Validation of the Wisconsin Brief Pain Questionnaire in a Multilingual South African Population

Article Outline

• Abstract
• Introduction
• Methods
  • Subjects
  • Study Instrument and Procedure
  • Statistical Procedure
• Results
  • Descriptive Statistics for the WBPQ
  • Reliability and Validity of the English Version of the WBPQ When Completed by Second Language English Speakers
  • Reliability and Validity of the WBPQ Translated into Three Indigenous African Languages
  • Construct Validity
• Discussion
• Acknowledgments
• Appendix 1. English Version of the WBPQ
• Appendix 2. isiZulu Version of the WBPQ
• Supplementary data
• References
• Copyright

Abstract 

Assessment of pain intensity and its effect on quality of life is important for proper management of pain, but no validated pain assessment tools that assess pain intensity and the interference pain has on daily life are available in indigenous South African languages. Therefore, the aim of this study was to validate translated versions of the Wisconsin Brief Pain Questionnaire (WBPQ) in South African HIV-positive patients. The WBPQ was translated into three indigenous South African languages, Setswana, isiZulu, and Xitsonga. We interviewed 452 ambulatory HIV-positive patients (327 urban and 125 rural patients) between the ages of 20 and 76 years old. Factor analysis to assess construct validity identified a two-factor structure (pain intensity and pain interference) for the isiZulu (n=132), Xitsonga (n=125), and Setswana (n=66) versions of the WBPQ, whereas a three-factor structure (pain intensity, mood interference, and activity interference) was identified for the English (completed by English second-language speakers, n=129) version of the WBPQ. Cronbach alphas, calculated to assess the reliability of the pain intensity and pain interference scales, were greater than 0.70 for all scales in all four versions of the WBPQ, showing internal consistency within the dimensions. These results provide evidence of validity for an easily administered questionnaire, which assesses pain intensity and pain interference, in three indigenous South African languages, and for English second-language speakers, in a population of South African HIV-positive patients.

Key Words: HIV-related pain, pain intensity, pain interference, reliability, validity

Back to Article Outline

Introduction 

Pain is one of the most frequent and debilitating symptoms in people infected with HIV.¹, ², ³ The prevalence of HIV-related pain ranges from 30% to 94%,², ⁴, ⁵ depending on disease severity and duration of infection. Despite physicians agreeing that HIV-related pain is a major complication of the disease, Laure et al.⁶ reported that 57% of HIV-positive patients who were in severe pain did not receive any analgesic treatment. When analgesic therapy was prescribed, it frequently was inadequate.⁷

Almost all studies that have investigated HIV-related pain have been conducted in first-world countries and on patient cohorts that consisted predominantly of males, Caucasians, homosexuals, and intravenous drug users.³, ⁷, ⁸ The only study on pain in HIV-positive patients in South Africa was of a limited nature, and focused on a highly selected group of patients who were under palliative care.⁵ Thus, there is a lack of knowledge on whether the prevalence of HIV-related pain is similar in a third-world environment, where the patient base is mainly women and blacks.⁹ Because pain is a subjective experience that is dependent on ethnicity¹⁰ and gender,¹¹ we believe that an authoritative study into HIV-related pain and its treatment is required in the South African population. To adequately assess pain and the effect it has on the daily life of South African HIV-positive patients, a validated instrument to obtain pain-related information in the languages frequently spoken in South Africa is required.

The methods most commonly used by South African clinicians to assess pain are the visual analog scale, and verbal and numerical rating scales.¹² Although these tools are easy to administer, they only measure pain intensity and not the effect of pain on the quality of life of patients. Assessment of pain intensity and the effect of pain on quality of life can be achieved by coadministering quality-of-life questionnaires, such as the Patient Evaluated Problem Score, with a pain rating scale.¹³ However, simple, easily administered questionnaires like the Brief Pain Inventory (BPI) and the Wisconsin Brief Pain Questionnaire (WBPQ) assess pain intensity, pain interference and treatment efficacy.

