Routine use of the Bath Adolescent Pain Questionnaire in a paediatric pain clinic

Abstract

Introduction:

Assessment of outcomes from health interventions are of increasing importance, primarily to identify effective and safe treatment, but also to justify funding decisions. The Bath Adolescent Pain Questionnaire (BAPQ) is a self-report questionnaire, validated in 11–18 year olds, assessing the impact of pain in multiple domains of adolescent life. The similarly validated Bath Adolescent Pain Questionnaire for Parents (BAPQ-P) uses the same domains as the BAPQ, assessing the functioning and development of the adolescent from the parents’ perspective.

Methods:

We conducted a prospective study, planning to routinely collect BAPQ/BAPQ-P data at initial assessment and 6 months later. All patients aged between 5 and 19 attending our chronic pain clinic for the first time between December 2009 and December 2014 were mailed BAPQ and BAPQ-P questionnaires before the first appointment and 6 months after the first appointment.

Results:

In total, 376 of 386 families returned questionnaires at time 0 and 96 after 6 months, 26% of those responded at time 0. We found statistically significant differences on patients’ BAPQ questionnaires from 0 to 6 months showing improvement in all domains. A different result was found on parents’ questionnaires where we only found a statistically significant difference on daily and emotional functioning. When comparing patient and parent questionnaires at 0 and 6 months, we found statistically significant differences between patients’ and parents’ questionnaires in the daily functioning and development domains.

Conclusion:

We believe BAPQ and BAPQ-P measurement proved useful tools to assess response to pain management input in adolescents over a 6-month period. Our experience and results suggest that these tools can, with appropriate administrative support, be used in routine clinical practice to assess patient outcomes. We also believe that BAPQ and BAPQ-P measurements have a utility to audit pain clinic activity and potentially a use in demonstrating beneficial outcomes to commissioners.

Introduction

IASP (International Association for the Study of Pain) currently defines pain as an unpleasant sensory and emotional experience associated with actual or potential tissue damage, or described in terms of such damage. Within this definition lies the concept that pain is not mere nociception, but rather its interpretation within our personality and past experiences. Chronic pain is recognised as pain that persists beyond normal healing time: pain is usually regarded as chronic when it lasts or recurs for more than 3–6 months.¹

Chronic and recurrent pain is prevalent in children and adolescents, with girls generally experiencing more pain than boys and prevalence rates increasing with age. A systematic review of epidemiological studies found median prevalence rates ranging from 11% to 38%.² Chronic pain causes maladaptive stress responses that include neuroendocrine dysregulation, fatigue, dysphoria and sleep disturbance. It also causes impairment of physical, emotional, psychological and social functioning.³ Children and young people with chronic pain that are not functioning well, frequently indicated by poor school attendance, benefit from being assessed and managed by interdisciplinary services with expertise in pain medicine, physiotherapy, psychology, occupational therapy and nursing.⁴

Assessment of outcomes from health interventions are of increasing importance, primarily to identify effective and safe treatment, but also to justify funding decisions. The NHS Outcomes Framework Indicators⁵ identifies five domains that should be considered. The NHS England service specification for specialised pain management services for children, adolescents and young people⁶ identifies several outcome measures that should be measured in Domains 2 (Enhancing quality of life for people with long-term conditions) and 3 (Helping people to recover from episodes of ill-health or following injury). One of these is the Bath Adolescent Pain Questionnaire (BAPQ).⁷

The BAPQ is a psychologically robust 61-item multi-dimensional self-report measure, validated in 11–18 year olds, assessing the impact of pain in multiple domains of adolescent life. It comprises between 7 and 12 questions in seven subscales: social functioning, physical functioning, depression, general anxiety, pain specific anxiety, family functioning and development. Some subscale scores are summed to produce four domain scores: daily functioning (physical and social), emotional functioning (depression, general anxiety and pain-related anxiety), family functioning and development. All questions score between 0 and 4, giving a total score between 0 and 244 (daily functioning 0–72, emotional functioning 0–80, family functioning 0–48, development 0–44): higher scores indicate more dysfunction. Eccleston et al.⁸ also developed and validated a similar questionnaire to assess the impact of chronic pain on adolescents from their parents’ perspective, the Bath Adolescent Pain Questionnaire for Parents (BAPQ-P). The BAPQ-P uses the same domains as the BAPQ, assessing the functioning and development of the adolescent from the parents’ perspective: scoring and potential scores are identical to the BAPQ.

