This is an author produced version of a paper published in Journal of Perioperative Practice. This paper has been peer-reviewed but does not include the final publisher proof-corrections or journal pagination.
Citation for the published paper:
Nilsson, Ulrica; Ericsson, Elisabeth; Eriksson, Mats; Idvall, Ewa; Bramhagen, Ann-Cathrine. (2019). Psychometric evaluation of the Post Hospitalization Behavior Questionnaire for Ambulatory Surgery and postoperative behvaior and recovery in children undergoing tonsil surgery. Journal of Perioperative Practice, vol. 29, issue 4, p. null
URL: https://doi.org/10.1177/1750458918782878
Publisher: Sage
This document has been downloaded from MUEP (https://muep.mah.se) / DIVA (https://mau.diva-portal.org).
1
Abstract
1
The study comprised a prospective, comparative cross-sectional survey in 143 children 2
undergoing tonsil surgery. Parents answered the Post Hospitalization Behavior Questionnaire 3
for Ambulatory Surgery (PHBQ-AS), and children answered the questionnaire Postoperative 4
Recovery in Children (PRiC). The PHBQ-AS had positive correlation with the PRiC and with 5
general health. On day 10 after surgery, up to 1/3 of the children still reported physical 6
symptoms (PRiC). No gender or age differences concerning the items of behavior (PHBQ-7
AS) were found. The quality of postoperative recovery (PRiC) in girls was lower, with higher 8
levels of nausea, dizziness, coldness, and headache compared to the boys. Children <6 years 9
of age reported higher levels of dizziness, and lower sleep quality and lower general health. 10
Keywords: Behavior, Children, Pain, Postoperative recovery, Tonsil surgery 11
2
Introduction
13
Postoperative recovery in children is an individual process with subjective experiences, 14
affecting daily life activities, physical and emotional comfort (Eriksson, Nilsson, Bramhagen, 15
Idvall, & Ericsson, 2017), and behavioral changes (Karling, Stenlund, & Hägglöf, 2007; 16
Wilson et al., 2016). Tonsil surgery is a frequently performed surgical procedure among 17
children, and behavioral disturbance has been reported as a common complication (Karling et 18
al., 2007; Sathe, Chinnadurai, McPheeters, & Francis, 2017; Stanko, Bergesio, Davies, 19
Hegarty, & Ungern-Sternberg, 2013; Wilson et al., 2016). The most frequent behavioral 20
changes were apathy and separation anxiety (Fortier, Del Rosario, Rosenbaum, & Kain, 21
2010; L. Kotiniemi, Ryhänen, & Moilanen, 1997; L. H. Kotiniemi, Ryhänen, & Moilanen, 22
1996), eating disturbances (Eriksson et al., 2017; Karling et al., 2007; L. Kotiniemi et al., 23
1997; L. H. Kotiniemi et al., 1996; Sathe et al., 2017), and sleep disturbance (Eriksson et al., 24
2017; Karling et al., 2007; L. Kotiniemi et al., 1997; L. H. Kotiniemi et al., 1996), “making a 25
fuss about eating,” and “temper tantrums” (Beringer, Segar, Pearson, Greamspet, & 26
Kilpatrick, 2014). Preoperative anxiety has been found to be a risk factor for, for example, 27
postoperative pain, sleeping problems, and eating problems after surgery (Kain, Mayes, 28
Caldwell-Andrews, Karas, & McClain, 2006). Young age (Fortier et al., 2010; Karling et al., 29
2007; Stargatt et al., 2006) and being a boy (Beringer et al., 2014) have also been reported as 30
risk factors for behavioral changes postoperatively. Even long-lasting problematic behavior 31
persisting for up to four weeks after surgery has been found in 16%–32% of children (Pearson 32
& Hall, 2017; Stargatt et al., 2006). Day-case tonsil surgery plays a prominent role in 33
pediatric practice in many countries (Hallenstål et al., 2017). 34
In 1966, Vernon, Schulman, and Foley developed a parent-rated instrument, the Post Hospital 35
Behavior Questionnaire (PHBQ), to quantify behavioral changes in children who undergo 36
3
hospitalization due to surgery or illness in the United States (Vernon, Schulman, & Foley, 37
1966). The PHBQ comprises 27 items in six subscales: general anxiety and regression, 38
separation anxiety, eating disturbance, aggression toward authority, apathy/withdrawal, and 39
anxiety about sleep. For each item, parents are asked to compare their child’s behavior before 40
hospitalization to the child’s current behavior (post-hospitalization) on a Likert-type scale 41
(Vernon et al., 1966). The PHBQ has been translated into Swedish, and its psychometric 42
properties were found to be acceptable (Karling, Stenlund, & Hägglöf, 2006). The Swedish 43
PHBQ includes 25 items. Two items, Does your child need a lot of help doing things? and Is 44
