- Email: [email protected]
Transcultural Reliability and Validity of the Japanese-Language Version of the Pelvic Girdle Questionnaire
Transcultural Reliability and Validity of the Japanese-Language Version of the Pelvic Girdle Questionnaire Asuka Sakamoto, PhD, Kenji Hoshi, PhD, and Kazuyoshi Gamada, PhD ABSTRACT Objective: The purpose of this study was to develop a Japanese version of the Pelvic Girdle Questionnaire (PGQ) and to confirm that the Japanese version of the PGQ (PGQ-J) was as valid as the original version. Methods: This study involves 2 phases: (1) a cross-cultural adaptation study and (2) a cross-sectional study. The English PGQ was translated referring to the cross-cultural adaptation study process. Forty healthy pregnant or postpartum Japanese women were recruited. Women with pelvic girdle pain (PGP) completed the PGQ-J and 5 other instruments. Internal consistency, construct validity, testeretest reliability, ceiling and floor effects, and discrimination validity of the PGQ-J were analyzed. Results: The PGQ-J showed high internal consistency with a Cronbach a of .968, and an interclass correlation coefficient of .79. The content validity showed a high positive correlation with the Oswestry Disability Index and Disability Rating Scale. Conclusion: The PGQ-J was reliable and valid with high internal consistency and content validity for assessing disability owing to PGP in Japanese pregnant and postpartum women. The PGQ-J is expected to facilitate research and clinical practice for PGP in Japan and contribute to the welfare of postpartum women. (J Manipulative Physiol Ther 2020;xx:1-9) Key Indexing Terms: Surveys and Questionnaires; Cross-Cultural Comparison; Cross-Sectional Studies
INTRODUCTION Postpartum pelvic girdle pain (PGP) is an overlooked issue in Japan even though over 70% of pregnant and postpartum Japanese women experience PGP.1 Pelvic girdle pain is defined as pain between the posterior iliac crest and the gluteal fold, particularly in the vicinity of the sacroiliac (SI) joint.2 Postpartum PGP commonly decreases or disappears within 12 weeks after the delivery.3,4 However, over 20% of postpartum women experience prolonged PGP for more than 6 months.2 Pelvic girdle pain affects women’s physical function, daily living, and social economy.5,6 Pelvic girdle pain is commonly diagnosed using magnetic resonance imaging,7 computed tomography,8 local anesthesia or block injection,9 specific pain Graduate School of Medical Technology and Health Welfare Sciences, Higashi Hiroshima City, Hiroshima, Japan. Corresponding author: Kazuyoshi Gamada, PhD, Graduate School of Medical Technology and Health Welfare Sciences, 555-36 Kurosegakuendai, Higashi Hirsoshima-shi, Hiroshima, 739-2631, Japan. (e-mail: [email protected]). Paper submitted May 13, 2018; in revised form July 17, 2018; accepted November 2, 2018. 0161-4754 © 2020 by National University of Health Sciences. https://doi.org/10.1016/j.jmpt.2018.11.019
provocation tests,10-12 or multiple-test scores.5,9 Because the mechanism and pathology of PGP differ from those of low back pain (LBP),13 European guidelines indicate that PGP and LBP should be classified separately owing to differences in their symptoms.2 In Japan, studies of PGP in pregnant and postpartum women have increased over the last decade, but the lack of a standardized self-assessment tool has limited our advances in this field. Symptoms of PGP can be measured using tools such as the visual analog scale, numerical rating scale, or McGill pain questionnaire. Limitations of physical activities owing to PGP and LBP have been assessed by the Oswestry Disability Index (ODI) or Disability Rating Scale.14-16 However, none of the above measurement tools were developed specifically for assessing symptoms of PGP; therefore, a validated measurement tool for PGP was needed. The Pelvic Girdle Questionnaire (PGQ) was developed and validated in 2012 as an assessment tool for PGP specifically in pregnant and postpartum women17 and is now widely used internationally to assess symptoms and physical limitations associated with PGP.13,18,19 Although the PGQ has been validated in Norwegian (PGQ-N), English (PGQ-E), Spanish (PGQ-S), and French-Canadian,20-23 there is no translated and validated Japanese version of the PGQ (PGQ-J). The original PGQ-N, translated PGQ-E, and PGQ in French-Canadian
2
Sakamoto et al The Japanese Version of PGQ
Journal of Manipulative and Physiological Therapeutics Month 2020
Fig 1. Flowchart of the process of translation and adaption of the PGQ. PGQ, Pelvic Girdle Questionnaire; PGQ-J, Japanese version of the Pelvic Girdle Questionnaire.
have shown high validity and reliability (internal consistency, Cronbach a: 0.88-0.94, intraclass correlation coefficients [ICCs]: 0.78-0.94).17,20,21 Therefore, it is considered possible to achieve a similar degree of reliability and validity for the PGQ-J. In Japan, many postpartum women experience persistent PGP, thus a standardized self-assessment tool is needed. Although English, Norwegian, and Spanish-language versions of the PGQ have been validated, a validated Japanese version has not previously been available. Therefore, the purpose of this study was to confirm the validity of a newly developed PGQ-J in comparison to previous cross-cultural adaptation studies of the PGQ. Our hypotheses in this study were as follows: (1) PGQ-J would demonstrate high validity, and (2) PGQ-J would be able to assess PGP and limitations of daily living in pregnant and postpartum women with PGP. By demonstrating the reliability and validity of the PGQ-J, a standardized self-assessment tool for PGP in Japan will become available for future studies and contribute to the global perspective on this problem.
METHODS Protocol This study protocol was approved by the local ethical review board at Seirei Christopher University (IRBN: 7045, 2017). Adaptation and validation of the PGQ-J was conducted through a descriptive cross-sectional study consisting of 2 phases: cross-cultural adaptation study and cross-sectional validation study. There are no conflicts of interest to declare.
Cross-cultural Adaptation Study Translation and cultural adaptation of the questionnaire was conducted by 8 specialists consisting of 2 obstetricians, 2 physical therapists, 2 Japanese English teachers at our university, and 2 linguists who are English and Japanese bilingual (Fig 1). The translation process was performed following the structured 5-phase procedure described by Beaton et al.24 In the first stage (forward translation), 2 translators took 2 separate English-to-Japanese translations of the questionnaire and synthesized the contents of these 2 documents into a single translation. In the second stage, 2 bilingual translators translated the first version of the PGQ-J back into English to develop the second version of the PGQ-J. In the third stage, the content validity of the second version was assessed by 8 experts including 2 physical therapists, 2 linguistics, 2 English teachers, and 2 obstetricians, and the third version was developed by making minor adjustments to the wording. In the fourth stage, a panel of experts examined the contents including quality and cultural adaptation and asked several questions on interpreting certain adjectives used in the PGQ-E. Because we were not able to determine the specific levels indicating the heaviness of loads or quickness of movements described in the PGQ-E, we asked the developer of the PGQ-N.17 After receiving answers to these questions, the fourth version was developed and comprehension was examined by a pilot test that included 20 Japanese women with PGP.