Methods

Sheffield Children’s Hospital provides a comprehensive outpatient paediatric pain management service. Children and adolescents with a wide range of painful conditions are seen: musculoskeletal, gastrointestinal, headaches and neuropathic pain. Most patients receive an individualised pain management approach, using a combination of medication and therapeutic rehabilitation with a therapist (occupational or physio). Patients are selected for psychological therapy depending on the perceived need at assessment or during ongoing pain management input.

BAPQ and BAPQ-P questionnaires have been routinely used at initial assessment for several years. Such patient-reported outcome measures are increasingly accepted as important tools to demonstrate effectiveness and justify service funding. A service evaluation to assess change in functioning over 6 months, using BAPQ/BAPQ-P data, was planned and registered with our Audit department; ethical approval was not required. We conducted a prospective study, planning to routinely collect BAPQ/BAPQ-P data at initial assessment and 6 months later.

All patients aged between 5 and 19 attending our chronic pain clinic for the first time between December 2009 and December 2014 were included in our study. Inclusion criterion was submitting the first BAPQ and BAPQ-P prior to clinic; exclusion criterion was not being assessed at 6 months time, either by not returning the questionnaire or by having been discharged within the first 6 months. Data were collected by mailed BAPQ and BAPQ-P questionnaires sent before first appointment and 6 months after the first appointment. Each questionnaire was reviewed by a single investigator and the score for each of the seven subscales was summated according to the original scoring systems: domain scores were calculated by summing the specified subscale scores.^9,10 The data collected were inserted into a SPSS 17.0 database along with gender and age (in years) at the time of first clinic appointment. Statistical analysis was performed using SPSS 17.0. To assess change over the 6 months, we computed the domain data as a difference between first and second questionnaires and did a one-sample t-test, using 0 as the target value (no difference at 6 months): a reduction in the score representing a positive improvement in function. This was made both for total BAPQ/BAPQ-P scores and for the four domains described originally (daily functioning, emotional functioning, family functioning and development).

As well as assessing change in individual patient and parent scores, we also looked for a difference between patient and parent perception in each domain. We were interested to know if the distribution of the BAPQ/BAPQ-P scores was the same between patient and parent, and to do so we conducted a paired T-test for each domain both at 0 and 6 months. We calculated the difference between patients and parents; a positive result reflecting higher scores, and more dysfunction, from the patients’ perspective.

We further grouped patients according to their age (child 0–10 years and adolescent 11–18 years) and by gender (male and female) in order to find if there was any difference in the evolution of their BAPQ/BAPQ-P scores during the 6 months. Because BAPQ questionnaires were validated in adolescents (11–18 years old), we compared both age groups assuming that there would be no difference between both group’s responses. We did independent two-tailed T-tests in order to compare both child and adolescent scores. An independent T-test was also used to compare male and female scores. We also compared the evolution of each group (age and gender) during the 6 months using independent two-tailed T-test applied to the difference between 0 and 6 months questionnaires in each domain.

A p value < 0.05 was used to determine statistical significance.

Results

In total, 376 of 386 families returned questionnaires at time 0 and 96 after 6 months, 26% of those who responded at time 0. One child and one adult did not provide data at either time 0 or 6 months. Two children and five parents did not provide data at time 0 and four parents did not provide data at 6 months. Age distribution at presentation (Table 1) demonstrated a mean of 12.8 years (range, 5–18 years). The cohort was predominantly female (74%). Twenty children (21%) returned questionnaires. Of the cohort who did not respond at 6 months, the mean age was 12.8 years (range, 2–17 years): females were 70% of non-respondents.

Table 1.

Age distribution.

Age	5	6	7	8	9	10	11	12	13	14	15	16	17	18
No.	1	1	4	1	6	7	9	8	12	18	18	7	3	1

Figure 1 shows the distribution of the difference in BAPQ/BAPQ-P scores, between the time of first appointment and 6 months into treatment.

Figure 1.

Box-plots showing the distribution of the difference in BAPQ/BAPQ-P scores.

We found statistically significant differences on patients’ BAPQ questionnaires from 0 to 6 months showing improvement in all domains. The biggest numerical improvement in the four domains was in emotional functioning with a mean difference of 5.65 (Table 2). A different result was found on parents’ questionnaires where we only found a statistically significant difference in daily and emotional functioning. There was no statistically significant difference on parents’ perceived improvement in family functioning and development (Table 3).

Table 2.

Patient BAPQ difference between 0 and 6 months.