The BPI, developed by Cleeland and colleagues,¹⁴ is widely used in the assessment of pain. The instrument measures the severity of pain and the impact of pain on quality of life of patients by asking patients to rate their pain intensity (pain at its worst, pain on average, and pain at its least during the last week, and pain right now) on a 0–10 numeric scale where 0 indicates “no pain” and 10 indicates “pain as bad as you can imagine.” The instrument also measures (on 0–10 numerical scales) how pain interferes with other aspects of patients' daily lives, namely mood, relations with other people, walking ability, sleep, general activity, enjoyment of life, and normal work. The BPI comes in a short form and a long form, with the short form containing the same eleven core questions on pain intensity and pain interference as the long form, but less patient history and details of the quality of the pain are included. Although the instrument was developed in English, the simplicity of the semantic structure of the BPI questions means that it has been translated and validated for use in the assessment of cancer pain¹⁵, ¹⁶, ¹⁷, ¹⁸, ¹⁹ and noncancer pain.²⁰, ²¹ Translated versions of the BPI have been used in South Africa in a study on cancer pain,²² but apparently without validation.

The WBPQ is a self-report questionnaire, developed by Daut and colleagues²³ before the development of the BPI, to assess pain intensity and pain interference in patients with cancer and other diseases.²³ Like the BPI, the WBPQ contains the same four core questions as the BPI that assess pain intensity, and the same seven core questions that assess pain interference. The rating of pain intensity in the WBPQ also uses a numerical rating scale anchored at 0 and 10, but the pain interference items are rated on a five-point scale, with 0 representing “no interference at all” and 4 representing “extreme interference.” In addition, the questions on pain intensity in the WBPQ ask patients to rate their pain intensity in the last month, not the last week. Although the underlying structure of the WBPQ and BPI are similar, and the WBPQ was developed before the BPI, the BPI has been more widely adopted than the WBPQ.

For the present study, we chose to validate the WBPQ for non–English-speaking South African populations. The WBPQ was selected because the underlying psychometric properties of the core questions of the WBPQ and BPI are very similar, the main difference being the time period over which patients assess each item, and the time taken to complete the long form of the BPI would be too onerous on our patients. The short form of the BPI was sufficiently brief, but does not provide the supporting questions included in the WBPQ.

There are 11 official languages in South Africa, with nine being indigenous African languages. Most people in South Africa, especially in the indigenous African population, are at least bilingual and frequently use more than one language in their homes.²⁴ Therefore, we set out to determine the validity of the pain intensity and pain interference items of translated versions of the WBPQ in South African patients, using rural and urban cohorts of HIV-positive patients. The questionnaire was translated into isiZulu, isiXhosa, Sepedi, Setswana, and Xitsonga. However, too few patients completed the isiXhosa and Sepedi versions to warrant evaluation of these two questionnaires. We also assessed the validity of the original English version of the WBPQ in second-language English speakers.

Back to Article Outline

Methods 

Subjects 

We recruited 327 ambulatory HIV-positive outpatients between the ages of 20 and 76 from the outpatient clinic for HIV-positive people at the Helen Joseph Hospital, Johannesburg, Gauteng, South Africa, and 125 ambulatory HIV-positive outpatients from a rural hospital, Tintswalo Hospital, Bushbuckridge, Limpopo, South Africa between March 2005 and July 2006. Of the 327 urban patients, 244 were female. Patients' CD4 counts ranged from 1 to 861 cells/mm³ and viral loads ranged from 380 to 123,000 copies/mL. In the rural hospital, 98 patients were female and 27 were male; the patients' CD4 counts ranged between 4 and 963 cells/mm³ and the viral load ranged from 25 to 357,000 copies/mL. The research protocol was approved by The University of the Witwatersrand Human Research Ethics Committee (Clearance Number: M041112), and all patients were informed of the study procedures, and if they wished to take part in the study, they signed a written consent form.