it difficult to get your child to talk to you?, loaded poorly on any factor (<0.4), and when this 45
was evaluated by an expert panel these questions were omitted (Karling et al., 2006). 46
In 2015, Jenkins and colleagues (Jenkins et al., 2015) reduce the number of items from 27 to 47
11, creating the Post Hospitalization Behavior Questionnaire for Ambulatory Surgery (PHBQ-48
AS). The PHBQ-AS demonstrated good internal consistency, reliability, and concurrent 49
validity (Jenkins et al., 2015). Since most of the tonsil surgeries among children are 50
performed on a day surgery basis (Alm, Stalfors, Nerfeldt, & Ericsson, 2017), we created a 51
modified version of the PHBQ-AS to test construct validity and internal consistency and to 52
measure behavioral changes in children undergoing tonsil surgery. 53
The instrument Postoperative Recovery in Children (PRiC) was recently developed and tested 54
in Sweden as a questionnaire aiming to measure self-reported postoperative recovery after 55
tonsillectomy in children aged 4–12 years. The results of the study provided evidence of the 56
reliability and validity of the PRiC as a measure of postoperative recovery among children 57
after tonsil surgery. Cronbach’s alpha for the total questionnaire was 0.83. The parents 58
reported in 59% of the cases that the children participated very much in answering the 59
questions (Bramhagen et al., 2016). However, continued psychometric testing of 60
4
questionnaires measuring postoperative recovery and post hospital behavior are needed, as 61
well as an assessment of these outcomes. 62
63
Purpose
64The study objectives were to (1) examine the psychometric properties of the modified 65
Swedish version of the Post Hospitalization Behavior Questionnaire for Ambulatory Surgery 66
(PHBQ-AS), (2) describe post-hospitalization behavior and postoperative recovery in children 67
undergoing tonsil surgery, and (3) explore any gender or age differences in connection with 68
these aspects. 69
Methods
70Study Design and Population 71
The study employed a prospective, comparative cross-sectional survey. A sample of 390 72
children undergoing tonsil surgery (tonsillectomy or tonsillotomy/partial tonsillectomy) and 73
their parents were invited to participate in answering one questionnaire each on day 10 after 74
surgery. The children were consecutively recruited from five different settings, four hospitals 75
and one private day surgery clinic, in Sweden in 2012–2014. The inclusion criteria were as 76
follows: healthy children without any chronic disease between the ages of 4 and 12 years, 77
who were selected for day surgery and whose parents could read, understand, and speak 78
Swedish. 79
Questionnaires 80
Post Hospitalization Behavior Questionnaire for Ambulatory Surgery 81
The modified version of the PHBQ-AS consisted of nine items, that is, the two items excluded 82
in the Swedish version of PHBQ was also excluded in the PHBQ-AS version. The two 83
5
excluded items were Is it difficult to get your child to talk to you? and Does your child need a 84
lot of help doing things? The nine items were answered by the parents on a five-point scale: 1 85
= much less than before, 2 = less than before, 3 = same as before, 4 = more than before, and 5 86
= much more than before. 87
Postoperative Recovery in Children 88
Postoperative recovery was measured with the PRiC, by the children themselves or with help 89
from the parents. The PRiC includes 23 items, 21 items about different aspects of recovery in 90
general and 2 items that are specific to tonsil surgery (ear ache and blood in the mouth). The 91
items concern the previous 24 hours and are assessed on a four-grade scale: 1 = not at all, 2 = 92
a little, 3 = much, and 4 = very much, where “very much” indicates the lowest level of 93
recovery. The PRiC also includes one item of a more general nature that addresses the 94
children’s present general health, to be answered with: 4 = very good, 3 = pretty good, 2 = 95
pretty bad, or 1 = very bad. 96
Background data were collected from the child’s medical journal using a protocol developed a 97
priori, concerning gender, age, surgical procedure, and type of anesthesia. 98
Procedure 99
The invited children and their parents received verbal and written information about the study. 100
At discharge from the hospital, the children received the PRiC questionnaire for assessing 101
postoperative recovery on the 10th day postoperatively. Their parents received the Swedish 102