	N	Mean	Standard deviation	T-test	95% confidence interval
					Lower	Upper
Total BAPQ	92	13.58	34.18	<0.001	6.50	20.65
Daily functioning	92	4.34	12.14	<0.001	1.82	6.85
Emotional functioning	93	5.65	15.03	<0.001	2.55	8.74
Family functioning	93	1.94	7.16	0.01	0.46	3.41
Development	92	1.80	7.49	0.02	0.25	3.35

BAPQ: Bath Adolescent Pain Questionnaire.

Statistical significance at p value < 0.05.

Table 3.

Parent BAPQ-P difference between 0 and 6 months.

	N	Mean	Standard deviation	T-test	95% confidence interval
					Lower	Upper
Total BAPQ	86	7.84	31.40	0.02	1.11	14.57
Daily functioning	86	3.91	11.61	<0.001	1.42	6.40
Emotional functioning	85	3.16	12.05	0.02	0.57	5.77
Family functioning	86	–0.07	7.13	0.93	–1.60	1.46
Development	83	0.66	7.05	0.39	–0.88	2.20

BAPQ: Bath Adolescent Pain Questionnaire.

Statistical significance at p value < 0.05.

When comparing patient and parent questionnaires at 0 and 6 months, we found statistically significant differences between patients’ and parents’ questionnaires in the daily functioning domain (both at 0 and 6 months) with patients having lower scores in daily functioning, suggesting better function, than their parents (Table 4) (p = 0.01; mean difference: −1.83, 95% confidence interval, CI: (−3.17, −0.49) and p < 0.001; mean difference: −1.92, 95% CI (−3.18, −0.66) at 0 and 6 months, respectively). We also found a statistically significant difference in development perception at 0 months with patients having scored higher than their parents, suggesting more impairment in the development domain p < 0.001; mean difference: 1.58, 95% CI (0.57, 2.60) (Table 4). This difference did not persist at 6 months’ evaluation.

Table 4.

Difference between patients and parents.

	N	Mean	Standard deviation	Sig. (two-tailed)	95% confidence interval
					Lower	Upper
Total BAPQ 0M	87	0.20	19.37	0.93	–3.93	4.32
Daily functioning 0M	87	–1.83	6.28	0.01	–3.17	–0.49
Emotional functioning 0M	87	–0.82	10.00	0.45	–2.95	1.31
Family functioning 0M	87	1.02	5.70	0.10	–0.19	2.24
Development 0M	84	1.58	4.67	<0.001	0.57	2.60
Total BAPQ 6M	89	–3.94	18.84	0.05	–7.91	0.03
Daily functioning 6M	89	–1.92	5.98	<0.001	–3.18	–0.66
Emotional functioning 6M	89	–1.94	9.40	0.05	–3.92	0.04
Family functioning 6M	90	–0.73	4.90	0.16	–1.76	0.29
Development 6M	89	0.53	5.66	0.38	–0.66	1.72

BAPQ: Bath Adolescent Pain Questionnaire.

Statistical significance at p value < 0.05.

Comparing child and adolescents groups showed no statistically significant difference in any of the domain scores apart from family functioning in adolescents at 6 months (Table 5). Comparing the two age groups evolution during the 6 months of treatment, we found statistically significant differences on total BAPQ score and daily functioning evolution, with children showing greater improvement than adolescents both in total BAPQ scores and daily functioning (Table 6).

Table 5.

Difference between children and adolescents.

	Age group	N	Mean	Standard deviation	T-test	Mean difference
Total BAPQ 0M	Children	18	117.72	38.54	0.38	7.91
	Adolescents	73	109.81	33.04	0.43	7.91
Daily functioning 0M	Children	18	35.72	12.76	0.113	5.48
	Adolescents	73	30.25	13.06	0.12	5.48
Emotional functioning 0M	Children	18	42.00	15.47	0.36	3.41
	Adolescents	73	38.59	13.62	0.40	3.41
Family functioning 0M	Children	18	16.33	8.66	0.64	–0.95
	Adolescents	73	17.29	7.42	0.67	–0.95
Development 0M	Children	18	23.67	8.15	0.85	–0.35
	Adolescents	72	24.01	6.51	0.87	–0.35
Total BAPQ 6M	Children	19	89.37	33.12	0.30	–10.92
	Adolescents	73	100.29	41.80	0.24	–10.92
Daily functioning 6M	Children	19	25.16	14.02	0.49	–2.57
	Adolescents	73	27.73	14.55	0.49	–2.57
Emotional functioning 6M	Children	19	31.68	14.31	0.60	–2.30
	Adolescents	74	33.99	17.63	0.56	–2.30
Family functioning 6M	Children	19	12.37	5.46	0.07	–3.43
	Adolescents	74	15.80	7.75	0.03	–3.43
Development 6M	Children	19	20.16	6.83	0.21	–2.52
	Adolescents	74	22.68	8.03	0.18	–2.52

BAPQ: Bath Adolescent Pain Questionnaire.