Study Instrument and Procedure 

We used the WBPQ, which consists of pain intensity and pain interference items similar to that of the BPI.²³ The WBPQ was forward translated from English into isiZulu, isiXhosa, Sepedi, and Setswana, the four most frequently spoken and understood languages in urban Johannesburg, and also into Xitsonga, the predominant language spoken in rural areas of Limpopo Province, South Africa, by bilingual translators. The translated questionnaires were back-translated into English by bilingual translators who had not seen the English version of the WBPQ. We compared the back-translated versions of the WBPQ to the original English version to check that the meaning of the questions had been maintained. When the meaning of a question was not maintained, a third translator was asked to verify the accuracy of the forward and back-translations, and if no suitable wording could be found, the question was removed from the questionnaire. The meanings of the following questions were not maintained and the questions were removed: Average pain (Question 7) from all the translated versions, and mood (Question 14) from the isiZulu version. (Copies of the full English and isiZulu versions of the WBPQ are presented in Appendices 1 and 2 of this article; the Xitsonga and Setswana versions are available in Appendices 3 and 4 in the online version at doi: 10.1016/j.jpainsymman.2007.10.020.) Only the data for the English-second language, isiZulu, Xitsonga, and Setswana versions of the BPQ are reported because the sample sizes were close to, or greater than 100, which is the minimum recommended sample size for factor analysis.²⁵ The sample sizes for the Sepedi and isiXhosa versions of the WPBQ were 35 and 34, respectively. Therefore, data from these 69 patients were not analyzed.

Because some of the patients were illiterate, we trained interviewers to administer the questionnaire to all patients. Interviewers were trained so as to standardize the interview technique, and they were instructed to only clarify the questions asked in the questionnaires and not to add any additional information if they were asked by the patients.

Statistical Procedure 

We used the Kruskal–Wallis test with Dunn's multiple comparison test to assess for differences between scores for the 11 pain intensity and pain interference items between the different language versions of the WBPQ. To determine the construct validity of the questionnaires, factor analysis with Varimax rotation was performed to determine the number of factors underlying the relationships found among items in the English and translated questionnaires. That is, we wanted to determine whether the pain intensity items (worst pain, least pain, average pain, pain now) and the pain interference items (mood, relations with other people, walking ability, sleep, normal work, and enjoyment of life) for each version of the WBPQ were grouped logically under factors representing the different dimensions of the WBPQ, namely pain intensity and pain interference. Factors were retained if their eigenvalues exceeded one, and it was possible to interpret the factors. The extent to which each factor represented the data was assessed by examining the amount of explained variance accounted for by the factors. The suitability of the grouping of items under each factor was assessed by examining the factor loadings for each item. If the loading for an item under a factor was greater than 0.4, the loading was considered meaningful.²⁶, ²⁷ Items that loaded on two factors, but highly on only one factor were considered as part of that factor. Items that loaded highly on more than one factor and the items that did not load on any factors were excluded as items within the questionnaire. All the removed items were not included in the reliability analysis. In addition, to confirm the criterion validity of the pain severity items on the WBPQ, we performed a Spearman's correlation between the scores patients gave for the “pain now” item on the WBPQ with their rating of their current pain intensity, assessed at the same time, using the Faces Scale. The Faces Scale consists of illustrated faces with expressions that range from a neutral face indicating “no pain” to a down-turned mouth indicating “worst pain ever.” The Faces pain scale has been used successfully to rate pain in a similar cohort of South African patients.¹²

To determine the reliability of the questionnaires, we computed coefficient alphas for the pain intensity and pain interference subscales in each questionnaire. Alpha values ≥0.7 were considered as evidence of acceptable internal consistency for the scale under consideration.²⁸ To determine the contribution each item made to the underlying construct of a subscale, coefficient alpha values for the subscales also were calculated when individual items in each subscale were deleted. Items that significantly compromised the underlying construct of a subscale (that is, the scale alpha value rose when the item was deleted) were removed.²⁹ To determine whether patients had difficulty using the 11-point rating scale used in the pain intensity subscale of the questionnaires, we also determined how many patients, within each language group, rated the intensity of their average pain, least pain, or pain now greater than their worst pain.¹⁶

Back to Article Outline

Results 

Descriptive Statistics for the WBPQ 

Table 1 shows the descriptive statistics for all the translated versions of the WBPQ and the English version (completed by English second-language speakers). In general, we found that the group of patients from the rural community (Xitsonga questionnaire) rated their “pain now” and pain interference items significantly greater than patients in the urban cohort of patients (English, isiZulu, and Setswana questionnaires). However, there were no significant differences in the rating of “least pain” and “worst pain” between questionnaires.

