4 level III nurseries in a large metropolitan area. Respondents were asked to rate the painfulness of 12 common bedside nursery procedures and how often pharmacologic and nonpharmacologic (comfort) measures are currently used and should be used for those procedures. Demographic data were also collected.
Results. Surveys were completed by 374 clinicians (80% response rate). Physicians and nurses believe infants feel as much pain as adults and that 9 of the 12 listed procedures are moderately to very painful. Neither pharmacologic nor comfort measures are believed to be used frequently, even for the most painful procedures. Physicians and nurses believe both pharmacologic and comfort measures should be used more frequently, but nurses believe comfort measures should be used more frequently than do physicians. Beliefs about infant pain and procedural pain were related to pain management preferences. Physicians' but not nurses' ratings were associated with significant personal pain.
Conclusions. Despite their beliefs that infants experience significant procedure-related pain, clinicians believe pain management for infants remains below optimal levels. Barriers to more consistent and effective pain management need to be identified and surmounted.
Pain has a long history of being undermanaged across the lifespan.1-4 However, it has been recognized as being particularly inferior for infants. Various reasons for poor pain management in infants have been cited. These include a belief that neural immaturity protects infants from pain,5 a reluctance to use anesthetic and analgesic agents in very young infants,6 and an absence of an alternative to verbal report,7 the criterion standard for adult pain assessment.8 However, ample data now indicate that the neurophysiologic basis for pain is established by the end of the second trimester of pregnancy,9 that infants respond to and remember early noxious stimuli,10-12 and that pain assessment tools13-16 and safe pharmacologic and nonpharmacologic methods to reduce pain are available for use in infants.17 Further, there have been strong recommendations from government and medical organizations for aggressive preoperative and postoperative pain management for infants.18,19 Despite all these advances, it is still not clear that pain is being adequately managed in infants20,21; this may be most notable with respect to pain associated with procedures performed outside the operating room. Many invasive procedures (eg, arterial and venous cutdowns, chest tube insertions, intravenous [IV] placements) are consistently performed at the bedside in neonatal intensive care units, and there is a growing interest among some pediatric surgeons to perform other surgical procedures (eg, patent ductus arteriosus ligation, Broviac catheter insertion) at the bedside to avoid transporting the infant to an operating room.22 In older children, procedure-related pain is believed to be among the most difficult type of pain to deal with, from both the patient's and the clinician's perspective.23,24 Recommendations of aggressive pain management for infants commonly do not include procedure-related pain.18,19
Previous reports indicated that effective pain management could be hindered by both physicians (who were less likely to order analgesics for infants and children as compared with adults) and nurses (who were less likely to administer the analgesics to infants and children as ordered, if at all).25,26 Although procedural pain management is the responsibility of the physicians and/or nurses who care for hospitalized infants, little is known about their beliefs and behavior regarding pain management, particularly with regard to procedural pain.
The purpose of this study was to examine physicians' and nurses' beliefs regarding infant pain and the painfulness of the procedures they commonly perform, and to examine what they believe is being done and should be done in their nurseries to reduce the incidence and severity of infant procedural pain. An additional objective was to try to identify individual differences among the clinicians that might influence their beliefs regarding pain management for newborn infants.
Subjects and Procedure
A questionnaire was developed and distributed to physicians and nurses working in a level II (N = 11) and/or a level III (N = 4) nursery in the St Louis metropolitan area. Questionnaires were distributed either by a supervisor within each nursery and collected the following day by a research assistant or by a research assistant who collected the completed questionnaire approximately 10 minutes later. Questionnaires were distributed only to those individuals who were working during a particular shift. At the time of distribution at a given site, the number of potential respondents that included nursing (staff, supervisor/manager, clinical specialist) and medical (resident, fellow, and attending physician) personnel was documented.