Cross-sectional Study Participants were recruited in Nagasaki, Hiroshima, Shizuoka, Tokyo, Kanagawa, and Iwate in Japan. Inclusion
Journal of Manipulative and Physiological Therapeutics Volume xx, Number xx
Sakamoto et al The Japanese Version of PGQ
Fig 2. Study protocol. PGQ, Pelvic Girdle Questionnaire. criteria were women aged 20 to 40 years reporting SI joint pain exceeding 4 points on the SI joint score.9 All of these women were either pregnant (after 36 weeks’ gestation) or postpartum (within 24 weeks). Exclusion criteria were caesarean section, pain other than PGP (only pain located in pelvic girdle area was included), severe disability, having a child with disabilities, difficulties in answering questions, difficulties in communication, mental illness, congenital anomaly, and systemic or internal disease. The selection criteria, instructions for clinical tests, and questionnaires were given to 3 physical therapists working in private practice. A total of 59 women were examined for PGP conditions using the SI joint score,9 palpation, and the Newton test. Forty women met the selection criteria (SI joint score of 4 or more and pain located around posterior pelvic or pubis) and participated in this study (Fig 2). Twenty participants participated in the testeretest reliability assessment by completing the same questionnaire between 7 and 10 days later. As a clinical assessment of PGP, the SI joint score proposed by Kurosawa et al9 was used in this study to discriminate between pain originating from the posterior ligaments and that from lumber disc herniation. This scoring system ranges from 0 to 9 and examines 2 pain locations (the posterosuperior iliac spine and groin) using a 1-finger test, pain while sitting on a chair, a provocation test, and 2 tenderness points (posterosuperior iliac spine and the sacrotuberous ligament). A score of 4 is considered the positive cutoff point for pelvic pain (sensitivity of 90.3%, specificity of 86.4%).9 Participants with a score of 4 or more were included in this study. Participants completed a comprehensive questionnaire, the PGQ-J, and a battery of self-reporting instruments including the following: (1) the ODI (for chronic low back pain, 10 items with a 6-point scale yielding a total score of 0 to 100
with 100 representing the greatest possible disability25), (2) the Disability Rating Scale (to assess physical disability in chronic pain shoulder, neck, and low back, 12 items with 100 representing the greatest possible disability),26 (3) the Fear Avoidance Beliefs Questionnaire (FABQ) activity subscale for patients with chronic low back pain (5 items with a 0- to 6-point scale yielding a total score of 0 to 24 with 24 representing the highest level of fear-avoidance belief),27 (4) the Pain Catastrophizing Scale for patients with chronic pain (13 items with a 0- to 4-point scale and the highest-level representing pain catastrophizing),28 and the 36-Item Short Form Health Survey (SF-36) (to assess general health using 8 items weighted with norm-based scoring and with a score above 50 considered normal).29,30 The PGP intensity was represented by a numerical rating scale, and the location of pain was marked on a drawing. The PGQ also assesses activity limitation and symptoms using a score of 0 to 3 (a 4-point scale) per item to yield a total score of 0 to 100, with a score of 100 representing the greatest possible disability.17 A well-organized instruction session was provided to answer all questions.
Statistical Analysis The sample size was calculated using G*Power 3.1.2 (Franz Faul, Universitat Kiel, Germany), and a minimum of 32 participants was required. Forty women with PGP participated in this study. Descriptive statistics of basic information were presented as the mean, standard deviation (SD), and range of CI at 95% (95% CI). The descriptive analysis of pain intensity was presented with median and range (minimum and maximum). Statistical analyses to examine the validity and to test the comparisons to other questionnaires were performed by the following: (1) internal
3
4
Sakamoto et al The Japanese Version of PGQ
Journal of Manipulative and Physiological Therapeutics Month 2020
Table 1. Participants’ Baseline Characteristics Characteristic
Total Sample (N ¼ 40)
TesteRetest Subsample (n ¼ 20)
Age (average [95% CI])
30.8 (29.6-32.0)
30.6 (28.4-34.8)
Pain duration (median [range])
12 w (3 wk-4 y)
12 wk (3 wk-4 y)
Pain located at 1 or 2 pelvic joints
19
10
Pain located at all 3 pelvic joints
2
2
Pregnant participants
18
7
Number of children (median [range])
1 (0-3)
1 (0-3)
High school
4
3
College
15
10
Bachelor’s degree
12
5
Master’s degree
3
2
34
19
Pain location (number of people)
Education level (last education level) Number of people
Work status Employed participants CI, confidence interval.
consistency, (2) ceiling and floor effects, (3) content validity (the ICC), (4) factor analysis (Kaiser-Meyer-Olkin test and Bartlett sphericity test), and (5) discriminatory ability (Kaiser-Meyer-Olkin test and Bartlett sphericity test) using receiver operating characteristic curves.
RESULTS Cross-cultural Adaptation Study: Testing of the Prefinal Version of the PGQ-J On assessment by 8 experts, the 25 items on the prefinal version of the PGQ-J demonstrated values greater than 0.70 on Aiken’s V index. Most participants completed the PGQ-J within 5 minutes. The median time for completion was 2 minutes 33 seconds (range 1 minute 29 seconds to 5 minutes 0 seconds). Although most participants thought most questions were easy to answer, 4 participants thought that items 6, 15, 16, 19, and 23 could not be answered because those items were not applicable. Two participants asked us about items 11, 12, and 15 to clarify the respective definitions of heavy or light objects and the speed of running. To conform to the meanings intended in the original PGQ, we contacted the author of the PGQ-N. We were advised that since the PGQ is a self-evaluation questionnaire, the individual respondent’s perceptions of weight and speed are the priority. Thus, if the participant feels the object is heavy, the object is considered heavy. We added an emphasis on
“self-evaluation” to the PGQ-J instructions. Descriptive data of the 25 items are summarized in Table 1.
Testing of the PGQ-J: A Cross-sectional Study The final version of the PGQ-J was completed by 40 women, including 15 pregnant and 25 postpartum women with a mean age (95% CI) of 30.8 (29.8-34.6) years. The results showed that 90% (36 participants) reported 1 or 2 pain locations, whereas 10% (4 participants) had 3 pain locations. Participants’ baseline characteristics are summarized in Table 2.
Ceiling and Floor Effects None of the patients in the overall PGQ-J achieved the minimum or maximum score, implying the absence of ceiling and floor effects. For each item, the range for the lowest possible scores was 9.5% (items 12, 21, 22) to 76.2% (item 14), and the range for the highest possible scores was 0% (items 5, 7, 11, 14) to 14.3% (items 3, 8, 24) (Table 2).
Internal Consistency
Internal consistency measured by the Cronbach a coefficient for items referred to activity subscale (1-20) and symptoms subscales (21-25) were .960 and .960, respectively (Table 3).
Journal of Manipulative and Physiological Therapeutics Volume xx, Number xx
Sakamoto et al The Japanese Version of PGQ
Table 2. Missing Data, Described Data of 25 Items and Internal Consistency Questions
No. of Participants With Missing Data
Average Score (SD)
% of Participants With Lowest Scores
1. Dress yourself
0
0.57 (0.75)
57.1
4.76
2. Stand for less than 10 min
0
0.57 (0.81)
57.1
4.76
3. Stand for more than 60 min
0
1.43 (0.93)
14.3
4. Bend down
0
1.10 (0.94)
28.6
9.52
5. Sit for less than 10 min
0
0.43 (0.68)
66.7
0
6. Sit for more than 60 min
0
1.38 (0.97)
14.3
19
7. Walk for less than 10 min
0
0.52 (0.81)
66.7
0
8. Walk for more than 60 min
0
1.29 (1.01)
19
9. Climb stairs
0
1.00 (0.95)
33.3
9.52
10. Do housework
0
1.05 (0.74)
19
4.76
11. Carry light objects
0
0.67 (0.80)
52.4
0
12. Carry heavy objects
0
1.57 (0.93)
13. Get up/sit down
0
0.95 (0.80)
28.6
4.76
14. Push a shopping cart
0
0.33 (0.66)
76.2
0
15. Run
0
0.95 (0.92)
38.1
4.76
16. Carry out sporting activities
0
1.10 (0.89)
23.8
9.52
17. Lie down
0
1.00 (1.00)
38.1
9.52
18. Roll over in bed
0
1.33 (1.06)
28.6
19. Have a normal sex life
0
0.81 (0.93)
47.6
4.76
20. Push something with 1 foot
0
0.67 (0.91)
57.1
4.76
9.52
% of Participants With Highest Scores
14.3
14.3
19
14.3
Pelvic Girdle Questionnaire symptom subscale score 21. Pain in the morning
0
1.33 (0.80)
9.52
9.52
22. Pain in the evening
0
1.38 (0.74)
9.52
4.76
23. Has your leg/have your legs given way?
0
0.62 (0.92)
57.1
24. Do you do things more slowly?
0
1.10 (1.00)
28.6
25. Is your sleep interrupted?
0
0.81 (1.03)
52.4
9.52 14.3 9.52
SD, standard deviation.