Table 1. Medians and Interquartile Range of Pain Intensity and Pain Interference Items for the Translated Versions of the Wisconsin Brief Pain Questionnaire

Variables	Language
	English (n=129)	isiZulu (n=132)	Setswana (n=66)	Xitsonga (n=125)
Intensity itemsa
Worst pain	5 (0, 7)	5 (3, 8)	4 (0.5, 8)	5 (0, 9)
Least pain	2 (0, 4)	3 (1, 5)	2 (0, 4)	3 (0, 5)
Average pain	3 (0, 5)	—	—	—
Pain now	0 (0, 1)	5 (0, 2)	0 (0, 5)	3 (0, 6)b, c
Interference itemsd
Mood	0 (0, 1)	—	0 (0, 1)	1 (0, 3)b, c
Relations	0 (0, 0)	0 (0, 0)	0 (0, 0)	0 (0, 2)b, c, e
Enjoyment of life	0 (0, 1)	0 (0, 0)	0 (0, 1)	1 (0, 2)b, c, e
Walking	0 (0, 1)	0 (0, 2)	0 (0, 0)	0 (0, 2)b, c
Normal work	0 (0, 2)	0 (0, 0.3)	0 (0, 0)	1 (0, 3)b, c, e
Sleep	0 (0, 1)	0 (0, 2)	0 (0, 1)	1 (0, 2)b, c

Average Pain was removed from the isiZulu, Setswana, and Xitsonga questionnaires because no appropriate translation of the question was achieved; Mood was removed from the isiZulu questionnaire for the same reason.

Reliability and Validity of the English Version of the WBPQ When Completed by Second Language English Speakers 

One hundred twenty-nine patients that were second-language English speakers completed the English version of the WBPQ. The majority of these patients spoke Zulus (32%) or South Sotho (16%) as a first language. Factor analysis revealed that the questionnaire had a three-factor structure (Fig. 1). Factor 1 included all the pain intensity items, Factor 2 included the pain interference items describing mood-related interference (mood, relations with other people, and enjoyment of life), and Factor 3 included interference items describing activity-related interference (walking, normal work, and sleep). The eigenvalues of the three factors were 2.56 for pain intensity, 1.86 for mood-related interference, and 1.63 for activity-related interference. The three factors explained 67% of the variance; factor loadings are shown in Fig. 1. The Cronbach alpha for pain intensity items was 0.88, 0.73 for mood related interference, and 0.80 for activity interference (Table 2). Omission of any of the items in each of the subscales did not negatively affect the underlying construct of the subscales (Table 2). Eight percent of the patients that completed the English version of the WBPQ rated average pain, least pain, and pain now greater than their worst pain.

• 
  • View Large Image
  • Download to PowerPoint

• Fig. 1 

  Factor loadings plots of the South African versions of the Wisconsin Brief Pain Questionnaire.

Table 2. A Summary of Item Analysis and Alpha Values if an Item is Omitted for the English and Translated Versions of the Wisconsin Brief Pain Questionnaire

	Language
Variables	English	isiZulu	Setswana	Xitsonga
Pain intensity items	α=0.88	α=0.80	α=0.86	α=0.84
Worst pain	0.83	0.69	0.77	0.81
Least pain	0.82	0.70	0.81	0.71
Average pain	0.83	—	—	—
Pain now	0.88	0.78	0.82	0.84
Pain interference items	α=0.82	α=0.73	α=0.80	α=0.94
Mood	0.81	—	0.71	0.93
Relation with others	0.81	0.68	0.78	0.93
Enjoyment of life	0.78	0.68	0.70	0.93
Walking	0.78	0.71	0.77	0.93
Normal work	0.77	0.64	0.77	0.93
Sleep	0.80	a	a	0.92
Mood interference items	α=0.73
Mood	0.68
Relations with others	0.68
Enjoyment of life	0.56
Activity interference items	α=0.80
Walking	0.70
Normal work	0.76
Sleep	0.73

Average Pain was removed from the isiZulu, Setswana, and Xitsonga questionnaires because no appropriate translation of the question was achieved; Mood was removed from the isiZulu questionnaire for the same reason.