Statistical significance at p value < 0.05.

Table 6.

Difference between 0 and 6 months grouped by age.

	Age group	N	Mean	Standard deviation	T-test	Mean difference	95% confidence interval
							Lower	Upper
Total BAPQ	Children	18	27.17	33.69	0.04	17.89	0.596	35.19
	Adolescents	72	9.28	32.87
Daily functioning	Children	18	10.06	11.16	0.02	7.51	1.35	13.68
	Adolescents	72	2.54	11.91
Emotional functioning	Children	18	9.67	17.56	0.19	5.28	–2.58	13.14
	Adolescents	73	4.38	14.37
Family functioning	Children	18	4.00	6.79	0.15	2.66	–0.96	6.27
	Adolescents	73	1.34	6.94
Development	Children	18	3.44	6.60	0.23	2.22	–1.44	5.88
	Adolescents	72	1.22	7.08

BAPQ: Bath Adolescent Pain Questionnaire.

Statistical significance at p value < 0.05.

When comparing genders, no statistically significant difference was found between groups (Table 7). When looking at the evolution during the 6 months, no statistically significant difference was found between boys and girls (Table 8).

Table 7.

Difference between male and female.

	Gender	N	Mean	Standard deviation	T-test	95% confidence interval
						Lower	Upper
Total BAPQ 0M	Female	69	109.65	35.13	0.48	–21.74	10.37
	Male	24	115.33	30.88
Daily functioning 0M	Female	69	29.96	13.66	0.07	–10.46	0.37
	Male	24	35.00	10.48
Emotional functioning 0M	Female	69	39.59	13.90	0.62	–4.91	8.18
	Male	24	37.96	13.91
Family functioning 0M	Female	69	16.81	7.78	0.45	–4.93	2.22
	Male	24	18.17	6.98
Development 0M	Female	68	23.63	7.17	0.74	–3.95	2.80
	Male	24	24.21	7.10
Total BAPQ 6M	Female	69	97.75	42.85	0.79	–16.31	21.34
	Male	25	95.24	33.41
Daily functioning 6M	Female	69	26.58	15.13	0.77	–7.72	5.76
	Male	25	27.56	12.69
Emotional functioning 6M	Female	70	34.39	17.33	0.26	–3.38	12.24
	Male	25	29.96	15.52
Family functioning 6M	Female	70	15.06	7.89	0.94	–3.65	3.36
	Male	25	15.20	6.61
Development 6M	Female	70	21.64	8.13	0.64	–4.60	2.85
	Male	25	22.52	7.83

BAPQ: Bath Adolescent Pain Questionnaire.

Table 8.

Difference between male and female between 0 and 6 months.

	Gender	N	Mean	Standard deviation	T-test	95% confidence interval
						Lower	Upper
Total BAPQ	Female	68	11.68	33.99	0.37	–23.42	8.86
	Male	24	18.96	34.85
Daily functioning	Female	68	3.41	11.63	0.22	–9.26	2.16
	Male	24	6.96	13.41
Emotional functioning	Female	69	4.97	15.32	0.47	–9.70	4.48
	Male	24	7.58	14.29
Family functioning	Female	69	1.61	7.12	0.46	–4.64	2.11
	Male	24	2.88	7.34
Development	Female	68	1.90	7.05	0.84	–3.19	3.90
	Male	24	1.54	8.76

BAPQ: Bath Adolescent Pain Questionnaire.

Discussion

The primary goal of this study was to evaluate reduction of pain impact on the daily life of a cohort of paediatric patients with chronic pain, regardless of the type of pain or the exact treatment plan that they followed. We assessed this cohort over a period of 6 months from start of treatment.

First, we acknowledge that the response rate for returning BAPQ and BAPQ-P questionnaires at 6 months was low at 26%. Similarly, an evaluation of the Cleveland Clinic Pediatric Pain Rehabilitation Program, incorporating BAPQ data, only reports 26% of families providing completed pre- and post-treatment questionnaires.¹¹ Poor response rate at follow-up is a recognised problem for postal questionnaires. Improving the response rate using postal questionnaires would require additional administrative time to allow telephone follow-up, a resource that was not available to us. Other methods of data collection are now available, particularly online. Nonetheless, a centrally funded national initiative to collect outcome data in Australasia (PaedePPOC) reports an end of care response rate of only a third using predominantly online data input.¹²