Construct Validity Spearman’s rank correlation coefficient showed a high correlation between the PGQ-J, ODI, and Disability Rating Index (DRI) (r ¼ .806-.902) (Table 3). There was a moderate correlation between the PGQ-J, Pain Catastrophizing Scale,
and physical function (SF-36) (r ¼ .601-.799). However, there was a low correlation between the PGQ-J and mental health (SF-36), social functioning (SF-36), role-physical, and role-mental (SF-36) (r ¼ .004-.393) (Table 4). Each score on the questionnaire is presented in Table 5.
5
6
Sakamoto et al The Japanese Version of PGQ
Journal of Manipulative and Physiological Therapeutics Month 2020
Table 3. Internal Consistency
Discriminate Validity Cronbach a
Item-Total Correlation
1. Dress yourself
0.96
0.81
2. Stand for less than 10 min
0.96
0.75
3. Stand for more than 60 min
0.96
0.76
4. Bend down
0.96
0.69
5. Sit for less than 10 min
0.96
0.90
6. Sit for more than 60 min
0.96
0.75
7. Walk for less than 10 min
0.96
0.85
8. Walk for more than 60 min
0.96
0.87
9. Climb stairs
0.96
0.82
10. Do housework
0.96
0.77
11. Carry light objects
0.96
0.70
12. Carry heavy objects
0.96
0.67
13. Get up/sit down
0.96
0.56
14. Push a shopping cart
0.96
0.59
15. Run
0.96
0.70
16. Carry out sporting activities
0.96
0.77
17. Lie down
0.96
0.81
18. Roll over in bed
0.96
0.64
19. Have a normal sex life
0.96
0.78
20. Push something with 1 foot
0.96
0.78
21. Pain in the morning
0.97
0.38
22. Pain in the evening
0.96
0.62
23. Has your leg/have your legs given way?
0.97
0.35
24. Do you do things more slowly?
0.96
0.73
25. Is your sleep interrupted?
0.96
0.79
Questions Activity subscale
Symptoms subscale
Factor Analysis Kaiser-Meyer-Olkin and Bartlett sphericity tests showed a value of 0.368; the sample size was not sufficient to carry out an exploratory factor analysis.
We compared responses of participants who were pregnant with those of postpartum women and between the 3 pain locations using receiver operating characteristic curves. For pregnant and postpartum women, the average areas (95% CI) under the curve were .345 (.252-.497) and .606 (.402-.810), respectively. Therefore, the PGQ-J discriminated between pregnant and postpartum women. In addition, the PGQ-J discriminated between participants with pain at 1, 2, or 3 locations. The areas under the curve for 1, 2, and 3 pain locations were .655, .413, and .500, respectively.
TesteRetest Reliability Twenty random subsample participants who completed the PGQ-J were included in testeretest reliability testing between 7 and 10 days later. Testeretest reliability of the entire PGQ-J, physical activity subscale, and symptoms subscale were ICC of 0.79 (P ¼ .011), 0.76 (P ¼ .011), and 0.87 (P ¼ .011), respectively.
DISCUSSION This study developed the PGQ-J and determined that the PGQ-J was valid. We found high internal consistency (Cronbach a ¼ .960) and construct validity for self-reporting instruments used for women with PGP. However, the PGQ is the only instrument that significantly discriminated participants between pregnant and postpartum women and among the number of pain locations. The present study found moderate testeretest reliability (ICC ¼ 0.79). Most of the items were understood by participants in pretesting, and they answered all items within 5 minutes. For items 11, 12, and 15, “Carry heavy or light objects” and “Running,” the weight and speed were unclear to some participants who asked us for definitions of heavy and light to give a clear answer. Those questions might have been influenced by Japanese personality characteristics, such as a serious and submissive disposition. We contacted the author of the original PGQ to ensure that the meaning of those items was appropriately explained. As a result, we were advised that the PGQ is a subjective evaluation of activity and symptoms. Therefore, we emphasized that “PGQ-J is based on self-evaluation” when instructing participants in the cross-sectional study. The PGQ-J showed a high correlation with the ODI and DRI and moderate convergent validity. Previous studies for the PGQ-E and PGQ-S showed high convergent and discriminatory validity for pregnant women with PGP. The transcultural adaptation study for the PGQ-S found high internal consistency (Cronbach a ¼ .96). The present study also showed high internal consistency (Cronbach a ¼ .96). The PGQ-J showed good correlation with the ODI and DRI, which assess functional disability owing to chronic pain. The results of this study were similar to those
Journal of Manipulative and Physiological Therapeutics Volume xx, Number xx
Sakamoto et al The Japanese Version of PGQ
Table 4. Correlation Among Questionnaires
Table 5. Descriptive Data of Questionnaires Total Sample (N ¼ 40)
Instruments
PGQ-J Total
PGQ-J Activity PGQ-J Symptom Subscale Subscale
ODI
.902 (.001) .853 (.001)
.730 (.001)
DRI
.730 (.001) .602 (.004)
.806 (.001)
FABQ
.393 (.058) .294 (.195)
.121 (.602)
PCS
.651 (.001) .594 (.005)
.493 (.023)
Instruments
Average Score (95% CI)
PGQ-J total
28.6 (28.4-28.8)
PGQ-J activity subscale
16.7 (16.6-16.8)
PGQ-J symptom subscale
4.7 (4.6-4.7)
DRI score
5.3 (5.3)
ODI score
10.6 (10.5-10.7)
FABQ score
16.3 (16.2-16.4)
PCS score
16.2 (16.1-16.3)
SF-36 General
.096 (.679) .051 (.825)
.269 (.239)
Physical function
.622 (.003) .601 (.004)
.349 (.121)
Role-physical
.373 (.096) .417 (.060)
.072 (.756)
Bodily pain
.443 (.044) .427 (.054)
.492 (.024)
Vitality
.504 (.020) .527 (.014)
.251 (.273)
Social functioning .263 (.249) .253 (.268)
.103 (.656)
Mental health
.091 (.696) .162 (.482)
.004 (.987)
Role-emotional
.342 (.129) .341 (.130)
.225 (.327)
SF-36 score
Values are correlation coefficient (P value). DRI, Disability Rating Index; FABQ, Fear-Avoidance Beliefs Questionnaire; ODI, Oswestry Disability Index; PCS, Pain Catastrophizing Scale; PGQ-J, Japanese-language version of the Pelvic Girdle Questionnaire; SF-36, 36-Item Short Form Health Survey.
of Grotle’s study.20 The FABQ activity subscale showed a low correlation with the PGQ-J in this study, similar to the findings of Grotle et al20 and Stuge et al.17 The FABQ activity subscale in our study was lower than those in previous studies.20,21 Therefore, mental conditions might not be reflected on the PGQ-J. The PGQ-J showed high testeretest reliability (ICC ¼ .79, P ¼ .011), which was similar to the results (a > .80) of previous studies.17,20,21 There were several advantages in this study. First, we developed a well-organized questionnaire and a guidebook to help participants fill out the questionnaire. Second, adequate instruments for assessing low back pain, such as the ODI and DRI, were used to examine content validity. The validated SI joint score9 was used to assess pelvic pain instead of a provocation test because a single provocation test has low reliability.10 The SI joint score is highly reliable for diagnosing PGP clinically, and participants with PGP could be adequately assessed using the SI joint score.
Limitations A limitation of our study was that this study did not use images such as the magnetic resonance imaging or computed tomography to rule out pain from other causes, such as
General
64.4 (64.2-64.6)
Physical function
80.4 (80.2-80.6)
Role physical
72.0 (71.6-72.3)
Bodily pain
30.8 (30.7-31.4)
Vitality
51.4 (51.2-51.6)
Social functioning
77.9 (77.6-81.1)
Mental health
70.0 (69.8-70.2)
Median (95% CI) unless specified. CI, confidence interval; DRI, Disability Rating Index; FABQ, FearAvoidance Beliefs Questionnaire; ODI, Oswestry Disability Index; PCS, Pain Catastrophizing Scale; PGQ-J, Japanese-language version of the Pelvic Girdle Questionnaire; SF-36, 36-Item Short Form Health Survey.
discogenic pain. The PGQ-J can be used to examine Japanese pregnant and postpartum women because the PGQ-J was highly valid and reliable for assessing PGP. Further studies are needed to compare the PGQ-J with commonly used Japanese self-reporting questionnaires, such as the Japanese Orthopedic Association Back Pain Evaluation Questionnaire.