Reliability and Validity of the WBPQ Translated into Three Indigenous African Languages 

isiZulu. One hundred thirty-two patients completed the isiZulu version of the WBPQ. The first three eigenvalues for the factor analysis were 1.90, 1.82, and 0.27. All the pain intensity items loaded under Factor 1 and all the pain interference items loaded under Factor 2, except for “sleep,” which did not load on either of the factors (Fig. 1). We, therefore, excluded “sleep” as an item from the pain interference dimension for the isiZulu version of the WBPQ. The two factors explained 69% of the variance; factor loadings are shown in Fig. 1. The Cronbach alpha was 0.80 for the pain intensity subscales and 0.73 for the pain interference subscale, after exclusion of “sleep” (Table 2). Nine percent of the patients that completed the isiZulu version of the WBPQ rated average pain, least pain, and pain now greater than their worst pain.

Setswana. The Setswana version of the WBPQ was completed by 66 patients. The first three eigenvalues from the factor analysis were 2.39, 2.22, and 0.22. The first two factors explained 73% of the variance and all the pain interference items loaded under Factor 1, except for “sleep” and all the pain intensity items loaded under Factor 2 (Fig. 1). Thus, “sleep” was excluded from analyses of the Setswana version of the WBPQ. After exclusion of “sleep,” the Cronbach alpha was 0.86 for pain intensity items and 0.80 for pain interference items (Table 2). Four percent of the patients who completed the Setswana version of the WBPQ rated average pain, least pain, and pain now greater than their worst pain.

Xitsonga. The Xitsonga questionnaire was completed by 125 patients. Factor analysis identified two factors, with the first three eigenvalues being 4.15, 2.35, and 0.12. All the pain interference items loaded under Factor 1 and all the pain intensity items loaded under Factor 2 (Fig. 1). These two factors explained 78% of the variance; factor loadings are shown in Fig. 1. The Cronbach alphas for the pain intensity and pain interference dimensions were 0.84 and 0.94, respectively (Table 2). No patients rated average pain, least pain, and pain now greater than their worst pain.

Construct Validity 

A significant, positive relationship was detected when we correlated the scores from the “pain now” item on the WBPQ against the scores patients gave using the Faces Scale. The R² for the correlation was 0.31 and the P-value was <0.0001.

Back to Article Outline

Discussion 

To allow us to assess pain severity and the effect pain has on the lives of HIV-positive patients in South Africa, we validated translated versions, and the original English version of the Wisconsin Brief Pain Questionnaire, in HIV-positive South Africans. We found that the questions in the English questionnaire, which was completed by English-second language speakers, could be grouped under a three-factor structure: pain intensity, mood interference, and pain interference. All the indigenous language questionnaires had a two-factor structure, with questions grouping under pain intensity and pain interference factors. However, in the isiZulu and Setswana versions of the WBPQ, “sleep” did not load on either of the two factors, shown by the loadings being less than 0.4 for both factors. Also, we were unable to provide a suitable translation of the meaning of average pain in any of the translated questionnaires and there was no suitable term to describe “mood” in isiZulu. Despite the omissions of these questions, there was good internal consistency within each factor for all questionnaires, and we believe that the translated versions of the WBPQ can be successfully used in a South African population. Correlations between the pain intensity items from the WBPQ and the pain rating scales from the Faces scale were performed to assess the criterion validity of the WBPQ. The correlation observed showed good agreement between the WBPQ and the Faces scale.