version of the PHBQ-AS, also to be answered on the 10th day postoperatively, along with 103
prepaid envelopes. No reminder was sent out. 104
Ethical Approval 105
6
The study was approved by the Research Ethics Committee in Uppsala (No. 2012/106). 106
Parents provided written informed consent for the study, and the children provided assent. 107
The children could either write their name or draw a picture on the consent form. The 108
researcher stated that children could withdraw from the study at any time. Data sheets were 109
stored in a locked cabinet and no identifying information was collected. 110
Statistics 111
Power Calculation 112
Based on the suggestions by Ferketich (1991)of a five-to-one ratio (five individuals per scale 113
item) for examining psychometric properties in a nine-item instrument, a minimum of 45 114
participants would be needed to conduct principal components analysis (Ferketich, 1991). 115
Therefore, our sample size provided sufficient power to conduct psychometric testing of the 116
Swedish modified version of PHBQ-AS. 117
Validity to evaluate the accuracy was assessed as follows: 118
1. Principal components analysis: Although the PHBQ-AS (Jenkins et al., 2015) was 119
developed during the time of data collection of the present study, a decision was made 120
to determine construct validity of the Swedish version of the PHBQ-AS, using a 121
principal components analysis with varimax rotation rather than presenting all 25 items, 122
that is, the Swedish version of the PHBQ. 123
2. Construct validity is the degree to which the scores of a questionnaire are consistent 124
with hypotheses based on the assumption that the questionnaire validity measures the 125
construct to be measured. For a correlation coefficient 0.3 < r < 0.7, moderate 126
correlation was assumed. 127
7
3. Hypothesis testing: To analyze construct validity, a priori hypotheses were set up, 128
hypothesizing that the PHBQ-AS would correlate moderately positively with the PRiC 129
and with general health measured with Spearman rank (rho) correlation coefficients. 130
Internal consistency (reliability) describes consistency and was assessed based on the 131
following: Cronbach’s alpha coefficient was calculated to determine the homogeneity of each 132
factor among the items in the PHBQ-AS. 133
Analysis of the Post Hospitalization Behavior Questionnaire for Ambulatory Surgery and 134
Postoperative Recovery in Children 135
1. Parent-reported behavioral changes, PHBQ-AS, are presented as frequencies and 136
proportions. 137
2. The self-reported items of the PRiC are presented with mean and standard deviation 138
(SD) and with the frequencies for all answers in each item. 139
3. Age was categorized dichotomously as younger children (<6 years) or older children 140
(≥6 years), guided by the mean age 6.5 years of the included population. Ages were 141
compared using the Mann Whitney U-test. 142
4. Differences between genders were analyzed with Mann Whitney U-test. 143
All statistical analyses were conducted with SPSS 23.0 (SPSS Inc., Chicago, IL, USA). 144
Descriptive statistics are presented as numbers and percentages, arithmetic means, and 145
medians. A p-value less than 0.05, two-tailed, was considered statistically significant. 146
Results
147On the 10th day after the surgical procedure a total of 37% (n = 143) of the included children 148
answered the PRiC, while their parents answered the PHBQ-AS. The mean age of the 149
responding children was 6.5 years (SD 2.3); 58% were girls (n = 82) and 42% boys (n = 61). 150
8
Half of the children underwent adenotonsillotomy (n = 60) or tonsillotomy (n = 7), and the 151
other half underwent adenotonsillectomy (n = 41) or tonsillectomy (n = 35). The main 152
indications for the procedures were airway obstruction/hypertrophic tonsils (n = 112), 153
infection/inflammation such as recurrent tonsillitis (n = 16), and “chronic” tonsillitis (n = 15). 154
The types of anesthesia were inhalation (78%, n = 112) and total intravenous anesthesia 155
(22%, n = 31). 156
Psychometric Properties 157
A principal components factor analysis with varimax rotation was conducted, using the nine 158
remaining items of the PHBQ-AS to determine whether the Swedish version was 159
unidimensional. The principal components analysis resulted in a one-factor solution 160
explaining 50.8% of the total variance (Table 1). The Cronbach’s alpha was 0.86. Construct 161