We are, of course, unable to speculate as to whether or not more responses would have altered our results, but the almost identical demographics of responders and non-responders in our study are reassuring. This study reports a service evaluation of routine data collection. We did not routinely collect other outcome data at this time and therefore do not have any data to compare with the BAPQ data. Collecting reliable data on school attendance, widely considered an important reflection of global function, would require consent from families to approach schools. We have not been able to implement this process thus far. Our clinical impression is that our pain service is generally effective in improving functioning and school attendance and this view is supported by the results of this study showing improvement in functioning on BAPQ data. Although discharge from our service was an exclusion criterion, in practice few patients were ‘cured’ or did not engage at all in the initial 6 months; most non-responders did not return a mailed questionnaire.

Information on normative BAPQ data and evidence of clinically significant improvement are lacking. A small retrospective study in 30 paediatric patients with chronic pain and joint hypermobility syndrome claims clinically significant improvement with BAPQ mean changes similar to those in our study.¹³ Clinical significance is claimed by using a calculation and the Cleveland data¹⁰ as a comparison group: we do not fully understand the methodology used by the authors. BAPQ data have been reported to show correlation with disability,¹⁴ social development¹⁵ and sensory modulation¹⁶ in adolescents with chronic pain. A study also reports BAPQ data as a predictor of physical and social functioning in adolescents with rheumatological conditions.¹⁷

We report the total BAPQ/BAPQ-P scores as well as the domain scores. The developers suggest that the total score is not clinically useful, but do not clarify this statement.⁹ We feel that as the total score provides a global assessment across four domains, it is of interest, but have limited our evaluation to the separate domains.

After 6 months’ treatment we found statistically significant patient improvement in all four domains of the BAPQs. The domain that showed greatest improvement, for both patients and parents, was the emotional domain, which may reflect acknowledgement that they are believed and that someone actually is aiding them forward. Not having felt believed by clinicians is a frequent report when paediatric patients and their parents are first assessed and relate their story in our clinic. Engagement with a believing and helpful, responsive, team may account for this early improvement in the emotional domain of patients and parents. Both patients and parents also reported improvements in daily functioning, presumably reflecting a combination of medication effects and engagement with a self-management model of pain management. Interestingly, patients’ perception of their function was better than their parents’ perception at both 0 and 6 months, but both assessed a significant improvement over the 6-month period. The reason for patients consistently rating their daily functioning better than their parents do is not obviously apparent.

Parental assessment of the family functioning and development domains showed no difference during follow-up. It is interesting that patients perceive an improvement in family functioning; this possibly reflects motivation and effort on behalf of parents and less explicit helplessness. Six months is a small period of time to expect major family changes; however, patients appear to be aware of improvement from their perception. Considering the development domain, it is also interesting to note the different perceptions of patients and their parents. Parents did not report any improvement over the 6-month period, but patients did. Patients significantly rated themselves as more affected in this domain at initial assessment compared to their parents; this difference disappears at 6 months. These findings appear to suggest that patients are overwhelmed at initial presentation and catastrophise with regard to school, friends and independence. It is also possible that parents are not fully aware of their child’s difficulties and that patients are not communicating their feelings of low self-esteem.

Grouping patients in two age cohorts, we found that children had higher improvement in the daily functioning domain than adolescents. We are not surprised by this finding as children tend to engage better and quicker into treatment than adolescents; we would expect that with a longer follow-up period this difference would tend to disappear. Care must be taken when looking at this difference because the number of adolescents far exceeds that of children and the BAPQ is not validated for ages under 11. We are not aware of any specific tools to measure chronic pain in children less than 11 years of age. We therefore chose to administer the questionnaires to families of all children attending our clinic, accepting that parental involvement in completion of the child questionnaire was likely in very young children.

Males and females showed no difference in their evolution during the 6-month follow-up. Once again the sample size was very different in the two groups with predominance of female patients.

We are aware that with posted questionnaires we could not reliably know whether the patients filled in their own questionnaires, or parents unduly influenced the patient responses. The differences in the responses suggest that there was not undue parental influence.

We believe BAPQ and BAPQ-P measurement proved useful tools to assess response to pain management input in adolescents over a 6-month period. In our cohort, some statistically significant improvements were shown, which concur with our perception of clinically significant improvement during involvement with our service.

We also believe that BAPQ and BAPQ-P measurements have a utility to audit pain clinic activity and potentially a use in demonstrating beneficial outcomes to commissioners. Our experience and results suggest that these tools can, with appropriate administrative support, be used in routine clinical practice to assess patient outcomes.