As we already have mentioned, we had to exclude some of the questions from the translated versions of the WBPQ because of a lack of suitable translations. For example, “mood” in isiZulu was excluded from the questionnaire because a single term in isiZulu is used to describe a person's mood, well being, and, sometimes, life. Similarly in all translated versions of the WBPQ, “average pain” was excluded because we could not identify a simple translation of the term “average pain” into the African languages we used. Typical translations of “average pain” included “medium pain,” “pain in between,” and “mild pain.” The exclusions of these items, especially “average pain,” from the questionnaires is unfortunate, and compromises the comprehensiveness of the questionnaires, but we do not believe that the exclusion of these items invalidates the use of the questionnaires even though we only assess the extremes of the intensity of pain experienced by patients, that is, their least pain and their worst pain. Also, like others who have validated the BPI, we also have had to exclude “sleep” as an item because of poor factor loadings.¹⁶, ²⁹

Coefficient alphas are used to identify if there are any errors during measurement. In our study, we have observed coefficient alphas that were in the range of those values calculated for BPI validations for other countries (Table 3). Although in our study we used the WBPQ, which has a narrower rating scale than the BPI for rating pain interference items, and the pain intensity items are rated over a month in the WBPQ and not a week as in the BPI, the core components of the BPI and the WBPQ appear to be comparable. Yet despite the similarities in the psychometric properties of the WBPQ and the BPI, by validating the WBPQ and not the BPI, we may have limited the opportunity for comparisons to be made between data collected from South African populations using the WBPQ and patients from other countries, whose data were collected using the BPI. Also, validating the BPI may have facilitated inclusion of South African study sites in multinational pain studies. Nevertheless, by validating the WBPQ, we have provided scientists and health care professionals with the first validated multidimensional pain assessment instrument for indigenous South African populations.

Table 3. Cronbach Alphas of Pain Intensity and Pain Interference Subscales of the Brief Pain Inventory and WBPQ in Different Countries and Languages

Country	Language	Pain Intensity (α)	Pain Interference (α)
BPI
China	Chinese	0.86	0.91
France	French	0.86	0.90
Germany	German	0.88	0.92
Greece	Greek	0.88	0.85
India	Hindi	0.89	0.91
Italy	Italian	0.78	0.78
Japan	Japanese	0.81	0.81
Korea	Korean	0.85	0.93
Malaysia	Malay	0.81	0.88
Norway	Norwegian	0.87	0.92
Philippines	Filipino (Tagalog)	0.80	0.86
Spain	Spanish	0.87	0.89
Taiwan	Taiwanese	0.81	0.89
USA	English	0.87	0.91
Vietnam	Vietnamese	0.85	0.93
WBPQ
South Africa	English (second language)	0.88	0.82
South Africa	IsiZulua	0.80	0.73
South Africa	Setswanab	0.86	0.80
South Africa	Xitsongaa	0.84	0.94

All South African values were obtained using the Wisconsin Brief Pain Questionnaire, where pain interference items were measured on a scale of 0–4, whereas all other pain intensity items were obtained using a 0–10-point scale as in the BPI.

Clinicians and researchers wanting to use the WBPQ should be aware that, like the BPI,¹⁹, ²⁹ there are no suitable methods for dealing with missing values in the WBPQ. However, we overcame this problem by using interviewers to administer the questionnaire to the patients, and we had no missing values. Also, like the BPI, we cannot be sure that the pain interference items on the WBPQ are assessing actual impairment caused by pain, or merely reflect changes in health-related quality of life. Lastly, there is a lack of normative data for the WBPQ.

In contrast to the translated versions of the WBPQ, which all had a two-factor structure, we observed a three-factor structure in the English questionnaire completed by English second-language speakers. Patients had a choice of being interviewed in English or in their home language, and patients who chose to conduct the interview in English, and not their home language, did not give reasons for their preference. Three-factor structures have previously been reported in validations of the Norwegian and Hindi versions of the BPI, but not for English versions of the BPI being completed by second-language English speakers.¹⁶, ¹⁹ Unlike the three-factor structure of the Norwegian translation of the BPI where “sleep” loaded as an item within the mood interference dimension,¹⁹ we found that “sleep” loaded as an item within an activity interference dimension. It is possible that our patients considered “sleep” as a position of lying down, and hence an interference on activity.