validity showed a low positive correlation between the PHBQ-AS and PRiC, rho = -0.25, 162
P = 0.003 and a moderate correlation between PHBQ-AS and general health, rho = -0.30, 163
P = 0.000. On item level, low positive correlations were found between seven PHBQ-AS 164
items and 13 PRiC items. The strongest correlation was found between the items PHBQ-AS 165
Does your child make a fuss about eating? and the PRiC items difficulty eating (rho = 0.31, 166
P = 0.000) and sore throat (rho = 0.30, P = 0.000). The highest frequencies of correlations 167
were found in the two PHBQ-AS items measuring fuss about eating and poor eating. No 168
correlations were found between two of the PHBQ-AS items, Does your child have bad 169
dreams at night or wake up and cry? and Does your child get upset when you leave him (or 170
her) alone for a few minutes?,and any of the PRiC items (Table 2). 171
Behavioral Changes, PHBQ-AS 172
The most frequently reported negative behavioral changes were fuss about eating (17%, n = 173
28) and having a poor appetite (17%, n = 28) (Figure 1). There were no significant differences 174
9
in younger children compared to children from 6 years and older in the PHBQ-AS or on item 175
levels. No differences were seen between genders. 176
Postoperative Recovery (PRiC) 177
The item general health was scored to be rather good, 3.7 (SD 0.5), and the majority of the 178
children had a relatively high quality of recovery on postoperative day 10, with 61%–98% 179
reporting no symptoms or difficulties. The most frequently reported symptoms were sore 180
throat (39%), difficulties with eating (28%), feeling sad (20%), and feeling cold (18%). 181
Twenty percent of the children reported difficulties with attending daycare or school (Table 182
3). Younger children reported more difficulties with dizziness, 1.3 vs 1.1 (P = 0.009), and 183
sleep 1.3 vs 1.1 (P = 0.042), compared to the older children. There were no differences in the 184
other 21 items. Girls suffered more from nausea, 1.1. vs 1.0 (P = 0.048); dizziness, 1.2 vs 1.0 185
(P = 0.009); coldness, 1.4 vs 1.1 (P = 0.001); and headache, 1.2 vs 1.0 (P = 0.002), compared 186
to the boys. There were no differences in the other 19 items. A difference in age was also seen 187
in the item general health, in which younger children assessed lower levels of health 188
compared to the older ones, 3.6 vs 3.8 (P = 0.021). No differences were observed between 189
genders. 190
Discussion
191The study examined a parent-reported questionnaire about post-hospitalization behavior, 192
PHBQ-AS (Jenkins et al., 2015), together with a child-reported questionnaire about 193
postoperative recovery, PRiC (Bramhagen et al., 2016), in a group of tonsil-operated children. 194
Our findings from the psychometric testing suggest that the modified Swedish version of the 195
PHBQ-AS has good construct validity and internal consistency. The Swedish version of 196
PHBQ-AS consists of 9 items, instead of 11 items as in the original PHBQ-AS (Jenkins et al., 197
2015), due to two items being excluded in the Swedish version of PHBQ (Karling et al., 198
10
2006). The Swedish 9-item version of PHBQ-AS cannot be compared with the original 199
American 11-item version of PHBQ-AS. We also assume that there could be cultural 200
differences between Sweden and the United States that also might influence items on the scale 201
and the psychometric properties. Assessment of construct validity should include testing 202
hypotheses that can demonstrate the proposed construct. We expected and found a positive 203
correlation between the PHBQ-AS and the PRiC and general health, because they measure 204
related concepts (behavior vs recovery/health), and children with behavioral changes are 205
expected, to some extent, to have lower quality of recovery and general health. Jenkins et al. 206
(2015) found a moderate positive correlation, r 0.49, between the PHBQ-AS and the 207
Functional Disability Inventory (FDI). The FDI is a 15-item instrument that assesses 208
limitations in psychosocial and physical functioning as a function of children’s physical 209
health (Walker & Greene, 1991). In the present study we also found some positive 210
correlations on item level between PHBQ-AS items and PRiC items. The strongest correlation 211
was noticed between the PHBQ-AS item Does your child make a fuss about eating? and 212
PRiC items difficulty eating (rho = 0.31, P = 0.000) and sore throat (rho = 0.30, P = 0.000). 213
Highest frequencies of correlations were found in the two PHBQ-AS items measuring fuss 214