CONCLUSION This study showed that PGQ-J is a valid instrument for assessing PGP disability in Japanese pregnant and postpartum women. The PGQ-J showed high internal consistency and content validity. This may encourage research and clinical application of PGP and international research. The PGQ-J can be used for individuals who understand the written Japanese language.
FUNDING SOURCES
AND
CONFLICTS
OF INTEREST
No funding sources or conflicts of interest were reported for this study.
7
8
Sakamoto et al The Japanese Version of PGQ
CONTRIBUTORSHIP INFORMATION Concept development (provided idea for the research): K.G., A.S. Design (planned the methods to generate the results): A.S. Supervision (provided oversight, responsible for organization and implementation, writing of the manuscript): K.G. Data collection/processing (responsible for experiments, patient management, organization, or reporting data): A.S. Analysis/interpretation (responsible for statistical analysis, evaluation, and presentation of the results): A.S. Literature search (performed the literature search): A.S. Writing (responsible for writing a substantive part of the manuscript): A.S. Critical review (revised manuscript for intellectual content, this does not relate to spelling and grammar checking): K.G., K.H.
Practical Applications
This cross-cultural adaptation study is the first study to establish the PGQ-J. The PGQ-J showed high reliability and content validity and was a valid instrument for assessing PGP disability in Japanese pregnant and postpartum women. The PGQ-J may be a useful questionnaire to encourage research and clinical application of PGP and international research for individuals who understand the written Japanese language.
ACKNOWLEDGMENTS The authors thank Dr. Takashi Tsukiyama and Ms. Hiroko Ito for supporting cross cultural adaptation study. Additionally, we thank Ms. Minako Yamagami, Mr. Yusuke Kubo, Mr. Takeshi Sugiura, Mr. Masashi Kawabata and Mr. Naoya Yano for supporting recruitment and data collection. We also thank associate professor Britt Stuge, who helped with our understanding of the English version of the PGQ.
APPENDIX A. SUPPLEMENTARY DATA Supplementary data to this article can be found online at https://doi.org/10.1016/j.jmpt.2018.11.019.
Journal of Manipulative and Physiological Therapeutics Month 2020
REFERENCES 1. Kusumi K, Kano N, Komatsu M. The relationship between the passage of time and factors associated with back pain in postpartum. J Japan Acad Midwif. 2007;21(2):36-45. 2. Vleeming A, Albert HB, Ostgaard HC, Sturesson B, Stuge B. European guidelines for the diagnosis and treatment of pelvic girdle pain. Eur Spine J. 2008;17(6):794-819. 3. Robinson HS, Eskild A, Heiberg E, Eberhard-Gran M. Pelvic girdle pain in pregnancy: the impact on function. Acta Obstet Gynecol Scand. 2006;85(2):160-164. 4. Albert H, Godskesen M, Westergaard J. Prognosis in four syndromes of pregnancy-related pelvic pain. Acta Obstet Gynecol Scand. 2001;80:505-510. 5. Stuge B. Pelvic girdle pain:examination, treatment, and the development and implementation of the European guidelines. J Assoc Chartered Physiotherapists Women’s Health. 2012; 111:5-12. 6. Gutke A, Lundberg M, Ostgaard HC, Oberg B. Impact of postpartum lumbopelvic pain on disability, pain intensity, health-related quality of life, activity level, kinesiophobia, and depressive symptoms. Eur Spine J. 2011;20(3): 440-448. 7. Reed K, Kochetkova I, Whitby E. Visualising uncertainty: examining women’s views on the role of Magnetic Resonance Imaging (MRI) in late pregnancy. Soc Sci Med. 2016;164: 19-26. 8. Herren C, Sobottke R, Dadgar A, et al. Peripartum pubic symphysis separationeCurrent strategies in diagnosis and therapy and presentation of two cases. Injury. 2015;46(6): 1074-1080. 9. Kurosawa D, Murakami E, Ozawa H, et al. A diagnostic scoring system for sacroiliac joint pain originating from the posterior ligament. Pain Med. 2017;18(2):228-238. 10. Arab AM, Abdollahi I, Joghataei MT, Golafshani Z, Kazemnejad A. Inter- and intra-examiner reliability of single and composites of selected motion palpation and pain provocation tests for sacroiliac joint. Man Ther. 2009;14(2):213-221. 11. Hidalgo B, Hall T, Nielens H, Detrembleur C. Intertester agreement and validity of identifying lumbar pain provocative movement patterns using active and passive accessory movement tests. J Manipulative Physiol Ther. 2014;37(2): 105-115. 12. Laslett M, Young SB, Aprill CN, McDonald B. Diagnosing painful sacroiliac joints: a validity study of a McKenzie evaluation and sacroiliac provocation tests. Aust J Physiother. 2003;49(2):89-97. 13. Hilde G, Gutke A, Slade SC, Stuge B, Hilde G. Physical therapy interventions for pelvic girdle pain (PGP) after pregnancy. Cochrane Database Syst Rev. 2016;11:CD012441. 14. Vianin M. Psychometric properties and clinical usefulness of the Oswestry Disability Index. J Chiropr Med. 2008;7(4): 161-163. 15. Khorsan R, Coulter ID, Hawk C, Choate CG. Measures in chiropractic research: choosing patient-based outcome assessments. J Manipulative Physiol Ther. 2008;31(5):355-375. 16. Holm I, Friis A, Storheim K, Brox JI. Measuring self-reported functional status and pain in patients with chronic low back pain by postal questionnaires. A reliability study Spine (Phila Pa 1976). 2003;28(8):828-833. 17. Stuge B, Garratt A, Jenssen HK, Grotle M. The pelvic girdle questionnaire: a condition-specific instrument for assessing activity limitations and symptoms in people with pelvic girdle pain. Phys Ther. 2011;91(7):1096-1108.
Journal of Manipulative and Physiological Therapeutics Volume xx, Number xx
18. Kovacs FM, Garcia E, Royuela A, Gonzalez L, Abraira V. Spanish Back Pain Research Network. Prevalence and factors associated with low back pain and pelvic girdle pain during pregnancy: a multicenter study conducted in the Spanish National Health Service. Spine (Phila Pa 1976). 2012;37(17): 1516-1533. 19. Ceprnja D, Chipchase L, Gupta A. Prevalence of pregnancyrelated pelvic girdle pain and associated factors in Australia: a cross-sectional study protcol. BMJ Open. 2017;7(11): e018334. 20. Grotle MG, Garratt AM, Krogstad Jenssen H, Stuge B. Reliability and construct validity of self-report questionnaires for patients with pelvic girdle pain. Phys Ther. 2011;92(1): 111-123. 21. Rejano-Campo M, Ferrer-Pena R, Urraca-Gesto MA, et al. Transcultural adaptation and psychometric validation of a Spanish-language version of the “Pelvic Girdle Questionnaire”. Health Qual Life Outcomes. 2017;15(1):30. 22. Girard MP, Marchand AA, Stuge B, Ruchat SM, Descarreaux M. Cross-cultural adaptation of the Pelvic Girdle Questionnaire for the French-Canadian population. J Manipulative Physiol Ther. 2016;39(7):494-499. 23. Girard MP, O’Shaughnessy J, Doucet C, et al. Validation of the French-Canadian Pelvic Girdle Questionnaire. J Manipulative Physiol Ther. 2018;41(3):234-241.