All patients we interviewed were at least bilingual, suggesting that the factor structure in our cohort was not determined by the language capabilities of patients, that is, monolingual or bilingual, as suggested by Saxena and colleagues.¹⁶ It also has been suggested that differences in factor structure in translations of the BPI could be due to variations in the conceptual meaning of items during the translation process,¹⁹ or in our case, completing the questionnaire in a second language. Although loss of meaning caused by translation of the questionnaire into another language could be a possible reason for differences in factor structure, in this study, back-translation of the translated questionnaires showed that the original meaning of questions had been maintained, so it is likely that the loss of meaning when completing the question in a second language may be responsible for the difference in factor structure we observed. Indeed, the factor structure of the BPI differed between the monolingual Hindi speakers completing a Hindi BPI, and bilingual Hindi and English speakers, who completed the English version of the BPI.

In conclusion, we believe that these translated and validated questionnaires can be best used to measure pain intensity and interference items in a population of HIV-positive South African patients.

Back to Article Outline

Acknowledgments 

The authors thank the interviewers, Ms. Edith Paledi, Ms. Thulile Makafola, and Ms. Zanele Msomi, and the staff and patients of Helen Joseph Hospital, Johannesburg and Tintswalo Hospital, Bushbuckridge for their invaluable assistance.

Back to Article Outline

Appendix 1. English Version of the WBPQ 

• 
  • View Large Image
  • Download to PowerPoint

• 
  • View Large Image
  • Download to PowerPoint

• 
  • View Large Image
  • Download to PowerPoint

• 
  • View Large Image
  • Download to PowerPoint

Back to Article Outline

Appendix 2. isiZulu Version of the WBPQ 

• 
  • View Large Image
  • Download to PowerPoint

• 
  • View Large Image
  • Download to PowerPoint

• 
  • View Large Image
  • Download to PowerPoint

• 
  • View Large Image
  • Download to PowerPoint

Back to Article Outline

Supplementary data 

Back to Article Outline

References 

1. Lebovits AH, Lefkowitz M, McCarthy D, et al. The prevalence and management of pain in patients with AIDS: a review of 134 cases. Clin J Pain. 1989;5:245–248
   • View In Article
   • MEDLINE
   • CrossRef
2. O'Neill WM, Sherrad JS. Pain in human immunodeficiency virus disease: a review. Pain. 1993;54:3–14
   • View In Article
   • Abstract
   • Full-Text PDF (1669 KB)
   • CrossRef
3. Singer EJ, Zorilla C, Fahy-Chandon B, et al. Painful symptoms reported by ambulatory HIV-infected men in a longitudinal study. Pain. 1993;54:15–19
   • View In Article
   • Abstract
   • Full-Text PDF (622 KB)
   • CrossRef
4. McCormack JP, Li R, Zarowny D, Singer J. Inadequate treatment of pain in ambulatory HIV patients. Clin J Pain. 1993;9:279–283
   • View In Article
   • MEDLINE
   • CrossRef
5. Norval DA. Symptoms and sites of pain experienced by AIDS patients. S Afr Med J. 2004;94(6):450–454
   • View In Article
   • MEDLINE
6. Larue F, Fontaine A, Colleau SM. Underestimation and undertreatment of pain in HIV disease: multicentre study. Brit Med J. 1997;314:23–28
   • View In Article
7. Breitbart W, Rosenfeld BD, Passik SD, et al. The undertreatment of pain in ambulatory AIDS patients. Pain. 1996;65:243–249
   • View In Article
   • Abstract
   • Full-Text PDF (719 KB)
   • CrossRef
8. Hewitt DJ, McDonald M, Portenoy RK, Rosenfeld B, Passik S. Pain syndromes and etiologies in ambulatory AIDS patients. Pain. 1997;70:117–123
   • View In Article
   • Abstract
   • Full Text
   • Full-Text PDF (112 KB)
   • CrossRef
9. Shisana O, Rehle T, Simbayi LC, et al. South African national HIV prevalence, HIV incidence, behaviour and communication survey. Cape Town: Human Sciences Research Council Press; 2005;33–36
   • View In Article