about eating and poor eating. However, we strongly suggest that the results from both PHBQ-215
AS and PRiC should be analyzed on item level, not as total scores. The reason behind this is 216
that we believe it is important to study each item separately when evaluating the child’s 217
postoperative behavior and recovery. To merge all items (symptoms and signs) into a sum 218
score can dilute or blur the results and thus diminish the external validity. 219
The present study also aimed to describe both post-hospitalization behavior and postoperative 220
recovery 10 days after tonsil surgery. The results showed that 61%–98% of the children 221
reported no symptoms in the different PRiC items and also scored general health as high. 222
However, 39% reported a sore throat and 28% had difficulties eating on day 10 after surgery. 223
11
This is in line with other studies showing that children experienced pain up to 14 days after 224
tonsil surgery (Stanko et al., 2013; Stewart et al., 2012). Regarding the eating factor in the 225
behavioral changes instrument (PHBQ-AS), the parents reported that only 15% had poor 226
appetite. One explanation for this discrepancy could be that the appetite was considered to be 227
normal, but the pain from the sore throat made it difficult for the children to eat. 228
Previous research has shown that pain medications are underused in pediatric populations 229
(Baugh et al., 2011). This underutilization can be attributed to several factors, including 230
parental concerns. A behavioral measurement can provide information to parents about the 231
distress and behavioral changes that children exhibit following surgery, indicative of pain. An 232
advantage of creating a short form of PHBQ is that it would increase the utility of the tool, 233
making it faster to complete and possibly increase compliance, with repeated use during the 234
child’s postoperative recovery. Pain and post-operative recovery are individual and subjective 235
experiences and should be measured with self-assessment in children, as is standard practice 236
in adults. The PHBQ-AS could assist parents and serve as a supplement to the children’s 237
recovery self-reports. Parents play a crucial role in pain management after hospital discharge, 238
particularly given the increases in day surgery and brief hospitalization. 239
We did not find any gender differences in behavioral changes or general health, but the PRiC 240
results revealed that girls reported more nausea, dizziness, coldness, and headache compared 241
to the boys. No differences were found in the other 19 items of the PRiC. Gender differences 242
reported in other studies include a higher incidence of postoperative nausea and vomiting in
243
girls, following outpatient tonsillectomy (Sadhasivam et al., 2015), and less postoperative
244
pain in boys (Chieng et al., 2013). A Swedish registry of 32,225 tonsil surgeries on children
245
reported no clinically relevant difference between genders in patient-reported pain-related 246
outcome measurements, except for days to regular food intake, which was somewhat higher in 247
the girls after tonsillectomy. One factor was that the girls tended to be older at the time of 248
12
surgery, which could influence patient-reported outcome measures for pain (Alm et al., 2017). 249
One possible explanation in our study is that the girls’ symptoms were related to dehydration 250
from the influence of diet in the postoperative recovery. 251
Our study showed that younger children scored lower in the item general health compared to
252
older children. Younger children reported more dizziness and more problems going to sleep
253
on day 10 after surgery. This is in line with other studies that have reported that younger
254
children are at higher risk of behavioral changes after surgery (Fortier et al., 2010; Karling et
255
al., 2007; Stargatt et al., 2006). Karling and colleagues (2007) considered that the items in the
256
original subscale of the PHBQ better reflect behaviors in younger children. There was no
257
difference in age in the item “sore throat” to explain untreated pain in our study. Alm and
258
colleagues (2017) found that older children generally reported higher pain in patient-reported
259
outcome measures. It might be speculated that older children need more support for coping
260
with the pain and that they also are more aware that the parents are responsible for the pain
261
treatment.