Sakamoto et al The Japanese Version of PGQ
24. Beaton DE, Bombardier C, Guillemin F, Ferraz MB. Guidelines for the process of cross-cultural adaptation of self-report measures. Spine (Phila Pa 1976). 2000;25(24): 3186-3191. 25. Fairbank JC, Couper J, Davies JB, O’Brien JP. The Oswestry low back pain disability questionnaire. Physiotherapy. 1980; 66(8):271-274. 26. Salen BA, Spangfort V, Nygren L, Nordemar R. The disability rating index: an instrument for the assessment of disability in clinical settings. J Clin Epidemiol. 1994;47(12):1423-1435. 27. Waddell G, Newton M, Henderson I, Somerville D, Main CJ. A Fear-Avoidance Beliefs Questionnaire (FABQ) and the role of fear-avoidance beliefs in chronic low back pain and disability. Pain. 1993;52(2):157-168. 28. Sullivan ML. The Pain Catastrophizing Scale: development and validation. Psychol Assess. 1995;7(4):524-532. 29. McHorney CA, Ware Jr JE, Raczek AE. The MOS 36-Item Short-Form Health Survey (SF-36): II. Psychometric and clinical tests of validity in measuring physical and mental health constructs. Med Care. 1993;31(3):247-263. 30. Tokuda Y, Okubo T, Ohde S, et al. Assessing items on the SF-8 Japanese version for health-related quality of life: a psychometric analysis based on the nominal categories model of item response theory. Value Health. 2009;12(4): 568-573.
9
INTRODUCTION Postpartum pelvic girdle pain (PGP) is an overlooked issue in Japan even though over 70% of pregnant and postpartum Japanese women experience PGP.1 Pelvic girdle pain is defined as pain between the posterior iliac crest and the gluteal fold, particularly in the vicinity of the sacroiliac (SI) joint.2 Postpartum PGP commonly decreases or disappears within 12 weeks after the delivery.3,4 However, over 20% of postpartum women experience prolonged PGP for more than 6 months.2 Pelvic girdle pain affects women’s physical function, daily living, and social economy.5,6 Pelvic girdle pain is commonly diagnosed using magnetic resonance imaging,7 computed tomography,8 local anesthesia or block injection,9 specific pain Graduate School of Medical Technology and Health Welfare Sciences, Higashi Hiroshima City, Hiroshima, Japan. Corresponding author: Kazuyoshi Gamada, PhD, Graduate School of Medical Technology and Health Welfare Sciences, 555-36 Kurosegakuendai, Higashi Hirsoshima-shi, Hiroshima, 739-2631, Japan. (e-mail: [email protected]). Paper submitted May 13, 2018; in revised form July 17, 2018; accepted November 2, 2018. 0161-4754 © 2020 by National University of Health Sciences. https://doi.org/10.1016/j.jmpt.2018.11.019
provocation tests,10-12 or multiple-test scores.5,9 Because the mechanism and pathology of PGP differ from those of low back pain (LBP),13 European guidelines indicate that PGP and LBP should be classified separately owing to differences in their symptoms.2 In Japan, studies of PGP in pregnant and postpartum women have increased over the last decade, but the lack of a standardized self-assessment tool has limited our advances in this field. Symptoms of PGP can be measured using tools such as the visual analog scale, numerical rating scale, or McGill pain questionnaire. Limitations of physical activities owing to PGP and LBP have been assessed by the Oswestry Disability Index (ODI) or Disability Rating Scale.14-16 However, none of the above measurement tools were developed specifically for assessing symptoms of PGP; therefore, a validated measurement tool for PGP was needed. The Pelvic Girdle Questionnaire (PGQ) was developed and validated in 2012 as an assessment tool for PGP specifically in pregnant and postpartum women17 and is now widely used internationally to assess symptoms and physical limitations associated with PGP.13,18,19 Although the PGQ has been validated in Norwegian (PGQ-N), English (PGQ-E), Spanish (PGQ-S), and French-Canadian,20-23 there is no translated and validated Japanese version of the PGQ (PGQ-J). The original PGQ-N, translated PGQ-E, and PGQ in French-Canadian
2
Sakamoto et al The Japanese Version of PGQ
Journal of Manipulative and Physiological Therapeutics Month 2020
Fig 1. Flowchart of the process of translation and adaption of the PGQ. PGQ, Pelvic Girdle Questionnaire; PGQ-J, Japanese version of the Pelvic Girdle Questionnaire.
have shown high validity and reliability (internal consistency, Cronbach a: 0.88-0.94, intraclass correlation coefficients [ICCs]: 0.78-0.94).17,20,21 Therefore, it is considered possible to achieve a similar degree of reliability and validity for the PGQ-J. In Japan, many postpartum women experience persistent PGP, thus a standardized self-assessment tool is needed. Although English, Norwegian, and Spanish-language versions of the PGQ have been validated, a validated Japanese version has not previously been available. Therefore, the purpose of this study was to confirm the validity of a newly developed PGQ-J in comparison to previous cross-cultural adaptation studies of the PGQ. Our hypotheses in this study were as follows: (1) PGQ-J would demonstrate high validity, and (2) PGQ-J would be able to assess PGP and limitations of daily living in pregnant and postpartum women with PGP. By demonstrating the reliability and validity of the PGQ-J, a standardized self-assessment tool for PGP in Japan will become available for future studies and contribute to the global perspective on this problem.
METHODS Protocol This study protocol was approved by the local ethical review board at Seirei Christopher University (IRBN: 7045, 2017). Adaptation and validation of the PGQ-J was conducted through a descriptive cross-sectional study consisting of 2 phases: cross-cultural adaptation study and cross-sectional validation study. There are no conflicts of interest to declare.
Cross-cultural Adaptation Study Translation and cultural adaptation of the questionnaire was conducted by 8 specialists consisting of 2 obstetricians, 2 physical therapists, 2 Japanese English teachers at our university, and 2 linguists who are English and Japanese bilingual (Fig 1). The translation process was performed following the structured 5-phase procedure described by Beaton et al.24 In the first stage (forward translation), 2 translators took 2 separate English-to-Japanese translations of the questionnaire and synthesized the contents of these 2 documents into a single translation. In the second stage, 2 bilingual translators translated the first version of the PGQ-J back into English to develop the second version of the PGQ-J. In the third stage, the content validity of the second version was assessed by 8 experts including 2 physical therapists, 2 linguistics, 2 English teachers, and 2 obstetricians, and the third version was developed by making minor adjustments to the wording. In the fourth stage, a panel of experts examined the contents including quality and cultural adaptation and asked several questions on interpreting certain adjectives used in the PGQ-E. Because we were not able to determine the specific levels indicating the heaviness of loads or quickness of movements described in the PGQ-E, we asked the developer of the PGQ-N.17 After receiving answers to these questions, the fourth version was developed and comprehension was examined by a pilot test that included 20 Japanese women with PGP.
Cross-sectional Study Participants were recruited in Nagasaki, Hiroshima, Shizuoka, Tokyo, Kanagawa, and Iwate in Japan. Inclusion
Journal of Manipulative and Physiological Therapeutics Volume xx, Number xx
Sakamoto et al The Japanese Version of PGQ
Fig 2. Study protocol. PGQ, Pelvic Girdle Questionnaire. criteria were women aged 20 to 40 years reporting SI joint pain exceeding 4 points on the SI joint score.9 All of these women were either pregnant (after 36 weeks’ gestation) or postpartum (within 24 weeks). Exclusion criteria were caesarean section, pain other than PGP (only pain located in pelvic girdle area was included), severe disability, having a child with disabilities, difficulties in answering questions, difficulties in communication, mental illness, congenital anomaly, and systemic or internal disease. The selection criteria, instructions for clinical tests, and questionnaires were given to 3 physical therapists working in private practice. A total of 59 women were examined for PGP conditions using the SI joint score,9 palpation, and the Newton test. Forty women met the selection criteria (SI joint score of 4 or more and pain located around posterior pelvic or pubis) and participated in this study (Fig 2). Twenty participants participated in the testeretest reliability assessment by completing the same questionnaire between 7 and 10 days later. As a clinical assessment of PGP, the SI joint score proposed by Kurosawa et al9 was used in this study to discriminate between pain originating from the posterior ligaments and that from lumber disc herniation. This scoring system ranges from 0 to 9 and examines 2 pain locations (the posterosuperior iliac spine and groin) using a 1-finger test, pain while sitting on a chair, a provocation test, and 2 tenderness points (posterosuperior iliac spine and the sacrotuberous ligament). A score of 4 is considered the positive cutoff point for pelvic pain (sensitivity of 90.3%, specificity of 86.4%).9 Participants with a score of 4 or more were included in this study. Participants completed a comprehensive questionnaire, the PGQ-J, and a battery of self-reporting instruments including the following: (1) the ODI (for chronic low back pain, 10 items with a 6-point scale yielding a total score of 0 to 100
with 100 representing the greatest possible disability25), (2) the Disability Rating Scale (to assess physical disability in chronic pain shoulder, neck, and low back, 12 items with 100 representing the greatest possible disability),26 (3) the Fear Avoidance Beliefs Questionnaire (FABQ) activity subscale for patients with chronic low back pain (5 items with a 0- to 6-point scale yielding a total score of 0 to 24 with 24 representing the highest level of fear-avoidance belief),27 (4) the Pain Catastrophizing Scale for patients with chronic pain (13 items with a 0- to 4-point scale and the highest-level representing pain catastrophizing),28 and the 36-Item Short Form Health Survey (SF-36) (to assess general health using 8 items weighted with norm-based scoring and with a score above 50 considered normal).29,30 The PGP intensity was represented by a numerical rating scale, and the location of pain was marked on a drawing. The PGQ also assesses activity limitation and symptoms using a score of 0 to 3 (a 4-point scale) per item to yield a total score of 0 to 100, with a score of 100 representing the greatest possible disability.17 A well-organized instruction session was provided to answer all questions.