10. Riley JL, Wade JB, Myers CD, et al. Racial/ethnic differences in the experience of chronic pain. Pain. 2002;100:291–298
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (103 KB)
    • CrossRef
11. Jackson T, Iezzi T, Gunderson J, Nagasaka T, Fritch A. Gender differences in pain perception: the mediating role of self-efficacy beliefs. Sex Roles. 2002;47:561–568
    • View In Article
12. Shipton EA. Pain: acute and chronic. Parktown: Witwatersrand University Press; 1993;11–19
    • View In Article
13. Pratheepawanit N, Salek MS, Finlay IG. The applicability of quality-of-life assessment in palliative care: comparing two quality-of-life measures. Palliat Med. 1999;13:325–334
    • View In Article
    • MEDLINE
    • CrossRef
14. Cleeland CS, Ladinsky JL, Serlin RC, Nugyen CT. Multidimensional measurement of cancer pain: comparisons of US and Vietnamese patients. J Pain Symptom Manage. 1988;3:23–27
    • View In Article
    • MEDLINE
    • CrossRef
15. Radbruch L, Loick G, Kiencke P, et al. Validation of the German version of the Brief Pain Inventory. J Pain Symptom Manage. 1999;18:180–187
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (219 KB)
    • CrossRef
16. Saxena A, Mendoza T, Cleeland GS. The assessment of cancer pain in North India: the validation of the Hindi Brief Pain Inventory-BPI-H. J Pain Symptom Manage. 1999;17:27–41
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (2000 KB)
    • CrossRef
17. Aisyaturridha A, Naing L, Nizar AJ. Validation of the Malay Brief Pain Inventory Questionnaire to measure cancer pain. J Pain Symptom Manage. 2006;31:13–21
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (171 KB)
    • CrossRef
18. Ger L, Ho S, Sun W, Wang M, Cleeland GS. Validation of the Brief Pain Inventory in a Taiwanese population. J Pain Symptom Manage. 1999;15:316–322
    • View In Article
19. Klepstad P, Håvard L, Borchgrevink PC, et al. The Norwegian Brief Pain Inventory Questionnaire: translation and validation in cancer pain patients. J Pain Symptom Manage. 2002;24:517–525
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (75 KB)
    • CrossRef
20. Tan G, Jensen MP, Thornby JI, Shanti BF. Validation of the Brief Pain Inventory for chronic nonmalignant pain. J Pain. 2004;5:133–137
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (93 KB)
    • CrossRef
21. Keller S, Bann CM, Dodd SL, et al. Validity of the Brief Pain Inventory for use in documenting the outcomes of patients with noncancer pain. Clin J Pain. 2004;20:309–318
    • View In Article
    • MEDLINE
    • CrossRef
22. Beck SL, Falkson G. Prevalence and management of cancer pain in South Africa. Pain. 2001;94:75–84
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (177 KB)
    • CrossRef
23. Daut RL, Cleeland GS, Flanery RC. Development of the Wisconsin Brief Pain Questionnaire to assess pain in cancer and other diseases. Pain. 1983;17:197–210
    • View In Article
    • Abstract
    • Full-Text PDF (951 KB)
    • CrossRef
24. Pan South African Language Board . Language use and language interaction in South Africa: A national sociolinguistic survey. Pretoria: Pan South African Language Board; 2000;2
    • View In Article
25. Kline P. An Easy Guide to Factor Analysis. London: Routledge; 1994;
    • View In Article
26. Wang XS, Mendoza TR, Gao S-Z, Cleeland CS. The Chinese version of the Brief Pain Inventory (BPI-C): its development and use in a study of cancer pain. Pain. 1996;67:407–416
    • View In Article
    • Abstract
    • Full-Text PDF (1147 KB)
    • CrossRef
27. Mystakidou K, Mendoza T, Tsilika E, et al. Greek Brief Pain Inventory: validation and utility in cancer pain. Oncology. 2001;60:35–42
    • View In Article
    • MEDLINE
    • CrossRef
28. Nunnaly JC. Psychometric theory. 2nd ed.. New York, NY: McGraw-Hill; 1978;
    • View In Article
29. Caraceni A, Mendoza TR, Mencaglia E, et al. Breve Questionario per la valutazione del dolore. (A validation study of an Italian version of the Brief Pain Inventory). Pain. 1996;65:87–92
    • View In Article
    • Abstract
    • Full-Text PDF (568 KB)
    • CrossRef