262
One limitation in the present study is the chosen follow-up time. The pain after tonsil surgery 263
increases a few days after surgery before gradually decreasing, and is often most pronounced 264
on days 3 to 5 after tonsillectomy (Ericsson, Brattwall, & Lundeberg, 2015) We might have 265
seen a stronger correlation between behavior (PHBQ-AS) and recovery (PRiC) if data 266
collection had been performed during that period. 267
268
Implications for Clinical Practice and Further Research
269
It has previously been shown that the family structures one-parent families (Karling et al.,
270
2007), two or more older siblings (Stargatt et al., 2006), area of residence, and parental
271
education (Karling et al., 2007) are risk factors for developing negative behavioral changes
13
postoperatively. This was not the focus of our study, but is perhaps something worth studying
273
in the future, together with other sociodemographic factors that possibly could influence
274
postoperative recovery and post-hospitalization behavioral changes.
275
This study is limited in terms of generalizability and whether the sample size is enough for a 276
subgroup analysis of gender and age. Furthermore, this study was conducted in Swedish- 277
speaking children and their parents. Further studies including all types of anesthesia and 278
surgeries should be conducted, as well as studies including non-Swedish-speaking participants 279
answering in their own languages. Another limitation to be acknowledged which may 280
influence the results, is that we had no information about if the children had any earlier 281
experiences of undergoing surgery or any siblings with experience of it. 282
283
The modified Swedish PHBQ-AS seems to be relevant for measuring parent-reported 284
behavioral changes in children undergoing tonsil surgery. However, there is a difference 285
between the child’s perspective and, for example, the health care system that defines what is 286
good for them (Nilsson et al., 2015). Children’s own views should be considered when 287
measuring children’s postoperative recovery and care (Bramhagen et al., 2016). Furthermore, 288
to ensure that important and relevant outcomes are measured in clinical practice and future 289
studies, a core outcome set developed specifically for this purpose would be highly desirable. 290
Such a core outcome set would also help limit heterogeneity in outcome reporting. There is a 291
lack of core outcome sets within pediatric anesthesia and surgery, both parent-reported and 292
self-reported among children (Pearson & Hall, 2017). We therefore suggest that parent proxy 293
PHBQ-AS reporting should be supplemented with children’s self-reported symptoms during 294
the postoperative phase, as assessed with the PRiC (Bramhagen et al., 2016) This would 295
identify both children with behavioral changes due to anesthesia and surgery and those 296
children who currently experience low-quality postoperative recovery, such as girls and 297
14
younger children, as well as measure differences in behavior and postoperative recovery in 298
clinical trials. The availability of the PHBQ-AS and the PRiC will be valuable in research and 299
clinical practice with children experiencing pain at home following medical care. 300
Conclusion 301
Our findings from the psychometric testing suggest that the modified Swedish version of the 302