The questionnaire consisted of a series of questions on pain and pain management for newborns through the first month of life. The questions were in reference to 12 procedures that are frequently performed in level II or III nurseries. An additional question asked respondents to compare the intensity of infant and adult pain. Responses were in a Likert scale format, ranging from 0 to 4 (Table1). Following common practice, the ratings were treated as being on an interval scale.
A final series of questions sought information about the respondents' primary nursery affiliation (level of intensiveness), current position (eg, board certified neonatologist, pediatric resident, staff registered nurse), number of years experience, gender, age, race, parental status, whether the respondent had undergone surgery, whether that surgery was a painful experience, whether the female respondent had undergone pregnancy, labor/delivery, cesarean section, and whether the respondent believed he/she had had a significant experience with pain, considering his/her own past medical history.
For each question in which the 12 procedures were rated (eg, painfulness of procedures, frequency of use of comfort measures), a repeated measures analysis of variance was computed. From this single analysis we were able to make comparisons both among procedures and between physician and nurse respondents with greater power than individual analyses would have allowed. Because of the large number of degrees of freedom and multiple comparisons, only the Fstatistics are reported. All were significant at the P= .0001 level, except where noted. For the comparison between infant and adult pain, a mean rating for physicians and for nurses was computed and these were compared using a t test. Correlational analyses were performed to examine relations among the ratings separately for physicians and nurses. Correlations were also computed between the ratings and demographic characteristics for physicians and nurses separately. All computations were made with SAS (SAS Institute, Cary, NC).27
A total of 467 questionnaires were distributed and 374 were completed. The response rate was 80% but varied among hospitals between 60% to 100%. The demographic characteristics of nurses and physicians are shown in Table 2.
Intensity of Infant Versus Adult Pain
Although most physicians (59%) and nurses (64%) believed that infants can feel the same amount of pain as can adults, about 27% believed infants feel more pain and only 10% believed infants feel less pain than do adults. The average ratings were 2.27 ± 0.67 for physicians (mean ± SD) and 2.21 ± 0.69 for nurses.
Painfulness of Procedures
Nine of the 12 procedures were rated as being at least moderately painful (rating ≥2). Both physicians and nurses perceived the 12 procedures as differing in their painfulness (F(11,4087) = 193.71; Fig1). Circumcision and insertion of a chest tube were rated as most painful (ratings ≥3.5) and tracheal suctioning, insertion of a gavage tube, and insertion of an umbilical catheter were rated as least painful (ratings ≤2.0).
Use of Pharmacologic Agents
Analgesic and anesthetic agents were believed to not be used frequently, even for the most painful of procedures, although there was some differentiation across procedures in usage (F(1,3797) = 132.21; Fig2). Physicians (mean = 0.91) rated the frequency of usage greater than did the nurses (mean = 0.49;F(1,360) = 35.72), but only for the insertion of a chest tube and for cutdowns did physicians' average ratings even reach the level of “often”.
Use of Comfort Measures
Comfort measures were also believed to be used not very often but more often than pharmacologic agents. However, there was much less differentiation between procedures for the use of comfort measures (F(11,3863) = 23.97; Fig3). Nurses (mean = 1.77) tended to rate their use as slightly higher than did physicians (mean = 1.52; F(1,367) = 2.90; P < .09).
Optimal Use of Pharmacologic Agents
Both physicians and nurses believed that pharmacologic agents should be used more frequently than they currently are, with the highest preference for circumcisions, insertion of chest tubes and cutdowns (F(11,3878) = 267.03; Fig4). Generally, physicians (mean = 1.73) and nurses (mean = 1.71) agreed about the desired frequency of usage (F(1,370)=.08; P= .77). However, nurses thought pharmacologic agents should be used more frequently for lumbar punctures and physicians thought they should be used more frequently for tracheal suctioning and intubations. Ratings were low for peripheral IVs, heelsticks, shots, suctionings, and gavage tube and umbilical catheter insertions.