Statistical Analysis The sample size was calculated using G*Power 3.1.2 (Franz Faul, Universitat Kiel, Germany), and a minimum of 32 participants was required. Forty women with PGP participated in this study. Descriptive statistics of basic information were presented as the mean, standard deviation (SD), and range of CI at 95% (95% CI). The descriptive analysis of pain intensity was presented with median and range (minimum and maximum). Statistical analyses to examine the validity and to test the comparisons to other questionnaires were performed by the following: (1) internal
3
4
Sakamoto et al The Japanese Version of PGQ
Journal of Manipulative and Physiological Therapeutics Month 2020
Table 1. Participants’ Baseline Characteristics Characteristic
Total Sample (N ¼ 40)
TesteRetest Subsample (n ¼ 20)
Age (average [95% CI])
30.8 (29.6-32.0)
30.6 (28.4-34.8)
Pain duration (median [range])
12 w (3 wk-4 y)
12 wk (3 wk-4 y)
Pain located at 1 or 2 pelvic joints
19
10
Pain located at all 3 pelvic joints
2
2
Pregnant participants
18
7
Number of children (median [range])
1 (0-3)
1 (0-3)
High school
4
3
College
15
10
Bachelor’s degree
12
5
Master’s degree
3
2
34
19
Pain location (number of people)
Education level (last education level) Number of people
Work status Employed participants CI, confidence interval.
consistency, (2) ceiling and floor effects, (3) content validity (the ICC), (4) factor analysis (Kaiser-Meyer-Olkin test and Bartlett sphericity test), and (5) discriminatory ability (Kaiser-Meyer-Olkin test and Bartlett sphericity test) using receiver operating characteristic curves.
RESULTS Cross-cultural Adaptation Study: Testing of the Prefinal Version of the PGQ-J On assessment by 8 experts, the 25 items on the prefinal version of the PGQ-J demonstrated values greater than 0.70 on Aiken’s V index. Most participants completed the PGQ-J within 5 minutes. The median time for completion was 2 minutes 33 seconds (range 1 minute 29 seconds to 5 minutes 0 seconds). Although most participants thought most questions were easy to answer, 4 participants thought that items 6, 15, 16, 19, and 23 could not be answered because those items were not applicable. Two participants asked us about items 11, 12, and 15 to clarify the respective definitions of heavy or light objects and the speed of running. To conform to the meanings intended in the original PGQ, we contacted the author of the PGQ-N. We were advised that since the PGQ is a self-evaluation questionnaire, the individual respondent’s perceptions of weight and speed are the priority. Thus, if the participant feels the object is heavy, the object is considered heavy. We added an emphasis on
“self-evaluation” to the PGQ-J instructions. Descriptive data of the 25 items are summarized in Table 1.
Testing of the PGQ-J: A Cross-sectional Study The final version of the PGQ-J was completed by 40 women, including 15 pregnant and 25 postpartum women with a mean age (95% CI) of 30.8 (29.8-34.6) years. The results showed that 90% (36 participants) reported 1 or 2 pain locations, whereas 10% (4 participants) had 3 pain locations. Participants’ baseline characteristics are summarized in Table 2.
Ceiling and Floor Effects None of the patients in the overall PGQ-J achieved the minimum or maximum score, implying the absence of ceiling and floor effects. For each item, the range for the lowest possible scores was 9.5% (items 12, 21, 22) to 76.2% (item 14), and the range for the highest possible scores was 0% (items 5, 7, 11, 14) to 14.3% (items 3, 8, 24) (Table 2).
Internal Consistency
Internal consistency measured by the Cronbach a coefficient for items referred to activity subscale (1-20) and symptoms subscales (21-25) were .960 and .960, respectively (Table 3).
Journal of Manipulative and Physiological Therapeutics Volume xx, Number xx
Sakamoto et al The Japanese Version of PGQ
Table 2. Missing Data, Described Data of 25 Items and Internal Consistency Questions
No. of Participants With Missing Data
Average Score (SD)
% of Participants With Lowest Scores
1. Dress yourself
0
0.57 (0.75)
57.1
4.76
2. Stand for less than 10 min
0
0.57 (0.81)
57.1
4.76
3. Stand for more than 60 min
0
1.43 (0.93)
14.3
4. Bend down
0
1.10 (0.94)
28.6
9.52
5. Sit for less than 10 min
0
0.43 (0.68)
66.7
0
6. Sit for more than 60 min
0
1.38 (0.97)
14.3
19
7. Walk for less than 10 min
0
0.52 (0.81)
66.7
0
8. Walk for more than 60 min
0
1.29 (1.01)
19
9. Climb stairs
0
1.00 (0.95)
33.3
9.52
10. Do housework
0
1.05 (0.74)
19
4.76
11. Carry light objects
0
0.67 (0.80)
52.4
0
12. Carry heavy objects
0
1.57 (0.93)
13. Get up/sit down
0
0.95 (0.80)
28.6
4.76
14. Push a shopping cart
0
0.33 (0.66)
76.2
0
15. Run
0
0.95 (0.92)
38.1
4.76
16. Carry out sporting activities
0
1.10 (0.89)
23.8
9.52
17. Lie down
0
1.00 (1.00)
38.1
9.52
18. Roll over in bed
0
1.33 (1.06)
28.6
19. Have a normal sex life
0
0.81 (0.93)
47.6
4.76
20. Push something with 1 foot
0
0.67 (0.91)
57.1
4.76
9.52
% of Participants With Highest Scores
14.3
14.3
19
14.3
Pelvic Girdle Questionnaire symptom subscale score 21. Pain in the morning
0
1.33 (0.80)
9.52
9.52
22. Pain in the evening
0
1.38 (0.74)
9.52
4.76
23. Has your leg/have your legs given way?
0
0.62 (0.92)
57.1
24. Do you do things more slowly?
0
1.10 (1.00)
28.6
25. Is your sleep interrupted?
0
0.81 (1.03)
52.4
9.52 14.3 9.52
SD, standard deviation.
Construct Validity Spearman’s rank correlation coefficient showed a high correlation between the PGQ-J, ODI, and Disability Rating Index (DRI) (r ¼ .806-.902) (Table 3). There was a moderate correlation between the PGQ-J, Pain Catastrophizing Scale,
and physical function (SF-36) (r ¼ .601-.799). However, there was a low correlation between the PGQ-J and mental health (SF-36), social functioning (SF-36), role-physical, and role-mental (SF-36) (r ¼ .004-.393) (Table 4). Each score on the questionnaire is presented in Table 5.