PHBQ-AS has good construct validity and internal consistency, and it reflects aspects of the 303
postoperative period measured by PRiC. Both instruments have their value and should be 304
implemented in the postoperative follow-up. Postoperative recovery seems to differ between 305
genders and ages, presented as lower quality of recovery in girls and younger children. 306
References 307
308
Alm, F., Stalfors, J., Nerfeldt, P., & Ericsson, E. (2017). Patient reportedoutcome of pain after tonsil 309
surgery: An analysis of 32,225 children from the National TonsilSurgery Register in Sweden 310
2009-2016. Paper presented at the Nordic Pediatric Pain Symposium 2017, Stockholm, 311
Sweden, March 30-31, 2017. 312
Baugh, R. F., Archer, S. M., Mitchell, R. B., Rosenfeld, R. M., Amin, R., Burns, J. J., . . . Li, K. K. (2011). 313
Clinical practice guideline: tonsillectomy in children. Otolaryngology–Head and Neck Surgery, 314
144(1_suppl), S1-S30. 315
Beringer, R. M., Segar, P., Pearson, A., Greamspet, M., & Kilpatrick, N. (2014). Observational study of 316
perioperative behavior changes in children having teeth extracted under general anesthesia. 317
Pediatric Anesthesia, 24(5), 499-504. 318
Bramhagen, A. C., Eriksson, M., Ericsson, E., Nilsson, U., Harden, S., & Idvall, E. (2016). Self-reported 319
post-operative recovery in children: development of an instrument. Journal of evaluation in 320
clinical practice, 22(2), 180-188. 321
Chieng, Y. J. S., Chan, W. C. S., Liam, J. L. W., Klainin-Yobas, P., Wang, W., & He, H. G. (2013). 322
Exploring influencing factors of postoperative pain in school-age children undergoing elective 323
surgery. Journal for Specialists in Pediatric Nursing, 18(3), 243-252. 324
Ericsson, E., Brattwall, M., & Lundeberg, S. (2015). Swedish guidelines for the treatment of pain in 325
tonsil surgery in pediatric patients up to 18 years. International Journal of Pediatric 326
Otorhinolaryngology, 79(4), 443-450. 327
Eriksson, M., Nilsson, U., Bramhagen, A.-C., Idvall, E., & Ericsson, E. (2017). Self-reported 328
postoperative recovery in children after tonsillectomy compared to tonsillotomy. 329
International Journal of Pediatric Otorhinolaryngology, 96, 47-54. 330
Ferketich, S. (1991). Focus on psychometrics. Aspects of item analysis. Research in nursing & health, 331
14(2), 165-168. 332
Fortier, M. A., Del Rosario, A. M., Rosenbaum, A., & Kain, Z. N. (2010). Beyond pain: predictors of 333
postoperative maladaptive behavior change in children. Pediatric Anesthesia, 20(5), 445-453. 334
15
Hallenstål, N., Sunnergren, O., Ericsson, E., Hemlin, C., Hessén Söderman, A.-C., Nerfeldt, P., . . . 335
Stalfors, J. (2017). Tonsil surgery in Sweden 2013–2015. Indications, surgical methods and 336
patient-reported outcomes from the National Tonsil Surgery Register. Acta Oto-337
Laryngologica, 1-8. 338
Jenkins, B. N., Kain, Z. N., Kaplan, S. H., Stevenson, R. S., Mayes, L. C., Guadarrama, J., & Fortier, M. A. 339
(2015). Revisiting a measure of child postoperative recovery: development of the Post 340
Hospitalization Behavior Questionnaire for Ambulatory Surgery. Pediatric Anesthesia, 25(7), 341
738-745. 342
Kain, Z. N., Mayes, L. C., Caldwell-Andrews, A. A., Karas, D. E., & McClain, B. C. (2006). Preoperative 343
anxiety, postoperative pain, and behavioral recovery in young children undergoing surgery. 344
Pediatrics, 118(2), 651-658. 345
Karling, M., Stenlund, H., & Hägglöf, B. (2006). Behavioural changes after anaesthesia: validity and 346