Optimal Use of Comfort Measures
Both physicians and nurses believed that comfort measures should be used much more frequently than they currently are, but there was no substantial differentiation among procedures as to the frequency with which comfort should be used (F(11,3918)= 28.05; Fig 5). Nurses rated the overall desired frequency higher (mean = 3.34) than did physicians (mean = 2.86; F(1,371) = 15.64). For circumcision, 84% of the nurses felt that comfort should always be used compared with only 68% of the physicians. For all of the procedures, at least 48% of the nurses felt that comfort measures should always be used. In contrast, there were 5 procedures where <40% of physicians felt that comfort measures should always be used: lumbar punctures (38%), insertion of gavage tubes (33%), tracheal suctioning (30%), insertion of umbilical catheter (30%), and endotracheal suctioning (26%).
The internal consistency of the ratings was examined by computing Cronbach's α for each question for which the procedures were rated, ie, procedure painfulness, how often are pharmacologic measures used, how often are comfort measures used, how often should pharmacologic measures be used, and how often should comfort measures be used. The observed alphas were .87, .80, .93, .82 and .92, respectively, demonstrating good reliability of the survey.
Relationships Among Ratings
To examine further differences among respondents we computed a mean rating for each of the questions. Table3 shows the intercorrelations between these mean ratings separately for nurses and for physicians. Those who thought that infants experienced more pain than adults were more likely to rate the listed procedures as painful. Those who perceived procedures as being more painful were more likely to prefer more frequent administration of both pharmacologic agents and comfort measures. Those who thought such measures were already being used more frequently believed even greater use was more desirable.
Care Giver Characteristics
Table 3 also presents correlations between demographic data and survey ratings. Physicians who were older or whose primary affiliation was a level III nursery rated the current use of pharmacologic agents as higher. Level III nurses also rated the desired use of pharmacologic agents as higher. Physicians who reported having had significant surgical pain or a significant personal experience with pain rated the painfulness of procedures as higher, current pharmacologic use as higher, and comfort measures as being more desirable. There were no important relationships between the ratings and any of the other demographic, medical, or pain history variables for the nurses.
Results from this study indicate that most neonatal clinicians believe infants experience pain equal to or greater than that experienced by adults. These beliefs are generally consistent with current data9 and confirm a previous report suggesting that attitudes have recently changed in the area of neonatal pain.28 Clinicians also rated the painfulness of most of the listed nursery procedures as being at least moderately if not very painful. Only tracheal suctioning, insertion of a gavage tube and insertion of an umbilical catheter were rated as being less than moderately painful. Together, these findings indicate that individuals who are at the bedside of newborn infants on a daily basis generally believe those infants are capable of experiencing pain and are routinely exposed to painful procedures. Indeed, the higher the clinicians rated the pain that infants could experience as compared with adults, the higher they rated the painfulness of the procedures the infants commonly undergo.
Despite this increased acknowledgment of infant procedural pain, clinicians generally believed that neither pharmacologic agents nor comfort measures were being used very often to manage infant pain. Even for those procedures they rated as most painful, neither pharmacologic (circumcision = 1.0, cutdown = 1.9, and insertion of chest tube = 1.8) nor comfort (circumcision = 2.1, cutdown = 1.7, and insertion of chest tube = 1.4) measures were believed to be used often. Physicians rated the overall use of pharmacologic agents as more frequent than did nurses and nurses tended to rate the overall use of comfort measures as more frequent than did physicians. These rating differences may reflect performance differences (ie, physicians are traditionally associated with the prescribing and administering of pharmacologic agents and nurses with administering comfort). However, nurses are often present at the bedside when physicians perform procedures and would, thus, observe the administration of anesthetics or analgesics and, likewise, physicians, while performing procedures, would be able to observe nurses administering comfort. Nurses traditionally perform 5 of the 12 procedures we listed: shots, gavage tube insertions, suctionings, heelsticks, and IV placements.