5
6
Sakamoto et al The Japanese Version of PGQ
Journal of Manipulative and Physiological Therapeutics Month 2020
Table 3. Internal Consistency
Discriminate Validity Cronbach a
Item-Total Correlation
1. Dress yourself
0.96
0.81
2. Stand for less than 10 min
0.96
0.75
3. Stand for more than 60 min
0.96
0.76
4. Bend down
0.96
0.69
5. Sit for less than 10 min
0.96
0.90
6. Sit for more than 60 min
0.96
0.75
7. Walk for less than 10 min
0.96
0.85
8. Walk for more than 60 min
0.96
0.87
9. Climb stairs
0.96
0.82
10. Do housework
0.96
0.77
11. Carry light objects
0.96
0.70
12. Carry heavy objects
0.96
0.67
13. Get up/sit down
0.96
0.56
14. Push a shopping cart
0.96
0.59
15. Run
0.96
0.70
16. Carry out sporting activities
0.96
0.77
17. Lie down
0.96
0.81
18. Roll over in bed
0.96
0.64
19. Have a normal sex life
0.96
0.78
20. Push something with 1 foot
0.96
0.78
21. Pain in the morning
0.97
0.38
22. Pain in the evening
0.96
0.62
23. Has your leg/have your legs given way?
0.97
0.35
24. Do you do things more slowly?
0.96
0.73
25. Is your sleep interrupted?
0.96
0.79
Questions Activity subscale
Symptoms subscale
Factor Analysis Kaiser-Meyer-Olkin and Bartlett sphericity tests showed a value of 0.368; the sample size was not sufficient to carry out an exploratory factor analysis.
We compared responses of participants who were pregnant with those of postpartum women and between the 3 pain locations using receiver operating characteristic curves. For pregnant and postpartum women, the average areas (95% CI) under the curve were .345 (.252-.497) and .606 (.402-.810), respectively. Therefore, the PGQ-J discriminated between pregnant and postpartum women. In addition, the PGQ-J discriminated between participants with pain at 1, 2, or 3 locations. The areas under the curve for 1, 2, and 3 pain locations were .655, .413, and .500, respectively.
TesteRetest Reliability Twenty random subsample participants who completed the PGQ-J were included in testeretest reliability testing between 7 and 10 days later. Testeretest reliability of the entire PGQ-J, physical activity subscale, and symptoms subscale were ICC of 0.79 (P ¼ .011), 0.76 (P ¼ .011), and 0.87 (P ¼ .011), respectively.
DISCUSSION This study developed the PGQ-J and determined that the PGQ-J was valid. We found high internal consistency (Cronbach a ¼ .960) and construct validity for self-reporting instruments used for women with PGP. However, the PGQ is the only instrument that significantly discriminated participants between pregnant and postpartum women and among the number of pain locations. The present study found moderate testeretest reliability (ICC ¼ 0.79). Most of the items were understood by participants in pretesting, and they answered all items within 5 minutes. For items 11, 12, and 15, “Carry heavy or light objects” and “Running,” the weight and speed were unclear to some participants who asked us for definitions of heavy and light to give a clear answer. Those questions might have been influenced by Japanese personality characteristics, such as a serious and submissive disposition. We contacted the author of the original PGQ to ensure that the meaning of those items was appropriately explained. As a result, we were advised that the PGQ is a subjective evaluation of activity and symptoms. Therefore, we emphasized that “PGQ-J is based on self-evaluation” when instructing participants in the cross-sectional study. The PGQ-J showed a high correlation with the ODI and DRI and moderate convergent validity. Previous studies for the PGQ-E and PGQ-S showed high convergent and discriminatory validity for pregnant women with PGP. The transcultural adaptation study for the PGQ-S found high internal consistency (Cronbach a ¼ .96). The present study also showed high internal consistency (Cronbach a ¼ .96). The PGQ-J showed good correlation with the ODI and DRI, which assess functional disability owing to chronic pain. The results of this study were similar to those
Journal of Manipulative and Physiological Therapeutics Volume xx, Number xx
Sakamoto et al The Japanese Version of PGQ
Table 4. Correlation Among Questionnaires
Table 5. Descriptive Data of Questionnaires Total Sample (N ¼ 40)
Instruments
PGQ-J Total
PGQ-J Activity PGQ-J Symptom Subscale Subscale
ODI
.902 (.001) .853 (.001)
.730 (.001)
DRI
.730 (.001) .602 (.004)
.806 (.001)
FABQ
.393 (.058) .294 (.195)
.121 (.602)
PCS
.651 (.001) .594 (.005)
.493 (.023)
Instruments
Average Score (95% CI)
PGQ-J total
28.6 (28.4-28.8)
PGQ-J activity subscale
16.7 (16.6-16.8)
PGQ-J symptom subscale
4.7 (4.6-4.7)
DRI score
5.3 (5.3)
ODI score
10.6 (10.5-10.7)
FABQ score
16.3 (16.2-16.4)
PCS score
16.2 (16.1-16.3)
SF-36 General
.096 (.679) .051 (.825)
.269 (.239)
Physical function
.622 (.003) .601 (.004)
.349 (.121)
Role-physical
.373 (.096) .417 (.060)
.072 (.756)
Bodily pain
.443 (.044) .427 (.054)
.492 (.024)
Vitality
.504 (.020) .527 (.014)
.251 (.273)
Social functioning .263 (.249) .253 (.268)
.103 (.656)
Mental health
.091 (.696) .162 (.482)
.004 (.987)
Role-emotional
.342 (.129) .341 (.130)
.225 (.327)
SF-36 score
Values are correlation coefficient (P value). DRI, Disability Rating Index; FABQ, Fear-Avoidance Beliefs Questionnaire; ODI, Oswestry Disability Index; PCS, Pain Catastrophizing Scale; PGQ-J, Japanese-language version of the Pelvic Girdle Questionnaire; SF-36, 36-Item Short Form Health Survey.
of Grotle’s study.20 The FABQ activity subscale showed a low correlation with the PGQ-J in this study, similar to the findings of Grotle et al20 and Stuge et al.17 The FABQ activity subscale in our study was lower than those in previous studies.20,21 Therefore, mental conditions might not be reflected on the PGQ-J. The PGQ-J showed high testeretest reliability (ICC ¼ .79, P ¼ .011), which was similar to the results (a > .80) of previous studies.17,20,21 There were several advantages in this study. First, we developed a well-organized questionnaire and a guidebook to help participants fill out the questionnaire. Second, adequate instruments for assessing low back pain, such as the ODI and DRI, were used to examine content validity. The validated SI joint score9 was used to assess pelvic pain instead of a provocation test because a single provocation test has low reliability.10 The SI joint score is highly reliable for diagnosing PGP clinically, and participants with PGP could be adequately assessed using the SI joint score.
Limitations A limitation of our study was that this study did not use images such as the magnetic resonance imaging or computed tomography to rule out pain from other causes, such as
General
64.4 (64.2-64.6)
Physical function
80.4 (80.2-80.6)
Role physical
72.0 (71.6-72.3)
Bodily pain
30.8 (30.7-31.4)
Vitality
51.4 (51.2-51.6)
Social functioning
77.9 (77.6-81.1)
Mental health
70.0 (69.8-70.2)
Median (95% CI) unless specified. CI, confidence interval; DRI, Disability Rating Index; FABQ, FearAvoidance Beliefs Questionnaire; ODI, Oswestry Disability Index; PCS, Pain Catastrophizing Scale; PGQ-J, Japanese-language version of the Pelvic Girdle Questionnaire; SF-36, 36-Item Short Form Health Survey.
discogenic pain. The PGQ-J can be used to examine Japanese pregnant and postpartum women because the PGQ-J was highly valid and reliable for assessing PGP. Further studies are needed to compare the PGQ-J with commonly used Japanese self-reporting questionnaires, such as the Japanese Orthopedic Association Back Pain Evaluation Questionnaire.
CONCLUSION This study showed that PGQ-J is a valid instrument for assessing PGP disability in Japanese pregnant and postpartum women. The PGQ-J showed high internal consistency and content validity. This may encourage research and clinical application of PGP and international research. The PGQ-J can be used for individuals who understand the written Japanese language.
FUNDING SOURCES
AND
CONFLICTS
OF INTEREST
No funding sources or conflicts of interest were reported for this study.