liability of the Post Hospitalization Behavior Questionnaire in a Swedish paediatric 347
population. Acta paediatrica, 95(3), 340-346. 348
Karling, M., Stenlund, H., & Hägglöf, B. (2007). Child behaviour after anaesthesia: associated risk 349
factors. Acta paediatrica, 96(5), 740-747. 350
Kotiniemi, L., Ryhänen, P., & Moilanen, I. (1997). Behavioural changes in children following day-case 351
surgery: a 4-week follow-up of 551 children. Anaesthesia, 52(10), 970-976. 352
Kotiniemi, L. H., Ryhänen, P. T., & Moilanen, I. K. (1996). Behavioural changes following routine ENT 353
operations in two-to-ten-year-old children. Pediatric Anesthesia, 6(1), 45-49. 354
Nilsson, S., Björkman, B., Almqvist, A.-L., Almqvist, L., Björk-Willén, P., Donohue, D., . . . Hvit, S. 355
(2015). Children’s voices–Differentiating a child perspective from a child’s perspective. 356
Developmental neurorehabilitation, 18(3), 162-168. 357
Pearson, K. L., & Hall, N. J. (2017). What is the role of enhanced recovery after surgery in children? A 358
scoping review. Pediatric surgery international, 33(1), 43-51. 359
Sadhasivam, S., Chidambaran, V., Olbrecht, V. A., Costandi, A., Clay, S., Prows, C. A., . . . Martin, L. J. 360
(2015). Opioid-related adverse effects in children undergoing surgery: unequal burden on 361
younger girls with higher doses of opioids. Pain Medicine, 16(5), 985-997. 362
Sathe, N., Chinnadurai, S., McPheeters, M., & Francis, D. O. (2017). Comparative Effectiveness of 363
Partial versus Total Tonsillectomy in Children: A Systematic Review. Otolaryngology–Head 364
and Neck Surgery, 156(3), 456-463. 365
Stanko, D., Bergesio, R., Davies, K., Hegarty, M., & Ungern-Sternberg, B. S. (2013). Postoperative pain, 366
nausea and vomiting following adeno-tonsillectomy–a long-term follow-up. Pediatric 367
Anesthesia, 23(8), 690-696. 368
Stargatt, R., Davidson, A. J., Huang, G. H., Czarnecki, C., Gibson, M. A., Stewart, S. A., & Jamsen, K. 369
(2006). A cohort study of the incidence and risk factors for negative behavior changes in 370
children after general anesthesia. Pediatric Anesthesia, 16(8), 846-859. 371
Stewart, D. W., Ragg, P. G., Sheppard, S., & Chalkiadis, G. A. (2012). The severity and duration of 372
postoperative pain and analgesia requirements in children after tonsillectomy, orchidopexy, 373
or inguinal hernia repair. Pediatric Anesthesia, 22(2), 136-143. 374
Walker, L. S., & Greene, J. W. (1991). The functional disability inventory: measuring a neglected 375
dimension of child health status. Journal of pediatric psychology, 16(1), 39-58. 376
Vernon, D. T., Schulman, J. L., & Foley, J. M. (1966). Changes in children's behavior after 377
hospitalization: some dimensions of response and their correlates. American Journal of 378
Diseases of Children, 111(6), 581-593. 379
Wilson, C. A., Sommerfield, D., Drake-Brockman, T. F., Bieberstein, L., Ramgolam, A., & Ungern-380
Sternberg, B. S. (2016). Pain after discharge following head and neck surgery in children. 381
Pediatric Anesthesia, 26(10), 992-1001. 382
383 384
1
Figure 1. Graph representing the frequency of negative behavioral changes (n = 140) on postoperative day 10.
28 28 20 16 12 10 10 9 7 F U S S A B O U T E A T I N G P O O R A P E T I T E G E T T I N G T O T R O U B L E S L E E P B A D D R E A M S U P S E T L E A V I N G D I F F I C U L T I N T E R E S T E D A N T R U M S T E M P E R D O I N G N O T H I N G U N I N T E R E S E D