Beliefs about the desired use of pain relief for infants were considerably different than those about their current use. Physicians and nurses reported that both pharmacologic and comfort measures should be used more frequently than they currently are. Those who believed pharmacologic agents were being used more often were more likely to think that pharmacologic agents should be provided more often; similarly, those who believed that comfort measures were being used more often were more likely to think that comfort should be provided even more often.
Individual differences in practice setting were associated with ratings. For nurses, working in a level III nursery was associated with higher ratings of procedure painfulness and the need for more frequent pharmacologic interventions. Although others29 have reported the opposite (ie, noncritical care nurses were more likely to identify pain and to rate the pain of infants as higher when compared with critical care nurses), those differences appeared to be related to the noncritical care nurses having attended a pediatric pain class. Physicians whose primary affiliation was a level III nursery were also more likely to rate the current use of pharmacologic agents as higher. Again, level III personnel are likely more familiar with the use of pharmacologic agents due to a higher incidence of invasive procedures.
Personal medical history also was associated with ratings. Among physicians, those who reported having had a significant pain experience or significant surgical pain were more likely to rate the average painfulness of procedures as higher and both the current and desired frequency of use of pharmacologic agents as higher. We did not observe a similar relation among the nurses. One other study has examined the association between personal pain experience and the ability to recognize or grade infant pain among nurses only and found no association.29 Because personal medical history appeared to influence physicians' beliefs about procedural painfulness and because those latter beliefs were significantly related to beliefs about optimal pain management, these associations should be examined further.
The discrepancy between beliefs concerning current and optimal pain management was striking. Concerns about overdose, addiction, and respiratory depression have been cited as reasons care givers are reluctant to administer analgesics and anesthetics; however, the actual incidence of these adverse effects appears very low.25Generally, physicians and nurses agreed about which procedures should receive more pharmacologic intervention but the procedures over which there were disagreements are of interest. Nurses thought that more pharmacologic pain relief should be provided for lumbar punctures, a procedure that nurses are traditionally present for to restrain the infant, but one which is performed by a physician. We have shown in a previous randomized, controlled trial that local anesthesia does not modify the physiologic response of acutely ill infants undergoing lumbar punctures as compared with the responses of infants who did not receive the anesthetic.30 Similarly, physicians thought more pharmacologic pain relief should be used for suctioning and gavage tube insertions, procedures that nurses typically perform without a physician and ones that were rated in the current survey as having relatively low pain. Because appropriate pharmacologic agents for these two procedures are not well-established, these disagreements may stem from the absence of a clearly identified, appropriate pharmacologic agent for a specific procedure. Alternatively, the disagreements may relate to a locus of control issue because it was the clinicians who do not typically perform the procedures who reported that more pain relief was desirable for those procedures.
Irrespective of which procedure was being considered, clinicians believed more comfort should be administered. Although these beliefs have high social desirability, little data exist to show that nonpharmacologic techniques effectively reduce procedure-related stress and pain in newborn infants or that particular soothing techniques are more effective than others.31 Field and Goldson32 showed that providing full-term and premature neonates a pacifier during heelstick procedures significantly reduced behavioral distress (eg, fussing, crying) but only those premature infants who were medically stable also showed an attentuation of physiologic changes during heelstick. Other reports suggest that reduction of behavioral distress by means of comfort measures may be deceiving; crying was significantly reduced during circumcision in full-term infants who were given a pacifier but the pacifier's effects on serum cortisol, a stress hormone, were not significant.33 Positive effects (eg, weight gain, improved clinical course, lowered stress hormones) of massage therapy have been reported from studies of premature infants but only in those who were medically stable.34 Als35 did report significant clinical and developmental effects in a small sample of acutely ill premature infants from an individualized behavioral care program that included non-nutritive sucking, supportive positioning, and steady parental touch; however, dozens of other interventions may have contributed to the results.