7
8
Sakamoto et al The Japanese Version of PGQ
CONTRIBUTORSHIP INFORMATION Concept development (provided idea for the research): K.G., A.S. Design (planned the methods to generate the results): A.S. Supervision (provided oversight, responsible for organization and implementation, writing of the manuscript): K.G. Data collection/processing (responsible for experiments, patient management, organization, or reporting data): A.S. Analysis/interpretation (responsible for statistical analysis, evaluation, and presentation of the results): A.S. Literature search (performed the literature search): A.S. Writing (responsible for writing a substantive part of the manuscript): A.S. Critical review (revised manuscript for intellectual content, this does not relate to spelling and grammar checking): K.G., K.H.
Practical Applications
This cross-cultural adaptation study is the first study to establish the PGQ-J. The PGQ-J showed high reliability and content validity and was a valid instrument for assessing PGP disability in Japanese pregnant and postpartum women. The PGQ-J may be a useful questionnaire to encourage research and clinical application of PGP and international research for individuals who understand the written Japanese language.
ACKNOWLEDGMENTS The authors thank Dr. Takashi Tsukiyama and Ms. Hiroko Ito for supporting cross cultural adaptation study. Additionally, we thank Ms. Minako Yamagami, Mr. Yusuke Kubo, Mr. Takeshi Sugiura, Mr. Masashi Kawabata and Mr. Naoya Yano for supporting recruitment and data collection. We also thank associate professor Britt Stuge, who helped with our understanding of the English version of the PGQ.
APPENDIX A. SUPPLEMENTARY DATA Supplementary data to this article can be found online at https://doi.org/10.1016/j.jmpt.2018.11.019.
Journal of Manipulative and Physiological Therapeutics Month 2020
REFERENCES 1. Kusumi K, Kano N, Komatsu M. The relationship between the passage of time and factors associated with back pain in postpartum. J Japan Acad Midwif. 2007;21(2):36-45. 2. Vleeming A, Albert HB, Ostgaard HC, Sturesson B, Stuge B. European guidelines for the diagnosis and treatment of pelvic girdle pain. Eur Spine J. 2008;17(6):794-819. 3. Robinson HS, Eskild A, Heiberg E, Eberhard-Gran M. Pelvic girdle pain in pregnancy: the impact on function. Acta Obstet Gynecol Scand. 2006;85(2):160-164. 4. Albert H, Godskesen M, Westergaard J. Prognosis in four syndromes of pregnancy-related pelvic pain. Acta Obstet Gynecol Scand. 2001;80:505-510. 5. Stuge B. Pelvic girdle pain:examination, treatment, and the development and implementation of the European guidelines. J Assoc Chartered Physiotherapists Women’s Health. 2012; 111:5-12. 6. Gutke A, Lundberg M, Ostgaard HC, Oberg B. Impact of postpartum lumbopelvic pain on disability, pain intensity, health-related quality of life, activity level, kinesiophobia, and depressive symptoms. Eur Spine J. 2011;20(3): 440-448. 7. Reed K, Kochetkova I, Whitby E. Visualising uncertainty: examining women’s views on the role of Magnetic Resonance Imaging (MRI) in late pregnancy. Soc Sci Med. 2016;164: 19-26. 8. Herren C, Sobottke R, Dadgar A, et al. Peripartum pubic symphysis separationeCurrent strategies in diagnosis and therapy and presentation of two cases. Injury. 2015;46(6): 1074-1080. 9. Kurosawa D, Murakami E, Ozawa H, et al. A diagnostic scoring system for sacroiliac joint pain originating from the posterior ligament. Pain Med. 2017;18(2):228-238. 10. Arab AM, Abdollahi I, Joghataei MT, Golafshani Z, Kazemnejad A. Inter- and intra-examiner reliability of single and composites of selected motion palpation and pain provocation tests for sacroiliac joint. Man Ther. 2009;14(2):213-221. 11. Hidalgo B, Hall T, Nielens H, Detrembleur C. Intertester agreement and validity of identifying lumbar pain provocative movement patterns using active and passive accessory movement tests. J Manipulative Physiol Ther. 2014;37(2): 105-115. 12. Laslett M, Young SB, Aprill CN, McDonald B. Diagnosing painful sacroiliac joints: a validity study of a McKenzie evaluation and sacroiliac provocation tests. Aust J Physiother. 2003;49(2):89-97. 13. Hilde G, Gutke A, Slade SC, Stuge B, Hilde G. Physical therapy interventions for pelvic girdle pain (PGP) after pregnancy. Cochrane Database Syst Rev. 2016;11:CD012441. 14. Vianin M. Psychometric properties and clinical usefulness of the Oswestry Disability Index. J Chiropr Med. 2008;7(4): 161-163. 15. Khorsan R, Coulter ID, Hawk C, Choate CG. Measures in chiropractic research: choosing patient-based outcome assessments. J Manipulative Physiol Ther. 2008;31(5):355-375. 16. Holm I, Friis A, Storheim K, Brox JI. Measuring self-reported functional status and pain in patients with chronic low back pain by postal questionnaires. A reliability study Spine (Phila Pa 1976). 2003;28(8):828-833. 17. Stuge B, Garratt A, Jenssen HK, Grotle M. The pelvic girdle questionnaire: a condition-specific instrument for assessing activity limitations and symptoms in people with pelvic girdle pain. Phys Ther. 2011;91(7):1096-1108.
Journal of Manipulative and Physiological Therapeutics Volume xx, Number xx
18. Kovacs FM, Garcia E, Royuela A, Gonzalez L, Abraira V. Spanish Back Pain Research Network. Prevalence and factors associated with low back pain and pelvic girdle pain during pregnancy: a multicenter study conducted in the Spanish National Health Service. Spine (Phila Pa 1976). 2012;37(17): 1516-1533. 19. Ceprnja D, Chipchase L, Gupta A. Prevalence of pregnancyrelated pelvic girdle pain and associated factors in Australia: a cross-sectional study protcol. BMJ Open. 2017;7(11): e018334. 20. Grotle MG, Garratt AM, Krogstad Jenssen H, Stuge B. Reliability and construct validity of self-report questionnaires for patients with pelvic girdle pain. Phys Ther. 2011;92(1): 111-123. 21. Rejano-Campo M, Ferrer-Pena R, Urraca-Gesto MA, et al. Transcultural adaptation and psychometric validation of a Spanish-language version of the “Pelvic Girdle Questionnaire”. Health Qual Life Outcomes. 2017;15(1):30. 22. Girard MP, Marchand AA, Stuge B, Ruchat SM, Descarreaux M. Cross-cultural adaptation of the Pelvic Girdle Questionnaire for the French-Canadian population. J Manipulative Physiol Ther. 2016;39(7):494-499. 23. Girard MP, O’Shaughnessy J, Doucet C, et al. Validation of the French-Canadian Pelvic Girdle Questionnaire. J Manipulative Physiol Ther. 2018;41(3):234-241.
Sakamoto et al The Japanese Version of PGQ
24. Beaton DE, Bombardier C, Guillemin F, Ferraz MB. Guidelines for the process of cross-cultural adaptation of self-report measures. Spine (Phila Pa 1976). 2000;25(24): 3186-3191. 25. Fairbank JC, Couper J, Davies JB, O’Brien JP. The Oswestry low back pain disability questionnaire. Physiotherapy. 1980; 66(8):271-274. 26. Salen BA, Spangfort V, Nygren L, Nordemar R. The disability rating index: an instrument for the assessment of disability in clinical settings. J Clin Epidemiol. 1994;47(12):1423-1435. 27. Waddell G, Newton M, Henderson I, Somerville D, Main CJ. A Fear-Avoidance Beliefs Questionnaire (FABQ) and the role of fear-avoidance beliefs in chronic low back pain and disability. Pain. 1993;52(2):157-168. 28. Sullivan ML. The Pain Catastrophizing Scale: development and validation. Psychol Assess. 1995;7(4):524-532. 29. McHorney CA, Ware Jr JE, Raczek AE. The MOS 36-Item Short-Form Health Survey (SF-36): II. Psychometric and clinical tests of validity in measuring physical and mental health constructs. Med Care. 1993;31(3):247-263. 30. Tokuda Y, Okubo T, Ohde S, et al. Assessing items on the SF-8 Japanese version for health-related quality of life: a psychometric analysis based on the nominal categories model of item response theory. Value Health. 2009;12(4): 568-573.
9