The current study indicates that clinicians' beliefs about the need for adequate pain management are more in tune with the scientific findings regarding infant pain than are their self-described pain management behaviors. To what extent similar relationships might describe pain management in other populations, such as the elderly and their care givers is not known. Further research to identify barriers between clinicians' beliefs and behavior regarding effective pain management is necessary. Clinicians who want to provide analgesia during painful procedures may still have few rational treatment choices available to them due to an absence of approved products for infants.21,36 Inadequate time to administer drugs during emergent procedures and the risk/benefit ratio of administering drugs for very brief procedures or for infants at risk for adverse consequences are common rationales that have been used to justify the failure to administer adequate pain relief. Lack of training about how to manage pain and lack of readily available information in the pediatric literature concerning pain management have also been suggested as contributing factors.37 Obstacles to nonpharmacologic interventions are likely similar but may additionally include factors such as resource allocation (eg, dedicating time to administer comfort measures) and social and/or psychological pressures that may dissuade clinicians from administering comfort. Identifying which of these or other factors may be operative will provide a focus for intervention studies to continue the momentum of bringing significant improvements to the management of pain in newborn infants.
This research was supported by a grant from the National Institute of Child Health and Human Development to Dr Porter.
The authors would like to express their gratitude to the nurses and physicians who participated in this study and to their respective hospitals for welcoming us.
- Received October 14, 1996.
- Accepted March 17, 1997.
Reprint requests to (F.L.P.) Department of Pediatrics, Washington University School of Medicine, One Children's Place, St Louis, MO 63110.
- IV =
- Harkins S
- ↵Flechsig P. Plan des menschlichen Gehirns. Leipzig, Germany: Veit & Co; 1983
- Hatch DJ
- ↵Schechter NL, Berde CB, Yaster M. Pain in infants, children and adolescents: an overview. In: Schechter NL, Berde CB, Yaster M, eds. Pain in Infants, Children and Adolescents. Baltimore, MD: Williams & Wilkins; 1993:3–10
- Anand KJS,
- Hickey PR
- ↵Porter F. Pain assessment in children: infants. In: Schechter NL, Berde CB, Yaster M, eds. Pain in Infants, Children and Adolescents. Baltimore, MD: Williams & Wilkins; 1993:87–96
- ↵Fitzgerald M, Anand KJS. Developmental neuroanatomy and neurophysiology of pain. In: Schechter NL, Berde CB, Yaster M, eds. Pain in Infants, Children and Adolescents. Baltimore, MD: Williams & Wilkins; 1993:11–32
- ↵Mathews JR, McGrath PJ, Pigeon H. Assessment and measurement of pain in children. In: Schechter NL, Berde CB, Yaster M, eds. Pain in Infants, Children and Adolescents. Baltimore, MD: Williams & Wilkins; 1993:97–112
- ↵Porter FL. Assessment and Management of Pain in the Newborn Infant. In: Oldham KT, Columbani PM, Foglia RP, eds. Surgery of Infants and Children: Scientific Principles and Practice. Philadelphia, PA: Lippincott-Raven; 1997:75–77
- American Academy of Pediatrics
- Agency for Health Care Policy and Research
- Yaster M
- ↵American Academy of Pediatrics, American College of Obstetricians and Gynecologists. Guidelines for Perinatal Care. 4th ed. Elk Grove Village, IL: American Academy of Pediatrics; In press
- Zeltzer LK,
- Altman A,
- Cohen D,
- LeBaron S,
- Munuksela EL,
- Schechter NL
- ↵SAS Institute, SAS/STAT Software. Changes and Enhancements through Release 6.11. Cary, NC: 1996
- Porter F,
- Miller JP,
- Cole FS,
- Marshall RE
- Birns B,
- Blank M,
- Bridger WH
- Field T,
- Goldson E
- Als H,
- Lawhon G,
- Brown E,
- Gibes R,
- Duffy FH,
- McAnulty G,
- Blickman JG
- Rana SR
- Copyright © 1997 American Academy of Pediatrics